ams-02 tracker thermal control system: development of new ... the ams tracker thermal control system...

UNIVERSITÀ DEGLI STUDI DI PERUGIA

Doctorate Course: NUOVI MATERIALI STRUTTURALI E FUNZIONALI

PER APPLICAZIONI SPAZIALI

XXII CYCLE

Scientific disciplinary sector: ING‐IND/04‐COSTRUZIONIESTRUTTUREAEROSPAZIALI

AMS-02 Tracker thermal control system:

development of new technologies for manufacturing of two-phase cooling system

PhD Candidate Elisa Laudi

Coordinator: Prof. Roberto Battiston Supervisor: Ir. Johannes van Es National Aerospace Laboratory (NLR), The Netherlands

ABSTRACT The AMS Tracker Thermal Control System (TTCS) is a two-phase cooling system, part of the Alpha Magnetic Spectrometer (AMS-02) experiment to be located on the International Space Station (ISS) truss. The AMS-02 is a space born detector for cosmic rays built by an international collaboration, lead by Nobel prize laureate S.C. Ting and will operate aboard the truss of the International Space Station (ISS) for at least 3 years, collecting several billions of high-energy protons and nuclei. The main goal is to search for cosmic antimatter, (that is for anti-helium nuclei primarily), for dark matter and lost matter. The heart of the AMS-02 experiment is the Silicon Tracker. It measures particle trajectories through AMS's strong magnetic field. Around the silicon planes detecting front-end electronics are located, providing the accurate measurements needed. In order to keep the Tracker stable in temperature the Tracker waste heat need to be collected and radiated to deep space. Therefore a dedicated thermal control system was required to meet the stringent electronics temperature stability. The TTCS is a mechanically pumped two-phase carbon dioxide cooling loop. Main objective is to provide accurate (< 3 K) temperature control and remove 144 W heat of the AMS-02 Tracker front-end electronics. The TTCS requirements, system design, development status and some typical test results have been described. . This dissertation deals with the description of the new technologies that have been development during the manufacturing and integration of a two-phase cooling system for a space application. The goal of the thesis is to give the basis for the potential spin-off of such application. In particular to what extend the system can be used as cooling system for high-power communication satellites, future scientific spacecraft requiring tight temperature control and AMS-like terrestrial particle detectors. This dissertation is subdivided in 6 chapters. The first chapter presents a general introduction to AMS-02 experiment, its main scientific goals and all its different sub-detectors In the second chapter, the main thermal control techniques are reviewed and the AMS-02 thermal control system discussed in detail. The on-orbit AMS- 02 thermal environments and the thermal requirements driving the thermal control system design are also presented.

The description of Tracker Thermal Control System (TTCS) design, the main components and its system and environmental requirements are introduced in the third chapter. The manufacturing and thermal test campaign of one of the TTCS subsystems, the TTCS condensers, have been reported in the fourth chapter. The general description and driving design parameter and the overall manufacturing sequence in Taiwan (AIDC) for the production of the two Engineering Models (EM), one Qualification Model (QM) and the four Flight Model (FM) have been explained. The thermal tests performed in China (SYSU) on the EM and QM condensers to qualify the components prior to start the flight production, has been also described. In the fifth chapter, the Tracker Thermal Control Box integration for the qualification model ( QM) and flight model (FM) and the environmental tests have been reported. In particular the vibration and thermo-vacuum qualification test for the FM TTCBs have been presented. Finally in the sixth chapter the TTCS component integrations at CERN has been described. In addition, the basic monitoring and controlling operation for the overall TTCS system have been underlined, for the further understanding of the data results during the Tracker operation on ground and on orbit.

Contents

i

Contents

L'indice è vuoto perché non stai utilizzando gli stili paragrafo selezionati nelle impostazioni del documento.

1

Chapter 1

1. The Alpha Magnetic Spectrometer on the

International Space Station

1.1. Alpha Magnetic Spectrometer 02

The AMS-02 experiment is a state-of-the-art particle physics detector being designed, constructed, tested, and operated by an international team of ~500 physicists and engineers from 16 different countries, led by the Nobel Laureate S.C.C. Ting since 15 years. The AMS-02 experiment will use the unique environment of space, outside the limitation imposed by Earth’s atmosphere, to advance knowledge of the universe and potentially lead to a clearer understanding of the universe origin. Specifically, the science objectives of the AMS are to search for antimatter (anti-helium and anti-carbon) in space, dark matter (90% of the missing matter in the universe), and to study astrophysics (to understand cosmic ray propagation and confinement time in the galaxy) [1].

AMS will operate aboard the truss of the International Space Station (ISS) for at least 3 years, collecting several billions of high-energy protons and nuclei.

A first version of the detector, known as AMS-01, flew aboard the Space Shuttle Discovery during the STS-91 mission (2-12 June 1998), collecting about hundred millions of cosmic particles. This trial mission confirmed the main ideas of the project and gave important suggestions for further development.

For the ISS mission, the detectors are slightly different in concept, achieving a higher resolution. In fact, AMS-02 will be an “improved” version of AMS-01. The solid magnet of the AMS-01 mission has been replaced by a more powerful Helium cooled super-conductive cryo-magnet in AMS-02.The introduction of the cryo-magnet does not only introduce additional magnet cooling, it also increases the thermal design complexity.

In the following is briefly described a review of the scientific goals of the experiment and the main components of the experimental apparatus.

Chapter 1

2

The Alpha Magnetic Spectrometer on the International Space Station

1.2. Scientific goal of AMS experiment

The AMS experiment has been conceived to address fundamental open

issues in the current cosmology and particle physics scenarios by measuring the composition and energy spectra of cosmic radiation:

• Existence of primordial anti-matter; • Nature of the dark-matter; • Existence of new states of matter;

The un-precedent accuracy of the AMS multi-channel measurements in an extended energy range (GeV-TeV) will also supply a powerful experimental basis in the high energy astrophysics domain, allowing to test and compare different models for the origin and propagation of the cosmic radiation in the galaxy.[2]

The AMS detector design has been driven by its ambitious scientific objectives: a ten-days engineering flight on the Space Shuttle DISCOVERY has been performed in the early phases of the project (1998) to verify key components of the instrument against the harsh environmental stresses to reach and operate in space. Scientific and technical results from this mission, subsequently referred to as AMS-01, have guided the design of the AMS-02 instrument which is currently in its final phase of integration in a clean room laboratory at CERN, the European Organization for Nuclear Research.

1.3. The AMS-02 payload overview

The Alpha Magnetic Spectrometer-02 (AMS-02) has been designed to operate for a minimum of three years on board of the ISS, at an altitude of 400 km with an orbit inclination of 51:7 with respect the Earth equatorial plane. The AMS-02 will be transported to the International Space Station (ISS) in the cargo bay of the Space Shuttle (Figure 11) for installation on the external truss of the ISS (Figure 12). Once on-orbit the AMS-02 will remain on the ISS for at least three operational years of data collection, and due to limited space shuttle flights, AMS-02 will not return to Earth and will remain on the ISS.

3

Chapter 1

Figure 11: AMS 02 in the Space Shuttle Orbiter Cargo Bay (from above looking

forward)

Figure 12 AMS-02 on the International Space Station Truss

4


High-energy particles, passing through AMS-02, will leave different signals in the different detectors constituting the apparatus: the nature and the kinematical properties of the traversing particle are then reconstructed by combining signals gathered in all the AMS-02 subsystems. Figure 1-3 presents a schematic drawing of the AMS-02 detector and some examples of the characteristic signals released by different particles in the various sub-detectors.[3]

Figure 1-3: AMS-02 Detector Signatures

The AMS-02 core is the largest super-conducting magnet ever conceived for

use in space: it will provide a 0.8 T magnetic field to allow the separation of matter and anti-matter charged components of the cosmic radiation by their different bending in traversing the magnetic field volume (~ 3m3). Eight layers of silicon microstrip ladders constitute the Tracker detector: it measures the particle trajectories in the magnetic field allowing to reconstruct the particle absolute charge (Z) and rigidity (R). The nature of the cosmic particles traversing the apparatus are determined by combining the measurement of the tracking spectrometer with those performed by the other sub-systems, each related to a different property of the particle:

• velocity and impinging direction of the particle are reconstructed by the Time-of-Flight (TOF) detector;

• highly relativistic particles are distinguished from non relativistic ones by the Transition Radiation Detector (TRD);

• the energy of electrons, positrons and high energy photons is measured by electromagnetic calorimeter (ECAL);

• background particles, traversing only part of the apparatus, are rejected

5

Chapter 1

by the use of Anti-Coincidence-Counters (ACC); • the precise measurement of the velocity and a redundant measurement of

the absolute charge value are performed in the Ring Imaging Cherenkov (RICH) detector;

• to resolve the source of detected high energy photons, the pointing direction of the AMS detector with respect to the stars will be measured with great accuracy by the Star Tracker (ST) equipment. The following is a top-level review of the major components of the AMS-02

flight hardware.

1.3.1. The cryogenic superconducting magnet

The superconducting magnet system for AMS-02 consists of a pair of large dipole coils together with two series of six smaller racetrack coils circumferentially distributed between them as depicted in Figure 1-4. The dipole coils are used to generate the majority of the transverse magnetic field which defines the AMS x-axis, reaching a maximum value of ~ 0.87 T at the magnet centre. The racetrack coils are included for minimizing the magnitude of the stray field outside the magnet and to minimize the magnetic dipole moment of the magnet system to avoid an undesirable torque on the ISS resulting from the interaction with the Earth magnetic field (0.27 Nm) [4].

The magnet will operate at a temperature of 1.8 K, cooled by a surrounding toroidal vessel of 2500 l of superfluid Helium. The cryogenic magnet will be launched at the operating temperature, while the field will be charged only after installation on the ISS. Because of parasitic heat loads, the Helium will gradually boil away throughout the lifetime of the experiment. After the project time of 3 years, the Helium will be used up and the magnet will warm up and no longer be operable. A complex cryogenic system has been developed in order to preserve the cryogenic temperature and to maximize the life of the superfluid Helium. The basic components are:

• a phase separator to separate the heated gaseous Helium from the superfluid (in zero-gravity there is no separation between the liquid and gas);

• a cooling system composed by four concentric shields enclosing the magnet and Helium vessel. The helium vapour is guided in these shields reducing dramatically the heat leak of the superfluid;

• four cryocoolers are added in order to reduce significantly heat at higher temperature;

• a mass Gauging system evaluates the mass of the helium fluid measuring the temperature variation occurring after a small heat pulse;

• two thermo-mechanical cooling pumps are used for cooling down the cryogenic system in the case of quenching or for charging and discharging of the whole apparatus.

6


The magnet, the Helium vessel and the cryogenic system are all enclosed in a toroidal vacuum tank (the Vacuum Case, VC) with inner diameter of 1.1m, outer diameter of 2.7 m and a length of the central cylinder surrounding the tracker of 0.9m. In Figure 1-4 the schematic drawing of the magnet components is presented.

Figure 1-4: Super conducting magnet coils (left) and component identification (right)

1.3.2. Unique Support Structure

The Unique Support Structure or USS is the backbone and basic frame

supporting the total weight of AMS: 7000 Kg. Even if this is not a difficult task in space due to the reduced effect of gravity, during the launch process the USS will have to support a threefold effective AMS weight during the Shuttle acceleration phase. In Figure 1-5 all AMS-02 detectors are integrated with the USS structure.

The USS is made of aluminium with different painting coverage and partially enveloped in Multi Layer Insulator (MLI) according to thermal requirements. The USS design is not only the result of requirements driven by mechanical robustness, but it also acts as integral part of the thermal control system.

Connection to the support structure of different subsystems required precise drilling of 17000 holes in it. The middle empty part of the USS is the space where the magnet vacuum case is attached: in fact, the magnet vacuum case itself is also considered integral part of the support structure.

7

Chapter 1

Figure 1-5:AMS-02 detectors assembled as in their final flight configuration on the

USS during the pre-integration phase

1.3.3. Silicon Tracker In combination with the Superconducting Cryomagnet, the Silicon Tracker

represents the centrepiece of the AMS-02 suite of detectors. The Tracker (Figure 16) consists of eight layers of double-sided silicon micro-strip detectors (ladders) on five support planes. Within the bore of the Cryomagnet three of the double sided planes will operate. The two outermost single sided planes are located at the entrance and exit of the Cryomagnet’s field volume. The spatial resolution will be better than 15µm in the Cryomagnet’s bending plane and 30µm perpendicular to that. All eight tracker planes together comprise 192 silicon ladders corresponding to an active area of about 6m2 of silicon and 200000 readout channels. The entire tracker electronics consume 800 W of power. All support planes are made of an aluminium honeycomb structure enclosed within thin carbon fiber skins: a larger diameter (d=1.4m) and higher density (ρ=0.032 g/cm2) characterize the two external planes with respect the internal ones (d=1 m, ρ= 0.016 g/cm2).

The adopted designed minimizes the amount of material traversed by the particle, thus leaving its trajectory unperturbed, guaranteeing at the same time the stiffness needed for the mechanical stress at launch. Particles traversing the silicon detectors will leave an energy deposit related to their absolute charge at a position which can be reconstructed within 10(30)µm precision in the bending (non bending) coordinate. Up to eight measurement points will be then available to reconstruct the particle trajectory and estimate its rigidity from its curvature in the magnetic field up to rigidities of few TV [5][6].

8


Figure 16: The silicon tracker: in the schematic drawing (left) can be distinguished the

detector sensitive area (light blue) and the front-end electronics (red) mounted vertically on thermal bars. Tracker Support Plane 2 with Ladders installed (right)

1.3.4. Transition Radiation Detector (TRD)

The role of the TRD (Figure 17) is to discriminate between electrons/anti-

protons and positrons/protons up to energies of 300 GeV. This is accomplished by detecting X-ray photons emitted by highly energetic electrons and positrons when they pass through a radiator. For heavier particles such radiation is strongly reduced. The radiation is detected in tubes filled with Xe and CO2 gas in an 80:20 ratio. Xenon gas ionizes very easily and is thus very sensitive to the passage of photons.[7]

The TRD detector is composed of 328 modules arranged in 20 layers supported by a conical octagon made of aluminum-honeycomb walls with carbon-fiber skins and bulkheads (Figure 17). To provide a 3D tracking, the lower and upper four layers are oriented parallel to the AMS-02 magnetic field while the middle 12 layers run perpendicular. Each modules contains:

• 20 mm of radiator made of polypropylene/polyethylene fiber fleece corresponding to 0.06 g/cm3. The large number of interfaces increases the probability of production of X-rays.

• 16 tube straws filled with a Xe:CO2 (80%:20%) gas mixture operating at 1600V (full avalanche regime). The Xenon-rich gasses have an high efficiency for X-rays detection. The TRD is located at the top of the experiment stack, just above the Upper

Time of Flight (UTOF). The octagon structure is supported by the M-Structure, which is mounted to the USS-02 at four locations on the upper USS-02 just above the VC interface.[8]

9

Chapter 1

Figure 17:Schematic of the TRD assembly (left) and detector integrated in AMS-02

1.3.5. Time of Flight (TOF) The TOF serves to: 1) be a fast trigger to the experiment for traversal of a

particle across the bore of Cryomagnet and Silicon Tracker, 2) distinguish between upward and downward travelling particles, and 3) measure the absolute charge of the particle. Particles that pass through the scintillators generate photons as they pass through the counter paddles, and these photons are detected by groups of two or three sensitive Photo Multiplier Tubes (PMT’s) on either end of the detector element, the counter paddles.

The TOF is composed of four planes of detectors, two atop the AMS tracker, two below as shown in Figure 18. Numbered from the top down, detector assemblies 1, 2, and 4 have eight detector paddles per plane and detector assembly 3 has ten. The pairs of detector assemblies are oriented 90° to each other. This configuration gives a 12 x 12 cm2 resolution for triggering particle events over the 1.2 m2 area the TOF covers. The measured time-of-flight resolution gives a distinction power between up-going and down-going nuclei >109. This performance is a mandatory requirement for the anti-matter detection.[9]

Figure 18: Time of Flight Counter Construction: Upper Tof( UTOF) on the left, Lower

Tof (LTOF) on the right

10


1.3.6. Ring Imagining Cerenkov Detector (RICH)

The RICH (Figure 19) is located near the bottom of the experiment stack,

below the Lower TOF and above the Electromagnetic Calorimeter (ECAL). The RICH is used in conjunction with the Silicon Tracker to establish the mass of particles that traverse the AMS-02. Functionally, the RICH is composed of three basic elements. The top layer, the Cerenkov radiator, is composed of silica aerogel and sodium fluoride blocks that serve as sources for the Cerenkov radiation generated by the passage of the high energy particles. The intermediate layer is the conical mirror and the PMT Structural interfaces make up the third layer.[2]

Figure 19: The RICH detector: schematic drawing (left) and the mirror integrated with

the PMT plane (right)

1.3.7. Electromagnetic calorimeter (ECAL) The ECAL (Figure 1-10) is a scintillating fiber sampling calorimeter that

allows precise, 3-dimensional imaging of the shower of smaller particles generated when a particle collides with the calorimeter. The calorimeter also provides a stand-alone photon trigger capability to AMS. The ECAL measures the energy of electrons, positrons and gamma rays up to one TeV. In Figure 1-10 is a general diagram of the ECAL.

The mechanical assembly has met the challenges of supporting the intrinsically dense calorimeter during launch and landing with minimum weight. The light collection system and electronics are optimized for the calorimeter to measure electromagnetic particles over a wide energy range, from GeV up to TeV. [1]

11

Chapter 1

Figure 1-10: ECAL general assembly view

1.3.8. Anti-Coincidence Counters (ACC)

The ACC is a single layer of scintillating panels that surround the AMS-02

Silicon Tracker inside the inner bore of the superconducting magnet. The ACC identifies particles that enter or exit the Tracker through the side. This provides a means of rejecting particles that have not passed through all the detectors and may confuse the charge determination if they leave “hits” in the Tracker close to the tracks of interest.

The ACC scintillating panels are fitted between the Tracker shell and the inner cylinder of the VC, which contains the Cryomagnet system. The ACC is composed of 16 interlocking panels fabricated from BICRON BC414 (Figure 1-11).

The panels are 8-mm thick and are milled with tongue and groove interfaces along their vertical edges to connect adjacent panels. This provides hermetic coverage for the ACC detection function around the Silicon Tracker. The panels are supported by a 33.46” tall x .78” diameter x 0.047” thick M40J/CE Carbon Fiber Composite Support Cylinder[1]

12


Figure 1-11: ACC location within the Inner Cylinder of the vacuum case

1.3.9. Star Tracker (ST)

The Space Station, which is large and fairly flexible, cannot measure its own position with a high degree of accuracy and thus cannot directly tell the AMS-02 where it is exactly and where it is pointing. To optimize science from the Tracker detector carried by AMS it is important to have the capability to determine accurately the position of the AMS payload at the exact time that an event occurs. To accurately determine its position, AMS carries a Star Tracker called AMICA (for Astro Mapper for Instrument Check of Attitude). AMICA is equipped with a pair of small optical telescopes (AMICA Star Tracker Cameras or ASTCs). The ASTCs are mounted to the upper Vacuum Case Conical Flange on opposite sides of AMS to increase the probability that one has a clear view of the stars (Figure 1-12).

13

Chapter 1

Figure 1-12: Star tracker mounting on the AMS-02

1.3.10. AMS-02 Electronics

AMS-02 contains numerous electronics boxes, some termed “Crates,” that

supply the necessary readout/monitor/control electronics and power distribution for each detector (Figure 113). The box nomenclature is generically x-Crate or xPD, where “x” is a letter designating the detector function, and “Crate” refers to the readout/monitor/control electronics box and “PD” refers to the Power Distribution box for that specific detector. Similarly xHV bricks provide high voltage for some detectors.

Additionally, electronics are mounted in the Power Distribution System (PDS), the Cryomagnetic Avionics Box (CAB), the Uninterruptible Power Supply associated with the CAB, and the Cryocooler Electronics Box (CCEB).[10]

Figure 113: Electronic crates location and definition

14

Chapter 2

2. AMS-02 Thermal control system

The thermal control subsystem (TCS) is an integral part of every spacecraft.

It's purpose is to maintain all the components of a spacecraft within their respective temperature limits. There are several different sources of thermal energy acting on a spacecraft; solar radiation, albedo, earth emitted infrared, and heat generated by onboard equipment. Therefore, the thermal control subsystem is different for every spacecraft.

Figure 21: Satellite thermal control environment. The most significant external heat source is the Sun, but we must also include reflected solar energy (albedo) and Earth

infrared in our calculation. The only way a spacecraft can get rid of the heat is by radiation it to space

In general, there are two types of TCS, passive and active. A passive system

relies on conductive and radiative heat paths and has no moving parts or electrical power input. An active system is used in addition to the passive system when passive system is not adequate, for example, on manned missions. Active systems rely on pumps, thermostats, and heaters, use moving parts, and require electrical

Chapter 2

15

AMS-02 Thermal Control System

power. Different types of components for both passive and active systems are described in sections 2.1 and 2.2.

Many factors influence the design and development of the thermal control system and each spacecraft TCS will have it's own unique set. Mission constraints, mission objectives, and the physical design of a spacecraft determines the inputs and outputs of the TCS interface.

1.1. Passive thermal control components

A passive system typically relies on conductive and radiative paths to transfer heat from the components to the radiator without using moving parts or an electrical power input. They are preferred because they are usually cheaper, lighter and less complex. However, the selection of the control system depends on the thermal requirements of the payload. Passive control technologies include surface finishes, insulation blankets, mounting interfaces, radiators, heat switches and phase change materials.

1.1.1. Surface Finishes Thermal control coatings surfaces have special radiation properties, such as

black and white paints, and gold, silver, and aluminum foils; they are very efficient and lightweight.[11]

Based on their solar absorbivity a and their emissivity e, thermal control surfaces fall under four basic categories: solar reflectors, solar absorbers, thermal reflectors, and thermal absorbers. Solar reflectors typically have a very low absorbance to emittance ratio (a/e). They reflect incident solar IR energy while absorbing and emitting far IR energy. Two examples are quartz second surface mirrors and aluminized Kapton. Solar absorbers absorb solar energy while emitting only small amounts of far IR energy. An example is black Kapton. As for thermal reflectors, they minimize heat transfer by reflecting both near and far IR. Examples include polished aluminium and gold plating. Conversely, thermal absorbers maximize heat transfer by absorbing both near and far IR. Black paint is a good example. The properties of common surface finishes are shown Figure 22.

16

Chapter 2

Figure 22: Properties of Common Finishes. The absorptivity and emissivity of typical

spacecraft finishes are shown. Note that a combination of finishes can be made to create the desired absorptivity to emissivity ratios.

1.1.2. Insulation

Multilayer insulation (MLI) and single-layer radiation shields are among the

most common thermal control elements on spacecraft. MLI blankets are used either to prevent excessive heat loss from a component or excessive heating from environmental fluxes or rocket plumes. Most spacecraft are covered with MLI blankets, with cut-outs provided for radiator areas to reject internally generated waste heat. Single-layer radiation barriers are sometimes used in place of MLI where a lesser degree of thermal isolation is required, since they are lighter and cheaper to manufacture.

Multilayer insulation is composed of multiple layers of low-emittance films with low conductivity between layers, as shown in Figure 23. The simplest construction is a layered blanket assembled from embossed, thin Mylar sheets (1/4 mil thick) with a vacuum-deposited aluminium finish on one side of each sheet. The embossing results in the sheets touching only at a few points, thereby minimizing conductive heat paths between layers. The layers are aluminized on one side only, so that the Mylar can act as a low-conductivity spacer. Higher-performance construction uses Mylar film metalized on both surfaces (aluminium or gold) with silk or Dacron net, Tissueglas paper, or “Super-Flock” whiskers as the low-conductance spacers. To complicate MLI design further each layer must be grounded to reduce the chance of electrostatic discharge.

17


Figure 23: Composition of a typical MLI blanket. Multilayer insulation blankets are

made of fairly sophisticated layers of low-emittance films with low conductivity between layers.

1.1.3. Radiator Most spacecraft waste heat is rejected to space by radiators. These occur in

several different forms, such as spacecraft structural panels, flat-plate radiators mounted to the side of the spacecraft, or panels that are deployed after the spacecraft is on orbit.

Whatever the configuration, all radiators reject heat by IR radiation from their surfaces. The radiating power is dependent on the emissivity of the surface and its temperature. The radiator must reject both the satellite waste heat plus any radiant-heat loads from the environment or other spacecraft surfaces that are absorbed by the radiator, as shown in Figure 24.

Figure 24: Radiator Energy Balance (no external blockage). Note that we must select

radiative properties of the spacecraft surface to achieve an energy balance among spacecraft internal dissipation, external heat sources and reradiation to space to obtain

the desired temperature

18

Chapter 2

Most radiators are therefore given surface finishes with high IR emissivity (> 0.8) to maximize heat rejection and low solar absorptivity (< 0.2) to limit heat loads from the Sun. Typical finishes include quartz mirrors, silvered or aluminized Teflon, and white paint.[12]

Passive structural radiator designs use the aluminium honeycomb panels of the payload as the radiator. To increase the transverse conduction of the honeycomb panel, the face sheets are sometimes thicker than what it would be needed to be for structural reasons. When passive radiators and heat sinks are not adequate, body mounted actively controlled radiators must be used. Typically heat pipes, looped heat pipes, pumped fluid loops, or capillary pumped loops are used to transfer heat from the component to the radiator (Figure 25).

Figure 25: Details of embedded heat pipe the AMS-02 main radiator.

1.1.4. Phase Change Materials The advantages that Phase Changer Materials (PCM) affords to designers

are four. First, they absorb large amounts of energy via latent heat of fusion without an appreciable temperature rise to the component. This makes them well suited for protecting high power dissipating components. Second, because the process is reversible, they are excellent at damping cyclical loading such as moving in and out of the Earth's shadow. When the satellite is in the sun, the PCM absorbs the additional heat load. When the satellite is in the Earth's shadow, the PCM releases the heat to the components. As a result, radiators can be sized for mean loads and not peak loads, which reduces the radiator area and mass. Third, PCMs help maintain thermal stability and ensure isothermal conditions. Because their temperature stays constant as they undergo the phase change, they are well suited for applications that require tight temperature control. Finally, they are completely

19


passive and very reliable. Thus, robust TCSs can be designed around the use of phase change materials.

1.1.5. Interface filler To increase conductance between the unit and a spacecraft structure

interface filler can be used. The most common filler material are Graphite (Sigraflex) and Silicone Elastomer (Cho-Term 1671). The first one is made of laminated graphite sheet, meaning that it acts as an electrical conductor, usually with a thickness of 0.25mm the second one is a silicone binder, filled with boron nitride particles, reinforced with fibreglass cloth, leading to an electrical isolator behaviour.

