{' l rev. no. of pace - lpi.usra.edu #298 - sum... · subpackage no. 2 thermal model nodes...

43
{' L Summary of ALSEP Subpackage No. 2 Thermal Control Design RECEIVED REV. NO. PAGE OF JOe pace bptemtl Dlvlelon 7 DATE 6 December 1969 This ATM presents the thermal design, analyses, and testing of the ALSEP Engineering and Prototype Subpackage 2 pallet and the RTG models and includes a summary of Flight transient and steady state temperatures for the Subpackage 2 equipment during ALSEP lunar and operation. Prepared by: Approved by:

Upload: nguyenkhanh

Post on 22-Jun-2018

216 views

Category:

Documents


0 download

TRANSCRIPT

{' L Summary of ALSEP Subpackage No. 2

Thermal Control Design

RECEIVED

REV. NO.

PAGE OF

JOe pace bptemtl Dlvlelon

7 DATE 6 December 1969

This ATM presents the thermal design, analyses, and testing of the ALSEP Engineering and Prototype Subpackage 2 pallet and the RTG models and includes a summary of Flight transient and steady state temperatures for the Subpackage 2 equipment during ALSEP lunar ~eployment and operation.

Prepared by:

Approved by:

: ; I ~

1 0 0

2.0

3.0

4.0

5.0

Summary of ALSEP Subpackage No. 2 Thermal Control Design

CONTENTS

INTRODUCTION

SUBP ACKAGE NO. 2 DESCRIPTION

THERMAL MODELS AND AN.ALYSES

3.1 STEADY STATE MODEL

3.2 TRANSIENT MODEL

3-3 SUBP ALLET, CABLE REEL AND ALHT CARRIER

3.4 FUEL TRANSFER TOOL

ANALYSIS AND TEST RESULTS

4.1 SUBP ACKAGE NO. 2 TRANSIENT ANALYSIS

4.2 CABLE REEL, UHT, SIDE, EQ.•JIPMENT

4.3 SUBPACKAGE NO. 2 DEPLOYED

4. 3-1 Proto A Thermal Vacuum Test Results

4.4 ASTRONAUT GLOVE TEMPERATURE DU"iUNG TRAVERSE

4.4.1 Maximum Astronaut Contact Times

SUMMARY AND CONCLUSIONS

NO. REV. NO.

ATM-821

PAGE i OF

DATE 6 December 1969

Page

1

1

6

6

10

13

13

17

17

17

23

31

35

Summary of ALSEP Subpackage No. 2 Thermal Control Design

1 oapeoe Syatem8 Dlvlelan

Figure

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

19

20

21

LIST OF FIGURES

Title

Lunar Surface Phase

Deployment of ALSEP 1

Stowed Configuration

Subpackage No. 2 Equipment

Barbell Carry

Traverse

Nodal Breakdown of RTG

Nodal Breakdown of Subpackage 2 Pallet

Pallet 2 Deployed

Experiment Mounting Provisions

Transfer Fuel

RTG Fuel Transfer

Fuel Handling Tool Transient Response

Temperature Decay of Fuel Capsule

Flight Subpackage No. 2 Transient Response

D-1A Subpackage No. 2 Transient Response

Flight Model Subpackage No. 2 Transient Response with SIDE and Tool Box Mounted

Flight Model Subpackage No. 2 Transient Response, Cable Spool and Mounting Struts

Subpackage No. 2 Analytical Mode, Deployed Steady State Temperatures

Subpackage No. 2 D-1A Test Results, Deployed Steady State Temperatures

Pallet Assembly No. 2 Proto Test Results, Deployed Steady State Temperatures

NO. REV. NO.

ATM-821

ii PAGE OF

DATE 6 Decembar 1969

2

2

4

4

7

7

8

8

9

14

14

15

16

18

19

20

21

22

24

25

26

' '

: : I ~

'Otlpac&

Summary of ALSEP Subpackage No. 2 Thermal Control Design

~ • .&'telns Division

LIST OF FIGURES (Cont 1 d)

Figure Title

22 Proto A Thermal Vacuum Test Summary - 1

23 Proto A Thermal Vacuum Test Summary - 2

24 Proto A Thermal Vacuum Test Summary - 3

25 Proto A Thermal Vacuum Test Summary- 4

26 Proto A Thermal Vacuum Test Summary - 5

27 Proto A Thermal Vacuum Test Summary - 6

28 Proto A Thermal Vacuum Test Summary - 7

29 Transient Deployment Response, Inner Surface Astronaut Glove, Logarithmic Interpolation

30 Transient Deployment Response, Inner Surface Astronaut Glove, Linear Interpolation

31 Mast/Carry Bar

HO. REV. NO.

ATM-821

PAGE iii OF

DATE 6 December 1969

Page

27

27

28

28

29

29

30

of 32

of 33

34

' l

: : . ~

Table

1

2

3

4

5

Summary of ALSEP Subpackage No. 2 Thermal Control Design

TABLES

Title

Summary of Radiative Properties on Surfaces in Subpackage No. 2 Thermal Model

Subpackage No. 2 Thermal Model Nodes

Thermal Model Resistors

Maximum Astronaut Contact Times

Inner Glove Safe Contact Time

NO. REV. NO.

ATM-821

PAGE iv OF

DATE 6 December 1969

3

5

11-12

36

37

' '

Summary of ALSEP S~bpackage No. 2 Thermal Control Design

NO.

ATl'-:1-821

REV. NO.

PAGE OF

DATE 6 December 1969

1.0 Introduction

This report summarizes the thermal analyses and design of the ALSEP Subpackage No. 2. Results of this ATM are based on the analysis, design, and testing of the Thermal/Mechanical Simulator (D-1A model), the analysis and design of the Prototype A and Flight 1 models, and the thermal testing of the Prototype A model. The thermal analysis of the Engineering D-1A model in the lunar environment, including deployment, is presented as well as the results from the D-1A tests (July 1967) and the Prototype A Thermal Vacuum Tests (December 1967). The thermal analysis of the Flight 1 design, including Subpackage No. 2 equipment, astronaut tools and cable reel, is presented and compared with the D-1A analysis. A thermal analysis of astro­naut glove inner surface temperatures is also presented for the time span corresponding to the astronaut traverse to deployment site.

This summary is primarily concerned with the transient phases of the ALSEP deployment during lunar surface operations. The deployment phases include the SNAP-27 fuel capsule transfer, the astronaut traverse to deploy­ment site, and the subsequent removal of the subpackage equipment at the deployment site (see Figure 1). The temperature responses of the deployed ALSEP Pallet ~SNAP-27 RTG as illustrated in Figure 2 also are given.

2.0 Subpackage No. 2 Description

Subpackage No. 2 components and equipment identification along with thermal coatings are listed in Table 1.