1.1.6. Heat Pipes, Capillary Pumped Loops (CPL) and Loop

Heat Pipes (LHP)

A heat pipe uses a closed two-phase fluid-flow cycle to transport large quantities of heat from one location to another without the use of electrical power. The heat pipe can be used to create isothermal surfaces or to spread out heat from a localized source uniformly over a larger area. Loop heat pipes and capillary pumped loops are more advanced cousins of the basic heat pipes that can provide constant temperature operation with varying heat loads under gravity or acceleration. [12]

A typical heat pipe consists of a working fluid, a wick structure, and an envelope as depicted in Figure 26.To illustrate how a heat pipe works, consider a simple horizontal heat pipe in equilibrium with an isothermal environment.. The liquid in the wick and the vapour in the vapour space are at saturation. If heat is applied to the evaporator, raising its temperature, liquid in the wick evaporates (removing some of the added heat), which depresses the meniscus in the evaporator since less liquid is present there. This process also raises the local vapour pressure, since it must be in saturation with the heated liquid in the wick. The difference between the increased curvature of the meniscus in the evaporator wick and the unchanged meniscus in the condenser wick causes a difference in capillary pressure sufficient to pull liquid from the condenser wick toward the evaporator. This replenishes the liquid in the evaporator wick. At the same time, heated vapour flows from the evaporator to the condenser, which is at a lower pressure. When this vapour comes in contact with cooler condenser surfaces, it condenses. This cycle is shown schematically in Figure 26.

20

Chapter 2

Figure 26: Heat pipes are very efficient in moving heat from one place to another on

the spacecraft. A schematic is shown on the left. On the right is an example of the most common wick/envelope design which consists of axial grooves in the wall of extruded

aluminium tubing.

Capillary pumped loops (CPL) and loop heat pipes (LHP) are a relatively new thermal control technology. Their main advantage is in the capability to transport heat loads as large as 1500 W over long distances with a low ensuing temperature drop[14]. Each system consists of an evaporator, wick system, condenser, transition lines, and either a reservoir or compensation chamber (CC). A schematic of a traditional loop heat pipe is presented in Figure 27. In CPLs, the reservoir is separate from the system, and fluid is not shared with the evaporator. LHPs differ from CPLs in that the CC is connected to the evaporator and shares liquid with it.

In CPLs, vapour generated in the evaporator flows to the condenser where it is condensed and subcooled. [15]The liquid flows back to the core via an optional bayonet whose purpose is to position any vapour voids near the coldest sections so that voids are minimized. In the core, liquid is pulled radially through the wick, is vaporized on the surface, and returns to the condenser.

For LHPs, the standard approach consists of an evaporator and compensation chamber assembly, condenser, and transfer lines. The evaporator consists of a wick structure encapsulated by a cylindrical case. The case is mounted to the heat source. A network of vapour removing channels is formed in the volume between the wick and the case. The CC shares liquid with the evaporator; its purpose is to ensure that liquid is always available to the evaporator and condenser in the operational temperature range. Despite sharing some similarities, CPLs and LHPs have their own advantages and disadvantages under various operational conditions. For example, a LHP can start, stop, and re-start at any time regardless of operating conditions; however, a CPL requires a difficult and sometimes time consuming process prior to start-up. The advantage that CPLs provide is tight temperature control, that it is typically not possible with a LHP. [16]

21


Figure 27: Schematics of a traditional LHP

1.2. Active thermal control components

Active systems maintain temperature through an active control scheme and are generally more complex, expensive, and heavier. They include heaters, pumped fluid loops, etc. Active systems are only used when thermal requirements cannot be efficiently met using simple passive systems.

1.2.1. Heaters Large variations in the environment and in component heat generation rates

can drive passive systems outside their temperature limits. In these cases, active systems, such as heaters, are required. The heater primary functions are to protect components for cold case conditions, to supply heat when components are not operational, and to warm components to minimum operating temperatures before they are turned on. Almost all heater systems have some sort of switch or control. This typically involves a relay that is commendable from the ground to enable or disable power to the heater, a fuse to protect the spacecraft from a short circuit, and, most commonly, a thermostat or solid-state controller to turn the heater on and off at predetermined temperatures.[12] Patch heaters are the most common type of heater used for thermal control. A patch heater consists of an electrically resistive element sandwiched between two sheets of a flexible electrically insulating material, such as Kapton. An example is shown in Figure 28.

22

Chapter 2

Figure 28: Kapton Film Strip Heater

1.2.2. Pumped fluid loops

When large heat loads must be transported over a large distance from one

point on a payload to another, pumped fluid loops (PFL) provide the most efficient means. Space based pumped fluid loops are similar to those used on Earth. The basic components include a pumping device, heat exchanger, radiator, and a working fluid. The fluid can either be expendable or nonexpendable. Expendable fluids are released to space when consumed; nonexpendable fluids are re-circulated in the system. As with ground systems, designers must be concerned with basic fluid flow equations, friction analysis, heat exchanger design, pump sizing and working fluid analysis. The overall complexity of the system is dependent on the requirements of the thermal control subsystem. Because of their mass and complexity, PFLs are mainly used on large payload such as the ISS, Shuttle, and the Mars Landers.

1.3. AMS-02 Thermal control system

The AMS-02 Thermal Control System (TCS) is being developed and designed by the AMS experiment team. During nominal operations on ISS, AMS-02 draws up to 2600 watts of power. This power must be dissipated as heat, while maintaining all components within their temperature limits and maintaining the Vacuum Case as cold as possible.

The payload also must be able to survive STS environments, handoff between STS and ISS, periods with no power (both during transfer and while berthed on ISS) and peak power excursions. Passive thermal design options are utilized as much as possible, but more complex thermal control hardware is required for some subdetector components to assure mission success. TCS specific hardware includes radiators, heaters, thermal blankets, heat pipes, loop heat pipes, optical coatings and a dedicated CO2 pumped loop system for Tracker cooling, object of this thesis and described in detail in the chapter 3. AMS-02 is designed such that passive thermal control is all that is required to sustain the payload safely through extended periods of power loss without hazard.

23


For these reasons, the Thermal Control System (TCS) is an integral part of the whole AMS-02 instrument and its detailed description will be given in the following. Environmental conditions and experimental requirements driving the TCS design will be first reviewed. Following a description of the passive and active technologies used for the TCS will be described.

1.3.1. AMS-02 thermal environments and requirements The primary external factor in the thermal environment is solar illumination.

This depends primarily on the angle between the ISS orbital plane and the direction to the Sun. The “beta” angle is illustrated in Figure 29(a). For the ISS orbital inclination (51.6°) and the tilt of the Earth’s axis (23.5°), the beta angle varies between −75.1° and +75.1°, modulated by the seasons as shown in Figure 29 (b).

For beta angles with magnitudes greater than 70° the orbit is entirely in sunlight, while for beta angles near 0° about 40% of the orbit is in the Earth’s shadow .For the majority of the time, the beta angle lies within −50° to +50°. The intensity of the solar illumination, or solar constant, also varies annually with the distance to the sun, from 1322 to 1424 W/m2 at closest approach.

Figure 29: Beta Angle Definition (a) and Variation (b)

Heat from the Earth is accounted for in two ways. The temperature of the

Earth as seen from space varies from 245 to 266 K. The albedo constant, which is associated with reflected solar radiation also varies from 20 to 40%.In addition, the thermo-optical properties (emissivity and absorptivity) of the exposed experimental surface change with time.

As summarized in Table 2-1 the extreme values of the main external parameters have been taken into consideration to define the average worst hot and cold conditions along the ISS orbit.

24

Chapter 2

Table 2-1: Parameters influencing external heat loads for the worst cases Internally to the experiment, the thermal environment is also complex: all

subsystems are characterized by a different power consumption/heat dissipation which reflects the highly specialized functions of each AMS-02 component. All of them are thermally interconnected by radiation or conduction through common support structures and cabling.[17],[18].

For any given element of the AMS-02 instrument two different temperature ranges are defined in the design phase and fix the thermal requirements for that element:

• The operative range: the sub-system temperature should always be kept within this range when it is active. Switch-off or active thermal control are applied if this requirement is not met.

• The non-operative or survival range: the sub-system temperature should always be kept within this range. Permanent damage of the system can occur if this requirement is not met, and active thermal control is applied to avoid trespassing of the limits. Table 2-2 summarizes operational/survival temperature limits for the AMS-

02 major elements as well as requirements, when applicable, for maximal temperature variations along the orbit or in the element volume. In fact, as the response of all detectors depends on the temperature, increased stability over time and volume can minimize the possibility of systematic shifts in the physics analysis.

Table 2-

2: Summary of thermal requirements of major AMS-02 elements

Elements Dissipation Non-operating

range

Operating range

Max ∆T per orbit

Max ∆T over

volume TRD 18.5 W -20°C,+40°C +10°C,+25°C 2°C

Tracker (silicon)

0 W -20°C,+40°C -10°C,+25°C 3°C 10°C

Tracker (hybrid)

144 W -20°C,+60°C -10°C,+40°C 3°C

Related Factors Worst hot cases Worst cold cases Solar constant 1424 W/m2 1322 W/m2

Albedo 0,4 0,2 Earth temperature 266,5 K 245,5 K ISS attitude 277,8 km 500 km Thermal-optical the properties

Values at the End of life (EOL)

Values at Beginning of life (BOL)

25


Cryocoolers 432 W -10°C,+40°C -10°C,+20°C

RICH 49.8 W -35°C,+60°C -30°C,+50°C 7°C 15°C ECAL 68.4 W -40°C,+40°C -20°C,+40°C 5°C 10°C TOF 8.25 W -40°C,+60°C -30°C,+55°C 5°C 10°C ACC 1 W -20°C,+40°C -20°C,+40°C

Electronics 1500 W -40°C,+80°C -20°C,+50°C

1.3.2. AMS-02 Active and passive thermal control components

1.3.2.1. Radiators

Most of the heat generated by AMS-02 is rejected to space via dedicated radiators (Figure 210) disposed above (Zenith) and on the two sides of the instrument (Ram and Wake). Ram and Wake are the two side of AMS-02 with respect to the flight direction: RAM is the side in the flight direction and WAKE is the opposite side. [19]

Figure 210: AMS-02 Radiators and thermal loads

Main WAKE and RAM radiators are designed to both dissipate heat from the electronics crates and provide their structural support. The crates, which are optimized to transfer heat to the radiator, are bolted directly to the honeycomb panel using inserts. A silicone based thermal interface filler, Chotherm 1671, is used to minimize the thermal resistance across this interface. They consist of a 25mm thick ROHACELL R core with 0.5mm thick 6061-T6 aluminium face sheets and imbedded heat pipes. A cross section is shown in Figure 211. The heat pipes are standard axial groove, made of aluminium 6063 and filled with high purity .ammonia. The outer surfaces of the radiator face sheets are painted with SG121FD

26

Chapter 2

white paint to optimize heat rejection. Portions of the crates and inner radiator surfaces are covered with MLI blankets to minimize heat rejection back to the vacuum case and to adjacent ISS payloads.

Figure 211: Main radiator Cross Section (left) and RAM main radiator het pipe loop

(right)

The Ram and Wake Tracker Radiators are designed to reject the heat

transported by the Tracker Thermal Control System (TTCS), a two-phase CO2 loop running from inside the Tracker to condensers mounted on the Radiators. A dedicated detailed description will be given in the chapter 3.

The zenith radiators consist of four separate panels, each designed to reject heat (up to 150 watts) transported via two Loop Heat Pipes (LHPs) from a single Cryocooler (Figure 212). The radiator panels are constructed with aluminium 2024 T81 face sheets (1.6 mm for the upper face, sheet and 0.3mm for the lower), with a 10 mm ROHACELL R core. The condenser portion of each Loop Heat Pipes is a 4 mm OD (3 mm ID) aluminium 6063 tube, which is brazed to the upper face sheet of the radiator along a path designed to optimally reject heat. The outer face of the Zenith Radiator is covered with Silver-Teflon to maximize heat rejection capability. An MLI blanket is used on the Zenith Radiator side facing the TRD in order to minimize the heat transfer to the detector. [19]

27


Figure 212: Zenith Radiator Panels

1.3.2.2. Multi-Layer Insulation (MLI) Blankets

AMS-02 will have numerous MLI blankets on various components and sub-detectors. Concepts for a few of the larger blankets are shown in Figure 213. Typical construction will include Beta cloth as the outermost surface, 5 to 20 layer of aluminized Mylar separated by Dacron scrim, and reinforced aluminized Kapton as an inner surface. All MLI blankets used on AMS-02 will meet or exceed the NASA requirements for grounding and venting and will be positively secured [19].

Figure 213: AMS-02 MLI Blankets

1.3.2.3. Heaters

28

Chapter 2

Most heaters on AMS-02 will be used to assure that electronics are sufficiently warm before they are turned on. These heaters are mounted on the Main Radiators at locations where the embedded heat pipes can conduct heat to the crates. When AMS-02 will receive power for the first time, thermostatically controlled heaters will warm up the Power Distribution System (PDS) crate to its minimum switch-on temperature. After the PDS will be turned on, it will enable other heaters to warm up other electronics. When switch-on temperatures will be achieved, heaters will be disabled (by the PDS) prior to turning on electronics. The PDS supplies 11 distinct 120 V heater circuits which may be disabled or enabled as needed.

Additional heaters that will be activated during normal operation include those for the RICH, ECAL, Lower TOF, TRD, TRD Gas System, Tracker Thermal Control System, CAB, High Voltage Bricks, and for the Warm Helium Supply. Heaters on the TTCS CO2 lines will be used to thaw frozen CO2 in the event of a loss of power while in a cold environment (Figure 214). Heaters on the Cryocoolers are used to heat them up to their minimum switch-on temperature and to start the Loop Heat Pipes [19][21].

Figure 214: Heaters on the TTCS CO2 lines and condenser plate to thaw frozen CO2 in

case of power loss.

1.3.2.4. Heat pipes

Passive thermal control of AMS-02 includes the use of various axial groove heat pipes. While AMS-02 heat pipes vary in terms of length and cross section, all are constructed of aluminium and filled with high-purity ammonia. The amount of ammonia in each pipe is so small that freezing poses no concern. As discussed in Section 2.1.3, heat pipes are embedded in both the Tracker and Main Radiators to help distribute heat. Besides radiators, heat pipes are also used in the Cryomagnet

29


Avionics Box (CAB) base plate, the TTCS control box and to minimize temperature gradients across one of the USS-02 joints [19].

1.3.2.5. Optics

Thermal optical properties of external AMS-02 surfaces play a critical role in the thermal control of the payload. Typically surface optical properties are selected to bias temperatures cold where needed. This is achieved by selecting coatings which have a low ratio of solar absorptivity a over Infra-Red (IR) emissivity e. Much of AMS-02 is covered and their emissivity with MLI blankets, which use a glass-fiber cloth (e.g. Betacloth) as the outer surface. The Main Radiators and Tracker Radiators are painted with SG121FD white paint, a very stable, low a/e coating similar to what is used on the ISS radiators (Figure 215).

The Zenith radiator, +/- X quadrants of the Vacuum case, portions of the USS-02, +Y face of the CAB, and various other small electronics are covered with a silver-Teflon film as in Figure 216 Figure 216 (typically 5 or 10 mil FEP over vapour deposited silver over vapour deposited Inconel with 966 acrylic adhesive). This film has the lowest ratio of a/e, but since it is highly specular, its use is limited to surfaces where it is absolutely needed. All exposed aluminium surfaces are anodized for corrosion protection. Except for a few exceptions (handrails, grapple fixtures) this is a clear anodize which keeps temperatures reasonably cool. Table 2-3 lists optical coatings and properties for all significant exposed surfaces (bolt heads, rivets, cable ties, etc. are not considered thermally significant). [19]

Figure 215:Main radiator RAM on the left (a) and WAKE on the right (b)white

painted with SG121FD

30

Chapter 2

Figure 216: Zenith radiator covered with Silver-Teflon film

Table 2-3: AMS-02 Surface Optical Properties

Beginning of Life (BOL) End of Life (EOL) Surface Optical Property Absorptivity

a Emissivity e Absorptivity a Emissivity e

Beta Cloth 0.,22 0,9 0,47 0,86 White Paint 0.,18 0,94 0,27 0,88 Aluminized polyimide

0.,14 0,05 0,14 0,05

Super Teflon 5 mil 0.,08 0,78 0,13 0,75 Super Teflon 10 mil 0,09 0,89 0,15 0,85 Mixed properties (on Magnet)

0.,16 0,80 0,32 0,77

Anodized aluminium (clear anodize)

0,35 0,84 0,77 0,81

RICH mirror 0,03 0,82 0,1 0,75 Black anodized 0,88 0,82 0,88 0,78 Gold anodized Al 0,59 0,84 0,64 0,84

1.3.2.6. Cryocooler cooling

Cryocooler cooling is achieved using two redundant propylene Loop Heat Pipes (LHPs) to collect and transport heat from each of the four cryocoolers to the

31


corresponding zenith-mounted, direct-flow radiator. The evaporator portion of each LHP is attached to a heat rejection collar on the cryocooler body (Figure 217). This bolted interface includes an Indium interface filler to minimize the thermal resistance. The LHP does not interface directly with the Cryomagnet pressurized systems. Heaters are used for Cryocooler start-up and to keep them above minimum storage limits .Each LHP is made primarily of stainless steel, with nickel wicks and high purity propylene as a working fluid. 3 mm stainless steel tubing runs to the edge of the radiator, where it is transitioned to aluminium tubing via a bimetallic joint. As mentioned in the previous section, this aluminium tubing is brazed to the upper aluminium skin of the zenith radiators. [19]

Figure 217: LHP Evaporators Mounted on Cryocooler: positioning on AMS (left) and

details of the evaporators (right)

32

Tracker Thermal Control System

3. Tracker Thermal Control System

1.1. General Overview

The AMS-02 Tracker Thermal Control System (TTCS) is a two-phase cooling system developed by NIKHEF (The Netherlands), Geneva University (Suisse), INFN Perugia (Italy), Sun Yat Sen University Guangzhou (China), Aerospace Industrial Development Company (Taiwan) and NLR (The Netherlands). The TTCS is a mechanically pumped two-phase carbon dioxide cooling loop. The main objective is to provide accurate temperature control of AMS-02 Tracker front-end electronics. An additional objective is to prove and qualify a two-phase pumped cooling system in orbit and collect operational data in micro-g environment over a period of three years.[20]

The objective of the cooling system is to collect the dissipated heat at the tracker electronics and transport the heat to two dedicated heat pipe radiators. One radiator is located at the WAKE (anti-flight direction) side and the other one at the RAM (flight direction) side of the AMS instrument.

The two-phase loop incorporates a long evaporator, picking up the heat from the multiple heat-input stations evenly distributed over the six Tracker silicon planes. The heat is transported to condensers mounted onto the heat pipe radiators. The liquid is transported back to the evaporator by means of a mechanical pump.

The heat producing elements, the tracker front-end hybrid electronics are situated at the periphery of the tracker silicon planes and are located inside the cryogenic magnet. A total of 144 Watt is produced at 192 locations (Figure 3-1) and an additional 6-10 Watt cooling capacity is required for additional electronics and the Star Tracker, which is also attached to the loop. The temperature requirements for the silicon waver and the hybrid front-end electronics are summarized in the following table:

Chapter 3

33

Chapter 3

Table 3-1: Tracker thermal requirement on the silicon wafers and hybrid circuit

Silicon wafer thermal requirements Hybrid circuit thermal requirements Operating temperature: Operating temperature: -10˚C / +25˚C -10˚C / +25˚C Survival temperature: Survival temperature: -20˚C / +40˚C -20˚C / +40˚C Temperature stability: Temperature stability: 3˚C per orbit 3˚C per orbit Max. allowed gradient between any silicon:

10.0˚C Dissipated heat: Dissipated heat: 2.0 W EOL 144 W total (±10%)

0.75 W per hybrid pair (S=0.47 W, K=0.28W)

The thermal design challenges of the TTCS for ASM-02 are:

• Compatibility with the existing Tracker Hardware. • Limited volume. • Multiple and widely distributed heat inputs up to 160 W. • Minimal temperature gradients of less then 1oC • Low mass budget (< 72.9 kg), low power budget (< 80 watt). • High reliability i.e. fully redundant system design. • Two radiators thermally out of phase.

34


Figure 3-1: AMS-02 Silicon Tracker Schematic

1.1.1. TTCS loop lay-out The main functionality of the TTCS loop is to transport heat dissipated by

the tracker electronics to radiators that radiate the heat to deep space. For reliability reasons, two redundant loops have been implemented. In

Figure 32 the layout of the primary TTCS-loop is given. The secondary loop is a hydrodynamic complete independent of the first one but has the same layout (Figure 33). By following the loop routing in the below schematics, the loop operation is explained. At the pre-heaters the working fluid temperature is lifted to the saturation temperature. The working fluid enters the evaporator with a quality slightly above zero, ensuring a uniform temperature along the complete evaporator. Due to the widely distributed front-end electronics, the evaporator consists of two parallel branches collecting the heat at the bottom and top side of the Tracker planes. At an overall mass flow of 2 g/s, the mean quality at the outlet of the evaporators is approximately 30%.

The two-phase flow of both branches is mixed and led through the heat exchanger where heat is exchanged with the incoming subcooled liquid. Behind the heat exchanger the two-phase line (red) is split. One branch leads to the condensers at the RAM heat pipe radiator and the other is lead to the condensers at the WAKE heat pipe radiator. At the radiators the heat is rejected to space. After the mixing point of the two radiator branches, the sub-cooled fluid passes the accumulator. By controlling the accumulator temperature the evaporator set-point temperature is controlled by Peltier elements (cooling) and heaters. The set point can be varied to avoid extreme sub-cooling or operation with liquid temperatures just below saturation at the inlet of the pump. A distinct amount of sub-cooling is required to

35

Chapter 3

avoid cavitation at the pump. Behind the pump the sub-cooled fluid is warmed up in the heat exchanger before it enters again the pre-heater section.

36


Figure 32:Schematic of the Tracker Thermal Control System Primary Loop

37

Chapter 3

Figure 33:Schematic of the Tracker Thermal Control System Secondary Loop

38


1.1.2. Concept selection The concept selection to come to the current design, was based on the

common technique of thermal control design for spacecraft components. Active thermal control system were selected , as being the possible solutions for the tracker thermal control; three designs were considered :

• Capillary pumped systems • Single phase pumped systems • Two-phase pumped systems

Capillary pumped systems consisting of loop heat pipes and heat pipes are widely used in satellite thermal control. Large advantage is therefore the flight heritage and experience with these systems. However, with the lay-out of the widely distributed heat input, no Loop Heat Pipe construction inside the Tracker was possible due to the limited size of LHP evaporators. Alternatively a two-stage approach could have been followed. However this would have meant implementing a complex HP structure inside the magnet to collect the heat at the 192 locations. A second connecting layer of heat pipes would have been needed to transport the heat outside the magnet and Tracker where the heat would have been collected by LHP’s transporting the heat to the Tracker radiators. This design was rejected as the amount of metal mass and hardware inside the magnet would be detrimental to the AMS-02 experiment.

A second natural solution was implementing a single-phase mechanically pumped loop, that requires little volume inside the Tracker and the heat at the 192 dissipating elements can be collected by the heat capacity of the working fluid on a tube routed along all heat sources. The collected heat is then transported to the radiators and rejected to deep space. Ammonia would have been the most promising candidate working fluid for such a system. However it was found that even with the maximum permitted tube diameter the temperature drop over the needed tube length was far above the required maximum of 10 ºC between two Tracker silicons. As straightforward solutions were not feasible a more dedicated system was needed.

A two-phase pumped loop has large advantages to account for the AMS TTCS design challenges. It provides an almost isothermal environment for all the 192 distributed electronics elements as the heat is collected by evaporating fluid inside a tube. The temperature rise will be in the order of 1 K. The tube diameter size will be small (<3 mm) compared to single phase systems and the required pumping power will be small (<10 Watt).

To fulfil the temperature gradient requirements inside the tracker the mass flow is relative low compared to the single-phase pumped system. Where the single-phase system uses the sensible heat, a two-phase system uses the latent heat and can therefore transport in the order of 100-1000 times more heat with the same mass flow.

It was decided to verify if boiling a working fluid inside a tube could deal with the temperature stability and limited volume requirements. Based on pressure

39

Chapter 3

drop calculations however this showed to be also not straightforward [20]. For instance the boiling of ammonia in a small diameter tube (3mm) along the Tracker induces such high pressure drops that the corresponding temperature drop along the 9m long tube would exceed by far the temperature stability requirement. The problem is the large vapour pressure drop as the ammonia vapour density is 1000x smaller than the liquid density. Only a working fluid with a small ratio between vapour and liquid density could fulfil the requirements. The only possible candidate is CO2 with a ratio of 1:10. The concept was first tested at NIKHEF, and further optimised and tested in a full scale breadboard at NLR.

Drawback and concern of both the mechanically pumped single-phase and two-phase systems is the presence of a pump. For reliability reasons redundant pumps are foreseen in each separate redundant loop.

A quantitative comparison between a single-phase and a two-phase pumped loop is given in Table 3-2 and based on the following data:

• System Properties general: o Dissipated heat = 154 W o Tubing: L= 73m (length), t= 1mm (thickness), stainless steel o Mean velocity in tubing = 1 m/s o System Pressure head ~ 1.6 bar o Efficiency pump ~ 35% o Carbon dioxide properties at 0 ºC

• System Properties 1-phase:Maximum temperature gradient in evaporator ∆T = 2 K

• System Properties 2-phase:Maximum vapour quality in evaporator X = 0.35 Table 3-2:Comparison between a two-phase and single-phase CO2 pumped loop

Single-phase Two-phase

Mass Flow 31 g/s 2.25 g/s

Pump Power 15 Watt 1.1 Watt

Mean tubing diameter (based on a 1m/s mean flow velocity)

6.3 mm 1.8 mm

Fluid mass 2.1 kg 0.2 kg

Tubing mass 13 kg 5 kg

Total mass 15.1 kg 5.2 kg

40


The table shows the mass benefit of the two-phase system over the single-phase system. Drawback of the two-phase pumped system is the lack of sufficient flight heritage and the presence of a pump. However the system easily fulfilled the envelope requirements and was more mass effective than the alternative solutions. The two-phase mechanically pumped loop was finally chosen as preferred concept. The main advantages over the single-phase system are the almost perfect isothermal operation, the low mass, the small volume, and the low pump power required.

1.1.3. Working fluid selection With a two-phase pumped loop as baseline the working fluids are limited to

fluids with a boiling temperature in the operating range of the pay-load. Other working fluid requirements are:

• Low liquid/vapour density ratio • Boiling temperature range: -10 ºC to +20 ºC • Temperature survival range –100 ºC to + 65 ºC • Safety: Non-toxic • Radiation resistant fluids • Working pressure

High vapour flow velocities induce considerable pressure drops in the evaporator part of the loop. Apart from the additional pumping power needed to circulate the flow, a pressure drop also induces a temperature gradient in the saturation temperature along the evaporator. Working fluids with high liquid/vapour density ratios are therefore unfavourable as they introduce either large diameter piping in the evaporator or cause large temperature drops. Another important issue is safety. Non-toxic working fluids are preferred in view of the amount of fluid required (approx. 2 litre) and the Space Shuttle safety requirements.