Subpackage No. 2 is shown in Figure 3 in the stowed configuration and consists of the SNAP-27 Generator Assembly, Pallet Assembly No. 2, the Apollo Lunar'Handling Tool Carrier, the Suprathermal Ion Detector Experiment (SIDE) plus additional equipment boxes and astronaut handling tools. The deployed configuration consists of only the RTG mounted on Pallet No. 2. Figure 4 illustrates the deployed configuration plus the position of the subpallet and the ALHT Carrier on the pallet. These latter assemblies are removed prior to completion of deployment with the removal point in the sequence of deployment varying with the deployment mode. The one man deployment concept calls for the removal of these items at the deployment site, while the two man deployment concept calls for the removal prior to fueling the Generator Assembly. The D-1A model utilizes the one may deployment concept, while both concepts have been considered for the Flight and the Proto A models.

Reference to the subpallet assembly in this ATM is defined as the assembly consisting of the SIDE, Antenna Gimbal Box, UHT and the subpallet.

The pallet becomes a base for the deployed Generator Assembly and pro­vides a conductive thermal radiation surface between the Generator Assembly and the lunar surface. The pallet also acts as a thermal shield between the fueled Generator Assembly and the astronaut during the transposition of

' '

ALSEP DEPLOYMENT

TASKS AS SOC lA TED

WITH LM

ALSEP DEPLOYMENT

TASKS AT EXPERIMENT

SITE

LUNAR SURFACE PHASE TRANSFER

FUEL

ATM-821 2 of 38

REMOVE PACKAGES PREPARE FOR TRAVERSE

EMPLACE

CENTRA, STATION ~

~ ~~'