Candidate fluids for the two-phase pumped system were among others; ammonia, carbon dioxide, freon-like fluids. Ammonia is the most common working fluid used in satellite cooling systems and has large flight heritage. However the large liquid/vapour density ratio in the order of 102-103 is unfavourable. Also the system safety is a point of concern for system integration and tests. Freon-like fluids also have the disadvantage of having a high liquid/vapour density ratios, in the order of 102. Other drawbacks are the limited radiation resistance of some fluids and the unknown behaviour of mixtures in micro-g environment. Carbon-dioxide finally, has a low liquid/vapour density ratio in the order of 1 to 10 ideal for low pumping power (< 0 W), allows small tube diameters (d< 3 mm), and induces only small temperature drops in the evaporator (<1 K). Additional advantages are the proven radiation resistance (nuclear power plant cooling) and the low toxicity of carbon dioxide. Although it has a high working pressure and therefore a high design pressure (160 bar) carbon dioxide is selected as the preferred working fluid as it

41

Chapter 3

combines the advantages of low toxicity, low temperature drop in the evaporator and low pumping power.

1.1.4. TTCS Hardware locations

TTCS hardware is widely distributed over AMS-02 as it transports the heat from the Tracker at the centre of AMS to the Tracker radiators located on the RAM and Wake top sides.

Figure 34:Location of TTCS Hardware

Two complete redundant systems are integrated. The Primary loop is

located on the Port side and the Secondary loop is located on the Starboard side. Each system consists of the five main components:

• Evaporators (2 per loop, one at the bottom and on at the top) • Tracker Thermal Control Box (TTCB) • Condensers (2 per loop) • Transport tubes to connect the components • Tracker Thermal Control Electronics (TTCE)

The heart of the TTCS loops are the Tracker Thermal Control boxes. In these TTCB’s all components to operate the TTCS loops are combined. The TTCB are connected the AMS Unique Support Structure (USS) on the Wake side. Inside the magnet the TTCS bottom and top evaporators located. These thin-walled (0.2

42


mm) and 3 mm diameter tubes run at the bottom and the top of the Tracker to pick up the heat and keep the Tracker stable (< 3K/orbit) in temperature.

Figure 35:Location of the Thermal Tracker Control Boxes with and without main

wake radiator

Figure 36:Detailed views of the TTCB locations

The heat is rejected at the Tracker RAM and Wake radiators by the TTCS condensers. Each loop has one condenser on RAM-side and one on Wake side. The heat is further spread over the complete Tracker radiators by ammonia heat pipes.

43

Chapter 3

Figure 37: Location of TTCS condensers RAM

Figure 38: Location of TTCS condensers RAM

The RAM and Wake condenser, bottom and top evaporators and the TTCB

of each loop are connected by transport tubes running along the conical flange and the Wake side Vertical Support Beams (VSB).

44


Figure 39: TTCS transport tube routing

The evaporator loops inside are connected to the transport tubes by hydraulic connectors to avoid welding needs to be done with a direct coupling via the tubes to the delicate Tracker electronics. The Secondary loop has additional hydraulic connectors at the box evaporator connections. These connectors are used to attach a mini-TTCS with cooling capacity during AMS-02 beam testing. This mini-TTCS is needed to provide enough cooling capacity. Finally, both loops are operated by the Tracker Thermal Control Electronics (TTCE) located on bottom the main Wake radiator.

Figure 310:Location of the TTCE Electronics box on the Wake radiator (inside view)

1.1.5. Principal functionality of the components

45

Chapter 3

In the following table, brief description for the main component of the

TTCS is given. Further details will be discussed separately in the next chapter of this thesis.

Table 3-3. Function of the TTCS components of the loop

Component Function Pump Transport the fluid through the loop Accumulator Regulate the evaporator temperature in the tracker

Account for the expansion of the working fluid Accumulator Peltier elements

Regulate evaporation set-point in all operation modes (cooling)

Accumulator heaters Regulate evaporation set-point in all operation modes (heating) Emergency accumulator heat-up in case liquid line temperature approaches saturation temperature (to avoid cavitation in pump)

Heat Exchanger Exchange heat between hot evaporator outlet and cold evaporator inlet. Reduction of pre-heater power

Evaporator Collect heat at the tracker electronics. The evaporation process provides the temperature stability required.

Condensers Remove the heat from the working fluid to the radiators. The condensing process makes the heat transfer effective.

Absolute Pressure Sensors Monitor the absolute pressure inside the loop Differential Pressure Sensor Monitor pump pressure head Pre-heaters Heat evaporator liquid inlet to saturation point Start-up heaters Additional heater for cold start-up (off during nominal

operation) Cold Orbit heater Additional heater to keep the condenser temperature

above CO2 freezing temperature (-55ºC) during cold orbits

Liquid line health heaters Heaters to defrost the condenser inlet and outlet lines after an AMS-02 power down

Dallas Temperature Sensors Monitor temperatures TTCS temperatures Pt1000 Temperature Sensors

Control accumulator temperature Control pre-heater on/off Monitor cold temperatures on radiator and liquid lines

1.1.6. TTCS Development Philosophy

46


For the development of a fully compliant TTCS with the system requirements, the approach of the project adopts a 3-model philosophy, namely Engineering model (EM),Qualification Model (QM) and Flight Model (FM). For cost reduction reason, in fact, the Breadboard Models (BBM) and Engineering Models (EM) and the Qualification Models (QM) and Flight Spares (FS) have been combined into on model, as depicted in Figure 311.

In general The BreadBoard Models are used support design decisions and to optimise (sub)system designs. Engineering Model (EM) is flight representative in form, fit and function, without full redundancy and hi-rel parts. The engineering models are used for functional qualification, except redundancy verification, failure survival demonstration and parameter drift checking. The EM is also used for final validation of test facilities and GSE and the related procedures.[26]

The Qualification Model (QM) is developed and constructed with the objective to confirm the systems perform satisfactorily in the intended environment with adequate margin. The QM will be put through environmental tests (thermal, vibration, etc.) to confirm this. The FM is the final model that will be built and constructed after the project has cleared the QM milestone. FM is also the actual final component that will be launched. Flight Model (FM)It is subjected to formal functional and environmental acceptance testing.

Figure 311: AMS TTCS Model Philosophy

After the completion of the performance tests on one of the EM

(sub)systems the manufacturing of the Qualification Model (QM/FS) was started. The QM/FS for the several components of the TTCS, fully similar to the Flight Model (FM), have been subjected to a qualification programme (EMC, Vibration & Shock, Thermal Vacuum) and are stored as Flight Spare. After the successful qualification of the QM/FS the Flight Model manufacturing has been started. The FM (sub)systems were subjected to a functional check prior to integration in the AMS overall system.

1.2. System requirements

BBM/EM

QM/FS

FM

• Breadboard Model (BBM) • Engineering Model (EM) • Qualification Model (QM) • Flight Spare (FS) • Flight Model (FM)

47

Chapter 3

1.1.1. Functional Performance

The main objective of the TTCS system is to keep the Tracker electronics within the required temperature limits. It transports the dissipated heat from the Tracker electronics to two dedicated radiators: one on RAM-side and one on the WAKE-side of AMS. Both radiators are thermally out of phase, meaning that the orbital load is either on WAKE or on RAM, but not on both radiators at the same time.

1.1.2. Design temperatures

The design of the TTCS for the temperature limits is based on the tracker electronics requirements, that are located near the silicon wafer plates. The details of the silicon wafer and hybrid circuits of the tracker temperature are given in Table 3-4. All the temperature values are defined as operating limits that the component must remain within while operating and as survival limits that the component must remain within at all times, even when not powered.

Table 3-4: Tracker Electronic Temperature Ranges

Silicon wafer thermal requirements Hybrid circuit thermal requirements Operating temperature: Operating temperature: -10˚C / +25˚C -10˚C / +25˚C Survival temperature: Survival temperature: -20˚C / +40˚C -20˚C / +40˚C

The tracker cooling system described above includes also various sensors

and actuators. The sensors include: Pt1000 thermistors, semiconductor thermal sensors, differential and absolute pressure sensors, and pump rotational speed sensors. The actuators are: resistive heaters, peltier heat pumps, and liquid pumps speed control. All this components should also stay in temperature range specified in the following table. Table 3-5: TTCS Electronic Temperature Ranges (consistent with pump electronics reqs)

TTCE thermal requirements for electronic parts in TTCS-P and TTCS-S box Operating temperature: -20˚C / +55˚C Survival temperature: -40˚C / +80˚C

48


The electronic control system TTCE (Tracker Thermal Control Electronics),

is located at the Wake-side on the main radiator and it is included in a crate that contains two redundant independent control systems, A and B. Also in that case the electronics in the crate should withstand the operating and survival temperature define in the below table.

Table 3-6: TTCE Electronic Temperature Ranges

TTCE thermal requirements in TTCE box Operating temperature: -20˚C / +55˚C Survival temperature: -40˚C / +85˚C

Finally, also the selected fluid should meet the following temperature requirements in the overall mission timeframe.

Table 3-7 TTCS Fluid Temperature Ranges

TTCS Fluid temperature ranges Operating Temperature loop (set-point): -20˚C / +25˚C Survival temperature: -120 ˚C / +65˚C Start-up temperature: -40 ˚C / + 30˚C (accumulator start-up temperature)

1.1.3. Redundancy concept and requirements

The AMS overall philosophy is the avoidance of any single-point of failure. The TTCS subsystem is therefore completely redundant. Two complete independent loops are fully equipped to fulfil the thermal control task for the Tracker electronics. In principle one subsystem is hot and the other cold (i.e. the subsystem will not be operating at the same time).

The philosophy is further that also no single-point of failure is present in one of the two systems. All critical mechanical components in the separate loops are therefore also redundant. A list of redundant components is given in the following table.

49

Chapter 3

Table 3-8: Component redundancy . The pt1000’s used for control are triple redundant while the monitoring Pt1000’s are redundant. The survival heaters are not part of the TTCS-system but are incorporated for completeness.

Component Redundancy per loop Pump 2 Accumulator 1 Accumulator Peltier elements 2 Accumulator heaters 2 Heat Exchanger 1 Evaporator 1 Condensers 2 Absolute Pressure Sensors 2 Differential Pressure Sensor 2 Pre-heaters 2 Start-up heaters 2 Dallas Temperature Sensors 2 Pt1000 Temperature Sensors 3 or 2 Cold orbit heaters 2 Liquid line health heaters 2 Tracker radiator heaters 2 (connected to PDS A-side and to

PDS B-side) TTCE 2

The same holds for the complete chain of components and electronics. The

TTCE electronics are also completely redundant and divided in an A and B electronics block. A block diagram is shown in Figure 312.

50


Figure 312: TTCS Electronics Block Diagram From Figure 312 and Table 3-8 it is shown that all electrical components

are redundant in as well the primary as the secondary loop. One of each component is attached to electronics A and the redundant component is connected to electronics B. To allow for the maximum use of all critical components it is decided to be able to operate A and B electronics simultaneously.

1.1.4. Power and mass budget The power budget allocated to the TTCS system during operation is 134

Watt, representing the mean power to be delivered to TTCS during operation. The Secondary Loop is an exact copy of the Primary Loop the power budget is therefore the same.

For what the mass budget concern, the estimated distribution is 69.9 kg. This is the below the initial allocated budget of 72.9 kg.

1.1.5. Proof and Burst Pressure

In a two-phase system the pressure, temperature and density are interrelated. The maximum design conditions are based on the notion that the entire system pressure should not exceed 160 bar pressure (Maximum Design Pressure MDP) and a maximum temperature (average over the entire system) of 65 °C. The maximum allowable fill rate (system density) is then directly determined from the Mollier diagram, seen in Figure 313 as the intersection between the blue line (constant temperature (65 °C) line) and the horizontal line of constant pressure.

In Figure 313 the Mollier diagram is seen for CO2. The intersection of the T=65°C curve (blue line) with the ρ = 592.39 [kg/m3] (red line) occurs at a pressure of 160 Bars:

• Max Design Pressure 160 [Bar]

51

Chapter 3

• Max Design Temperature 65 [°C] • Max Design Density 592.39[kg/m3]

The above pressure, density and temperature relations are based on the notion of constant temperature over the entire system. If the temperature of a part of the loop is below the 65 °C the required volume to contain the CO2 at 160 Bars is smaller than the actual loop volume. This implies that the other parts of the loop can get warmer without exceeding the 160 Bars in pressure. This is approach is used for the safety analysis.[20]

Figure 313: Mollier diagram

Although different maximum temperatures are used for different parts of the loop the MDP in a two-phase is for all components the same. The design proof pressures are summarised in the below table and are based on the general design rules request from the ASM collaboration. For the condensers a special pressure regime is valid. The condensers are located on the Tracker radiators. In case of an AMS-02 power down these radiators can decrease to temperature well below the CO2 freezing temperature (-55°C). Therefore the condensers are designed such that

52


they can withstand the highest pressure build up during thawing (3000 bar) with still frozen inlets and outlets.

Table 3-9: TTCS MDP, proof and burst pressures

TTCS design pressures TTCS components MDP [bar] Proof pressure

[bar] Burst pressure [bar]

Evaporator tubing 160 240 640 Tubing in TTCS-P-box & TTCS-S box

160 240 640

TTCS-components - pumps - APS - DPS - Accumulator - Condenser manifolds - Hydraulic connectors

160 240 400

Condenser 3000 6000 12000

It is experimentally verified [29] that the pressure build-up in the condenser design during thawing perfectly follows the melting line (Figure 314). In Figure 6-1 a typical freezing and thawing cycle of the test is indicated in the CO2 P-T diagram shown. At starting point A the ambient pressure in the test set-up is set by controlling the temperature of the accumulator. Going from A to B the condenser is cooled to a temperature whereby CO2 freezes, while the pressure is kept constant. After a certain time at point B the setup has settled in temperature. Condenser and feed lines are now frozen. Next, the condenser is warmed up again. At point C the CO2 in the condenser has reached the melting temperature and tries to change into liquid and expand. However, the feed lines are still frozen, trapping the melting CO2. While temperature still increases, pressure starts to build up. When the feed lines finally thaw at point D, there is a sudden pressure drop as the overpressurized mixture of solid and liquid CO2 can now push away the liquid in the feed lines and take up more volume.

The highest pressure is based on the maximum Tracker radiator temperature after AMS-02 power down. The condenser design is also tested for proof and burst.

53

Chapter 3

The verified burst pressure (10,000 bar) is agreed with NASA safety and uses a safety factor of 3.33.

Figure 314 Maximum design pressure plotted along CO2 melt line

1.1.6. Leak tightness requirements and sealing

The overall leak tightness is based on a CO2 mass loss of 30g in 5.5 years

based on average operating pressures on earth and in space. The required helium leak rate equivalent is calculated to verify the leak tightness and subdivided in leak budgets of TTCS components and subassemblies The requirements for the assemblies are less strict then for the individual components. This is done for practical reasons in view of the number of test connectors necessary during verification. The leak budgets are shown below per component, subassemblies and TTCS-system. Table 3-10 Maximum allowed He leak rate, complete integrated TTCS and ‘integrated

parts’ level

Complete TTCS Maximum allowed He leak rate (vacuum method)

Complete TTCS 1x10-7 mbar.l/s Integrated parts

Component box 4x10-8 mbar.l/s

Condensers + manifolds + feed/return lines

3x10-8 mbar.l/s

54


Evaporator + feed/return lines + hydraulic connectors

3x10-8 mbar.l/s

The verification is performed with the Helium vacuum test method as this is

the best way to quantify leaks. An additional only qualitative check is performed with the Helium sniffer method on the TTCS working pressure. This is to detect leaks only arising at high pressures. Prime suspects are the hydraulic connectors used in the design. The verification of subassemblies so far (TTCB’s and condensers) showed vacuum method leak rate values better then the requirements.

The system shall be closed with pinch and a second closure. Welding is not allowed to not damage the connected Tracker Electronics. The total closure shall have the same leak tightness as hydraulic connectors used in the loop. The verification of the last closure (pinch and connector) of the system will be performed with a CO2 mass spectrometer as verification with any Helium test method is not possible. The schematic set-up is shown in the following figure.

Figure 3-15: Schematic leak detection set-up for pinch and closure

1.1.7. Cleanliness The TTCS-loop element and the TTCS components shall not contaminate

the system working fluid. Metallic particles are not allowed. The maximum number of non-metallic particles in a 100 ml sample shall be as follows and is equivalent to MIL-STD-1246 C class 100:

• > 100µm none • 100 µm 5 max • 50 µm 50 max • 25 µm 200 max • 10 µm 1200 max • 5 µm no limit

The cleanliness of components shall be verified by detection. Tubes and connectors have been procured clean and kept clean by working according to procedures. In addition all TTCS materials were chosen to be compatible with the

55

Chapter 3

working fluid Carbon Dioxide (CO2) and Isopropyl Alcohol (IPA) which was selected as cleaning material. Materials were also procured to avoid degradation at vacuum conditions, with the overall TTCS requested to be able operating in high vacuum (1*10-6 mbar).

1.1.8. Orientation The TTCS-system shall be able to operate during ground testing. For that

reason the loop split point and pre-heaters have been located such that no phase separation could occur during ground testing. The pump also was dimensioned such that the additional gravitational pressure head could have been performed. The same for the accumulator, that was requested to be able to provide full operational performance in all orientations during ground testing condition

1.3. Environmental Requirements

The TTCS is part of the AMS-02-mission. Therefore the TTCS has to be able to survive all environmental conditions during this mission. The mission is subdivided in the following mission phases:

• Storage (TTCS non-operating) • Ground testing conditions (TTCS operating/functional check) • Launch conditions (TTCS non-operating) • Shuttle bay (TTCS non-operating) • Transfer from shuttle to ISS truss (TTCS non-operating) • ISS-truss (TTCS start-up, TTCS-operating, and TTCS non-operating)

During all mission phases the TTCS must survive the environmental conditions. The environmental conditions comprise:

• Thermal environment requirements • Vibration and shock requirements • EMC and EMI requirements • Radiation requirements

In the following sections the environmental conditions are detailed. For the thermal environment the conditions are detailed per orbit and per mission phase. For all other environmental requirements the worst case requirements are given. When appropriate standards are/will be used a reference to the to be used standards is given.

1.1.1. Thermal Environmental Requirements/Orbital data

56


The TTCS is part of the AMS-experiment and will be transported by the Space Shuttle to the International Space Station ISS. Main interface of the TTCS with the orbital environment are the Tracker radiators. In this subsection first a summary is given of orbital data. Out of these conditions the worst case operational and non-operational conditions are selected and defined in separate subsections.

The extreme conditions are defined by the following table: Table 3-11. Summary of extreme thermal conditions

Extreme hot condition highest TTCS radiator temperatures

Extreme cold condition lowest TTCS radiator

temperatures high α (solar absorption) low α (solar absorption) Worst case optical

properties low ε (IR emission coefficient) high ε (IR emission coefficient)

Solar/Albedo/Earth heat load

Largest Lowest

Power dissipation during hottest/coldest part of orbit

Maximum Lowest

In Figure 316 a schematic of two extreme space station orbits is presented. Main change in orbital data is due to the change in beta angle (angle between the ISS orbital plane and the sun vector (earth centre to sun).

Figure 316: ISS estreme space station orbits

57

Chapter 3

Figure 317: Beta Angle Variation (top); distribution of time (days) spent at different

beta angles along the ISS orbit (bottom)

In Figure 320 the relative time-share of the different orbits is shown in a

histogram. Beta angles of +75 and –75 are scarce and most time AMS (on ISS) will view orbits around beta-angles of –30 and +30. In Figure 318 typical orbital interface data is presented. The full orbital data consists of: 1. MERAT temperatures of the RAM and WAKE Tracker radiators. 2. View temperatures at the back side of the radiator 3. Orbital loads (sun solar earth) on the RAM and WAKE Tracker radiators 4. Orbital loads at the back (inside) of the RAM and WAKE radiators 5. Orbital loads on the TTCB’s 6. USS I/F temperatures 7. Tracker I/F temperatures

58


Figure 318: Typical orbital data for hot orbit Beta+75-15-20-15

59

Chapter 3

The figure shows the MERAT temperatures and orbital loads on the RAM

and WAKE Tracker radiator and the TTCB-P start-up radiator. The MERAT temperatures are the effective temperatures of the radiator surroundings, excluding the orbital heat load. Each Tracker radiator is divided in four parts and for each part a MERAT temperature is given. Each node has a rectangular trapezoidal shape, whose dimensions are:

• For the lower, smaller nodes: 0.29 m2 • For the upper, bigger nodes: 0.32 m2

The nodal distribution for the RAM-radiator is shown in Figure 319

Figure 319: Nodal distribution on the RAM-radiator

The above-presented orbital data are used to calculate the temperatures of

the radiator. In non-operating cases the heat exchange with the environment is small. Therefore the Tracker radiator environment temperatures are not influenced by the radiator heat exchange. However during operation the TTCS dumps 154 Watt in the environment, thereby heating the environment. The given orbital data are therefore too optimistic. To account for this mutual interference the radiator temperatures during operation have been re-calculated in co-operative effort between NLR and CGS. The sequence is shown in Figure 320. Experience has learned that after one full iteration the radiator temperatures converge and the solution is reliable to use for design.

60


Figure 320: Calculation sequence to evaluate Tracker radiator temperatures

In the following table, the most important cases for the thermal

environmental requirement have been summarized, taking into account the different heat load of the TTCS hardwares , the orbits defined as Tracker radiator hottest and coldest orbits through the definition of the beta angles. Each cases will lead to different system design impacts, according to the specific operation and non operational phases [17]

61

Chapter 3

Table 3-12: Hot and cold case definition for both operational and non operational phases

Hot case operational

Cold case operational

Hot case non-operational

Cold case non-operational

Tracker heat load 144 Watt 144 0 Watt 0 Watt Star Tracker heat load 10 Watt 0 Watt 0 Watt 0 Watt

TTCS Pump dissipation 15 Watt 10 Watt 0 Watt 0 Watt

Beta angle case +75-15-20-15 +75-15+00-15

-75_+15_0-15 -75_0_0-15 -75-15_0+15 +75+15+15+15

+75-15-20-15 +75-15+00-15

-75_+15_0-15 -75_0_0-15 -75-15_0+15 +75+15+15+15

Design impact To define the minimum set-point in hot cases

To size the cold orbit heater to avoid condenser freezing during cold orbits.

To define the maximum non-op. environmental temperature of the Tracker radiators.

To size health heaters on the Tracker radiators. To size the accumulator to have in cold conditions fluid still present.

1.1.2. Vibration and Shock requirements

The TTCS-P, TTCS-S-box and the TTCE box should be able to withstand the vibration and shock requirements for Space Shuttle launch and transportation.

The vibration and shock requirements testing for the TTCE electronics wasaccording to Environmental requirements for AMS Tracker Electronics, CAEN Aerospace S.r.L.. TTCE vibration testing was performed at in CSIST in Taiwan.

The TTCB-component boxes were subjected to Minimum Workmanship Level Vibration testing. The profiles can be found in the following figure and table. The testing was performed at the SERMS laboratory in Italy.

Table 3-13:Minimum Wokmanship level for AMS-02

62


Figure 3-21:AMS-02 Random vibration Spectrum

1.1.3. EMC/EMI requirements The requirement for the EMC/EMI are dictated from the MIL-STD-461E .

The test has to be conducted about susceptibility: CS01, CS02, CS06, conducted emissions: CE01, CE03, Radiated susceptibility:

RS02: RS03 and radiated emissions: RE02. The verification of EMI/EMC for TTCE was performed at CSIST in Taiwan. The TTCB verification was performed at the CEM laboratory in Terni Italy.

63

TTCS Condenser: manufactruing and thermal tests

4. TTCS Condenser: manufacturing and

thermal tests

1.1. Condenser design overview

The main function of the condensers is to dump the collected heat to the Tracker RAM and WAKE radiators.

Figure 4-1: Condenser location on the heat pipe radiators

The vapour will condense at the set-point temperature. When all vapour is condensed, the liquid will be sub-cooled below the saturation point (set-point). For pump safe operation a minimum sub-cooling of 5 °C is required. Each loop has two parallel condensers, one on the WAKE Tracker radiator and one on the RAM Tracker radiator. The location of the condensers on the heat-pipe type radiators is shown in Figure 4-1.The condensers are attached to the heat pipes. The heat pipes will distribute the heat further over the radiator in axial direction. A detailed drawing of the radiator interface is shown in Figure 4-3

Chapter 4

64

Chapter 4

1.1.1. Design drivers Due to the fact that CO2 can freeze and the radiator can become extreme

cold (-120°C) the condenser design is not straightforward. The main design drivers are:

• Freeze proof design in cold orbit in accordance with NASA safety requirements for pressurised systems

• Cover a temperature range of -120 ºC to +65 ºC (critical for connection between Inconel tubes and aluminium base plate)

• Heat transfer capability in hot orbit • Small to moderate pressure drop through the condenser • Fit on the Wake and RAM tracker radiators

1.1.2. Design rationale heat transfer

In order to optimise the heat transfer capability to the radiator the design

needs look after sufficient heat transfer area and good thermal coupling between the condenser tubing and the radiator.[29]

The sufficient heat transfer area is realised by a number of parallel tubes. In total 7 parallel condenser tubes can be accommodated on each HP flange. The area can further be maximised by the length along the HP flanges.

The thermal coupling between the fluid and the radiator plate is built up of several contributions.

Main design challenge is to connect the Inconel condenser pipes to the aluminium base plate. The connection will be performed with MASTERBOND EP21TDC-2LO glue in order to cope with the CTE-difference between Inconel and aluminium. In order to show the feasibility of the connection with glue CTE testing was performed. NLR designed a test sample and test set-up and showed the feasibility.

1.1.3. Design rationale freezing

The major design challenge was to cope with the so-called freezing problem. In fact the freezing problem is a melting problem. In case of a full AMS power shutdown the temperature of the condenser section drops below the freezing temperature of CO2 (-55 ºC) down to minimum temperatures of -120 ºC. In case the condenser heats up in an un-controlled manner liquid CO2 can be present in enclosures surrounded by solid parts. Rising temperatures can then induce high pressures. This was a potential safety risk. The design solution chosen for this problem is as follows.

• Freezing is allowed in condenser part of the tubing

65


• This condenser part will be freeze proof up to maximum melting temperatures induced by the environment (-5 ºC) [25]

• It will be shown that the rest of the TTCS tubing will not be below the freezing point [26] of CO2 during the mission. So the normal TTCS MDP is valid in all sections except the condenser section. It has been also shown [27] that the MDP (Maximum Design Pressure) in

the condenser tube is 3000 bar based on a maximum condenser temperature of -5 º C. It was also shown that a small diameter Inconel 718 tube (din = 1.0mm, dout = 3.0 mm) can withstand this pressure using a safety factor of 1.5 for yield and 4.0 for burst. Based on this results a detailed design was made and further tests have been performed on the manufactured component, as describe din the following paragraphs.

1.1.4. Condenser Design Finally Inconel hardened 718 tubing was chosen with Do = 3.15±0.05 mm,

Di = 1±0.2 mm which can withstand the 3000 bar MDP. The design consists of seven parallel condenser tubes meandering over a base plate. The base plate is bolted with 98 bolts to the Tracker radiator.