DEPLOYMENT FOR ALSEP 1

FIGURE 1

~~-3o_O_IT_B~AS~E~D~ON~--~-~-~~~~;~~~~-~---~--~-~

~~ LM ASCENT BLAST

~~~~SOLAR WIND SPECTROMETER

MAGNETOMETER

N><s w

-o COLD CATHODE I ON GAGE FIGURE 2

1 .

2.

3.

4.

6.

8.

10.

11.

Summary of ALSEP Subpackage No. 2 Thermal Control Design

TABLE 1

Summary of Radiative Properties of Surfaces Subpackage No. 2 Thermal Models

Surface Description 0( f.

Pallet Assembly No. 2 (top/bottom) .2 ·9

Apollo Lunar Hand Tool (ALHT) . 25 .85

Antenna Gimbal Box . 25 .85

Fuel Cask Handling Tool (FHT) .25 .85

Cask Dome Removal Tool (CDRT) .25 .85

Carry Bars .25 .85

RTG Cable Reel . 25 .85

Universal Handling Tool (UHT) . 25 .85

RTGSurfaces .85 .85

Subpallet .2 ·9

Suprathermal Ion Detector Experiment (SIDE)

.25 .85

NO. REV. NO.

ATM-821

PAGE 3 OF

DATE 6 December 1969

On

Comments

Z~93 White Paint

Sulfuric Anodized

3M401 White Paint

3M401 White Paint

3M401 White Paint

3M401 White Paint

3M401 White Paint

3M401 White Paint

Iron Titinate

3M401 White Paint

S-13G White Paint

ANTENNA

STOWED CONFIGURATION

SUPRA THERMAL ION DETECTOR

ANTENNA

ATM-821 4 of 38

ALHT TOOL CARRIER

AIMING MECHANISM

SOLAR WIND SPECTROMETER

PASSIVE SEISMIC EXPERIMENT

UNIVERSAL HANDLING TOOL

SUPERTHERMAL ~ ~ ION DETECTOR J ~~ ,---------

ANTENNA GIMBAL BOX

1 I I I I I I I I

SUBPALLET

RADIOISOTOPE THERMOE;LECT RIC GENERATOR

PALLET ASSY NO. 2

SUBPACKAGE NO. 2

FIGURE 3

ALHT TOOL CARRIER

FIGURE 4

Summary of ALSEP Subpackage No. 2 Thermal Control Design

NO.

ATM-821

PAGE 5

REV. MO.

OF ospace

Systems Divlelon DATE 6 December 1969

Node Number

1

2

3

4

5

6-7

8-9

10

11

12-13

14-21

22

23

TABLE 2

Thermal Model N·Jdes

Total Thermal Capacitance

Node Description (BTU/°F)

Fuel 0.555

Capsule 0.350

Hot Frame 0.631

Thermopile and Insulation 0.411

Cold Frame 3.003

Outer Shell (symmetric about GA axis) 1.086

Fins (4 fins/node) 1.550

Capsule End Plate 0.151

RTG Bottom Plate (hermetic seal, lower) 0.030

RTG/Pallet Interface (mounting lug and inserts) 0.042

Pallet (4 top and 4 bottom) 0.658

Lunar Surface oo

Space ro

: -: I ;

Summary of ALSEP Subpackage No. 2 Thermal Control Design

NO.

ATM-821

PAGE 6

REV. NO.

OF

DATE 6 December 1969

ALSEP to the deployment site. The RTG is blocked by the other assemblies during this latter phase for the one man deployment concept. No blockage occurs with the two man deployment concept. These are illustrated in Figures 5 and 6.

The fueled SNAP-27 Generator Assembly consists of the fuel capsule assembly, the hot frame, thermopile and insulation, cold frame, outher shell and fins, and the hermetic seal and mounting feet. Basically, the Generator Assembly is a series of concentric cylinders with insulated ends, the inner cylinder being the hot frame. The outer cylinder is the cold frame and outer shell, hereafter referred to simply as the cold frame. Eight radial heat rejection fins are attached to the cold frame. The cap­sule is the innermost cylinder of the RTG, facing the hot frame. The thermopile (Thermoelectric couple assembly) and insulation occupy the space between the hot frame and the cold frame. The annulus at each end is a hermetic seal, the lower one forming the bottom of the Generator Assembly. The capsule end plate forms the top of the Generator Assembly. The hot frame receives radiant energy directly from the capsule and con­ducts most of this to the thermopile and insulation. The thermopile con­verts a portion of this thermal energy into electrical energy. The remain­ing thermal energy is conducted from the thermopile and insulation layers to the cold frame. The cold frame rejects this heat, plus solar heating, primarily by conduction to the fins and then by radiation to space and the lunar surface.

3.0 Thermal Models and Analyses

3.1 Steady State Model- Nodes

The steady state model for Subpackage 2 consisted of 23 nodes. The nodal arrangement was divided into eight (8) nodes for the pallet, nine (9) nodes for the Generator Assembly, two (2) nodes for the RTG/Pallet inter­face, and two (2) nodes for the environment. Table 2 lists the number of nodes, node description, and thermal capacitance. Nodes were also assigned for the fuel element and the capsule.

Figures 7 and 8 show the locations of the RTG nodes and the pallet nodes respectively, while Figure 9 illustrates the deployed state.

The steady state analysis assumptions were as follows:

1. RTG surface emittance = 0.85

2. Pallet surfaces coated with thermal control paint having an ex.; f- of • 20/.85

3. No conductive coupling between the pallet and the lunar surface

BARBELL CARRY

• ALLOWS ALL EQUIPMENT TO BE CARR I ED BY ONE MAN IN ONE TRAVERSE

• SUITCASE HANDLES FOR TWO-MAN OR BACKUP CARRY MODE

• Gl VES GOOD BALANCE & VIEW OF FEET

• EQUIVALENT EARTH WEIGHT~ 35 LB

TRAVERSE

A'I'N-821 7 of 38

• MAY BE SET DOWN TO REST

• CARRY BAR LATER USED AS ANTENNA MAST

FIGURE 5

• COMMANDER CARRIES INSTRUMENT SU BPALLET & TOOL CARRIER

• COMMANDER LEADS & PICKS ROUTE

• LM PI LOT CARRIES ALSEP BAR BELL

• REST, AS NECESSARY

FIGURE 6

Figure 7. Na'da1 Breakdown of RTG

RTG CUTAWAY

(1) FUEL---

(3 > HOT FRAME

(2)

ASSEMBLY

(6-7) OUTER CASE (5 > (COLD FRAME)

ASSEMBLY

NOTE: Node numbers in parenthesis

A--- Nodes 12 and 13 on Pallet under RTG

B --- Nodes 14 and 15 near RTG cable reel

C ---Nodes 16 and 17 near crew tools

D --- Nodes 18 and 19 beneath ALHT carrier

E --- Nodes 19 and 21 under SP/2 subpallet

I'

END PLATE (10)

OF FUEL CAPSULE

LUG

THERMOELECTRIC COUPLE ASSEMBLY (

4)

HEAT REJECTION (8-9)

FINS

( 11)

OD

oE

Figure 8. Nodal Breakdown of Subpackage 2 Pallet.

ATM-821 8 of 38

Outer Shell & Cold Frame Ass 1y.

~----"'-,

RTG/PAL Interface

" '

Figure 9.

Capsule End Plate

Pallet Under RTG

Outer Pallet

Pallet 2 Deployed

'-Oll>

o:£ 1-1)~ w 1\) CD-"

: : I •

if

1 oapace

Summary of ALSEP Subpackage No. 2 Thermal Control Design

NO. REV. HO.

ATM-821

PAGE 10 OF

Systems Dlvlelon DATE 6 December 1969

4. All boxes and equipment, except RTG, removed.

5. RTG axis perpendicular to the lunar surface.

3.2 Transient Thermal Models - Nodes

The transient thermal models of D-1A and the Flight (one man), with Subpallet Assembly and the Apollo Lunar Hand Tool (ALHT) Carrier mounted, consisted of the steady state nodes plus 6 nodes for each equipment box and/or experiment. The 6 nodes correspond to the 6 sides of each box or experiment. The Flight (two man) transient model was the same configura­tion as the steady state model. Figure 5 shows the one man deployment con­cept and Figure 6 illustrates the two man deployment.

All transient analyses assumed the following:

1. Surface emissivities and solar absorptivities per the steady state model.

2. No conductive coupling between the pallet and the lunar surface.

3. Surface of boxes facing RTG, ~ = 0.1 for D-1A, and E = 0.85 for Flight model (one man).

4. Box/equipment surfaces not facing RTG coated with a white thermal coating with an()(/~= .25/.85.

5. RTG axis parallel to lunar surface.

6. Corner of pallet closest to RTG oriented closest to lunar surface.

7. Bottom of the Subpackage No. 2 pallet radiating to a lunar sur­face temperature of 100°F.

8. Initial SP2 temperature at 160°F, except capsule assembly.

9. Capsule assembly at lunar landing temperatures.

Conduction resistors were calculated from configuration geometry and the material thermal conductivities. Thermal capacitances were calculated from material properties and nodal weights. Radiation resistors were de­rived from a Gebhart Absorption analysis wherein perfectly diffuse surfaces were assumed. Configuration factors for the Absorption program were de­rived with the BxA Confac II program. Surface areas were obtained from the configuration geometry, and emissivities were obtained from the particular surface properties. Solar heating was calculated for each surface, taking into account solar absorptivity, solar angle, shadowing and reflection from

Summary of ALSEP Subpackage No. 2 Thermal Control Design

TABLE 3

NO. REV. NO.

ATM-821

PAGE 11 OF

DATE 6 December 1969

Pallet 2 Steady State Thermal Model Resistors

Conduction Resistors

Node to Node

1 - 2 1 - 3 1 - 9 1 - 17 1 - 22 2- 4 2 - 9 2 - 17 2 - 23 5 - 6 5 - 20 5- 22 5 - 7 5 - 8 6 - 21 6 - 23 7 - 8 7 - 18 8 - 19

18 - 19 18 - 20 19 - 20 21 - 20 9 - 13 1 - 12 2 - 12

12 - 14 9 - 16

13 - 16 14 - 15 16 - 17

Radiation Resistors

Node to Node

1 - 3 1 - 4 1 - 5 1 - 7

R, hr-F/Btu

0.564 0.077

23.64 0.0018 1.47 0.077

23.64 0.0018 1.47 1.52 0.0266 0.865 3.16 2.70 0.0266 0.830

10.24 0.0196 0.0233

10.24 3.16 2.67 1.52

114.0 3.0 3.0 3.96

28.15 0.0614 0.067 0.0614

a"~ A Btu/hr 0 R4

o . 3242 x 1 o-9 0.3529 x 1o-11 0.2837 X 10-9

0 0.2393 X 10-1

Comments

Outer shell facing experiments to connecting nodes.

1 Outer shell away from experiments to connecting nodes.

t Pallet under RTG facing experiments to connectin~ nodes.

i Pallet under RTG away from experi­ments to connecting nodes. Outer pallet under SIDE to connecting nodes. Outer Pallet under ALHT to Bottom Pallet bottom to connectin nodes.

RTG bottom to Hot Frame. Outer shell to capsule and plate.

Capsule end plate to capsule. RTG bottom to Thermopile and insulation. Hot frame to Thermopile and insulation. Capsule to Fuel. Thermopile and insulation to Cold Frame.

Comments

Outer shell facing experiments to fins, pallet RTG, lunar surface and space.*

S~ary of ALSEP Subpackage No. 2 Thermal Control Design

TABLE 3 (Cont'd)

MO. REV. MO.

ATM-821

PAGE 12 OF 38

DATE 6 December 1969

Pallet 2 Steady State Thermal Model Resistors

Radiation Resistors

Node to Node

1 - 8 1 - 9 1 - 10 1 - 11 2 - 3 2 - 4 2 - 6 2 - 9 2 - 10 2 - 11 3 - 4 3 - 5 3 - 6 3 - 7 3 - 8 3 - 9 3 - 10 3 - 11 4 - 5 4 - 6 4 - 7 4- 8 4 - 9 4 - 10 4 - 11 5 - 6 5 - 7 5 - 8 5 - 9 ) - 10 5 - 11 6 - 9 6 - 10 6 - 11 7 - 10 7- 11 8 - 10 8 - 11 9 - 11

12 - 11

(f" tq. A Btu/hr 0R4

0.1483 X 10-10 0.4077 X 10-11 0.2731 X 10-9 0.6539 X 10-9 o. 3525 x 1 o-11 0.3238 X 10-9 0 • 2837 X 10-9 0 • 4077 X 10-11 0.2956 X 10-9 0.6705 X 10-9 0.4430 X 10-9 0.3268 X 10-9 0.3899 X 10-11 0.8169 x 1o-10 o. 114 3 x 1 o-9 o.4748 x 1o-11 0.1594 X 10-8

8 0.2459 X 10-0.3899 X 10-11 0.3268 X 10-9 0.8863 X 10-1 2 o • 1240 x 1 o-11 0.4747 X 10-

811

0.1765 X 10-0 • 2594 x 1 o-8 0.8572 x 1o-11 0.1359 x 1o-11 0.1335 x 1o-11 0.1134 X 10-9 0.2064 x 1o-10 0.6069 X 10-9 0.1134 X 10-9 0.2264 x 1o-10 0.6084 X 10-9 0 • 1739 X 10-

811

0.2094 X 10-0.1888 X 10-11 0.1082 X 10-8 0.1739 x 1o-1o 0.2940 X 10-9

Comments

Outer shell facing experiments to fins, pallet RTG, lunar surface and space. i~

t . Outer shell away from experlments to fins, pallet, RTG lunar surface, and space.*

1 Fins facing experiments to fins, pallet, RTG, lunar surface, and space.*

l Fins from experiments to fins, pallet, RTG, lunar surface, and space.

l Pallet under RTG facing experiments to pallet, RTG, lunar s;ll'face, and space.*

! Pallet under RTG away from experiments to RTG, lunar surface, and space.*

Outer pallet under SIDE to lunar surface and space.* Outer pallet under ALHT to lunar surface and space.* RTG bottom to space.* Capsule end plate to space.*

i~Evaluated using Gebhart absorption factor technique

Summary of ALSEP Subpackage No. 2 Thermal Control Design

NO.

ATM-821

PAGE 13

REV. NO.

OF ,epace

Systems Division DATE 6 December 1969

other surfaces. Contact conductance was estimated as 50.0 Btu/hr ft2 F for high contact interfaces, 5.0 Btu/hr ft2 F for medium contact interfaces, and 1.0 Btu/hr ft2 F for loose contact interfaces. Nodes were connected with the applicable resistors, listed in Table 3, and assigned thermal cap­acitance v~lues, initial temperatures and heat loads corresponding to each case under study. The Bendix Thermal Analyzer Program was utilized on the IBM 360 Computer to solve for both transient and steady state temperatures.

3.3 Sub-Pallet, Cable Reel and ALHT Carrier

The sub-pallet and ALHT Carrier were analyzed concurrently with the Pallet Assembly No. 2. Emissivity of the sub-pallet was assumed to be 0.85 for all surfaces, while the ALHT Carrier was analyzed for both 0.1 and 0.85 emissivities for the surface facing the RTG. The remaining surfaces of the ALHT Carrier were assumed to have an emissivity of 0.85.

The cable reel was analyzed as two nodes, the spool and the mounting struts. Emissivity was assumed to be 0.85 for all surfaces. No conduction to the pallet, cable, or between the mounting struts and the spool was assumed.

Figure 10 illustrates areas where the Universal Handling Tool (UHT) will be used. This tool will have approximately the same transient response as the cable reel if the sub-pallet is mounted on the Subpackage 2 during traverse.

3.4 Fuel Transfer Tool

The FTT was analyzed as 4 nodes with a constant temperature source node. The four nodes were:

1. Disc facing source

2. Fingers contacting source

3. Handle attach adapter for 1 and 2 above

4. Handle (center of node at handle C.G.)

Heat transfer is by radiation from the source to the disc, conduction from the source to the fingers, conduction from the disc and fingers to the attach adapter, and conduction from the adapter to the handle. Radiation to the environment and solar heating were not considered for this analysis in order to study worst case conditions. See Figures 11 and 12 for a summary of the fuel transfer task.

Figure 13 presents the results of a transient analysis on the Fuel Transfer Tool. The heat source was the Capsule End Plate at a constant

EXPERIMENT MOUNTING PROVISIONS ATM-821 14 of 38

LSM

TRANSFER FUEL • PREPARE CASK

RETRIEVE LOCK-PIN LANYARD, PULL PIN, RETRIEVE CASK-TILT LANYARD, ROTATE CASK TO DES IRED ANGLE, REMOVE DOME

• PREPARE SUBPACKAGE 2

RETRIEVE MAST SECTIONS & TOOLS, PLACE IN NEW LOCATIONS, ROTATE SUBPACKAGE TO TRANSFER POSITION, PREPARE FUEL TRANSFER ASSEMBLY TOOL

• FUEL RTG .

ENGAGE TOOL WITH CAPSULE, LOCK TOOL I

FIGURE 10

CAPSULE TO RELEASE FROM CASK, WITHDRAW !"""'-------("

CAPSULE, CARRY TO RTG, RELEASE TOOL I CAPSULE TO LOCK IN RTG

FIGURE 11

--' ... =

......

ATM-821

15 o

f 38

N

.......

~ o·

400

350

300

250

~ ~ ;:J ~

< ~ ~ 200 11. ~ ~

~

150

100

FUEL TRANSFER TOOL TRANSIENT RESPONSE INITIAL CONDITIONS: LUNAR LANDING TEMPERATURE CAPSULE END PLATE AT 800°F

DISC

HANDLE C.G.

~---------------------------------------------------------------------------------------------~~

0 2 4 6 8 10 12 14 16 18 20

ELAPSED TIME FROM CONTACT WITH CAPSULE END- MINUTES

FIGURE 13

0'\~

~~ 1\)

w -I. (X)

Summary of ALSEP Subpackage No. 2 Thermal Control Design

NO.

ATM-821

PAGE 17

REV. NO.

OF

DATE 6 December 1969

temperature of 800° F. Surface emissivity of the disc was assumed to be 0.25. Radiation to other surfaces of the FTT was assumed negligible. The disc exceeded 250° F within 2 minutes after ccntact of FTT with the Capsule End Plate. The remainder of that end of the FTT exceeded 250° F within 9 minutes. Temperature rise at the handle C.G. was approximately 2° F during the first 10 minutes.

Figure 14 shows the temperature decay of the capsule and end plate between removal from the GLFC and RTG insertion. The decay is very high for the first few minutes after removal. Thus the temperatures, beyond the first 2 minutes, shown in Figure 13, are conservative.

The surface of the disc may have a high emissivity, in which case FTT temperatures will be considerably higher. However, present FTT con­tact time with the capsule end is approximately one minute. An analysis of the FTT, with t= 0.9 for the disc, resulted in a disc temperature of 250° F with the remainder of the FTT below 200° F for one minute contact time.

4.0 Analysis and Test Results

4.1 Subpackage No. 2 Transient Analyses

Figures 15 through 17 present time-temperature predictions for Sub­package 2 during traverse to the deployment site. Figure 15 is the tran­sient analysis for the Flight (two man) model. The others are models with boxes and/or equipment mounted during traverse. RTG temperatures are approximately the same for all three cases. Pallet surfaces under the RTG are considerably cooler for the Flight (two man) model than for the other two cases. Cold frame temperatures exceed 250° F within 12 minutes from time of fueling, and all external RTG temperatures exceed 250° F within 25 minutes. Maximum pallet temperatures exceed 250° F in 40 to 50 minutes from fueling, depending on the model.

Test results indicate thermocouple location temperatures, while analytical temperatures are average nodal temperatures. Thus analytical results indicate approximately the average temperatures under the RTG and for the remainder of the pallet.

4.2 Cable Reel and Equipment

Figure 18 presents the results of a transient analysis on the Cable Reel during traverse. The mounting struts exceed 250° F in apgroximately 30 minutes after RTG fueling while the cable spool exceeds 250 F in approximately 50 minutes. Temperatures in the deployed mode will be less than for the traverse, but the time to reach 250° F during the deployment mode will be no more than 5 minutes greater than for traverse.

1400

1200

~ 0

1000 ~

r:il ~ p E-< ~

800 ~ r:il p. ::;E r:il E-< 600

400

200

0

0

TEMPERATURE DECAY OF FUEL CAPSULE GLFC REMOVAL TO RTG INSERTION

CAPSULE

2 4 6 8 10 12

ATM-821 18 of 38

14

ELAPSED TIME AFTER REMOVAL FROM GLFC - MINUTES

FIGURE T4

700

600

500 I \

I ' ~ 400 0 I

~

~ ~ 0

300 I ~ <r: ~ ~ I ~ ~ ~ 200 ~

L .. ~

l~ 100

0 0

/

__...,-

10

FLIGi .AODEL

SUB-PACKAGE NO. 2 TRANSIENT RESPONSE RTG FUELING TO DEPLOYMENT SITE INITIAL CONDITIONS- LUNAR LANDING TEMPERATURES 20° SOLAR ANGLE TO RTG & PALLET

CAHSULE·END.PLATE

COLD FRAME ~ ~

~ FINS

/ ------------- ----PAL/RT_G... INTERFACE

~ -------- ------ PALLET .UNDER FINS

~----- ~ OUTER PALLE'T'

20 30 40 50 60

ELAPSED TIME - MINUTES CAPSULE INSERTED AT TIME= 0

FIGURE 15

-":t> '-0 1-3

~t 1\)

w--" CX>

700

600

500 \ ~ 0

~

I ril 400 P:: p ~ <X; P:: ril 300 ~ ~ ril ~

200 I _/"'

100

0

0

"-....

~~

10

D-lA MODEL (REFLECTIVE BOXES)

SUB-PACKAGE NO. 2 TRANSIENT RESPONSE RTG FUELING TO DEPLOYMENT SITE INITIAL CONDITIONS- LUNAR LANDING TEMPERATURES 20° SOLAR ANGLE TO RTG & PALLET

CAPSULE END PLATE

------ ----- FINS

~-------

20 30 40 50

OUTER PALLET & BOXES (MAX)

60 I\) :X:. 05e 0 I

ELAPSED TIME - MINUTES - CAPSULE INSERTED AT TIME= 0 H) (X)

[\) LV-' co

FIGURE 16

~ 0

~

~ ~ ::> E--1 ~ ~ IZl ~ ~ ~ E--1

700

600

500

\ 400

I

300

200 I

100

0

0 10

FLIGHT MODEL WITH SIDE & TOOL BOX MOUNTED (BOTH NON -REFLECTIVE) SUB-PACKAGE NO. 2 TRANSIENT RESPONSE RTG FUELING TO DEPLOYMENT SITE INITIAL CONDITIONS - LUNAR LANDING TEMPERATURES 20° SOLAR ANGLE TO RTG & PALLET

CAPSULE END PLATE

----- COLD FRAME

_.,.---- FINS

PALLET UNDER FINS

>:c EQUIPMENT ANDTOOL BOX {MAX)

OUTER PALLET

20 30 40 50 60

ELAPSED TIME -MINUTES- CAPSULE INSERTED AT TIME= 0

1\);:t> ~1-3

~

~&, f\)

w~ co

>:c Equipment consists of SIDE, PSE Mount, Antenna Gimbal and Experiment Handling Tool FIGURE ~i7

~ 0

~

~ p::; ::> E-l <t: p::; !:il Po. ~ !:il E-l

350

300

250

200

150

100

0 10 20

FLIGHT MODEL