Figure 4-2.Condenser assembly 3D model

66

Chapter 4

Figure 4-3: Detailed radiator drawing with condenser locations shown

67


The condenser manifolds will finally merge the 7 capillary tubes to one

TTCS transport tube. The location of the manifolds is located on the upper vacuum case joints as indicated in Figure 4-4.The manifold is a brazed connection and is brazed in one go at the same time of the hardening heat treatment of the Inconel 718. The manifold also includes a simple wire mesh filter to avoid blockage of the small condenser tubes by contamination Figure 4-5.

Figure 4-4: Location of the condenser manifolds on AMS-02

Figure 4-5: Condenser manifold detail: the 7 condenser tubes (left), the filter

positioning (centre) and the manifold connector tube (right) to be welded to the TTCS transport tubes

68

Chapter 4

Figure 4-6Condenser manifold locations detail

The 7 capillary tubes are routed along the dedicated radiator struts (Figure 4-4 and Figure 4-6). The total condenser tube length is approximately 3.3 m, considering 2.49 m of condenser tube length embedded in the condenser plate and 0.45 m the length from the manifold to the base plate “entrance”.

The condenser manufacturing has been performed at AIDC Taiwan based on a NLR design and with NLR technical support and all under the local supervision of INFN. The main manufacturing steps will be described in the next section.

1.1.5. Liquid line health heaters Around the inlet and outlet tubes of the condenser the so-called liquid line

health heaters are wrapped. These heaters are used to defrost the liquid inlet and outlet after a AMS-02 power down. In that case the condensers are frozen and part of the inlet and outlets too. To avoid liquid is created in the condenser plates right between the frozen inlet and outlet the liquid health heaters are switched on first. This will melt the CO2. After this the Tracker radiator heaters can be switched on. Detailed information on the TTCS heater design can be found in the TTCS heater specification [28].Tests at NLR have been performed to understand and design the appropriate MLI to be put around those particular section of tubes.

69


Figure 4-7 Liquid line health heaters wrapped around the condenser inlet and outlet

1.2. Manufacturing in AIDC ( Taiwan)

The Aerospace Industrial Development Corporation (AIDC), previously known as the Aero Industry Development Centre, was founded in1969l and is a state owned aerospace company based in Taichung which developed the AIDC Ching-kuo aircraft. Government support of the past 40 years has enabled AIDC to establish a talented human resource base dedicated to the aviation industry and has well-equipped AIDC with the expertise and capability in aircraft system integration, aircraft development, parts manufacturing, aircraft assembly, testing and verification. AIDC's excellent achievements have outstripped its competitors in the Asia-Pacific region and have won itself recognition from the global aerospace community as a valuable supplier. The Taichung section was choose as suitable place of the TTCS component manufacturing based on the knowledge in airframe structure and systems, research and testing, avionics parts manufacturing and testing.

1.1.1. General manufacturing sequence

70

Chapter 4

The company has been asked to manufacture 2 EM, 1 QM and 4 FM

condensers. Basically the main manufacturing sequence can be summarized in the following process. Modification in the design, material and process development have been procured while producing the EM , QM and FM units, according also to the experience achieved each time from the previous built condensers and from the test results obtain in parallel in the SYSU laboratory in China.

Figure 4-8. Bending tool (right) and bent tubes (top right) compared to 3-D model

tubes (left)

The first step was the bending operation of the 7 tubes, different in material

and dimension for the EM and QM-FM configuration, as described in the following table. According to the bending requirement and the complicated shaped of the tubes, both the conventional and the NC bending machines couldn’t be used in that case. AIDC manufactured specially designed manual bending tool (Figure 4-8).

71


Table 4-1. Major differences in the condenser engineering, qualification and flight models

EM QM FM

Tubes material SS 316L Inconel 718 Inconel 718 Dimension tubes 3mm 3.19mm 3.19mm Manifold housing material SS 316L SS 316L SS 316L Heater installation no yes yes Nutplate installation no MS21052−L06 GFP6010L-6A

Each tube had undergo a leak check (with He) to see if any damage during

the bending operation and successively cut to the exact length, fowling the specification from the NLR drawing. After that cleaning operation of the inside and outside of the tubes were performed to meet the cleanliness requirement of the overall TTCS system.

The condenser bottom and top plate, the bottom and strain reliefs and the manifold housing, connect and bracket , as well as the transport and brazing jig, have been manufactured at INFN of Perugia and sent for the assembly to Taiwan.

Figure 4-9: Condenser component manufactured in Italy at INFN-sez. Perugia.

After that, the tubes were ready to undergo the brazing process, to have

strongest connection possible between tubes and the manifold. Brazing is a method of joining two pieces of metal together with a third,

molten filler metal. The joint area is heated above the melting point of the filler metal but below the melting point of the metals being joined; the molten filler metal flows into the gap between the other two metal pieces by capillary action and forms a strong metallurgical bond as it cools. Of all the methods available for metal joining, brazing may be the most versatile. Brazed joints have also great tensile strength, they are often stronger than the two metals being bonded together. That is

72

Chapter 4

the reason why this kind of process has been selected of the joint of the condenser manifold.

The process has been conducted on capillary Inconel 718 tubes (AMS 5590 E) (for the QM and FM model, while SS was sued for the EM), the SS manifold parts and a brazing alloy material AMS4787 (Gold-Nickel Alloy).the brazing of the manifold was followed by a (re-)solution process and a hardening process. In fact, the Inconel 718 was available in AMS 5590E and to finally come to the required strength the following steps had to been performed.

• Brazing of manifold and tubes • (Re)-solution of Inconel 718 according to AMS 5589 D • Return to ambient • Hardening of Inconel 718 according to AMS 5589 D

As the re-solution process needs a high cooling down gradient steps couldn’t be combined. The two separate temperature profiles are shown in Figure 4-10. An He-leak tests in between steps has also been performed necessary.

Figure 4-10:Temperature profile brazing and re-solution to AMS 5589D ( left) and temperature profile for strain hardening according to AMS 5589D (right)

Several test has been performed on manifold samples to understand

correctly the right parameters, as the right braze material quantity and the gap between the tubes and the manifold holes. The criteria for a good brazing results was to avoid any contamination of the brazing material into the filter, to have a complete gap filled by the braze material and to let some material visibly coming out from the tubes side. To achieve also this last requirement, the Inconel 718 tubes have been a nickel plated, to increase the flowability of the alloy on the different tube material (Inconel of the QM-FM) compared to the EM one (Stainless Steel).in the Figure 4-11 the visual inspection of the EM brazing sample is shown,

73


confirming the successful brazing process for the EM condenser and, after additional test, also for the QM and FM units.

Figure 4-11: Visual inspection of the EM test sample brazing manifold to check filter and gap between tubes and manifold holes; final result on the QM and FM manifolds

To check that the brazing process was conducted according to the

specification, a proof pressure test (at 240 bar) and subsequent helium leak check have been carried out, showing that the product could be proceed further I the manufacturing process.

At this point, the most challenging step for all the manufacturing was performed: the gluing process. The selected glue, Masterbond EP21TDC-2LO has the special property to withstand to low temperature, that was one of the requirement for its selection. The challenge was in the limited working time (90minutes) that was the main driving parameter in all the gluing process and steps. The overall sequence was spread into 7 days (including curing time of the glue) and for each condenser the following sub-steps have been performed:

• Surface treatment of he condenser plates and of the tubes before the gluing, to increase the surface adhesion of the component and the glue

74

Chapter 4

• The glue was first put on all the pocket of the bottom plate, then the tubes have been pushed inside and a complete layer of glue was placed over the tubes (Figure 4-12)

Figure 4-12: the glue is first placed in the bottom plate (left) then the bent tubes are place inside(centre) and an additional layer of glue is spread over the bottom plate

(right).

• A thin layer of glue was applied on the inner surface of the top plate, only

after the installation of the 98 nutplates (only on the QM and FM condensers) that will permit the connection to the tracker radiator., the same process ahs been applied also for the strain reliefs bottom and top, with embedded in between the SS/Inconel 718 tubes. Figure 4-13

Figure 4-13: the glue on the top plate (left) , nutplates installation all over the surface

(right) and strain relief gluing.

• The overall assembly ahs been put after each step into a vacuum bag, to

reduce the curing time foreseen for the Masterbond glue. Only on the QM and FM condenser flight heaters has been attached on the surface f the top plate to complete the overall process as foreseen in design. Figure 4-14

75


Figure 4-14: assembly into vacuum bag (left); heater installation on top plate (right)

• At this point the condensers have been installed to the transport and brazing jig, designed and manufacture at INFN of Perugia, and shipped to SYSU where both the EM and QM condenser have undergone thermal test (as described in the next section) to check the gluing performance (Figure 4-15) .The four unit FM have been shipped to CENR to be finally integrated in the AMS-02 detector.

Figure 4-15: condensers installed in the transport and brazing jig, also compare do

the 3-d model ( top); EM condenser in the climate chamber at SYSU (bottom left) and QM condenser at SYSU for thermal test campaign (bottom right)

76

Chapter 4

1.3. EM condenser thermal test in the EM loop in SYSU (China)

Once the manufacturing of the EM condensers has been completed and

before starting the QM-FM productions, thermal tests have been conducted at the Sun Yat-Sen University. In this laboratory, a complete TTCS EM set-up has been built (Figure 4-16), with both primary and secondary loops, in a way as close to the AMS-02 setup as possible, with all the tubing dimensions (length, inner diameter, etc.) the same as those of the flight model (FM).

Figure 4-16. TTCS Engineering Model Test Set-up schematic

The reason was to test the performance of the TTCS in both µ-g and 3D ,

respectively to check the behaviour of the loop in simulated environments of the space orbits and of the TVT in ESTEC. In such a loop, the previous breadboard model of the condensers have been replaced with two EM ones, with the main object to test the performance of the EM condenser themselves. This overall objective can be subdivided in the following sub-goals:

• To measure the leak tightness of the components • To measure the thermal conductance between the condenser tube and the

cold-plate interface that is in contact with the condenser via a layer of Sigraflex (i.e. measure the thermal performance of the interfaces)

• To thermal cycle the condenser to verify it can withstand the extreme temperatures in orbit

• To measure, after the thermal cycling, the thermal conductance between the

77


condenser tubes and the cold-plate interface to detect any performance degradation

• To measure the leak tightness of the condenser to verify the manifolds withstand the temperature profiles. The main test sequence for the performed tests can be summarized as it

follows: 1. A preliminary helium leak test of both condensers, EM1 and EM2, to check

that all the welding and coupled areas have not been damaged during transport and were performed according to specification [23]

2. Filling of the loop with CO2 (according to the procedure described in this document [24]

3. The conductance test for Condenser EM1 and successively for the Condenser EM2

4. A further helium leak check (without CO2) for condenser EM1 and EM2 to see detect if any changes on the hardware after the tests [23]

5. Thermal cycling of only condenser EM1 and check again the with helium leak detection [23]

6. Re-filling the loop with CO2 [24] 7. Conductance test of Condenser EM1

1.1.1. Description of hardware under test The EM condensers, already jointed to the transportation jig (Figure 4-17)

have been attached by means of 98 fasteners to a cold plate (blue, at Tcp, in the schematic of Figure 4-19), that has the main objective to simulate the thermal behaviour of the flight tracker radiator of AMS.

Figure 4-17 : Engineering Model Condensers

78

Chapter 4

Figure 4-18 Structure of cold-plate

Figure 4-18 illustrates the flow direction of liquid N2, and the heaters embedded in the cold-plate are parallel with the liquid tubes.

A layer of flight Sigraflex sheet (pink in the same figure, with thermal conductance of κ sigraflex) has been put in between the condenser bottom plate and bottom cold plate, as in the real flight configuration.

In the following schematic, Figure 4-19, it is shown the detail of the connection between the condenser and the cold plate. A Figure 4-19: Structure of condenser/cold-platethermal insulation (black) has been put all round the two test items, the condenser top plate is in light blue, at temperature of Tctp , and the Masterbond thermal glue (green) in direct contact with the stainless steel tubes (grey, at Ttube) is in between the top plate and the bottom plate (with thermal conductance of κt-b). Another layer of Masterbond thermal glue where the tube have been embedded, between the bottom plate and the tube, as been considered with a thermal conductance of κ glue, and finally the bottom plate (light blue, at Tcbp), and

the cold plate heater (red, with heating power of ) are shown.

79


Figure 4-19: Structure of condenser/cold-plate for the EM condenser tests

In Figure 4-20 the schematic of the TTCS loop is given. The pink line represents the 2-phase flow, while the blue one is the liquid flow. During the EM1 test the cold orbit heater will heat the return flow of CO2 from evaporator into two-phase, while the cold plate 2 will cool the condenser EM2 by liquid N2 to lower enough the temperature.

A dedicated control interface has been developed at the SYSU to check all the parameters during the different test. Figure 4-21 shows the loop schematic as reported in the Labview software interface, where the evaporators (on the right), the accumulator and heat exchanger (in the centre) and the two condensers (on the left) are monitored trough the value of temperatures and pressure.

Figure 4-20: TTCS loop for TTCS condenser performance test.

80

Chapter 4

Figure 4-21: Loop lay-out developed at SYSU with Labview S/W to control the loop

parameters

1.1.2. Principle of thermal conductance performance To measure such thermal conductance at reasonable accuracy, the

temperature of the condenser and that of the cold plate should as uniform as possible. To meet this requirement, a test scheme is proposed below, that the heaters in a cold plate are used as heat source, and a two-phase flow in condenser tubes, which keeps the flow at the same temperature, is used as a heat sink. This is based on the assumption that the heat transfer from the condenser to the cold plate is the same as that from the cold plate to the condenser. The thermal conductance of the condenser (κcd) should be measured by measuring the temperature of the cold plate (Tcp) and that of the two-phase flow (T2f) through the condenser tubes, respectively;

and also the heat flux ( ) from the cold plate to the two-phase flow. Based on equation (1), one can calculate the κcd:

(1) where T2f=Tin=Tout (see Figure 4-19) for the two-phase flow through the condenser. There is one scheme to calibrate the heat leak. To heat the cold plate and thus the condenser without CO2 flowing through; to simplify the evaluation of the heat leak, we assume that the thermal conductance between the cold plate and the environment is the same as that between the condenser top plate and the environment. For the

81


scheme, different heating power ( , set from 5W to 20W) is applied to the cold plate each time, and the stable temperatures of the cold plate and the condenser top plate were recorded. In this case, the temperature difference between the condenser top plate and the cold plate could be ignored (to be checked by the test results). Thus the overall thermal conductance between the condenser/cold-plate and the environment (κleak) is obtained by :

(2) and then the thermal conductance between the cold-plate and the environment (κcp-

en) and that between the condenser top plate and the environment (κctp-en) are given by κcp-en=κctp-en=κleak /2. Thus:

(3) To reduce such heat leak during the measurement (with two-phase flow through the condenser), one can adjust the temperature of the environment (bench-top chamber) (Ten) as close to that of the cold plate as possible (i.e., Ten=Tcp). In this way, the heat leak is close to zero. κsigraflex can be obtained by

(4) In a general case where the environment temperature is not equal to the average temperature of the cold plate and the condenser, the κcd is given by

(5) Similarly, we can get the thermal conductance of the glue (for two-phase flow, Ttube=T2f)

(6)

where (If the measured temperature difference between the condenser top plate and the bottom plate is small, equation (7) can be simplified further.)

82

Chapter 4

For the total thermal conductance between the cold plate and the condenser tubes, in a general case:

(7) By adjusting the environment temperature to the average temperature of the cold plate and the condenser top plate during the measurement, i.e., Ten=(Tcp+Tctp)/2, the total heat leak to the environment can be ignored. Thus:

(8)

1.1.3. Test preparation and definition of test cases

A set of thermocouples have been mounted onto the condensers (Figure 4-22) and the cold-plates, after the installation of the condensers and the cold plates on the transportation jig.

Figure 4-22: Thermocouple distribution and numbering on top condenser EM1. The

same distribution and numbering is applied to EM2.

83


The surfaces of the condensers and the cold-plates, the inlet and outlet tubes and the supports have been insulate to decrease as much as possible the heat exchange with the environments. After that, the condensers and coldplates have been put into the bench-top chamber, and connected to the test loop. Successively, a check of the thermocouples, (EM1, EM2, CP1, CP2, environment) data acquisition system, cold plate heaters (CP1, CP2) before the tests were done, to prevent any error during the test running. Finally it was executed a pressure test and leak test on the connectors from the condenser to the loop, to verify that the test items were completely sealed and the filling with C02 was done. In the following table the definition of the test cases are summarized:

• Condenser EM1 was tested. At this time, the cold plate 1 (CD1) is the test cold plate while cold plate 2 (CP2) is the cooling cold plate. The condenser EM2 was used to remove the heat generated by cold plate one (CP1); this is done by cooling the CO2 with enough subcooling to achieve single phase CO2 in the return line, when merging with the two-phase CO2. Cases 1 to 4 are defined for the condenser EM1 test at this test configuration (Table 4-2)

• Condenser EM2 was tested and condenser EM1 used instead to remove the heat from cold plate two (CP2). Cases 5 to 8 are defined for the condenser EM2 test at this test configuration. (Table 4-3)

• The overall performance of condenser EM1 and EM2 have been compared to assess the symmetry of the two condensers, glued following slightly different sequence procedure.

• Heat leak test without CO2 was performed for condenser EM1 and EM2 respectively.

• Thermal cycling for condenser EM1 from +20°C to -130°C, at a heating rate or cooling rate of 2°C /min, with resident time of 20mins. In total ten cycles have been required.

• After the thermal cycling, again a performance test for the condenser EM1 was carried out to compare the performance before and after the thermal cycling (Case 9 to 12, with the same parameter of the cases described in Table 4-2)

Table 4-2. Definition of test cases for the Performance Test for Condenser EM1

84

Chapter 4

Table 4-3. Definition of test cases for the Performance Test for Condenser EM2

1.1.4. Test result for condenser thermal conductance with

horizontal manifold

Pressure test, Cleaning the loop and filling with CO2 for EM condenser test

The loop was filled with N2 gas at 32 bars and after 11 hours no pressure drops were observed, meaning that the set up of the test was well performed.

After that the loop was cleaned fluxing CO2, until the requirement for purity of CO2

(P vacuum/PCO2) < 0.01% ( purity of C02) was reached. Only two cleaning were needed to obtain the final value 1x10-4 . After that the filling of EM loop was started, according to a different procedure in respect the one foreseen for the micro-g TTCS EM test loop [24], because for this test it was not needed the accuracy requested for micro-g test. So the loop was filled with 845 g ( 586g/L), while for the real micro-g loop it is expected a value of 450 g/L.

Thermal conductance measurement EM1 and EM2

Case Tsp (°C) Tbig chamber (°C)

Pcoh+ (W)

Pheating (W)

FRtotal (g/s)

Ten (°C) Tcp2 (°C)

1(9) -20 -25 25∼ 40±1 2∼ -18±2 －40

2(10) -20 -25 60∼ 100±2 5∼ -16±2 －40

3(11 15 10 25∼ 40±1 2∼ 17±2 －5

4 (12) 15 10 50∼ 160±2 4∼ 19±2 －20


Pcoh+ (W)

Pheating (W)

FRtotal (g/s)

Ten (°C) Tcp2 (°C)

5 -20 -25 25∼ 40±1 2∼ -18±2 －40

6 -20 -25 60∼ 100±2 5∼ -16±2 －40

7 15 10 25∼ 40±1 2∼ 17±2 －5

8 15 10 50∼ 160±2 4∼ 19±2 －20

85


In all the conductance tests, because on the loop no flow meter was put, the DPS value (in mbar) was used to set the flow rate close to the requested value (2 g/s, 4 g/s and 5g/s), knowing that the accuracy of the indirect measure could be affected by a 20% error. Not only cold orbit heater and pre-heater but also evaporator heaters and start up heaters were used to control that two phase was at the entrance of the condenser, properly regulating their power. The heaters in the cold plate are used as heat source while the two-phase CO2 flowing through the condenser tubes is used as heat sink to provide a uniform temperature distribution along the condensers. In al test result also errors were considered, because of thermocouple accuracy (+-0.3) following these equations to estimate the error:

• K=P/ΔT , the total conductivity estimated as [W/m]

• ,the error on the total conductivity, where P is the heating power added to the TEST cold plate and ΔT is the temperature difference between the CO2 flow and the test condenser;

• P=U*I ; • ΔP= ΔU*I+ ΔI*U=(1%*U)*I+(2%*I)*U;

• ΔT= , that is the average temperature of the TEST cold plate;

• δ (ΔT)=0.6°C , the accuracy of thermocouples

In EM1, during the test, different values were used to let the tolerance of thermocouple (0.6) to be less than the difference in temperature obtained between condenser and cold plate. For this reason, case 1 and case 2 were performed again. An additional case was considered (4*), in the hot condition, to check the behaviour of condenser with higher power heating. In Table 44, the set parameter values that were applied, used for the EM1 condenser conductance test, in agreement with the requested values as described in Table 4-2.

Table 44. Applied values during the EM1 conductance test and sequence


Pcoh+ Pevap+Ppre+Pstart

(W)

Pheating (W)

DPS (mbar)

Ten (°C)

Tcp2 (°C)

1 -15 -20 55,3 28,77 244 -18 -42

2 -15 -20 112,35 72.69 1036 -16 -42

4 15 0 181,5 158,86 648 18 -45

86

Chapter 4

3 15 0 179,19 83,57 387 18 -45

4* 15 0 199,79 203,24 839 18 -45

2 (rep.) -20 -23,3 77,8 104,12 1298 -18 -48,2

1 (rep.) -20 -23,3 89,51 50,08 693 -18 -48,2

The following results were obtained (Table 45), calculating the conductance (GL or K) as a ratio of the P applied to the coldpate and the delta T (defined between the inlet temperature of condenser and the temperature on top of the cold plate). The first case was not taken into account in the further analysis, because the thermocouple accuracy affected the results. Table 45. Conductance value K (W/m) for condenser EM1

For what the EM2 concerns, the following tables (Table 45 and Table 4-6) describe, as for the EM1, the real test parameter used. To have results useful for a comparison between EM1 and EM2, the same conditions of EM1, in terms of power and temperature, were applied to EM2 conductance test.

Table 4-6. Applied value during the EM2 conductance test and sequence

Case Tsp (°C)

Tbig chamber (°C)

Pcoh+ Pevap+Ppre+Pstart (W)

Pheating (W)

DPS (mbar)

Ten (°C)

Tcp2 (°C)

6 -20 -23 127,1 103,46 1134 -18 -48

5 -20 -23 127,1 57,9 860 -18 -48

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K) 1 -15°C 28 n.a 1 rep -20 °C 50 ±2 42,3±22,8 2 -15°C 73 ±2 68,1+40,34 2 rep -20 °C 104 ±3 38+±9,4 3 +15°C 83 ±3 60,7±28,2 4 +15°C 158±5 66,3±18,6 4* +15°C 203 ±6 65,5±14,5

87


8 15 -10 228,52 203,34 739 +18 -30

7 15 0 250,81 123,71 642 +18 -30

Table 4-7. Conductance value K (W/m) for condenser EM2

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K) 5 -20 58 ±2 50,6 ± 27,6 6 -20 104 ±3 37 ± 8,8 7 +15 104 ±3 69,3 ± 29,5 8 +15 203 ±6 63,8 ± 13,8

Comparing the value obtained from conductance test of EM1 and EM2 (Table 48), it is possible to state that at the same condition of set point and with the same value of heating power applied to the cold plate, no relevant difference are observed in the value of total conductance. In the cold cases, both condensers present a lower conductance value in respect the hot cases.

Table 48. Comparison test result for total conductance EM1 & EM2

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K) 1 rep -20 50 ±2 42,3±22,8

2 rep -20 104 ±3 38+±9,4 3 +15 83 ±3 60,7±28,2

EM1

4* +15 203 ±6 65,5±14,5 5 -20 58 ±2 50,6 ± 27,6 6 -20 104 ±3 37 ± 8,8 7 +15 104 ±3 69,3 ± 29,5

EM2

8 +15 203 ±6 63,8 ± 13,8 Heat leak measurement without CO2 in the loop

After the conductance tests of EM1 and EM2, a check on heat leak on the loop was performed to evaluate if the results could have been be affected by the insulation of the loop. The measurement of heat leak for condenser/cold-plate two was performed at the same time as for that of condenser/cold-plate one.

88

Chapter 4

Among all the cases foreseen in the procedure [23], only one was performed because the results obtained were good enough to state that no leaks were in the loop. The test was performed leaving inside the loop 1.8 bar of CO2. Different power was put on EM1 and EM2 cold plate, 40w and 20w respectively. After stabilization, the T on cold plate was 47°C for EM1 and 17°C for EM2, leading to a kleak value of 0.7 W/K for EM1 and 0.74 W/K for EM2. Appling these values in the following equation a comparison with the previous results demonstrate that no significant heat leak was in the TTCS EM loop.

Table 49: Conductivity results with and without heat leak

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K)

κ total + δ K (W/K) without heat leak

1 rep -20 50 ±2 44 ±23 42,3±22,8 2 rep -20 104 ±3 38 ±9 38+±9,4 3 +15 83 ±3 63±29 60,7±28,2

EM1

4* +15 203 ±6 66 ±15 65,5±14,5 5 -20 58 ±2 50 ±27 50,6 ± 27,6 6 -20 104 ±3 36 ±9 37 ± 8,8 7 +15 104 ±3 69 ±29 69,3 ± 29,5

EM2

8 +15 203 ±6 63 ±14 63,8 ± 13,8 Thermal cycling

According to the requirement of the Thermal Control system group, a thermal cycling on the EM1 condenser was performed (like the one already performed at NLR to check the property of the Masterbond glue). Each cycle has been performed according to the following test condition: setting the Cold plate temperature of EM1 at -130°C with a cooling rate of 2°C/min and waiting 20 minutes for the stabilization. Then the EM1 cold plate was set at +20 °C with 2°C/min heating rate and always waiting 20 minute of dwell time. A total of ten cycles has been performed, as reported in the following Figure 4-23.

89


Figure 4-23: Thermal cycling of condenser EM1.The top cold plate has been in contact

with insulation material for all the ten cycles.

Thermal conductance after thermal cycling: measurement of EM1

After the thermal cycling test, the complete EM loop was filled. This filling was different from the previous one, because now for the real EM TTCS loop test it was strictly needed to know the real volume and mass of CO2 to put in the loop. The choice to fill complete the loop at this point instead of waiting the end of the EM condenser tests was to gain one filling operation, in view of the subsequent EM –g and 3-D test foreseen in the SYSU. For the thermal conductance test after thermal cycling, the same procedure as described before, was repeated but only for EM1.The following cases were chosen to be compared with the EM1 before thermal cycling.

Table 410: Applied value during the EM1 conductance test after thermal cycling and

sequence.

Case Tsp (°C)

Tbig chamber (°C)


Pheating (W)

DPS (mbar) Ten (°C) Tcp2 (°C)

12 15 0 206,84 204,24 863,4 18 -45

10 -20 -23 77,87 103,12 1298 -18 -48

9 -20 -23,3 89,51 50,08 693 -18 -48,2

90

Chapter 4

The comparison between the performance of condenser before and after thermal cycling is summarized in the following table:

Table 411: Conductivity results before and after thermal cycling

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K) 1 rep -20 50 ±2 42,3±22,8 2 rep -20 104 ±3 38+±9,4

EM1 before t.c.