SUB-PACKAGE NO. 2 TRANSIENT RESPONSE CABLE TEMPERATURES DURING TRANSPORT TO DEPLOYMENT SITE INITIAL CONDITIONS - LUNAR LANDING TEMPERATURES 20° SOLAR ANGLE TO RTG AND PALLET

MOUNTING STRUTS

30 40 50 60

ELAPSED TIME -MINUTES - CAPSULE INSERTED AT TIME== 0 FIGURE 18

1\)~

1\)~ 0 I ~CP

1\) \..A.J __,. CP

: : . . Summary of ALSEP Subpackage No. 2

Thermal Control Design

NO.

ATM-821

PAGE 23

REV. NO.

OF

DATE 6 December 1969

Maximum traverse temperatures of the SIDE, Antenna Gimbal, UHT, and the non-reflective ALHT Carrier mounted to the Pallet Assembly No. 2 are shown in Figure 17. Traverse temperatures for reflective boxes mounted to Pallet Assembly No. 2 are shown in Figure 16. None of the temperatures exceed 200° F within 45 minutes after RTG fueling. UHT temperatures for the one man deployment may be taken from Figure 18.

The above parts are coated with thermal paint having a low 0(/~ ratio (.20/.85). Therefore, the temperatures will not exceed 250° F when the parts are removed from the fueled Subpackage No. 2. The equilibrium temperature for an Q(/€ = .20/.85 vertical surface with a solar angle of 40° is 140° F.

4.3 Subpackage No. 2 -- Deployed

Figures 19 through 28 present the Subpackage No. 2 temperature pro­files during Lunation. Figure 19 temperatures represent the analytical model, and others show D-1A and Proto Thermal Vacuum Test Results. The o~ mode of heat transfer assumed to exist between the bottom surface of the pallet and the lunar surface was radiation for all three cases. Analysis and test results showed that the RTG temperature changes are primarily a function of solar heating, being only slightly affected by pallet tempera­ture changes. Analysis has shown a RTG temperature drop of 5° F for lunar noon and a drop of 14° F for lunar night resulting from the addition of high coupling between the bottom surface of the pallet and the lunar sur­face. The pallet temperatures follow the lunar surface temperature more closely for the latter case than for the analytical case shown.

Pallet and RTG maximum temperatures are:

Component Lunar Noon, OF Lunar Night, OF

Capsule 1380 1350

Hot Frame 1100 1050

Cold Frame 480 440

Pallet under RTG 425 250

The pallet is nearly isothermal from the top surface to the bottom surface, and isothermal planes are approximately concentric with the longi­tudinal axis of the RTG.

1400

1200

1000

800

·~ 0 600

~ ex; ::> E-i <t: 400 ex; ~ ~ ~ ~ E-i 200

0

-300

SUB-PACKAGE NO. 2 ANALYTICAL MODEL DEPLOYED STEADY STATE TEMPERATURES ONE LUNATION PERIOD

LUNAR SURFACE----u

CAPSULE

HOT FRAME

THERMO PILE

CAPSULE END PLATE

OUTER SHELL NS

O'UTER PALLET

ATM-821 24 of 38

NIGHT 0 30 60 90 120 150 180

SOLAR ANGLE - DEGREES FROM SUNRISE

FIGURE )9

1400

1200

1000

800 ~

0

t:il tx: :;:J

600

E-4 <!:; tx: t:il 0;

400 ~ t:il E-4

200

0

-300

SUB-PACKAGE NO. 2 - D-1A TEST RESULTS DEPLOYED STEADY STATE TEMPERATURES

HOT FRAME

ROOT OF FINS

PALLET UNDER FINS

OUTER PALLET

LUNAR SURFACE

NIGHT 0 30 60 90 120 150

SOLAR ANGLE - DEGREES FROM SUNRISE

180

ATM-821 25 of 38

FIGURE 2.()

PALLET ASSEMBLY NO. 2 -PROTO TEST RESULTS DEPLOYED STEADY STATE TEMPERATURES: °F UPPER TEMPERATURE AT EACH POINT IS LUNAR NOON LOWER TEMPERATURE IS LUNAR NIGHT

226 s -18

238 s 2

227 s 8

214 s -14

BOTTOM SIDE

s 323 151

325 s 150

S329 '141

s 333 182

366 328 s

19, 190

s 333 170

TOP SIDE

s 248 42

s 325 137

s 337 177

6) 240 48

ATM-821 26 of 38

FIGURE 21

0 0

0 "'

0 0

0 ::r

0 oo .--.a

""' LLa -a

0 N

w a: ::::> t-o a:: a a: w~ CL L. w r-

0 0

FIGURE 22

BENDIX RER~SPRCE SYSTEMS DIVISI~N

ATM-821 27 of 38

PR~GRRM RLSEP TEST N~. S1 TEST ORTE 12/15/67 ZER~ TIME 000000 TEST TITLE PR~T~ R SYSTEM THERMAL VRCUUM TEST SUMMARY

~ 102 PAL 2 1CP SURFACE UNDER R1G ~ 104 PAL 4 1CP SURFACE UNDER R1G

i i

~ !03 PAL 3 1CP SURFACE UNDER R1G ~!OS PAL 5 TCP SURFACE UNDER RTG

-------+-----+-+---------------l----1~--e----- j----- -----l----- -----

---+----+-1--fV---------1------~ _j_ ___ -------++--1~ -+----+-----1 ' ~ I ~

II '-----

0

0 0

0 ~

0 0

0 ::r

0 0

0 ("")

- 0 oo .--<0 ¥N

U....o -a

0

w a: ::::> t­O:: a a: a

~0 L. w r-o

0

0 ~

I

0 0

0 N

10. O')

10. oo

~I !

I I

i I I 24.00 48.00 qn_oo 12o.oo 144.oo 168.00 19~.00 216.00 240.0

T I 1"1E (H~URSJ

FIGURE 23

BENDIX RER~SPRCE SYSTEMS DIVISI~N PR~GRRM RLSEP TEST N~. 51 TEST ORTE 12/15/67 ZERD TIME 000000 TEST TITLE PR~T~ R SYSTEM THERMRL VRCUUM TEST SUMMRRY

~ 101 PAL 1 TCP SURFACE UNDER RTG ~ l 06 PAL 6 TOP SURFACE. RIGHT FRCNT --l ~ 107 PAL 7 TOP S U f!,(R C£ . .L E F 1fTiltrr.J T <!> 108 PAL 8 TOP SURFACE. LEFT REAR

( J

~-----

- .F ~ I ,l$l._ ~ 'j( -- ~ I

.I

~1 fJ I -~ I

I~ ----~--

ft -·- --- --

IGN I }. ~

~

1,;:::! N

~If ~

r- LUNAR----1 t---LU1 AR---1 NOON Nl( HT

t; 24.00 48.00 72.00 95.00 LGG.Qn 144.00

T I ME UiCT\.J'R.5Jl 168.00 192.00 216.00 240.C

FIGURE .24

BENDIX REROSPRCE SYSTEMS DIVISION

ATM-821 28 of 38

PROGRAM RLSEP TEST NO. S1 TEST DRTE 12/15/67 ZERO TIME 000000 g TEST TITLE PROTO A SYSTEM THERMAL VACUUM TEST SUMMARY ~~-------------------------------------------------------------------------------

0 0

~ 109 PAL 9 BOTTOM SlOE UNDER PAL 2 ~Ill PAL 11 BOTTOM SlOE UNDER PAL 4

~ 110 PAL 10 BOTTOM SlOE UNDER PAL 3 ~ 112 PAL 12 BOTTOM SlOE UNDER PAL 5 l

~+-------,--------,-------,-------,·-------·-T-------,--------r-------,--------,-------~

~~ l ____ _j_ __ ___jl___--J~~~=!!.F* ----~~~ f -1

----t----t----P---J-+-----· __ J I

i

I . . -+ -- ~--·- . LL.o -a

0 1--· N w

·--- ------ ------j

a: :::::J t-o a:,p a: I w<;:

I

+--------t----------+-1\1-----+--- --- . - r----- ... -------+-----+----t-- ·- --·-

I 0... .L ! :

w t-

0 I 0

0 f------ -+ --- ·--~------+------1------ -

1

fLUNAR_,j j II ,,OON I I --1--- --- ---- -- - :. - - - t--

0 0

0 -

I

I I l ~ I .

~ i I

1

I

I 'o. DO 24.00 48.00 72.00 96.0r:J 120.10 14'-!.00 168.00 1!'12.00 216.00 240.C

TIME (HOiJ,-'1::.,)

FIGURE 2S

BENDIX AEROSPACE SYSTEMS DIVISION 0 0

PROGRAM RLSEP TEST NO. S1 TEST DATE 12/15/67 ZERO TIME 000000 TEST TITLE PROTO R SYSTEM THERMAL VRCUUM TEST SUMMARY

~-.-------------------------------------------- ------------------------------

0 0

~ 113 PAL 13 BOTTOM SlOE UNDER PAL 6 ~ 115 PAL IS BOTTOM SlOE UNDER PAL 8

g+-------,---~---.------,--------.--

- 0 ~ oo ~ ;~+-------~-------+-----.B~~~e-~J~~~---

LL.o -o

~ 114 PAL 14 BOTTOM SlOE UNDER PAL 7

~ ~+-------t-------+--~~f,--'1-- 1------ . -t----t----t---- ~ --+ --+---- - -·-----'

~o ~ I '-"'- ·~ -,_ /

a:o f!. A ~o+--------t-----+---H.~-----+---------~~4•'----=:~~~~?j[-·------t--------j-------t--------

~-o ~ LUNAR J : I-= LUiJAR (I 0 NOON~ ! t---NICI-JT~ ' 0 i -;--t-------~-----JI:---t--------+-------1------- t----- ---- ---r--·----t------t------1

ATM-821 29 of 38

0 0

0 lfJ

0 0

lfJ N

0 - 0 0 ~0

0

"'-

0 0

FIGURE 26

BENDIX REROSPRCE SYSTEMS DIVISION PROGRRM RLSEP TEST NO. Sl TEST DRTE 12/IS/67 ZERO TIME 000000 TEST TITLE PROTO R SYSTEM THERMRL VRCUUM TEST SUMMRRY

[!! 88 RTG I HIH FRAME. F]N 5 (T) 89 RTG 3 HIH FRAME. FJN 7-fJN 8 I

<!I 90 RTG 4 Hl'll FRAME. FJN 5 <!> 91 RTG 5 Hl'll FRAME. FJN 2-FJN 3

I

J...\ _11 ·-a~-,=Y I .,-4:h Pl ,.-/"= -= \\_ --- -----

I I

I I I

-~--r--I I

--- ~---- -- -i---------j--------------

I i I I

"'"R~T~,t~ --- ·-~- - ----1----- '----·-

~ I

-<?"d I

NOON ,H~ T

lk. r- --. ------r --f ,. ----

,Jl -- -- ·----~~ ----- ' ·-- ----- ---

~ I I

I I I

"b. 00 24.00 48.00 72.00 95.00 120.00 144.00 !68.00 192.00 216.00 240 .r T; :-~>:: (1-if'J'mJJ

FIGURE 2.7

BENDIX AEROSPACE SYSTEM~ JIVISION PROGRAM RLSEP TEST NO. Sl TEST DATE 12/15/67 ZERO TIME OOOOOQ

0 0 TEST TITLE PROTO R SYSTEM THERMRL VACUUM TEST SUMMARY ~.--------------------- ------------------- --·

FJN I. Tl'lP SLI'll 93 RTG 9 FJN

0 0

[!!

<!I

0+------­~

- 0 oo

92 94

RTG 7 RTG 10 FJN 5. Tl'lP

r:-w

(T)

SLI'll <!> 95 RTG 11 FJN ~

~ 1 -- - ---

I I

1. Bl'l111'lM SLI'll

5. MJODLE SLIH

! I

----------1

';g+-----+--------+---+--+----+-1--11 --- ----+----T+-----+------

LLo

-o I : w a:

~-~------+------+-+---+----~------- -- _____ , ___ _

, i I

t---------: ------ f--- --1

I I

~g I : i ~~-~------+------+--!---+----+--------- -+1-1\.r--------+-------j------+------l

:w :+------+---l~~~~~---f--1--L-::_t_:_-1 __ ~ ;:: ~~_-_j - ------1-----1---------1

:~~~~--~~~-~~, ~--~~~ 'o.oo 24.00 48.00 n.oo 95.oo 120.00 144.00 168.oo 192.00 2t6.oo 240.0

TIME (HOURSJ

0 0

0 lf)

0 0 . 0 ~

..... 0 oo .--<0 :;j(('f')

LLo ~a

0 ('\J

w a:

.::J

t-o O:o a: . w~ Q_

FIGURE 28

BENDIX AER~SPACE SYSTEMS DIVISI~N PR~GRAM ALSEP TEST N~. S1 TEST DATE 12/15/67 ZER~ TIME 000000 TEST TITLE PR~T~ A SYSTEM THERMAL VACUUM TEST SUMMARY

[!] 96 R1G 13 ~U1ER SHELL FIN PER C9 97 RTG 14 ~UTER SHELL FIN 2-FIN 3

~ 98 R1G 15 ~UTER "'O[lC.T...L T ll'l 8. Lei .-lEA ~ 99 R1G 16 Pe!WER HERDER TEMP

~ "' "' m 5 l\ (

(r ~ ..... £ ~ ~

"' "' ~ ~ ~ ~

~ ... .&

~ ~ ~ ~ ~ ~

------

- - - 0 .L w t-

- - -- ' \.

0 0

0

0 0

0 .- L_

'o. oo

l~ f ~

--------

24. o-o-----~

~ I-LUNAR~ NOON ~LUN

NIG ~--4

l l 72.00- 96.00 120.00 144.00 168.00 192.00 216.00

TIME (H~URSJ

I

I

LU:t> 0:£ OJ ~())

1\) LV--'-())

240.(

: ; . . Summary of ALSEP Subpackage No. 2

Thermal Control Design

NO.

ATM-821

PAGE 31

REV. MO.

OF 38

. . roapace Systems Dlvllllon DATE 6 December 1969

The temperature results of the Proto A Thermal Vacuum Test for the integrated SNAP-27 RTG/Pallet 2 configuration were near nominal. Gener­ator steady state hot and cold frame maximum temperatures were 1190° F and 440° F for lunar day operation and 1145° F and 380° F for lunar night at an alectric fuel capsule setting of 1400 watts. Pallet 2 equilibrium surface temperatures ranged from 225 to 375° F during lunar noon to -35 to 180° F for lunar night.

Figures 21 through 28 illustrate the temperature distribution on Subpackage No. 2 for Proto A. The temperature of the Pallet 2 surface below the RTG ranged from a maximum of 375° F during lunar noon to 177° F during lunar night compared to predicted temperature limits of 394° F and 187° F, respectively. Lunar morning, noon, and night are indicated in Figure 22. The differences between pretest predictions and the Prototype test results are attributed to the lower RTG power setting of 1400 watts instead of the 1500 watt RTG dissipation used in the analysis.

Near the end of the lunar night, the RTG power was increased to 1500 watts. The average hot and cold frame temperatures reached 1109° F and 392° F, respectively. This increase in power caused an increase in temperature across Subpackage 2 and is shown in the temperature plots at the end of lunar night.

4 .• 4 Astronaut Glove Temperature During Traverse

Figures 29 and 30 present analytical temperature responses for the inner surface of the astronaut glove during traverse. The thermal proper­ties of the glove were derived by interpolation of test data presented in the reference. This reference is an MSC report on tests to determine the temperature rise of the inner surfaces of the astronaut suit and glove. Tests were conducted with the outer glove surface in contact with various constant temperature sources.

0 No data were presented for source temperatures below 250 F. The data were presented in the form of inner surface temperature vs. time, for constant source temperatures. The slopes of these curves are then temperature rise per unit time.

The slope of the time temperature curve was assumed zero at 70° F. The temperatures shown in Figura 29 are the result of logarithmic inter­polation of the slopes from test data. Figure 30 presents the glove inner surface temperatures with linear interpolation of the slopes from the test data. The source temperatures used for both cases were the transient tem­peratures of the pallet surface under the RTG for Flight (two man) config­uration. The source temperature will be somewhat lower due to the thermal resistance between the pallet and the astronaut contact area on the carry bar (see Figure 31). Since linear interpolation of the slopes is worst case this will be assumed pending information as to the thermal properties of the glove at temperatures other than those presented in the reference.

140

130

P-i 0

fil 120 ~

~ < Jt: fil 110 ~ ~ fil f-i

fil > 100 0 ....:l a p::; fil z 90 z 1-1

80

70

0 10

TRANSIENT DEPLOYMENT RESPONSE INNER SURFACE OF ASTRONAUT GLOVE THERMAL RESISTANCE OF GLOVE OBTAINED BY LOGARITHMIC INTERPOLATION OF MSC TEST DATA

20 30 40 50 60 70 80 90 100

ELAPSED TIME FROM RTG FUELING - MINUTES

FIGURE 29

w:x> [\):£ o I ~co

[\) VJ-" co

140

130

0~ 120

~-

~ ::J ~

-< ~ 110 ~ ~ ~ ~ ~

~ 100

:> 0 ~ ()

~ ~ 90 z z H

80

70

0 10

TRANSIENT DEPLOYMENT RESPONSE INNER SURFACE OF ASTRONAUT GLOVE THERMAL RESISTANCE OF GLOVE OBTAINED BY LINEAR INTERPOLATION OF MSC TEST DATA

20 30 40 50 60 70 80

ELAPSED TIME FROM RTG FUELING - MINUTES

90 100

FIGURE 30

LU::t> LU~ 0 I ~(X)

1\) lAJ _,. (X)