4* +15 203 ±6 65,5±14,5 9 -20 50 ±2 47,9±18 10 -20 104 ±3 57,6±21,2

EM1 after t.c.

12 +15 203 ±6 66,7±15,3

Because the results present a big range of conductivity values, due to the thermocouple accuracy and other instrument errors, a comparison of the condenser temperature distribution before and after thermal test was performed (see Appendix B for temperature value distribution). This demonstrate that the performance of the condenser is not affected by the thermo cycling.

1.1.5. Test results for condenser thermal conductance with vertical manifold

The condenser conductivities above were tested under the condition of

horizontal condenser manifold. However, maybe there existed a problem of different flow rate in each pipe attributed to the gravity after going through the horizontal condenser manifold, indicating that gas-phase might be gather in the above pipes while liquid-phase in the below pipes, which might affect the efficiency of the condenser.

Hence the orientation of the condenser manifold has been changed to test the EM1 conductivity and re-evaluated the efficiency of the condenser.

91


Figure 4-24: horizontal (left) and vertical (right) condensers manifolds

Thermal conductance measurement EM1 with vertical manifold

The measurement EM1 with vertical manifold was the same as the measurement EM1 with horizontal manifold. The following table are listing the test case condition considered and results.

Table 4-12 Applied test values during the EM1 conductance test

Case Tsp (°C)

Tbig chamber (°C)


Pheating (W)

DPS (mbar) Ten (°C) Tcp2 (°C)

1 -20 -24 50 ±0.8 696.9 -18 -50 1

2 -20 -25 102 ±2.8 1307.9 -18 -50 2

3 15 10 202.7 ±10 832.4 18 -45 3

Table 413-EM1 conductivity with vertical manifold

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K) 1 -20 °C 50 ±0.8 57±40 2 -20 °C 102 ±2.8 48.6±14 3 +15°C 202.7 ±10 73.2±16

92

Chapter 4

Error analysis

When direct parameter was tested the random error need to be taken into account and the temperature of cold-plate top and inlet and outlet of the pipes were direct parameters in this test. While indirect parameter was tested, error propagation need to be considered according to statistics, the heating power and conductivity were two indirect parameters in this test. Therefore based on the error equations we could get the random error and propagated error.

(9)

(10)

Where is the random error and Is the system error, and in this test the thermocouple uncertainty is 0.3 °C

Table 414 Temperatures random error

δ (Tcp) δ (T2f) Overall δ (Tcp)

overall δ (T2f)

δ (Tcp-T2f) δ (Tcp-T2f)/(Tcp-T2f)

0.33 0.14 0.44 0.33 0.55 0.64 0.36 0.09 0.47 0.31 0.56 0.27 0.56 0.18 0.64 0.35 0.73 0.26

(11) , were the current and voltage respectively while the uncertainties of them are

2% and 1%. Table 415: Calibrated heating power error

i1 u1 i2 u2 2%*i1 1%*u1 2%*i2 1%*u2 δ P P δ P/P 1.79 13.9 1.8 14 0.04 0.14 0.04 0.14 0.81 50.08 0.02 2.55 19.9 2.57 20 0.05 0.20 0.05 0.20 2.86 102.15 0.03 3.6 28 3.6 28.3 0.07 0.28 0.07 0.28 10.26 202.68 0.05

The propagated error to conductivity could be got by

(12)

93


Table 416: Error propagation on conductivity

δ (Tcp-T2f)/(Tcp-T2f) δ P/P Conductivity(W/k) δ K δ K/K(%) 0.64 0.02 57.48 37.95 0.66 0.27 0.03 48.66 14.47 0.30 0.26 0.05 73.24 22.96 0.31

Comparing the EM1 conductivity of manifold horizontal with that of manifold vertical in the following Table 417, no obvious differences existed between them. It demonstrated that the orientation of the manifold didn’t affect the condenser conductance significantly, as shown also on the temperature comparison of the thermocouples in Appendix A.

The final conclusion after all the EM condenser test was that the production of the QM model could be started in the AIDC facility, using the flight material and design foreseen.

Table 417 EM1 conductivities comparison with horizontal and vertical manifolds

Case Tset point(°C) P + δ P (W) κ total + δ K (W/K) 1 -20 50 ±0.8 57±37

2 -20 102 ±2.8 48.6±14

EM1 Manifold vertical

3 +15 202.7 ±10 73.2±22 9 -20 50 ±2 47,9±18 10 -20 104 ±3 57,6±21,2

EM1 Manifold horizontal

12 +15 203 ±6 66,7±15,3

1.4. QM condenser thermal test in the EM loop in SYSU (China)

The QM condenser has been manufactured following a slightly different

gluing procedure, changed according to the EM experiences. Considering also that the material used were different form the EM and closer to the final flight configuration, it was necessary another test campaign. As for the EM, the thermal performance of the glue between the condenser tubes and the condenser plate, and the thermal conductance between condenser and cold-plate, before and after thermal cycling. were checked.

In addition, a freezing temperature cycle test was performed to evaluate the condenser performance when CO2 freezing encountered. The main test sequence for the performed tests can be summarized as it follows:

94

Chapter 4

1. Perform helium leak test of QM condenser 2. Perform the glue-performance test of condenser QM in the condition

with silicon gel plate and heaters 3. Conductance test 4. Empty the loop, seal it to avoid water vapor. 5. Perform heat-leak test (without CO2) of condenser QM 6. Carry out thermal cycling to the QM condenser with the cold plate. 7. Fill the loop 8. QM condenser thermal conductance test (after thermal cycling). 9. Perform the glue-performance test of condenser QM in the condition

with silicon gel plate and heaters 10. Perform top plate heater test, record T-profile 11. Perform a freezing and defreezing cycle

1.1.1. Description of hardware under test

To avoid the disconnection of the two EM condensers form the previous set up,

the QM condenser was connected to TTCS Heat exchanger in place of one of the two evaporators (both disconnected from the loop) and the transportation tube of the other is blocked by a valve. A by-pass flow meter was used to measure the flow rate through the QM Condenser (see Figure 4-25 ).

Figure 4-25: .TTCS loop for QM condenser performance test.

Only for the glue performance test, the condenser QM was not connected to the

cold plate, but to a silicongel plate as shown in Figure 4-26: the thermal insulation (black), foil heater (red) on the condenser top plate (light green) and silicongel,

95


thermal glue (green) between the top plate and the bottom plate (light green), condenser tube (grey), silicongel plate (blue), and stainless plate(yellow).

Figure 4-26 Structure of QM condenser and silicongel plate

Figure 4-27: The structure of the QM condenser and silicongel

The silicongel plate was mounted onto the condenser by 98 bolts. For each bolt, a running torque and a seating torque has been recorded to avoid the damage to the flight nut-plate.

In Figure 4-28 are shown the five types of foil heaters mounted on the silicongel to simulate the heat resource in the QM condenser glue performance test. The specification of the foil heaters are listed in Table 418.

96

Chapter 4

Figure 4-28 :Type distribution of foil heaters on silicongel plate

During the test, the QM condenser is supposed to be heated equally and hence every part of the condenser experience the heat flux as uniform as possible with the available heaters. The condenser is divided into several subsections according to the foil heaters distribution in order to calculate the appropriate voltage introduced to each type of heaters. The temperature range of the heaters (non-operational:-120 °C to + 100 °C and operational: -40 °C to + 60 °C ) have been selected such that they can withstand the low temperature expected during the freezing test.

Table 418 Specifications of the foil heaters for QM glue test

Type Width (mm) Length (mm) Resistance (Ohm) HK5161 12.7 101.6 78.4 HK5162 12.7 152.4 52.3 HK5163 25.4 25.4 157 HK5164 25.4 50.8 78.4 HK5165 25.4 76.2 52.3

97


Figure 4-29: Subsection of the QM Condenser with voltage chains definition for the

glue test heater

As demonstrated in Figure 4-29, during QM glue performance test, the heaters of type HK5161 and HK5162 were linked in series as chain A, and then voltage supply was VA, to each parallel branch. Similarly, type HK5165 and HK5164 together were linked as chain B, and the voltage supply was VB. Heater HK5163 was standing alone and the power was Vc, as chain C.

Table 419: Specification of power supply for the three chains

Type Voltage (V) Bmp（A） Power(W) Total P Chain A 50 2.68 133.90

Chain B 50 1.15 57.39 Chain C 28 0.54 14.99

206.26

During the QM condenser top plate heater test, the foil heaters are mounted to simulate the heat resource, as shown in Figure 4-30. From number 1 to number 7, they are on the QM condenser top plate, only number 8 is on the top strain relief. The specification of the foil heaters are listed in the following Table 420:Specifications of the top plate heaters Table 420.

98

Chapter 4

In that test, only heaters from number 1 to number 4 have been used to supply voltage V to each parallel branch, with power supply specifications listed below as reference (Table 421)

Figure 4-30:Heater distribution on condenser top plate (left) and voltage chain for the

top plate heater test (right)

Table 420:Specifications of the top plate heaters

Number Width (mm) Length (mm) Resistance (Ohm) 1 6.4 264.2 50.38

2 6.4 264.2 50.48

3 6.4 264.2 50.69

4 6.4 264.2 50.22

5 6.4 264.2 49.98

6 10.7 58.4 7.78

7 10.7 58.4 7.80

8 6.4 264.2 50.66

99


Table 421:Specification of power supply for top plate heater chain

Number Voltage (V) Bmp（A） Power(W) Total P 1 35 0.69

24.30

2 35 0.69

24.27

3 35 0.69 24.17 4 35 0.70

24.40

98

The condenser have been connected to the silicon gel plate through bolts, as in the real flight conditions, to simulate the real contact between condenser and the tracker radiator. So the torque applied to each bolt needed to be measured by a torque wrench a recorded. The torque value to be applied is composed of two parts:

• the running torque is the value during bolting before tightening • the seating torque, which is defined from mechanical analysis. Both of them should be measured in lubricate condition, meaning that a

lubricant has been applied on the threads of the bolts. The running torque reference values of the screws of the QM condenser was within a range from 1 to 6 lb-inch, and for seating torque is 4.5 lb-inch ±0.25 lb-inch. For condenser and cold-plate connection the only difference was in the seating torque value that should been in the range of 2.5 lb-inch ±0.25.

100

Chapter 4

Figure 4-31:Example of bolt numbering on the QM condenser

1.1.2. Principles of QM condenser performance evaluation

Principle of the glue performance evaluation

To evaluate the glue performance and to detect possible glitch of condenser glue performance, the condenser temperature should have been as uniform as possible, so that any temperature variation caused by glue damage might have been obvious. To meet this requirement, the test described test set up set up have been adopted.: heaters on the silicongel plate as heat source and the two-phase flow in the condenser tubes used as a heat sink (the two phase keeps the flow at the same temperature from inlet to outlet)

The temperature difference of the condenser top plate and silicongel has been used as an indicator of the thermal conductance of the QM condenser: several pairs of thermocouples have been installed on condenser top plate and silicongel as described in the next section. Each end of the thermocouple on the condenser top plate reads the temperature representing the vicinal area, so does that for the ones on silicongel. By calculating the temperature difference(ΔTi) between one pair of thermocouple on the condenser top plate and the corresponding one on silicongel, it has been possible to obtain information of the glue performance at that area.

101


Thermal cycling as for the EM condenser test has been conducted to evaluate the glue performance after the glue has been subjected to extreme conditions.

After thermal cycling, the ΔTi have been measured again, and compared to the previous case, to detect if any damage by the thermal cycling, and thus a decreasing on the thermal performance have been occurred. Principle of the thermal conductance test

Thermal conductance has been evaluated as for the EM condenser performance test, as described in section 4.3.2 QM condenser freezing and defreezing test

QM condenser freezing and defreezing test was carried out in order to evaluate the condenser performance in the freezing and defreezing condition.

When the environment temperature (the cold-plate temperature in this test) drops to the point below the freezing point (for CO2 this is -56.6°C), the carbon dioxide inside the loop comes into solid phase, initially at certain coldest parts of the condenser, and eventually proceeds to the rest area including condenser inlet and outlet. When the temperature rises, if the temperature exceed the freezing point, the solid CO2 will be heated up and return to liquid phase.

Figure 4-32:Condenser cool down sequence (left) and heat up sequence (right)

Similar to the freezing process, this phase change will happen in certain hottest parts of the condenser and it is quite possible that this area will not be the inlet and outlet tubes, in consideration of their limited heat transfer area. As a result, when the condenser is heated too rapidly, the CO2 inside the tube area, located inside the

102

Chapter 4

condenser plate, will first defreeze, gradually changing into liquid, expanding its volume.

Meanwhile, if the condenser inlet and outlet temperature are still below the freezing point and with CO2 still in solid state, they will block the expanding CO2 and eventually lead to a rapid increase in absolute pressure. It may damage the condenser if the local pressure inside condenser tube reaches some certain critical value (maximum design pressure).This value is equal to the highest pressure which can be induced by the highest radiator temperature from the space environment (-5 °C).

Although the design is capable of handling this high pressure, a special heating mechanism was proposed to prevent this from happening by specifically heating up the inlet and outlet tube. The QM condenser freezing and de-freezing test was planned to verify such mechanism.

Figure 4-33:Condenser with inlet outlet heater indication

It means that when the CO2 inside the condenser tube is getting closer to

freezing point in the heating up process, the inlet and outlet temperature (Tin and Tout) needs to be inspected and ensured to be above the freezing point. This can be done, turning on the heaters of condenser inlet and outlet simultaneously when the environment temperature (that will be the cold plate temperature) begins to arise. When the cold plate temperature reaches -80°C, if Tin and Tout are already above the freezing point with certain margin (such as 5°C), the CO2 can expand freely

103


along the tube and the de-freezing process could be safe; if Tin and Tout are still below the freezing point, the heating power of cold plate will be stopped, waiting for Tin and Tout to go above -50°C.

1.1.3. Test preparation definition of test cases and sequence

In order to evaluate the glue performance between the tubes and the condenser plate, 40 sensing points of insulated thermocouple have been installed on the interface of the foil heaters and silicon gel. During glue performance test, there are 32 thermocouple on the condenser top plate heater, in the corresponding position.

Additional 8 thermocouples (3 at the condenser inlet tubes, 3 at the outlet tubes and 2 to inspect the temperature of the exposed tube) have been included. The distribution of the sensing points of thermocouples is given the following figures.

Figure 4-34:Thermocouples distribution on silicongel (left) and on the condenser top

plate

Also in that cases, the surfaces of the condensers and the cold-plates, the inlet and outlet tubes and the supports have been insulate to decrease as much as possible the heat exchange with the environments. After that, the condensers and coldplates have been put into the bench-top chamber, and connected to the test loop. Test cases and test analysed have been summarized in the Table 4-22 listed as following:

Table 4-22:Test cases for the glue performance test of QM condenser

104

Chapter 4

Case Tsp (°C) Twalk-in (°C)

Ppreheater (W) Pheating (W) FR (g/s) Tbench-top (°C)

1 15 15 8 200 2 5

2 15 15 8 150 2 5

3 -20 -20 8 200 2 -35

4 -20 -20 8 150 2 -35

1.1.4. Test results Glue performance test

The heaters attached to the silicongel have been used to heat the condenser while CO2 was flowing through, with a flow rate of 2g/s.The temperature difference of the corresponding area on the QM condenser before thermal cycling and after thermal cycling have been compared. The heater and the thermal couple distribution are shown as follows:

Figure 4-35:Type and distribution of foil heater on silicongel

As previously described, the thermocouples on the condenser top-plate are

almost in the same area of the silicongel ones, but with different label indices. In the data processing, as shown in Appendix B, those the numbers have been converted into the corresponding indices of the silicongel and the result are listed in the tables of the appendix.

105


Test results for each case are listed from Table B-1 to Table B4. is it possible to notice that some temperatures are much higher than others on the QM condenser, as for example those of 19, 27, and 35.This phenomenon may result from the fact that the foil heaters in such areas (FHK5164) have a higher heating power (heat flux) than the others.

Table 423:Heater type and heating power density during the glue performance test

Heater type Heat flux

FHK5161, FHK5164 0.889watt/cm2 FHK5163 0.774watt/cm2

FHK5162, FHK5165 0.395watt/cm2

If the glue is ruptured during thermal cycling, the thermal conductance property

and thermal contact with the condenser plate may be damaged. As a consequence, the heat transfer coefficient from the condenser bottom plate to the condenser tube and the condenser top-plate should be reduced, resulting to a high temperature difference (delta T) between the top plate (or the flowing two-phase CO2) and bottom plate. And hence result in a high, positive difference of the delta Ts before and after thermal cycling.

From the test results shown above, the delta T between the bottom and the top plate changes little before and after thermal cycling. However, the change is rather unobvious and even negative in many districts, hence the conclusion that the glue ruptured did not happen during the thermal cycling. Condenser THERMAL conductance test

The thermal conductance is a ratio of the heat flux and the temperature difference:

(12) with formula (12), and based on the test data, it has been possible to calculate the thermal conductance in each case (Table 4-22).

The results of the thermal conductance test before and after thermal cycling are listed in Table 424mand

Table 425 respectively. Since the heat leak to the surrounding is quite small (details will be described below), the heat leak effect has been neglected in the conductance calculation.

106

Chapter 4

Table 424:Thermal conductance before thermal cycling

Case Tset (°C)

T2f (°C)

T_cold-plate (°C)

P (W)

κ total (W/K)

δ K

1 15 15.8 19.0 204.2 63.8 32% 2 15 15.7 18.1 153.8 64.1 34% 3 -20 -18.3 -14.7 204.2 56.7 30% 4 -20 -18.4 -15.8 153.5 59.0 34%

Table 425:Thermal conductance after thermal cycling

Case Tset (°C)

T2f (°C)

T_cold-plate (°C)

P (W)

κ total (W/K)

δ K

1 15 15.8 18.7 203.9 69.0 33% 2 15 15.7 17.8 152.7 70.7 38% 3 -20 -18.3 -14.9 204.2 59.6 33% 4 -20 -18.5 -15.9 152.2 59.9 36%

If the glue is ruptured during thermal cycling, the thermal contact between the condenser plate and the cold plate may be damaged, and the thermal conductance may be obviously lower than the one before thermal cycling.

The thermal conductance after thermal cycling doesn’t change obviously from, and even a little higher than those before. In addition, the temperature in the same district does not differ notably (see Appendix D), indicating that the test case has been re-achieved after thermal cycling. Therefore, the glue performance is sustained and the contacted thermal conductance doesn’t get worse because of thermal cycling.

During the experiment, the temperature difference between the cold-plate and the environment is about 3°C. Based on the heat leak test result, in both cases the heat leak flux is relatively small (0.7W/K*3K=2W) in comparison with the heating power (above 150W). Accordingly, it was possible to neglect the heat leak contribution in the calculation of thermal conductance. Thermal cycling test

The thermal cycling for QM condenser from 20°C~ to -130°C has been performed, with at a heating rate or cooling rate of 2°C/min, and resident time of 20 minutes for a total of ten cycles.

107


Figure 4-36:Cold plate temperature profile during thermal cycling

Condenser freezing and defreezing test

Freezing test for QM condenser from 20°C~ -130°C, at a cooling rate 2°C/min and Defreezing from -130°C ~ -20°C at a heating rate of 1°C /min. have been carried out in order to verify the designed strength of the condenser tubes by freezing and de-freezing of the CO2 inside the tubes. Figure 4-37 presents the whole cycle of freezing and de-freezing of the QM condenser.

In the process of de-freezing, the condenser top-plate heaters were used with 50W as the heating source. Due to the heat leak rate from condenser to the ambient (approximately 0.7W/K), the overall heating power was up to 120W in the lowest temperature situation, but decreased with the increase of the condenser temperature. To ensure adequate heating power, the temperature of the big chamber has been adjusted, when the temperature of condenser plate was increasing and already above the melting point.

The temperature profiles, during the de-freezing process, of the condenser top-plate, the inlet and the outlet tubes are shown in Figure 4-39 . All the condenser top-plate is colder than the inlet and the outlet tubes (since the inlet and the outlet are kept being heated with 15.6W), indicating that the CO2 melts in the inlet and the outlet tubes well before it does inside the condenser tubes. As a result, the expansion of the melted CO2 may not encounter obstacles of solid CO2, and the pressure inside the condenser tube is below the safety threshold. This verifies the condenser validity under de-freezing condition.

108

Chapter 4

Figure 4-37:Freezing and de-freezing temperature profile

Figure 4-38:Heating power profile during de-freezing

109


Figure 4-39:Temperature distribution during defreezing

At the beginning of defreezing process, the inlet and the outlet heaters (totally 15.6W), and the top-plate heaters (38W) have been turned on. The temperature of condenser plate (solid CO2) started to increase. To avoid the melting of CO2 inside the condenser before that inside the inlet and the outlet tubes, which may lead to the quick increase of pressure in the condenser, the top-plate heaters have been shut down when the condenser plate temperature was close to the melting point of CO2, until the inlet and the outlet temperatures were above the melting point, when the top-plate heaters were turned on again. This was able to prevent the block of the CO2 expansion during melting by the solid CO2 inside the condenser tubes.

The condenser top-plate is colder than the inlet and the outlet tubes, indicating that the CO2 melts in the inlet and outlet tubes before it does inside the condenser tubes. As a result, the expansion of the melted CO2 does not encounter obstacles and the pressure inside the condenser tube is below the safety threshold, hence verifying the condenser validity under freezing and de-freezing condition.

All the EM and QM test campaign for the condenser have been successful, leading to the conclusion that the flight model manufacturing could have been started following the design and the gluing procedure foreseen for the QM condenser.

110

Chapter 5

5. TTCB QM and FM: manufacturing and

qualification tests

1.1. Tracker Thermal Control Box (TTCB) design overview 1.1.1. Thermal Tracker Control Box lay-out The heart of the TTCS loops are the Tracker Thermal Control boxes. In

these TTCB’s all components to operate the TTCS loops are combined. The TTCB are connected the AMS Unique Support Structure (USS) on the Wake side, the primary is at PORT and the secondary is located at the STARBOARD side.

Figure 5-1:Primary TTCS box on port side of AMS

Chapter 5

111

TTCB QM and FM: manufacturing and thermal tests

Figure 5-2:Secondary TTCS box on starboard side of AMS

The box contains the following components (Primary box numbering):

• Pumps located on the start-up radiator (PMP1aP, PMP1bP) • 1 Pump electronics controller box • 1 Accumulator (ACCUP) • Accumulator control heaters Flight (HTR4aP, HTR4bP) • Accumulator Peltier elements Flight (TEC1aP, TEC1bP • Accumulator control heaters Ground (non-flight) (HTR8aP, HTR8bP) • 1 Heat exchanger (HX) with 2 integrated Start-Up Heaters (SUP) (HTR5aP,

HTR5bP) • Pre-heaters (HTR1aP, HTR1bP, HTR2aP, HTR2bP) • 2 Cold Orbit Heaters (COH) (HTR10aP, HTR10bP) • 2 Absolute Pressure Sensors (APS1aP, APS1bP) • 2 Differential Pressure Sensors (DPS1aP, DPS1bP) • Pt1000 temperature sensors • Control Pt1000’s • (Pt1NaP, Pt1LaP, Pt1RaP,........ Pt5NaP, Pt5LaP, Pt5RaP) • (Pt1NbP, Pt1LbP, Pt1RbP,........ Pt5NbP, Pt5LbP, Pt5RbP) • Monitor Pt1000’s • Pt6aP, Pt6bP..... Pt11aP, Pt11bP • Dallas sensors (total 26 per box) • LSS TTCB heaters (HTR9aP, HTR9bP) • Thermostats (total 20 per box)

Most TTCB components are located on the TTCB base plate under an aluminium cover.

112

Chapter 5

Figure 5-3:TTCB-Primary final design and schematic of MLI wrapped around the

TTCB

The box cover and base plate sides are wrapped in Multi Layer Insulation

(MLI) to insulate the components from the environment. Titanium (thermally insulating) washers are used to reduce also the heat leak to and from the USS. The TTCS pumps are the single components not located on the base plate. The pumps are located on a special start-up radiator.

This start-up radiator radiates to the back side of the main wake radiator providing a lower temperature then the TTCB I/F with the USS. This is needed to increase the orbital time window for normal (liquid) TTCS start-up. The pump temperature should therefore be lower then the accumulator temperature and the CO2 critical temperature (+33 °C).

The detailed box assembly is shown in Figure 5-4 and Figure 5-5 for respectively the Primary box and Secondary box. More detailed information on the box can be found in the TTCB drawing packages provided by NLR. In Figure 5-6and Figure 5-7 pictures of the integrated boxes are shown. In the following sections the loop components are presented and the designs elucidated.

113


Figure 5-4:TTCB-P box assembly (NLR detailed design)

Figure 5-5:TTCB-S box assembly (NLR detailed design)

114

Chapter 5

Figure 5-6:TTCB-P box assembly on theUSS simulator and in the vibration frame

Figure 5-7:TTCB-S box assembly front and back side

115


1.1.2. Pump assembly The most critical component in the loop is the pump. The pump provides the

fluid flow in the TTCS. Each TTCS-loop is equipped with two pumps for redundancy reasons. The location of the pump electronics controller box and the pumps is shown below. The pump electronics controller box is located near the pumps to minimise the lengths of the pump cabling (“dirty” high frequency signals).

Figure 5-8:Pump Electronics Controller Box and Pumps in TTCB-P 3d model

Figure 5-9:Pump Electronics Controller Box in as integrated in the flight TTCB-P Pump design

Pacific Design Technologies (Santa Barbara, USA) developed the TTCS pump. The designed pump is a single-stage centrifugal pump and based on the PDT Mars Pathfinder pump. The pump housing is an all-welded design to cope with the high design pressures and strict leak tightness requirements. The pump performance

116

Chapter 5

curves are rather flat. This is due to the fact that the pump has to operate in low flow rate area (1-4 g/s) with rather high pressure heads 100-850 mbar. In terrestrial application a gear pump would be selected. In view of the life time TTCS selected a centrifugal pump.

Figure 5-10:PDT Pump Engineering Model (QM and FM are similar)

System curve

The system curve as defined in the NLR breadboard loop are shown in Figure 5-11 The calculated performance curve is shown in Figure 5-12. The nominal flow rate is 2 g/s.

Figure 5-11:TTCS system pressure drop (experimental data with TTCS Breadboard)

The blue line shows the curve for nominal operation with 144 Watt dissipated power on the loop. The black line shows the pressure drop with one

117


closed condenser. The red line is the result of an experiment with totally unbalanced evaporator branches. This means all dissipated heat (144 Watt) is put in one branch. Down to a flow of 2.7 ml/s no dry-out occurred in the parallel branch. Based on this data the pump curve requirements were defined as:

• minimum flow rate 1 ml/s and 150 mbar • maximal flow 4 ml/s at 850 mbar

The performance curve as tested by PDT shows the pump can deliver enough pressure head.

Figure 5-12:Pump performance curves (QM Pump)

Pump Specifications

The pump specifications are shown in the below design drawing Figure 5-13. Additional to this, the pump should also:

• Be compliant with the AMS-02 environmental requirements • Capable of operation in a magnetic field (140-1000 Gauss) (Hall-effect

sensors) • Be able to start-up in supercritical-vapour conditions

118

Chapter 5

Figure 5-13:Pump specification as defined in the assembly drawing from PDT

The EM and QM pump have been tested in the full loop breadboard loop in

Zhuhai (China). The mass flow and pressure head for nominal environmental conditions are shown below.

Figure 5-14:Pump performance in EM Primary Loop

More detailed pump requirements can be found in [30]and [31].

119


1.1.3. Accumulator The accumulator is the control room of the TTCS-loop. In the accumulator

the evaporation temperature in the Tracker is set. Each loop is equipped with one accumulator. The set-point control components (Heaters, Peltiers) are redundant for each separate loop.

Figure 5-15:TTCS Accumulator with integrated Peltiers and heaters

Additional ground test heaters are located on the bottom side of the Primary

accumulator. For the Secondary TTCB the ground control heaters are located on the top. This is due to the orientation of the TTCB’s during the AMS-02 system testing in the Large Space Simulator (LSS) at ESA-ESTEC Noordwijk (The Netherlands). The ground control heaters are used in hot conditions when the wick structure is not capable to re-wet the wick around the heat pipe.

Figure 5-16:TTCS Accumulator Primary bottom view (left) and TTCS Accumulator

Secondary top view (right)

120

Chapter 5

Accumulator Functions

The main function of the accumulator can be summarized as following: • To regulate the evaporator temperature in the tracker, • To account for the expansion of the working fluid, • To account for the liquid front changes in the condenser during operation

(including quality changes in condenser lines). The Accumulator Peltier elements instead should regulate evaporation set-

point in all operation modes (cooling) while regulation of evaporation set-point in all operation modes for heating there are dedicated accumulator heaters. Finally, emergency accumulator are foreseen to heat-up in case liquid line temperature approaches saturation temperature (to avoid cavitation in pump) Set-point control

The objective of the accumulator is to control the evaporator set-point and therefore the “cold plate’ temperature of the Tracker electronics. The principle is based on the property that a pressurised (closed) system with liquid and vapour has the same saturation temperature and pressure everywhere in the loop (neglecting flow pressure drop differences). In the TTCS accumulator vapour and liquid are present and is the main two-phase part in the loop. By changing the accumulator saturation temperature it is possible to change and set the evaporation temperature in the evaporator.

The actual set-point temperature control is performed by heating or cooling of the accumulator. The cooling will be performed by two Peltier elements Melcor CP 1.0-127-05 L 2 in series. The heating is done by wire heaters attached to a heat pipe heating the centre of the accumulator. Account for volume changes due to temperature changes

Apart from the set-point control the accumulator also accounts for the volume changes (thermal expansion) of the working fluid (CO2) average temperature. The CO2 liquid density-changes from the lowest average (non-operating) temperature to the highest average (non-operating) temperature have to be taken care-off by the accumulator. At the lowest temperature the accumulator should still have liquid CO2 in the accumulator and at the highest temperature the accumulator should be able to cope with the extra liquid volume CO2.

121


Account for volume changes during operation

An accumulator in a two-phase system differs fundamentally from an accumulator in a single-phase system. A single-phase loop accumulator has to account only for the thermal expansion. In a two-phase other phenomena are present and have influence on the accumulator operation.

Most important phenomenon is the change in vapour volumes present in the condenser. As the condenser temperature change during orbit the so-called condenser front (i.e. the front were all fluid in the condensers is pure liquid) changes also. The changes in the condenser front immediately have effect on the accumulator level.

• Increasing radiator temperatures/larger vapour volumes in condenser

In case of an increasing temperature in the condensers the amount of vapour in the condenser increases. As vapour density is smaller than liquid density the same mass in the loop takes more volume, resulting in a liquid flow from the loop to the accumulator. The accumulator should be able to account for this volume.

• Decreasing radiator temperatures/smaller vapour volumes in condenser

In case of a decreasing temperature in the condensers the amount of vapour in the condenser decreases. This results in a liquid flow from the accumulator to the loop. The accumulator should be able to account for this volume.

Also the decreasing and increasing volumes in the accumulator cause fluid in the accumulator to condense and evaporate. The temperature control should be able to account for this changes. Accumulator design and specifications

Although some Russian two-phase systems have flown, little experience is available with two-phase accumulators in space. The main design challenges of the required CO2-two-phase system in space were:

• Keep liquid at the entrance of the accumulator to avoid vapour from the accumulator to enters the pump,

• Keep liquid attached to the wall where controllers heat the accumulator (avoid dry-out of wick material),

• Pressure resistant up to the pressures present in the CO2-systems 160 MDP.

122

Chapter 5

Figure 5-17:TTCS Accumulator assembly

Based on TTCS system design information NLR defined the accumulator

size requirements and concept design [32].The China Academy of Space Technology (CAST) translated the concept design in a detailed design. Development tests were defined in close co-operation between NLR, NASA and CAST. INFN made the bracket design and SYSU performed the thermal safety calculations as defined by NLR.

• Liquid at the entrance of the accumulator

As the accumulator has to operate in space, one of the main challenges is to keep the liquid present at the connection with the liquid line. It has to be avoided that vapour CO2 enters the loop as it will damage the pump. To keep the liquid “attached” to the entrance of the accumulator a wick structure is used around the liquid the entrance pipe. On top of that a fan structure is used. The fan structure contains enough liquid CO2 to provide the largest possible liquid request in case of a Tracker electronics shutdown. In that case all vapour in the evaporator condenses and additional liquid is needed to fill the gaps.

Figure 5-18:Liquid inlet pipe with mesh

• Liquid attached to the accumulator wall

To be able to control the set-point in the accumulator heat must be exchanged with the liquid in the accumulator. It is therefore of vital importance that the liquid is attached to the wall in micro-g conditions to avoid dry-out. This design

123


challenge is tackled by using a heat pipe to heat the internal CO2 wick structure. By using a heat pipe no dry-out can occur as the heat provided to the heat pipe will condense on the coldest spots inside the accumulator. As soon as a part of the CO2-wick will dry out no more heat (condenses) at that location of the heat pipe. Evaporated CO2 is replenished by liquid from the surrounding wick.

• High design pressures

A special CO2-related design challenge is the relative high design pressure of CO2. The accumulator structure has to deal with these pressures without loosing the connection between the wick structure and the container wall. More detailed information on requirements and design can be found in [32].

1.1.4. Heat Exchanger The heat exchanger is used to reduce the pre-heater power during nominal

operation. The heat is exchanged between the exit and the inlet of the evaporator. The heat dissipated by the Tracker electronics is re-used for pre-heating in cold orbits. The maximum required pre-heater power is therefore reduced to 50 Watt (25 Watt per branch). In Figure 5-19 the reduction of pre-heater of required pre-heater power is shown by introduction of a heat exchanger. The modelling is performed with a SINDA-FLUINT thermo-hydraulic model, calculating two-phase flow in the plumbing on one side and single phase flow at the other side. The periodic change in temperature is due to the environmental (orbital) heat flux boundary conditions. The simulation is performed with a low-efficiency heat exchanger and therefore only indicative for the relative reduction of power.

Figure 5-19:Pre-heater power with and without heat exchanger (modelled with low-

efficiency heat exchanger)

124

Chapter 5

Based on the simulations a heat exchanger was designed, build and tested in the NLR breadboard loop. The technical implementation is shown in Figure 5-20, a plate type heat exchanger with a cold passage for the subcooled inlet and one hot passage for the evaporator outlet. A plate heat exchanger is chosen for the high pressure design, the low pressure drop, and the leak tight design. The contribution of the heat exchanger to the overall system pressure head is negligible and in the order of 5 mbar at the two-phase side and 1 mbar at the liquid side.

Figure 5-20:Heat Exchanger preliminary flight design (plate type HX with 36 plates)

and 2-phase flow path

The heat exchanger has 36 plates with 18 passages for two-phase flow and 18 passage for single-phase flow at the subcooled side. The flow directions are such that the two-phase flow enters from the top (resulting in a different orientation for Primary HX and Secondary HX). The single phase flow path exits around the mantle of the stacked plates. After leaving the heat exchanger this flow will enter the evaporator. (Figure 5-20)

Figure 5-21:Heat Exchanger Engineering Model

125


Figure 5-22:Heat Exchanger Engineering Model details

Heat exchanger design

The main design challenges for the heat exchanger are: • High pressure design (solved by choosing Inconel 625 as construction

material) • Leak tight design (all welded design) • Operation in micro-g

For the latter the number of heat exchanger plates is over-dimensioned. The measured heat exchanger performance is excellent. This resulted in a reduction from the pre-heater power to almost zero. Only in extreme cold cases 16 Watt (2 x 8W) pre-heater power is needed. More information on the heat exchanger design can be found in [33].

1.1.5. Pre-heaters

The function of the pre-heaters is to heat the sub-cooled liquid to saturation temperature (i.e. set-point) before it enters the evaporator. Each evaporator branch is equipped with its own pre-heater (see Figure 5-23) to avoid phase-separation at the split-point to the evaporator branches. This design allows therefore ground testing in normal AMS-orientation. The pre-heaters wire heaters are redundant and soldered to the evaporator branch tubing on a small copper structure. The material is stainless steel, with an outer diameter of 05.mm and power supply of 28 V DC. The control is simple on/off. The maximum required pre-heater power is 8 Watt per evaporator branch. The main design challenges for the pre-heaters are the connection with the TTCS-tubing. More details can be found in the TTCS heater document [28]

126

Chapter 5

Figure 5-23:Pre-heater assembly 3D model and real manufacturing

1.1.6. Start-up heaters Next to the pre-heaters also start-up heaters are implemented in the design.

Function of the start-up heaters is to heat the flow to the evaporators above –20 °C to avoid cooling the Tracker electronics during start-up in cold orbit conditions.

During start-up deeply cooled liquid (-40 °C) is pumped from the radiator to the Tracker. The start-up heater is able to lift the cold CO2 to –20 °C. The start-up heater is controlled with a simple on/off control. The required start-up heater is calculated to be 50 Watt and is located on the mantle Heat Exchanger. This is possible because the liquid flow exits the heat exchanger along the mantle. The start-up wire heaters are soldered to the Heat Exchanger mantle and is protected by thermostats to avoid overheating and melting of the solder.

1.1.7. Cold Orbit heater The cold orbit heater is introduced in the design to avoid freezing of the

condensers in cold orbit. The Sinda-Fluint modelling showed a 60 Watt heater was provides enough power to avoid freezing [34]. The cold orbit heater is a soldered design. A thermostat is implemented to avoid unnecessary overheating of the heater assembly.

Figure 5-24:Cold orbit heater assembly 3D model and real manufacturing

127


1.1.8. Absolute pressure sensor The function of the absolute pressure sensors is to monitor the TTCS system

pressure. As the TTCS is a two-phase system the pressure is also a direct measurement of the saturation (set-point) temperature. Main specification are:

• Range: 0 - 100 bar. • Accuracy: 0.5% FS. • Total mass: < 400 g. • Maximum dimensional envelope: h x w x b = 100 x 100 x 100 mm3. • Power consumption: < 1 W. • Space qualified

Figure 5-25:APS Assembly 3D model and real manufacturing

The APS is manufactured by CETC in China. The brackets are designed by

INFN and NLR and built in AIDC.

1.1.9. Differential Pressure Sensor The function of the DPS is to monitor the pressure drop over the pump. This

pressure drop monitors the pump health: • Extreme low values will indicate dry-running (pumping vapour) or pump

failure, • Extreme high values will indicate obstruction of the loop flow.

The differential pressure also measures the loop mass flow. This value is not accurate as it is an indirect measurement and depends on the set-point temperature and the condenser temperature (environment). However terrestrial calibration of orbital conditions can give acceptable accurate results. The measured value for the mass flow will not be used in flow control, only as health indicator of the pump. The main design challenges for the DPS is the low differential pressure at extreme high system pressure, considering that the main specification are:

• Range: 0 - 1000 mbar. • Accuracy: 0.5% FS. • Total mass: < 200 g. • Maximum dimensional envelope: h x w x b = 100x100x100 mm3. • Power consumption: < 1 W.

128

Chapter 5

Figure 5-26:APS assembly 3D model and real manufacturing

1.1.10. Temperature sensors

The TTCS uses two types of temperature sensors: • Dallas DS 1820 sensors. • Pt1000’s.

Dallas sensors are only used for monitoring. The Pt1000’s are used for all low-level control options and for temperature monitoring on locations where the Dallas sensors are outside their temperature range.

1.2. Heat exchanger manufacturing and TTCB integration in AIDC

1.1.1. QM and FM HX manufacturing AIDC company has been asked also to manufacture the 2 flight model and

one qualification model of the Heat exchanger, with the successively integration into the TTCB. In the main manufacturing sequence, the major role was played from the brazing and orbital welding process.

For what the brazing concerns, all the metal material of HX brazing parts are Inconel 625, while the brazing material was made of B-NI-2(AMS4777), cut in

129


simple circular shapes. The T-profile used to braze the HX was similar to the one used for the condenser manifold brazing. After the first brazing, helium leak test has been performed to check the integrity of the connection done.

Figure 5-27:manufacturing of HX plates, brazing assembly and Helium leak check of

the brazed component

According to the NASA document for welding requirement [36], before performing the flight welds, a test qualification campaign has to be done.

For the HX, this weld procedure was followed : • Weld qualification prior to assembly

o Identification of optimum weld parameters o Weld qualification

• Weld re-qualification (if necessary) • Weld procedure during assembly

o Pre-weld o Flight welding

The weld method used was the orbital welding in which an electrical arc is established between a tungsten electrode and the part to be welded. To start the arc, a high voltage is used to break down the insulating gas between the electrode and the part. Current is then transferred through the electrode to create an electrode arc. The metal to be welded is melted by the intense heat of the arc and fuses together either with or without a filler material.

130

Chapter 5

Figure 5-28:Typical orbital welding head (as the one used for the HX housing welding)

and orbital welding process explanation (right)

The weld type is only one, with dimension of 56.5 mm OD x 3.25 mm wall (as shown in Figure 5-29), made by a connection between two part of Inconel 625 of the same batch material. For this weld type a weld qualification programme is needed.

All welds are classified as class B according to NASA document [36].TTCS is a pressurised system and therefore class B requirements and methods for pressurised components are applicable. Weld qualification

A weld qualification consists of the following steps: • Identification of optimum weld parameters • Weld qualification

The weld qualification is performed on a total of two weld samples. After agreement with NASA it was decided only two samples at nominal setting are necessary for weld qualification. This is agreed in view of the few flight welds needed (2 FM and 1 QM). All samples shall be send to NASA where they will be subjected to:

• Visually inspection to the Class B acceptance criteria • Liquid penetrant or magnetic particle inspection to the Class B acceptance

criteria

131


Figure 5-29:HX FM primary orbital welding step

The flight hardware welds shall be made with the nominal power setting.

Common is to use a range of input settings with a variation in the order of ±10%, but this is no hard requirement. Using such a range is to deal with power fluctuations in the welding apparatus between the low and high limits and to accommodate a wider range of thickness variations.

These wide variations are not present in the few accurate machined parts for the HX welds. Therefore it is decided in close consultation with NASA to perform qualification on two qualification samples (Q1 and Q2) at nominal settings. After successful qualification and approval the 3 flight welds (1QM and 2 FM HX’s) are performed. Determination of the nominal weld parameters

After some unsuccessful tests performed with DC current, pulsed current was chosen as the best solution to achieve a good welding result. The NASA guidelines for pulsed current were followed, to find the optimum parameters (reported below) to achieve acceptable repeatability and weld quality. Definitions:

• Peak Current = high current set-point • Background Current = low current set-point • Pulsed current ratio = ration of high current to low current set-points; to

be typically 3:1 • PPS = pulses per second (Hz); the pulse rate • Pulse width = Percentage of time the weld program stays at the peak

current setting in any one given cycle. The balance of the time will be spent at the low current setting.

• Average current = (% time of weld cycle)(peak current) + (% time of weld

cycle)(background current)

132

Chapter 5

• Weld cycle = period including one time segment at peak current + one time segment at background current

• IPM = Inches per minute; linear travel rate of the welding arc • RPM = revolutions per minute

Applying the above definition to the HX configuration the following starting setting were found (based on the above - INCO 625, 56.5mm O.D., 3.25mm wall thickness:

• Assume it will be a single pass weld, trying to limit high temperature excursions

• Peak current = 0.7 x (3.25/0.0254) = ~90 amps • Background current = 90/3 = 30 amps • Pulse width = 35% at peak, 65% at background • 6 pps (selected arbitrarily from experience) • Select 3.0 ipm (76 mm/min) - Tube circumference = ~177 mm, rotational

speed will then be 0.43 rpm (i.e., 76/177) • Total time for weld schedule will include the start-up dwell time,

circumference traverse time, and an area of overlap to fully consume your start-up area. The dwell time at the start should be such that to just achieve full penetration through the wall thickness, and then travel (rotational motion) will start. This amount of time will be added to the entire schedule time. The parameters were defined and result of the qualification campaign is

shown in next figures.

Figure 5-30:HX samples performed for penetration acceptance criteria

133


Figure 5-31:HX samples for qualification and the flight and QM welds on the HX

housing ( right)

1.1.2. QM and FM TTCB integration

One also the QM and FM heat exchanger have been available, al the TTCB integration was started following the main assembly sequence and the assembly drawing package provided y NLR. The integration could be seen from two different point of view:

• Welding of all the tubing to connectors inside the TTCB • Bolting of all the component to the base plate and structure of the TTCB.

Welding process was done following the qualification process required form NASA and finally for each box 54 welds were identified, adding up to a total of 162 welds for 2 FM and one QM box. The welds were done with two different methods and can be divided into nine types of welds. As the box is containing components for some welds different batches of metal are used. The welds from different batches are considered to be a different material. For connections with different materials a separate weld qualification is needed. After a check of the components and tube materials nine type of welds were indentified.

The overall installation sequence of the component into the TTCB can be summarized as following:

• Pump controller • APS • DPS • Accumulator • Preheater • Cold orbit heater • Pumps • Connector plate and cover to base plate • Base plate and side plate to USS simulator

134

Chapter 5

In the following figures are shown some example of the installation of the components, ending with the complete assembly and integration of the two FM units.

Figure 5-32: Pre-heater implemented in the TTCB-P

Figure 5-33:Cold orbit heater implemented in TTCB-P

Figure 5- 34:Implemented DPS

135


Figure 5-35:Final integration of the TTCB-Primary FM and f TTCB-Secondary FM

into transport frame and USS simulator

1.3. FM TTCB qualification tests in Terni

1.1.1. General qualification test process Space instrumentation is operated in hard environmental conditions, being

subject to severe mechanical and thermal stresses along its life. Nevertheless, it is required to function with a very high level of reliability to minimize, when possible, any intervention after launch. Key points to reach full reliability are:

• A careful design: detailed thermal and mechanical stress analysis of the instrument and its subsystems are performed in order to fulfil all the different requirements that characterize each mission phases;

• An extensive campaign of qualification tests verifying the system functionality in the expected environmental conditions. Correlation among measured thermal/mechanical behaviour of the items under test with the model predictions constitute a validation of the adopted design;

• Redundancy for the most critical parts (electronics, control systems). A payload and each single part of a payload must undergo a qualification

procedure before the final integration and launch. Aim of the qualification campaign is to certify that the hardware (and software, as applicable) design is suitable and conformed to the specification requirements.

136

Chapter 5

The qualification program for a payload usually includes functional and environmental tests (vibration test, shock test, electro magnetic compatibility thermal vacuum test) as shown in Figure 5-36.

Figure 5-36:Typical flow of qualification testing for payload and components: they

are qualified by a series of functional tests and exposure to environmental conditions

The sequencing in Figure 5-36. is based on a combination of the order in

which the environments are encountered during flight and the purpose to perceive defects as early in the test sequence as possible. The thermal and vibration qualification of a payload and its subcomponents constitutes an important step in the qualification process.

Figure 5-37:the SERMS laboratory overview

For the TTCB, the qualifications test have been performed in Terni, at the

SERMS facility (Figure 5-37) [22]The central part of the 500 m2 hangar surface is occupied by the vibration laboratory. An electrodynamic shaker connected to a 2x2m2 slipping table allows to perform dynamical tests within a frequency range between 5 Hz and 3000 Hz for a maximum applied force of 80 KN (Figure 5-38).

137


Figure 5-38:The SERMS electro dynamical shaker (left) and the space simulator

(right)

Up to 100 accelerometers and 128 strain gauges can be used to monitor in real time and to record for further analysis the response of the device under test. The area in front of the vibration laboratory is free of permanent structures to allow the handling of cumbersome objects to be tested: a 12.5 tons crane allows their moving from the hangar entrance to the vibration area. Two interconnected clean rooms insure the proper handling of electronics and instrumentation undergoing thermal tests in vacuum. A space simulator suitable for the test of small size satellites and instruments to be qualified in absence of atmosphere is located in the first clean room (class M6.5, 4x5x3,5 m3).

As shown Figure 5-38, the space simulator consists in a thermo-vacuum cylindrical chamber, 2100 mm length and 2100 mm inner diameter, where a pressure of 10-7 mbar can be reached in empty chamber conditions. Temperature is controlled in the range [-70°C, +125°C] and monitored within the chamber volume with PT100 thermal sensors. Using conductive and/or radiative coupling, it is possible to cycle the item under test within a temperature range comparable to that expected for the instrument during its operating life. Up to 64 PT100 and 128 strain gages can be used to continuously record temperature and mechanical stresses on the devices under test.

Thermal tests at ambient pressure can be carried in three thermal chambers located either in front of the clean area or in the second clean room (class M5.5, 4x4x3.5 m3) if needed. Thermal gradients of 1.5, 2, 15°C/min can be applied in the different chambers within volumes of 1-2 m3 along a temperature range of -70, +180°C.

The thermal chambers are also used for thermal tests in vacuum of small sized objects, as electronic boards: the device under test is put in a pressure tight container inside the chamber and an external turbo-molecular pump is used to reach the desired vacuum level.

138

Chapter 5

1.1.2. Vibration test of FM TTCBP and TTCBS

Prior to further integration into the overall AMS, the flight TTCB primary and secondary have been subjected to a vibration tests. The objective of the test is to demonstrate the TTCB’s can withstand Minimum Workmanship Level Vibrations. Before and after the test functional checks are performed to compare the system health prior and after tests.

The test is successful when the following requirements are fulfilled: • No visual damage of the test article is found • No significant discrepancies between pre- and post sine sweep curve

response • All mechanics frequencies are above 50Hz • Functional check before and after show no discrepancies

1.1.1.1. Test set up

The general view of the test setup for the test set-up is shown in Figure 5-

39. For each sub-test the TTCB has been mounted over the fixture matching the vibration axis of the component with the slip table axis, as shown below.

Many sensors (channels or CH) have been used during this sub-test. They were fixed on the TTCB by cyanoacrilic glue. The accelerometer to control the table have been always be located on the interface plate with the vibration table. The position will be changed according to the test axis to be performed.

Two other sensors have been applied to measure the response in the orthogonal directions. As the base plate is the most representative part the locations proposed were as close to the base plate as possible. An additional 3-axis sensor was located on or close to the pumps so the vibration levels of the most critical component were monitored. This was done to be sure the pumps are not overstressed (Figure 5-40).

The TTCB’s a subjected to Minimum Workmanship Level Vibration testing. Before and after each axis random vibration test a sine sweep is performed to characterise to characterize the TTCB response curves. The Sine sweep definition is as follows: Sine sweep from 5 to 1000 Hz – 0,2G – scan rate 1 oct/min

139


Figure 5-39: Location of the control sensor on the I/F plate for the first and second axis

( left) and for the third axis ( right)

Figure 5-40:Example of positioning of the several control sensors on the X-axis, Y-axis-

and z-axis .

1.1.1.2. Test graph and results

The TTCB primary and secondary flight model has been tested at S.E.R.M.S. in random vibration in May 2009, and the planned tests have been executed according to the test procedure and the required input levels have been verified. All the control parameters have been continuously monitored and recorded during the test. In this section, some graphs show the measured quantities during the whole test period All the test objectives have been fulfilled. In particular, the test has demonstrated that:

• in all performed sub-tests (single axis) no damage has been reported. • all sub-tests have been normally completed. • no discrepancies have been noted between curves • the frequencies of the mechanics are well above 50hz • after all the relevant functional showed no discrepancies before and after

the tests

140

Chapter 5

Figure 5-41:Profile of random vibration spectra for some of the control sensors

141


1.1.3. Thermo-vacuum test of FM TTCBP and TTCBS

Prior to further integration into the overall AMS, the flight TTCB primary

and secondary have been subjected to a thermal vacuum test. During the tests the item under test has been put in a vacuum <1*10-5 hPa (mbar) and the environmental temperature (shroud and heat sink) varied to maximum and minimum non-operating and operating temperatures. Functional tests have been performed prior to, during, and after the thermal vacuum test.

1.1.1.1. Test set up

The test set-up is for primary and secondary TTCB is shown schematically in Figure 5-42 with also the indication of the number of thermal sensors and their positions.

The test set-up in the thermal vacuum chamber is shown in Figure 5-43 for TTCB–primary and in Figure 5-44 for TTCB-secondary while the temperature profile adopted in the test is schematically presented in Figure 5-45

Figure 5-42:Test set-up schematic for both primary and secondary TTCB

142

Chapter 5

The TTCB and TTCE are located inside the thermal vacuum chamber. The TTCB is connected electrically to the TTCE and hydraulically to a reference liquid flow meter. As the reference flow meter cannot withstand a vacuum environment, it has to be located outside the chamber. However, as the TTCS working fluid (CO2)

caloric heat pick-up capability is low ( ), the external flow meter and tubing have to be thermally controlled. Otherwise the CO2 liquid leaving the chamber will vaporize during its flow outside the controlled thermal vacuum environment. Vapour entering the pumps shall be avoided not only because in the long term it may damage the pumps, but also because the TTCS normal operation is with sub-cooled liquid entering the pumps. Vapour entering the liquid flow meter will cause erratic flow readings. On top of that, a safety measure in the TTCS control prevents the pumps from running in a vapour condition, by increasing the accumulator set-point.

As with the envisaged method of using a commercial freezer outside the chamber, still heat will leak from the environment into the working fluid at the chamber feed through and part of the flow meter exit tubing, an additional simple heat exchanger is connected just at the TTCB liquid inlets. This will ensure that the working fluid is always conditioned at the required temperature.

Figure 5-43:TTCB-primary Test set-up in SERMS Thermal Vacuum Chamber

143


Figure 5-44:TTCB-secondary Test set-up in SERMS Thermal Vacuum Chamber

Figure 5-45:Test temperature profile.

A system to command the TTCE has been accommodated. Furthermore an

additional data-acquisition system was required to gather the environmental data (shroud and cold plate temperature and pressure) as well as some working fluid temperatures and mass flow.