~~~"""'"" rr~~~.; : II ,- I

II .· '·.·.1. II : I II r... I ,, ' ....... I ,- ...

,' .. r I _....J

MAST/CARRY BAR

BOTIOM SURFACE OF SUBPACKAGE l

BOTIOM

EDGE

SURFACE OF

TAPER EXAGGERATED

PALLET

HANDLE

FIGURE 31

LUlt> ~~

~fu 1\)

LU--" ():)

: : I ~

Summary of ALSEP Subpackage No. 2 Thermal Control Design

MO.

ATM-821

35 PAGE

RI!V. MO.

OF

1'08pace " ~&tame Division DATE 6 December 1969

The reference defines "pain" as an inner glove temperature of 115°F. Therefore, maximum contact time is defined herein as the time required for the inner surface to reach a temperature of 115° F when initially at 70° F.

Table 4 presents predicted safe contact times for various contact temperatures.

Inner glove data is from the reference report.

5.0 Summary and Conclusions

Steady state deployed temperatures for Subpackage No. 2 do not exceed maximum RTG performance temperatures. Ollter shell temperature range is 430-490° F with no conductive coupling to the lunar surface and 415 - 485° F with conductive coupling. Conductive heat leak into the pallet from the RTG was 60 to 80 watts. Pallet temperature changes have little effect on RTG temperatures.

An elapsed time of 30 minutes from RTG fueling to the last contact with Subpackage No. 2, or parts mounted on Subpackage No. 2 during traverse, appears to be the safe deployment period at this time. This number contains many safety factors, and could possibly be lengthened after further analysis and testing. However, it is approximately double the present deployment period, including a four· minute rest during traverse. Therefore, this number and the present state of analysis is considered sufficient for the present.

The maximum contact times for the FTT are a minimum of 5 times the anticipated FTT contact time (approximately one minute). All tools and equipment (except UHT and cable reel assembly) mounted on Subpackage No. 2 during traverse will be below 200° F at 30 minutes after RTG fueling. All tool contact times are less than one minute, as is the removal time for the sub-pallet and deployment of the cable. Therefore, no thermal hazard is seen during the use of any tools or removal of equipment.

Astronaut glove maximum contact times are summarized in Tables 4 and 5. Contact of the astronaut's glove with surfaces at or under 250° F appears safe for 18 minutes maximum, but the maximum contact time drops sharply for sur­faces above 250° F. The maximum contact time for a 400° F source is less than 2 minutes and is under one minute for sources above 600° F. Contact with a 300° F source is safe for approximately 4 minutes. These times could be used for the present, but more data on the glove is needed if a more exact analysis is desired.

In conclusion, the thermal control design of Subpackage No. 2 and astronaut tools is more than adequate to ensure against thermal hazards to the astronaut during RTG fueling and deployment. This, of course, assumes no contact with the capsule, GLFC, or RTG and is based on existing Task Sequence times.

Su~ry of ALSEP Subpackage No. 2 Thermal Control Design

TABLE 4

Maximum Astronaut Contact Times

Pallet Assembly No. 2

Initial Contact in Minutes from RTG fueling

Maximum Contact Time Minutes

Cable Reel Assembly and UHT

Initial Contact in Minutes from RTG fueling

Maximum Contact Time for Spool

Maximum Contact Time for Mounting Struts

Fuel Transfer Tool

0

34

15

43

39

15 30

17

30

25 19

20 7

8

60

13

4

60

5

Initial Contact with FTT when FTT Contacts Fuel Capsule End Plate

Surface FTT Maximum Contact Times -- Minutes

Disc (~H = 0.25) 10

Fingers 15

Handle End 20

MO.

ATM-821

PAGI!

REV. MO.

38 OF

DATE 6 December 1969

75

4

/ ,/'""-"'~-

"08pace

Summary of ALSEP Subpackage No. 2 Thermal Control Design

NO.

AT.M-821

PAGI 37

RIV. NO.

OF

DATI 6 December 1969

TABLE 5

Inner Glove Safe Contact Time

Maximum Contact Time

Linear interpolation of above slopes (zero slope at 70° F) resulted in the following:

Slope, Maximum Contact Time Source Temperature, OF ~/Min Minutes Seconds

150 1 . 11 40.60 2240 200 1.81 24.90 1494 250 2.50 18.00 1080 300 10.23 4.40 264 350 18.00 2.50 150 400 25.70 1. 75 105 450 28.52 1.58 95 500 31.35 1.44 86 550 34.17 1. 32 79 600 36.00 1. 25 75 650 45.00 . 99 59 700 55.00 .82 49

I ' \ t}.

: : I ~

( Summary of ALSEP Subpackage No. 2

Thermal Control Design

NO.

ATM-821

PAGE

RIV. NO.

OF

DATE 6 December 1969

REFERENCE

T. J. Ballentine, MSC Crew Systems Division, "Heat Transfer Measurement on Extravehicular Mobility Unit Composites at Various Temperature Levels," 28 November 1967.