1.1.1.2. Test preparation

144

Chapter 5

A shortcut tubing has been installed for the top and bottom evaporator in

and outlets. Extreme care was taken to avoid fouling of the tubing. The shortcut tubing and fittings were all cleaned (electro polished stainless steel tubing). The condenser return lines must be connected together and the condenser feed lines must be connected together. Both connected lines must be connected –via the pressure drop generator/heat exchanger to the flow meter outlet and flow meter inlet respectively. The pump inlet liquid will be subcooled by this external heat exchanger and the liquid leaving the TTCB will also be cooled by the external heat exchanger. The latter is required because heat will be dissipated during functional testing and vapour entering the liquid flow meter is to be avoided.

After installing the tubing, leak tightness was done to be checked to prevent too high working fluid losses or chamber depressurization problems. The Helium sniffer method is not allowed, as it is unclear how to remove the Helium, once it is inside the TTCB system. If He persists inside, subsequent leak tests may become problematic.

the total test item volume including test tubing shall be determined by measurement. During the volume measurement the complete test set-up shall be in thermal equilibrium, prior to depressurization of the thermal vacuum chamber. The system shall be filled such that the liquid level in the accumulator is sufficient to submerge the heat pipe during all operational tests. The final configuration of the test set up for the two boxes is shown in the following figure.

Figure 5-46:Positioning of the TTCB primary (left) and secondary (right), the TTCE

and the NLR plate (pressure drop generator).

1.1.1.3. Thermal sensors positioning

145


Temperature sensors (e.g. Pt-100’s) shall be installed at the locations as

shown in . All temperature readings and flow meter reading have been recorded using a data-acquisition system that is synchronized with de TTCE read-outs as well as the thermal vacuum chamber housekeeping data (pressure and temperatures). In more detail, a total of 11 Thermal Sensors (TS) were monitoring the environmental conditions of the test :

• 3 TS were placed on the 3 cold plate: Ch5 on the lower Cold plate, Ch6 on the middle cold plate and Ch 7 on the upper cold plate.

• 1 TS was placed on the shroud (Ch4) • 2 TS were placed on the TTCE crate foot (Ch8 and Ch9); these sensors have

been used as TRP (Temperature Reference Points) for the TTCE crate. • 1 TS was placed on the NLR plate (ch10) • 4 TS were placed on the TTCB: Ch16 on TTCB interface plate, Ch17 on

TTCB base plate, Ch18 on TTCB start up radiator and Ch19 on USS simulator Additional 20 Thermal Sensors have been installed on the TTCB primary

(and equally in the TTCB secondary) and all the stuff (tubes and freezer) used for the test:

• 14 Sensors have been placed on the TTCB box and tubes inside the chamber;

• 6 sensors have been placed on freezer wall and tubes outside the chamber; Two of these sensors have been used as Temperature Reference Points

during the test. After positioning, all sensors have been tested to verify possible failures after installation. In the following figures are presented the some of the sensors after positioning.

Figure 5-47:TS on upper cold plate (Ch7) and on TS on TTCE crate foot (Ch9). TRP2.

146

Chapter 5

Figure 5-48:TS on TTCB primary base plate (Ch17) and Sensor placed on the pump in

tube.

Figure 5-49:Sensor placed on the heat exchanger out tube and sensor placed on the

TTCB secondary cover (right side)

1.1.1.4. Test graph and results

147


The TTCB primary flight model has been tested at S.E.R.M.S. in the Thermal Vacuum Chamber (TVC) during the period May 23th – June 3th 2009, while The TTCB secondary flight model during the period June 24th – June 28th 2009.

The test has been performed according to the test profile shown in the previous section under the continuous supervision of NLR and SYSU representative, which operated the TTCB primary and secondary during the switch-on/switch-off and functional test phases.

All the control and environmental parameters in the TVC have been continuously monitored and recorded during the test. The TTCB primary and secondary temperatures have been monitored in 20 locations and their values recorded during the whole test period

In this section, the graphs summarizing the evolution vs. time of all measured quantities during the whole test period, are reported. All the test objectives have been fulfilled. In particular, concerning the temperature range and requirements, the test has demonstrated that:

• The minimum operative temperature conditions have been met whenever the TTCB P and S have been powered in a radiative environment at -40 °C in average

• The minimum non operative temperature conditions have been met whenever the system has been switched off in a radiative environment at -40°C in average

• The maximum operative temperature conditions have been met in a radiative environment at 20°C.

• The maximum non operative temperature conditions have been met in a radiative environment at +65°C. After this tests, all the data will be used for the correlation of the TTCS a

thermal model and with the TVT that will be performed in ESTEC. .

148

Chapter 5

Figure 5-50: Pressure profile of the test.

Figure 5-51:Chamber temperature profile.

149


Figure 5-52 :TTCE temperature profile.

Figure 5-53:TTCB TRP temperature profile.

150

Chapter 5

Figure 5-54:TTCB start up radiator and Accumulator temperature profile.

Figure 5-55:TTCB internal sensors temperature profile.

151


Figure 5-56:TTCB Temperature Reference Points and Chamber (shroud and cold

plate) temperature profile.

152

Chapter 6

6. TTCS integration in AMS-02 and

operational data results

The elements of the AMS-02 detector have been assembled as shown in Figure 6-1, where “RAM” refers to the ISS direction of flight, “WAKE” is opposite, “port” and “starboard” are transverse. The assembly took place at a dedicated facility installed at CERN, with support from the entire collaboration. After all elements have been integrated, the detector will be calibrated in particle beams at CERN and then hipped to the ESA Large Space Simulator at ESTEC, Noordwijk, the Netherlands. to undergo a complete thermal vacuum test.

After the instrument has been completely checked out it will be partially disassembled for transport and shipped to the Kennedy Space Center. There, after reassembly, it will be subjected to tests to verify that it is compliant with the NASA mechanical and electrical interfaces for both the space shuttle and the space station. The experiment will then be installed in the shuttle and launched to the space station.

In the overall AMS assembly sequence the TTCS component have been installed in different time, as highlight I the following general assembly integration steps: 1. USS and Magnet 2. TTCS transport tubing 3. Anticoincidence Counters (ACC) 4. TTCS Condensers 5. Tracker , TTCS evaporator and Star Tracker 6. TTCS TTCB primary and secondary 7. Upper Time of Flight (UTOF) and Transition Radiation Detector (TRD) 8. Tracker radiator installation 9. Integration of the Lower Parts (RICH, ECAL, LTOF) 10. Main Radiators & Electronics 11. ISS Hardware

Detail of the TTCS integration at CERN will be given in the next section.

Chapter 6

153

TTCS Integration in the AMS-02 and operational data results

Figure 6-1: AMS-02 Mechanical Integration at CERN.

1.1. TTCS subsystem integration

1.1.1. Transport tubing installation Transport tubing have been completely design by INFN Perugia, according

to the requirements from NLR and the AMS collaboration, owing that tubes had to been interfaced with many subdetector of AMS.

Tubes of 4 mm x 0.7 mm have been procured from Microtubes.inch, made of SS316L and delivered already clean to fulfil the cleanliness requirement of TTCS system. At the beginning all tubes were pre-bent off line, using a manual bending machining. Also in this case tolerance to be achieved were acceptable with such tool. A series of brackets, mainly made of AL 7075 T7351, with different shapes and dimensions, have been designed and them manufactured to hold in position the tubes in their complicate paths along the AMS detector (Figure 6-2)

154

Chapter 6

Figure 6-2:Transport tubing bracket positioning and definitions overview

After the bending and installation operation, orbital welding offline and

online have been performed. The TTCS transport tubing contains a total of 42 welds, as described in Figure 6-3 . The welds are done with only one methods and can be divided into 4 types of welds.

Figure 6-3:Transport tubing welding joint overview

155


A welding qualification campaign, as already described for the TTCB and heat exchanger in the previous chapter, has been conducted to guarantee the welding process according to NASA specification. In the next figure, example of the final installation of the transport tubing in the AMS detector is shown. An helium leak check has been done to discover eventual non conformance of the welding joints, before access has been limited by a layer MLI, put around all the tube surface.

Figure 6-4:Transport tubing integration in AMS detector: brackets and welding joint

1.1.2. Condenser integration

The four FM units, as shipped from AIDC, needed additional off-line operation, before the integration into AMS. First hey have been inspected (visual damage, helium leak check and heater check) to detect any anomalies caused from the transportation. Then they were fit checked with the tracker radiators to align the integration jig manufactured for this step in the overall assembly sequence. At that time, the last component has been installed to complete the overall condensers manufacturing, like TS on surface of the top plate, and the wire heaters running along the tubes to the manifold, as shown in Figure 6-5.

156

Chapter 6

Figure 6-5:Condenser fit check on the tracker radiator ( left) and wire heaters and

thermostat installation ( right)

The condensers have been attached to the USS thanks to the rod structure and bracket, foreseen to support the loads of the units and of the integrations jig. A thin layer of Sigraflex, the interface material between the condenser and the radiator to increase the thermal conductivity, has been put in place just before the position of the trackers radiators.

Figure 6-6:Condenser and integration jig positioning(top left), Sigraflex installation

(top right) and tracker radiator positioning (bottom)

157


Once the alignment has been guarantied, all the bolts (98 for each unit) has been tighten to final torque, according to value required and recorded in the dedicated procedure. Finally, welding operations always according to NASA welding requirement, were performed at the manifold side to connect the transport tubing to the condenser tubes loop.

Figure 6-7:Schematic of the condenser connection to the tracker radiator by means of

98 screws for each unit.

1.1.3. Evaporator integration into Tracker

As shown is the overview of the evaporator lay-out, the inner diameter of the evaporator is 2.6 mm and the total length is 10 m. Heat collected at the inner tracker planes is transported by thermal bars to the top and bottom evaporator ring. The main design challenges for the evaporator have been the limited volume (small diameter, small thickness piping) and welding of evaporator rings (performed by an orbital welding procedure from Nikhef).

The evaporator ahs been pre-integrated and welded in the tracker, and only after the tracker installation into AMS, it was possible the connection and integration to the TTCS loops, by means of special connectors positioned on the top and bottom of the tracker. Details are shown in the next figure.

158

Chapter 6

Figure 6-8:Top evaporator details and connection to the TTCS transport tubing

1.1.4. TTCB integration

As for the FM condenser, also the TTCB flight unit were shipped form

Taiwan to CERN to be finally integrated into AMS-02 detector. Several tests and visual inspection have been performed to check damages from transportation.

The installation was made by means of bolt connection to the unique support structure of AMS (USS) applied the torque range required and recorded in the dedicate procedure. First TTCB primary and then secondary have been jointed to the USS.

Figure 6-9:3D model of the TTCB installation into the AMS USS and details of the

flight installation.

159


Next step has been the identification of the tubes to be connected to the transport tubes, in order to complete all the TTCS tubing routing enclosure (Figure 6-10). As for the condensers, the orbital welding method has been chosen and qualification tests have been performed.

Once all the welds have been completed, the TTCBs underwent health check, helium leak detection to discover eventual leak and anomalies on the welding and joints. Finally the both loops have been filled with CO2 and pinched, sealing definitively the primary and secondary loop. MLI was put all around the TTCB as required form the thermal control system design. This permitted the starting of the cooling test of the tracker, to check if the two systems, primary and secondary, were correctly working.

Figure 6-10:Schematic of the TTCB welding joints and detail of the TTCS transport

tubing to TTCB overlap before welding operation

1.2. TTCS monitoring, control and operational data results

1.1.1. TTCS operational rules for TTCS start-up and operation

Two systems for TTCS commanding and monitoring are foreseen. For on-ground TTCS testing operations at CERN, ESTEC, and KSC a TTCS Ground Support Equipment is used. The in-flight operations are done via a TTCS Ground Operations and Monitoring system.

As the telemetry downlink has limited bandwidth, it is desirable to have the possibility for "autonomous on-board" of the non-schedulable TTCS start-up related procedures, under control of on-board software implemented in the JMDC. The defined procedures will be implemented (at least partially) in software and executed "semi-autonomously on board" by the JMDC. The current baseline for routine TTCS in-flight operations is to use the JMDC command list in view of the limited required bandwith of only 183 bits/second.

The TTCS can be started up and operated in different configurations, thanks to its redundancy:

160

Chapter 6

• - TTCE-A powered on, controlling Primary Loop using "a" components • - TTCE-A powered on, controlling Secondary Loop using "a" components • - TTCE-B powered on, controlling Primary Loop using "b" components • - TTCE-B powered on, controlling Secondary Loop using "b" components

So, first it must be selected which TTCE (A or B) is powered-on, next it is selected which Loop (Primary or Secondary) is to be started-up.

It is noted that closed loop operation of each control loop can be disabled/enabled and heaters may be operated ON/OFF "manually" via sending of the appropriate Write Data Type containing the correct bits settings for the intended operation. In-flight the commands are sent to the JMDC, which further forwards them to the TTCE.

After powering-on of a TTCE, a TTCS loop configuration (Primary A or B or Secondary A or B) can be started-up further via sending Write Data Type requests to the TTCS. Start-up of the TTCS does require continuous monitoring of TTCS on-board statuses (mainly temperatures) by the TTCS operator and verification of intermediate decision criteria, as it has to be checked that certain measured TTCS temperatures and statuses satisfy well defined criteria, before the next commands may be sent.

At CERN a control and monitoring system for the TTCS, called TTCS Ground Support Equipment, has been implemented. The Graphical User Interface has been split over two screens, given below, see Figure 6-11 and Figure 6-12. This system will also be used at ESTEC and KSC.

Figure 6-11:TTCS control and monitoring Graphical User Interface

161


Figure 6-12:TTCS Loop control and monitoring Graphical User Interface. Operational configuration selection

First it has to be selected which TTCE (A or B) is powered on, next it is to be selected which Loop (Primary or Secondary is to be started up.

The TTCE power-on selection is as follows: IF (TTCE-A ==HEALTHY) power-on TTCE-A ELSE IF (TTCE-B == HEALTY) power-on TTCE-B END The Loop selection is done as follows: IF (Primary Loop == HEALTHY) start-up Primary Loop ELSE IF (Secondary Loop == HEALTHY) start-up Secondary Loop. END

For the first-time in-flight start-up it is to be assumed that TTCE-A== HEALTHY and Primary Loop == HEALTHY. Exceptional cases in which P and S loop equipment are operated simultaneously shall be dealt with on a case by case basis and operations should be performed under full ground monitoring and control.

162

Chapter 6

Power-on of a TTCE

A TTCE may be powered-on at any time. Powering-on a TTCE can not endanger the TTCS system. After TTCE power-on the relevant 28V power switches are still OFF Default values for TTCS loop control parameters are given.

In summary the status at TTCE power-on is: • the 28 V power switches are OFF • all on/off control loops are DISABLED • pi control of the accumulator is DISABLED • all health-guards /alarms and alarm actions are ENABLED • the low pump speed alarm should be active as the pump power is OFF, the

pump speed set point= 0rpm, and hence the pump is not running. • FG ==1 so accumulator flight heater (FAC) will be used. • Temperature readings will depend on the actual system temperature

situation.

Tracker normal operating temperature range

The operational temperature range of the Tracker electronics is: -10 ºC to +25 ºC. The temperatures experienced by the Tracker are the evaporator temperatures, measured by Pt04 and Pt05 at the evaporators' entrances and the Dallas sensors distributed along the evaporator tubing These measured temperatures should be close to the accumulator set-point and accumulator temperature during steady state operation.

Tracker survival temperature range

The survival temperature range of the Tracker electronics is: -20 to 40 ºC.

Accumulator normal operating set-point temperature range

The operational temperature range of the Tracker electronics is: -10 ºC to +25 ºC The allowed minimum and maximum of the normal operating temperature range of the accumulator set-point is consistent but not equal to the normal operating temperature range of the Tracker electronics, i.e.: -20 ºC to +25 ºC.

The accumulator temperature is measured by Pt01.The accumulator temperature should be close to the accumulator set-point during steady state operation. The accumulator temperature may differ from the accumulator set-point after a set-point change, e.g. due to ground command or CAV_alarm action.

Pump inlet temperature cavitation margin during normal operation

163


As the suction pressure at the inlet of a running pump is below the

accumulator pressure, the pump inlet liquid temperature must be below the accumulator temperature (= saturation temperature) to prevent CO2 boiling in the running pump inlet and hence cavitation in the pump. To this end a cavitation margin is to be employed, with default value 5ºC, i.e. when the pump is running or started, the pump inlet temperature should be at least 5ºC below the accumulator temperature.

Given the maximum allowed operational temperature of the accumulator, the maximum allowed normal operational temperature of the pump inlet liquid is: maximum accumulator temp – cavitation margin = 25 – 5 = 20 ºC (default values).

The pump inlet temperature is measured by Pt02.

Accumulator set-point for automatic start-up; + 20 ºC

For an automatic start-up procedure by the JMDC, an accumulator set-point of 20ºC is chosen. The rationale is as follows. The start-up set-point of the accumulator should be above the pump inlet temperature, measured by Pt02, to prevent cavitation. At start-up the TTCS box has approximately a uniform temperature, which depends on the temperature of the support structure the box is mounted to. The liquid at the pump inlet has a temperature which is slightly below the box temperature, thanks to the small pump radiator. In order to be able to switch on the pump without the danger of cavitation the accumulator set-point and actually measured temperature must be above the box temperature. Therefore automatic start-up an accumulator set-point of +20ºC is chosen and a maximum allowed box temperature at start-up of 19 ºC is chosen A set-point of +20ºC is in 90% of the time (but not always) higher than a TTCS box temperature of 19ºC.

Hence, there remains a risk that during automatic start-up the box temperature (max allowed during automatic start-up is +19 ºC) is close to the start-up set-point + 20ºC and hence there is only a cavitation margin of 1ºC.

It has been decided to allow automatic start-up in such a situation. The rationale to allow continuation of automatic start-up in this situation is that the pump can suck in boiling CO2 for a short time span without harm. Normally, after some time cold liquid from the condensers will enter the pump and boiling -if present- will stop. Furthermore, the TTCS health-guards are all enabled after TTCE power-up and hence – if this situation occurs- the cavitation margin health guard will force the accumulator set-point higher -maximum is +25ºC- which also alleviates the situation, although the temperature reaction of the accumulator is rather slow.

So, after some time the pump inlet temperature (Pt02) should return well below the accumulator temperature (Pt01). If this does not occur probably there is not enough flow, possibly because the liquid at the pump inlet is boiling. So, if the Pt02 does not drop a certain margin below the accumulator temperature Pt01 after

164

Chapter 6

TBD seconds, the JMDC should stop the pump and signal this situation to the ground.

Maximum pump inlet temperature allowed at automatic start-up

As made plausible above the maximum pump inlet temperature allowed at automatic start-up is set to 19ºC.

Maximum temporary accumulator set-point

The used cooling medium CO2 becomes supercritical above +31ºC. The pump is not designed to handle supercritical CO2 during extended time periods. So, the pump in cooling loop should not be started-up nor be operational if the accumulator temperature is above +31ºC. As this temperature is also above the maximum operating temperature of the tracker, an accumulator set-point above + 25ºC (but below +31ºC) may only be used temporarily and if the Tracker is NOT ON.

1.1.2. TTCS operating at CERN: data results

After the closing and sealing of the TTCS lops, several tests on the tracker cooling system have been carried out, using the monitor and control interface described above.

Figure 6-13:Data results on the firs operation test on the TTCS at CERN.

165


The first test was done at beginning of November 2009, as showed in the following graph. Looking also to the dallas sensor positioning inside the tracker (Figure 6-14 and Figure 6-15), it is possible to recognize, the:

• the tracker temperature (A),reaching asymptotically 31°C • the temperature coming out from the loop (F),where it is noticeable the

change form the single phase (growing temperature) to two-phase (temperature fixed at the et point of 26°C)

• the temperature of the loop at the centre ( E) • the temperature of the loop at the entrance (G), slightly arising. • The pump temperature (Pt02), with a non linear behaviour, strongly

dependent on the environment (clean room) variation

Figure 6-14: Positions of dallas sensor on the upper evaporator

Figure 6-15:Positions of dallas sensor on the lower evaporator

It was shown that in an hot clean room environment (21°C) at CERN, the

Tracker can survive roughly two hours of operation switching off. The results have

166

Chapter 6

been improved adding some external conditioning o the tracker radiator surface, as in shown in the graphs below, that represent also the state of art of the AMS monitor interface available for the tracker rand TTCS operation check at CERN.

Further tests will be conducted during the Beam test and LSS, to check the operation of the TTCS also in vacuum condition.

Figure 6-16:AMS monitoring interface available to check real time the status of the

tracker and of TTCS

7. Conclusion

The AMS-02 Tracker Thermal Control System (TTCS) is a two-phase cooling system developed by NIKHEF (The Netherlands), Geneva University (Suisse), INFN Perugia (Italy), Sun-Yat Sen University Guangzhou (China), Aerospace Industrial Development Company (Taiwan) and NLR (The Netherlands). The TTCS is a mechanically pumped two-phase carbon dioxide cooling loop. The main objective is to provide accurate temperature control of AMS-02 Tracker front-end electronics. An additional objective is to prove and qualify a two-phase pumped cooling system in orbit and collect operational data in micro-g environment over a period of three years.

The objective of the cooling system is to collect the dissipated heat at the tracker electronics and transport the heat to two dedicated heat pipe radiators. One radiator is located at the WAKE (anti-flight direction) side and the other one at the RAM (flight direction) side of the AMS instrument.

The two-phase loop incorporates a long evaporator, picking up the heat from the multiple heat-input stations evenly distributed over the six Tracker silicon planes. The heat is transported to condensers mounted onto the heat pipe radiators. The liquid is transported back to the evaporator by means of a mechanical pump.

The heat producing elements, the tracker front-end hybrid electronics are situated at the periphery of the tracker silicon planes and are located inside the cryogenic magnet. A total of 144 Watt is produced at 192 locations and an additional 6-10 Watt cooling capacity is required for additional electronics and the Star Tracker, which is also attached to the loop.

The selling points of mechanically pumped two-phase systems are: • The possibility to locate the temperature control at a distant location (up to

100 meters) far away from the to be cooled item or instrument. • The possibility to provide thermal control for distributed dissipative

elements • High temperature stability and isothermal behaviour • Reliable straightforward operation and start-up

Special for CO2 is the low liquid/vapour density ratio resulting in low evaporator pressure drops providing the possibility to implement tight thermal control in extremely confined areas.

Chapter 7

168

Conclusions

Figure 71:Schematic of the Tracker Thermal Control System Primary Loop

An overview of the TTCS system and the system design considerations

have been presented in this dissertation. It is shown that a CO2 two-phase mechanically pumped loop is capable of cooling the widely distributed Tracker front-end electronics. The TTCS system design, the development status and some typical test results have been described.

Thermal cycling test on EM and QM condensers have been conducted to evaluate the glue performance after the glue has been subjected to extreme conditions. This demonstrated that the performance of the condenser is not affected by the thermo cycling. QM condenser freezing and de freezing test was carried out in order to evaluate the condenser performance in the freezing and de freezing condition. All the EM and QM test campaign for the condenser have been successful, leading to the conclusion that the flight model manufacturing could have been started following the design and the gluing procedure foreseen for the QM condenser.

Prior to further integration into the overall AMS, the flight TTCB primary and secondary have been subjected to a vibration test. The test has demonstrated that in all performed sub-tests (single axis) no damage has been reported and the frequencies of the mechanics were well above 50hz.After all, the relevant functional checks showed no discrepancies before and after the tests.

169

Chapter 7

The flight TTCB primary and secondary have been subjected also to a thermal vacuum test. They have been put in a vacuum condition <1*10-5 hPa (mbar) and the environmental temperature (shroud and heat sink) varied to maximum and minimum non-operating and operating temperatures. Functional tests performed prior to, during, and after the thermal vacuum test ha showed that TTCBs have met the temperature range requirement for both operation and non operation conditions.

The final TTCS test will be the flight model testing with connected flight Tracker radiators during AMS-02 thermal vacuum testing in ESA’s Large Space Simulator.

Currently the Flight Model TTCB component boxes, the FM condensers and FM evaporators are qualified and being integrated onto the AMS-02 experiment. Launch is planned for July 2010.

. The TTCS performances resulted in the implementation of TTCS-like system in the Vertex Locator (Velo) instrument of the Large Hadron Collider (LHCb) by NIKHEF. This 2.5 kW system is successfully implemented and tests showed more than satisfactory results. Currently the two other large CERN experiments ATLAS and CMS are considering a similar system for the upgrade of their inner detectors. This requires upscaling with a factor 1000 compared to the TTCS. The current commercial Spacecraft (S/C) are running to their thermal limits meaning that the radiator area is too small to provide enough cooling capacity. Recent requests of customers will force the industry to provide deployable radiators. Firstly this will be done with capillary pumped cooling systems but for S/C requiring stable temperatures or for future very dissipative telecommunication payloads (Q>9kW), two-phase mechanically pumped systems are advantageous and a competitive alternative to complex LHP-HP networks. The first application of a two-phase mechanically pumped loop is foreseen to cool an active antenna, requiring tight and stable temperature control. This is expected not earlier than 2015.

170

Bibliography

Bibliography [1] AMS document. The Construction of the Alpha Magnetic Spectrometer

(AMS) for the International Space Station. 2004. [2] B. Alpat. Alpha Magnetic Spectrometer (AMS-02) experiment on

theInternational Space Station (ISS). NUCLEAR SCIENCE AND TECHNIQUES, Vol.14, No.3, 2003.

[3] R.Battiston on behalf of the AMS Collaboration. The Anti Matter Spectrometer (AMS-02): a particle physics detector in space. 29th International Cosmic Ray Conference Pune (2005) 10, 151-172, 2005

[4] B.Blau et al. The superconducting magnet system of the alpha magnetic spectrometer AMS-02. Nuclear Instruments & methods in physics research-section A, A518(2004) : 139-142, 2004.

[5] J.Alcaraz et al. The alpha magnetic spectrometer silicon tracker: Performance results with protons and helium nuclei. Nucl.Instrum.Meth.A593 :376-398; 2008, 2008.

[6] B. Alpat et al. Charge determination of nuclei with the AMS-02 silicon tracker. Nucl.Instrum.Meth.A540 : 121-130; 2005, 2005.

[7] Kirn Th. The AMS-02 TRD: A detector designed for Space. Frascati Physics Series, Vol. XXV, 2001.

[8] S. Gentile on behalf of the AMS-02 TRD Group, J.Burger. The Performance of the AMS-02 TRD. The 28th International Cosmic Ray Conference, 2003.

[9] D.Casadei et Al. The AMS-02 Time Of Flight System. Nuclear Physics B, 113(2002)pp.133-138, 2002.

[10] G.Ambrosi.-AMS-02 Sub-Detector Electronics. AMS-02 webpage,http://ams.cern.ch/AMS/Electronics/, 2007.

[11] S. Cogan- Solar Absorptance and Infrared Emittance Modulation with Thin-Film Electrochromic Coatings: Application to Thermal Control. 7th Annual Spacecraft Thermal Control Technology Workshop, 1996.

[12] David G- Spacecraft Thermal Control Handbook Volume I: Fundamental Technologies. The Aerospace Press, El Segundo, CA, 2nd ed edition, 2002.

[13] L. Grob- Parametric Study of Variable Emissivity Radiator Surfaces. STAIF-2000, 2000.

[14] Theodore D. Swanson. NASA Thermal Control Technologies for Robotic Spacecraft. Applied Thermal Engineering, Vol. 23, pp.1055-1065, 2003.

171

Bibliography

[15] Tarik Kaya and Jentung Ku. Experimental Investigation of Performance Characteristics of Small Loop Heat Pipes. 41st Aerospace Sciences Meeting and Exhibit, 2003.

[16] Robert D. Karam. Satellite Thermal Control for System Engineers.AIAA, 1998.

[17] C. Vettore. AMS-02 TCS, Thermal model general description. CGS document, AMS-02-TN-CGS-006, 2004.

[18] I.Corradino-AMS-02 Thermal requirements and interface definition.CGS Document, AMS-02-RQ-CGS-001, 2007.

[19] T. D. Martin. Phase II flight safety data package for the Alpha MagneticSpectrometer-02 (AMS-02). NASA document JSC 49978A, 2007.

[20] J. van Es- TTCS System Design Description-AMSTR-NLR-TN-05-Issue03 [21] C. Vettore. AMS-02 120VDC and 28VDC heaters description. CGS

Document AMS-02-TN-CGS-007, 2006. [22] S.Borsini. The SERMS laboratory: a research and test facility for space

payloads and instrumentation. ICES 2007-01-3022, 2007 [23] J.F. Ding,Z.H.He-Leak detection for TTCS EM by He mass spectrograph,

TTCS-SYSU-GS-TN-003-1.0,2006. [24] X.H.Sun, Z.H.He-CO2 filling procedure for TTCS, TTCS-SYSU-GS-TN-

002-3.0 2007 [25] TTCS Radiator & Condenser Simulation Results, AMSTR-SYSU-SIMU-PR-

005-1.0. [26] AMS-02 Tracker Thermal Control System (TTCS) Cold Environment

Temperatures. ESCG-4470-06-TEAN-DOC-0032 [27] G. van Donk, R. van Houten, R. Annes, J. van Es- TTCS Condenser Freezing

Test Report AMSTR-NLR-TN-039 Issue03 [28] J. van Es , M.Bsibsi-TTCS Heater Specification AMSTR-NLR-TN-043

Issue 4.0. [29] J. van Es,G. van Donk- TTCS Condenser Design AMSTR-NLR-TR-045,

issue 1.1 in February 2006 [30] The revised requirements for the pumps of the AMS Tracker Thermal

Control System (TTCS)- AMSTR-NL-TN-010 Issue 04 [31] PDT pump proposal, TP-5059-2, “AMS Tracker Thermal Control System

Pump” - TP-5059-2, Revision B, 03-SEP-2004 [32] J. van Es,T. Zwartbol- TTCS Accumulator Specification- AMSTR-NLR-TN-

18-Issue03 [33] J. van Es,G. van Donk , G. Elbertsen- TTCS Heat Exchanger Design-

AMSTR-NLR-TN-053 Issue 1.0 [34] D-C.Mo,ZC.Huang,C.Vettore-TTCS Thermal Analysis Result-TTCS-SYSU-

SIMU-PR-003 Issue 1.0

172

Bibliography

[35] J. van Es,G. van Donk -TTCS component box thermal vacuum test plan AMSTR-NLR-PR-060 issue_1.1

[36] Daniel J. Rybicki Process Specification for Automatic and Machine Arc Welding of Steel and Nickel Alloy Hardware- PRC-0010, Rev. C

[37] T. Zwartbol,J.van Es- TTCS Commanding, Monitoring and Control AMSTR-NLR-TN-062 ISSUE 1.0 NOVEMBER 2009

[38] T. Zwartbol,J.van Es TTCE software user requirements document- AMSTR-NLR-TN-034 Issue 3

[39] T. Zwartbol, J.van Es- AMS-02 Interface Control Document Electrical and data- AMSTR-NLR-TN-24-Issue 4.0

173

Appendix A

Appendix A: Temperature distribution of EM condenser before and after thermal cycling

Table A1: Temperature distribution on the condenser and temperature difference (Ts=15°C): CASE 4* VS CASE 12

Before thermal cycling After thermal cycling Heating Power: 203W Heating Power: 204W DPS=840mbar DPS=812mbar

Thermocouple Tcondenser(℃) Thermocouple Tcondenser(℃)

Delta T (℃)

No. 1 15.5 No. 1 15.1 0.4 No. 2 15.5 No. 2 15.1 0.4 No. 3 15.6 No. 3 15.1 0.5 No. 4 Outflow No. 4 Outflow Outflow No. 5 15.7 No. 5 15.2 0.5 No. 6 17.6 No. 6 17.7 -0.1 No. 7 17.6 No. 7 17.7 -0.1 No. 8 17.6 No. 8 17.6 0.1 No. 9 17.5 No. 9 17.4 0.2

No. 10 17.2 No. 10 17.0 0.2 No. 11 16.9 No. 11 16.7 0.2 No. 12 16.7 No. 12 16.6 0.2 No. 13 16.7 No. 13 16.6 0.1 No. 14 16.7 No. 14 16.5 0.2 No. 15 16.9 No. 15 16.8 0.1 No. 16 17.2 No. 16 17.0 0.2 No. 17 17.0 No. 17 16.8 0.2 No. 18 17.1 No. 18 16.9 0.2 No. 19 17.2 No. 19 16.9 0.3 No. 20 17.1 No. 20 16.8 0.2 No. 21 17.4 No. 21 17.3 0.1 No. 22 17.3 No. 22 17.2 0.1 No. 23 17.4 No. 23 17.2 0.1 No. 24 17.2 No. 24 17.1 0.1 No. 25 17.3 No. 25 17.2 0.1 No. 26 17.3 No. 26 17.3 0.0 No. 27 17.2 No. 27 17.1 0.1 No. 28 17.2 No. 28 17.2 0.1 No. 29 17.2 No. 29 17.3 -0.1 No. 30 17.2 No. 30 17.4 -0.1 No. 31 17.2 No. 31 17.0 0.2 No. 32 17.3 No. 32 17.1 0.2 No. 33 17.0 No. 33 16.9 0.1

174

Appendix A

No. 34 17.0 No. 34 16.8 0.2 No. 35 17.0 No. 35 17.0 -0.1 No. 36 18.5 No. 36 18.5 0.0 No. 37 15.2 No. 37 15.2 0.0 No. 38 15.3 No. 38 15.2 0.1

Table A2: Temperature distribution on the condenser and temperature difference (Ts=-20°C): CASE 2 REP VS CASE 10



Delta T (℃)

No. 1 -19.0 No. 1 -18.9 -0.2 No. 2 -19.0 No. 2 -18.8 -0.1 No. 3 -18.6 No. 3 -18.9 0.3 No. 4 Outflow No. 4 Outflow Outflow No. 5 -18.5 No. 5 -18.8 0.3 No. 6 -16.7 No. 6 -17.6 0.9 No. 7 -16.7 No. 7 -17.6 0.8 No. 8 -16.3 No. 8 -17.6 1.3 No. 9 -16.3 No. 9 -17.7 1.4

No. 10 -16.2 No. 10 -17.8 1.6 No. 11 -16.0 No. 11 -17.9 1.9 No. 12 -16.2 No. 12 -18.1 1.8 No. 13 -16.4 No. 13 -18.1 1.6 No. 14 -16.7 No. 14 -18.1 1.4 No. 15 -16.9 No. 15 -18.0 1.1 No. 16 -16.6 No. 16 -17.9 1.3 No. 17 -16.9 No. 17 -18.0 1.1 No. 18 -17.0 No. 18 -18.0 1.0 No. 19 -16.6 No. 19 -17.9 1.3 No. 20 -16.8 No. 20 -18.0 1.2 No. 21 -16.8 No. 21 -17.5 0.7 No. 22 -16.8 No. 22 -17.6 0.9 No. 23 -16.8 No. 23 -17.7 0.9 No. 24 -17.1 No. 24 -17.8 0.7 No. 25 -16.7 No. 25 -17.8 1.1 No. 26 -17.0 No. 26 -17.8 0.8 No. 27 -17.2 No. 27 -17.9 0.7 No. 28 -17.0 No. 28 -17.9 0.9

175

Appendix A

No. 29 -17.1 No. 29 -17.8 0.7 No. 30 -17.1 No. 30 -17.8 0.7 No. 31 -17.4 No. 31 -17.7 0.3 No. 32 -17.7 No. 32 -17.8 0.1 No. 33 -17.7 No. 33 -17.7 (0.0) No. 34 -17.8 No. 34 -18.1 0.3 No. 35 -17.7 No. 35 -18.0 0.2 No. 36 -16.6 No. 36 -17.2 0.6 No. 37 -19.1 No. 37 -19.3 0.1 No. 38 -19.1 No. 38 -19.2 0.2

Table A3:Temperature distribution on the condenser and temperature difference (Ts=-

20°C): CASE 9 VS CASE REP 1



Delta T (℃)

NO. 1 -18,8422 NO. 1 -19,42 0,576

No. 2 -18,7018 No. 2 -19,33 0,633

No. 3 -18,8293 No. 3 -19,48 0,648

No. 4 overflow No. 4 overflow overflow

No. 5 -18,5909 No. 5 -19,40 0,811

No. 6 -18,0393 No. 6 -18,82 0,778

No. 7 -17,78 No. 7 -18,78 1,003

No. 8 -17,7099 No. 8 -18,81 1,100

No. 9 -17,7393 No. 9 -18,90 1,163

No. 10 -17,5026 No. 10 -18,90 1,393

No. 11 -17,5855 No. 11 -18,87 1,281

No. 12 -17,7264 No. 12 -19,05 1,323

No. 13 -17,6988 No. 13 -19,05 1,350

No. 14 -18,1594 No. 14 -19,11 0,952

No. 15 -18,0763 No. 15 -19,08 1,007

No. 16 -17,9358 No. 16 -18,98 1,040

176

Appendix A

No. 17 -18,322 No. 17 -19,10 0,777

No. 18 -18,0513 No. 18 -19,07 1,020

No. 19 -18,0282 No. 19 -19,07 1,041

No. 20 -18,0837 No. 20 -19,10 1,019

No. 21 -18,0387 No. 21 -18,70 0,659

No. 22 -17,9654 No. 22 -18,87 0,907

No. 23 -18,221 No. 23 -18,89 0,670

No. 24 -18,2613 No. 24 -18,90 0,636

No. 25 -18,0571 No. 25 -18,86 0,800

No. 26 -18,4081 No. 26 -18,93 0,525

No. 27 -18,2238 No. 27 -18,90 0,679

No. 28 -18,232 No. 28 -18,95 0,714

No. 29 -18,2769 No. 29 -18,89 0,614

No. 30 -18,1073 No. 30 -18,89 0,780

No. 31 -18,5908 No. 31 -18,83 0,242

No. 32 -18,6826 No. 32 -18,91 0,229

No. 33 -18,5963 No. 33 -18,73 0,129

No. 34 -18,8515 No. 34 -19,10 0,253

No. 35 -18,5633 No. 35 -19,01 0,450

No. 36 -18,0202 No. 36 -18,72 0,699

No. 37 -19,5615 No. 37 -19,63 0,073

No. 38 -19,2654 No. 38 -19,68 0,415

Table A4: Temperature distribution on the condenser and temperature difference

(case1) with the two manifold configurations.

Horizontal manifold Vertical manifold Heating Power: 50W Heating Power: 50W Thermocouple Tcondenser(℃) Thermocouple Tcondenser(℃)

Delta T (℃)

No. 1 -19.42 No. 1 -19.21 -0.21 No. 2 -19.33 No. 2 overflow overflow No. 3 -19.48 No. 3 -19.16 -0.32 No. 4 overflow No. 4 overflow overflow

177

Appendix A

No. 5 -19.40 No. 5 -19.21 -0.19 No. 6 -18.82 No. 6 -18.40 -0.42 No. 7 -18.78 No. 7 -18.40 -0.38 No. 8 -18.81 No. 8 -18.57 -0.42 No. 9 -18.90 No. 9 -18.52 -0.38

No. 10 -18.90 No. 10 -18.77 -0.13 No. 11 -18.87 No. 11 -18.61 -0.26 No. 12 -19.05 No. 12 -18.63 -0.42 No. 13 -19.05 No. 13 -18.74 -0.31 No. 14 -19.11 No. 14 -18.84 -0.27 No. 15 -19.08 No. 15 -18.91 -0.17 No. 16 -18.98 No. 16 -18.69 -0.29 No. 17 -19.10 No. 17 -18.91 -0.19 No. 18 -19.07 No. 18 -18.94 -0.13 No. 19 -19.07 No. 19 -18.60 -0.47 No. 20 -19.10 No. 20 -19.01 -0.09 No. 21 -18.70 No. 21 -18.50 -0.20 No. 22 -18.87 No. 22 -18.58 -0.29 No. 23 -18.89 No. 23 -18.61 -0.28 No. 24 -18.90 No. 24 -18.64 -0.26 No. 25 -18.86 No. 25 -18.67 -0.19 No. 26 -18.93 No. 26 -18.66 -0.26 No. 27 -18.90 No. 27 -18.67 -0.23 No. 28 -18.95 No. 28 -18.76 -0.19 No. 29 -18.89 No. 29 -18.76 -0.13 No. 30 -18.89 No. 30 -18.66 -0.22 No. 31 -18.59 No. 31 -18.63 0.04 No. 32 -18.68 No. 32 -18.66 -0.02 No. 33 -18.59 No. 33 -18.68 0.09 No. 34 -18.85 No. 34 -18.73 -0.12 No. 35 -18.56 No. 35 -18.63 0.07 No. 36 -18.02 No. 36 -18.25 0.23 No. 37 -19.56 No. 37 -19.38 -0.17 No. 38 -19.26 No. 38 -19.42 0.16

178

Appendix A

179

Appendix B

Appendix B: Temperature distribution of QM condenser before and after thermal cycling Table B-1:Case 1: 200W,15°C Corresponding Temperature difference before and after

thermal cycling

Silicongel T number

Condenser T number

delta T Before

delta T After

Variation

40 14 3.3 3.1 -0.2 39 15 2.2 1.9 -0.4 38 16 2.3 2.1 -0.2 37 17 -0.1 -0.2 -0.1 36 18 5.9 5.2 -0.7 34 19 13.0 11.9 -1.1 33 13 1.9 2.3 0.4 32 22 2.1 2.2 0.1 31 21 0.0 -0.1 0.0 30 20 5.9 5.4 -0.4 28 23 4.8 4.9 0.2 27 12 2.3 2.5 0.1 26 24 1.9 1.7 -0.2 25 25 0.0 0.0 0.0 24 26 4.9 4.1 -0.9 22 27 12.3 11.9 -0.4 21 11 2.3 1.7 -0.6 20 30 2.1 2.2 0.1 19 29 -0.1 0.0 0.0 18 28 6.4 5.7 -0.6 16 31 5.7 4.9 -0.8 15 10 2.3 2.0 -0.3 14 32 2.1 1.5 -0.6 13 33 0.0 0.1 0.0 12 34 6.4 6.4 0.0 10 35 12.9 12.8 -0.1 9 9 1.7 1.6 -0.1 9 8 2.6 2.7 0.2 8 37 2.0 2.0 0.1 6 36 6.2 6.3 0.1 3 4 3.8 4.2 0.4

180

Appendix B

Table B2: Case2: 150W,15°C Corresponding Temperature difference before and after thermal cycling

Silicongel T number

Condenser T number

delta T Before

delta T After

Variation

40 14 2.3 2.1 -0.2 39 15 1.5 1.3 -0.3 38 16 1.5 1.4 -0.1 37 17 -0.1 -0.2 -0.1 36 18 4.2 3.6 -0.6 34 19 9.1 8.4 -0.7 33 13 1.4 1.6 0.3 32 22 1.4 1.5 0.0 31 21 -0.1 -0.1 0.0 30 20 4.1 3.8 -0.3 28 23 3.3 3.4 0.1 27 12 1.6 1.7 0.1 26 24 1.3 1.2 -0.1 25 25 0.0 0.0 0.0 24 26 3.5 2.9 -0.6 22 27 8.6 8.4 -0.3 21 11 1.6 1.2 -0.4 20 30 1.4 1.5 0.1 19 29 -0.1 -0.1 0.1 18 28 4.5 4.0 -0.5 16 31 3.9 3.4 -0.5 15 10 1.6 1.3 -0.2 14 32 1.4 1.0 -0.4 13 33 0.0 0.0 0.0 12 34 4.6 4.5 0.0 10 35 9.1 9.1 0.0 9 9 1.2 1.2 -0.1 9 8 1.8 2.0 0.2 8 37 1.4 1.5 0.1 6 36 4.4 4.5 0.1 3 4 2.7 3.0 0.3

181

Appendix B

Table B3:Case3: 200W,-20°C Corresponding Temperature difference before and after thermal cycling

Silicongel T number

Condenser T number

delta T Before

delta T After

Variation

40 14 3.4 3.1 -0.3 39 15 2.6 2.0 -0.6 38 16 2.7 2.1 -0.6 37 17 -0.2 -0.3 -0.1 36 18 7.3 5.8 -1.5 34 19 13.7 12.3 -1.4 33 13 2.2 2.6 0.4 32 22 3.0 2.7 -0.3 31 21 0.0 0.0 0.0 30 20 8.0 6.7 -1.3 28 23 4.9 5.2 0.3 27 12 3.7 3.7 0.0 26 24 2.6 2.4 -0.2 25 25 0.1 0.1 0.0 24 26 6.0 4.5 -1.5 22 27 14.1 12.6 -1.5 21 11 3.7 2.3 -1.5 20 30 2.8 2.7 -0.1 19 29 -0.3 -0.2 0.2 18 28 6.6 5.9 -0.7 16 31 5.7 4.9 -0.8 15 10 3.7 3.2 -0.6 14 32 2.2 1.5 -0.7 13 33 0.1 0.1 0.1 12 34 7.7 6.7 -1.0 10 35 14.8 13.6 -1.2 9 9 3.8 3.2 -0.6 9 8 4.5 4.2 -0.3 8 37 2.6 2.3 -0.3 6 36 6.7 6.7 0.0 3 4 4.1 4.6 0.5

182

Appendix B

Table B4: Case4: 150W, -20°C Corresponding Temperature difference before and after thermal cycling

Silicongel T number

Condenser T number

delta T Before

delta T After

Variation

40 14 2.5 2.2 -0.3 39 15 1.9 1.4 -0.5 38 16 2.0 1.5 -0.5 37 17 -0.3 -0.3 0.0 36 18 5.4 4.1 -1.3 34 19 9.8 8.8 -1.0 33 13 1.6 1.9 0.2 32 22 2.2 2.0 -0.3 31 21 -0.1 0.0 0.0 30 20 6.0 5.0 -1.0 28 23 3.5 3.7 0.2 27 12 2.8 2.6 -0.2 26 24 1.9 1.7 -0.2 25 25 0.0 0.1 0.1 24 26 4.5 3.3 -1.2 22 27 10.3 9.1 -1.2 21 11 2.8 1.6 -1.2 20 30 2.0 1.9 -0.1 19 29 -0.3 -0.1 0.2 18 28 4.8 4.2 -0.6 16 31 4.0 3.4 -0.6 15 10 2.8 2.3 -0.4 14 32 1.6 1.0 -0.6 13 33 0.1 0.1 0.0 12 34 5.7 4.8 -0.9 10 35 11.0 9.8 -1.2 9 9 2.9 2.6 -0.3 9 8 3.3 3.2 -0.1 8 37 2.0 1.6 -0.4 6 36 5.0 4.8 -0.2 3 4 3.0 3.3 0.3

183

Appendix C

Appendix C: Temperature distribution of QM condenser before and after thermal cycling during thermal conductance test

Table C1: Case 1 during thermal conductance test

Condenser top T before thermal cycling

Condenser top T after thermal cycling

Thermocouple Tcondenser(℃) Thermocouple Tcondenser(℃) 1 15.9 1 15.9 2 15.9 2 15.9 3 15.9 3 15.9 4 16.1 4 16.1 5 16.0 5 16 6 16.0 6 16.1 7 16.0 7 16 8 17.2 8 17.2 9 17.1 9 17

10 17.4 10 17.4 11 17.5 11 17.4 12 17.4 12 17.2 13 17.2 13 17.1 14 16.9 14 16.7 15 16.8 15 16.7 16 16.8 16 16.7 17 16.9 17 16.8 18 16.9 18 16.8 19 17.4 19 17.3 20 17.2 20 17.1 21 17.3 21 17.2 22 17.1 22 17 23 17.1 23 17 24 17.1 24 17 25 17.2 25 17.1 26 17.3 26 17.1 27 17.9 27 17.8 28 17.4 28 17.3 29 17.2 29 17.2 30 17.3 30 17.2 31 17.2 31 17.1 32 17.1 32 17.1 33 17.1 33 17

184

Appendix C

34 17.3 34 17.2 35 17.6 35 17.5 36 17.1 36 17.1 37 17.0 37 17 38 15.7 38 15.6 39 15.7 39 15.6 40 15.7 40 15.6




Thermocouple Tcondenser(℃) Thermocouple Tcondenser(℃) 1 15.8 1 15.7 2 15.8 2 15.8 3 15.8 3 15.8 4 15.9 4 15.9 5 15.9 5 15.9 6 15.9 6 15.9 7 15.9 7 15.8 8 16.8 8 16.7 9 16.7 9 16.6

10 17 10 16.9 11 17 11 16.9 12 16.9 12 16.8 13 16.8 13 16.7 14 16.6 14 16.4 15 16.5 15 16.4 16 16.5 16 16.4 17 16.6 17 16.5 18 16.6 18 16.5 19 17 19 16.8 20 16.8 20 16.7 21 16.9 21 16.7 22 16.8 22 16.6 23 16.7 23 16.6 24 16.8 24 16.6 25 16.8 25 16.7 26 16.9 26 16.7

185

Appendix C

27 17.4 27 17.2 28 16.9 28 16.8 29 16.8 29 16.8 30 16.8 30 16.8 31 16.8 31 16.7 32 16.7 32 16.7 33 16.7 33 16.7 34 16.8 34 16.8 35 17.1 35 17 36 16.7 36 16.7 37 16.7 37 16.6 38 15.6 38 15.6 39 15.6 39 15.6 40 15.6 40 15.5




Thermocouple Tcondenser(℃) Thermocouple Tcondenser(℃) 1 -18.0 1 -18.0 2 -18.0 2 -18.0 3 -18.0 3 -17.9 4 -17.8 4 -17.8 5 -17.9 5 -17.8 6 -17.8 6 -17.8 7 -17.9 7 -17.9 8 -16.3 8 -16.2 9 -16.8 9 -16.7

10 -16.0 10 -15.9 11 -15.9 11 -15.9 12 -16.0 12 -16.1 13 -16.2 13 -16.3 14 -16.7 14 -16.7 15 -16.8 15 -16.8 16 -16.8 16 -16.8 17 -16.6 17 -16.6 18 -16.7 18 -16.6 19 -16.1 19 -16.1 20 -16.2 20 -16.2

186

Appendix C

21 -16.2 21 -16.2 22 -16.5 22 -16.4 23 -16.5 23 -16.5 24 -16.5 24 -16.4 25 -16.4 25 -16.3 26 -16.3 26 -16.3 27 -15.6 27 -15.6 28 -16.2 28 -16.2 29 -16.3 29 -16.2 30 -16.3 30 -16.3 31 -16.3 31 -16.3 32 -16.5 32 -16.4 33 -16.5 33 -16.4 34 -16.3 34 -16.3 35 -16.1 35 -16.1 36 -16.6 36 -16.6 37 -16.6 37 -16.6 38 -18.5 38 -18.6 39 -18.5 39 -18.6 40 -18.5 40 -18.6



Thermocouple Tcondenser(℃) Thermocouple Tcondenser(℃) 1 -18.2 1 -18.2 2 -18.2 2 -18.2 3 -18.2 3 -18.2 4 -18.1 4 -18.1 5 -18.1 5 -18.1 6 -18.1 6 -18.1 7 -18.2 7 -18.2 8 -16.9 8 -16.8 9 -17.3 9 -17.2

10 -16.6 10 -16.6 11 -16.5 11 -16.5 12 -16.7 12 -16.7 13 -16.8 13 -16.9 14 -17.2 14 -17.2 15 -17.3 15 -17.3 16 -17.3 16 -17.3 17 -17.1 17 -17.2

187

Appendix C

Table C4:Case 4 during thermal conductance test

18 -17.2 18 -17.1 19 -16.7 19 -16.7 20 -16.8 20 -16.8 21 -16.8 21 -16.8 22 -17.0 22 -17.0 23 -17.1 23 -17.1 24 -17.0 24 -17.0 25 -17.0 25 -16.9 26 -16.9 26 -16.9 27 -16.4 27 -16.4 28 -16.8 28 -16.8 29 -16.9 29 -16.9 30 -16.9 30 -16.9 31 -17.0 31 -16.9 32 -17.1 32 -17.0 33 -17.1 33 -17.1 34 -16.9 34 -16.9 35 -16.7 35 -16.8 36 -17.1 36 -17.2 37 -17.2 37 -17.2 38 -18.7 38 -18.8 39 -18.7 39 -18.8 40 -18.7 40 -18.8

188

Appendix C

189

Appendix D

Appendix D: List of Acronyms AMS Alpha Magnetic Spectrometer ISS International Space Station FM Flight Model QM Qualification Model EM Engineering Model BBM Bred Board Model AIDC Aerospace Industrial Development Corporation NLR National Aerospace Laboraroty SYSU Sun Yat-Sen University I/F Interface MLI Multi Layer Insulation DAQ Data Acquisition EGSE Electronic Ground Segment Equipment EOL End Of Life VSB Vertical Support Beam UPS Uninterruptible Power Supply HV High Voltage Brick PMT Photo-Multiplier Tube EIB Ecal Intermediate Board TS Temperature Sensor LTOF Lower Time Of Flight detector UTOF UPPER Time Of Flight detector TRD Transition Radiation Detector RICH Ring Imaging CHerenkov detector ECAL Electromagnetic CALorimeter detector PDS Power Distribution System CCEB Cryo Cooler Electronic Box CDD Cryo Dump Diodes TCS Thermal Control Subsystem TVC Thermal Vacuum Chamber TVT Thermal Vacuum Test TVTB Thermo Vacuum Thermal Balance Test LSS Large Space Simulator TRP Temperature Reference Point MERAT Mean Effective Radiation Temperature TTCS Tracker Thermal Control System HX Heat Exchanger APS Absolute Pressure Sensor DPS Differential Pressure Sensor TTCB Tracker Thermal (Control) Component Box TTCB-P Tracker Thermal (Control) Component Box Primary

190

Appendix D

TTCB-S Tracker Thermal (Control) Component Box Secondary TTCE Tracker Thermal Control Electronics

ams-02 tracker thermal control system: development of new ... the ams tracker thermal control system...

Documents