engineering and development division

. ·-~' .I .. ;

' "

AGN

r

Engineering and Development Division

AEROJET-GENERAL NUCLEONICS INDUSTRIAL REACTOR (AGNIR)

REACTOR PHYSICS TESTS

AN-1527 September 1966 r----------------·- --------- - ~

- NOTICE - E COPY THE ATTACHED FILES ARE OFFICIAL RECORDS OF THE DIVISION OF DOCUMENT CONTROL. THEY HAVE BEEN CHARGED TO YOU FOR A LIMITED TIME PERIOD AND MUST BE RETURNED TO THE RECORDS FACILITY BRANCH . 016. PLEASE DO NOT SEND DOCUMENTS CHARGED OUT THROUGH THE MAIL. REMOVAL OF ANY PAGE(S) FROM DOCUMENT FOR REPRODUCTION MUST BE REFERRED TO FILE PERSONNEL.

JlEGU.lATORY DOCKET TILE COPI DEADLINE RETURN DATE

RECORDS FACILITY BRANCH NUCLEON I CS

IFORNIA

l-

AGN

ENGINEERING AND DEVELOPMENT DIVI.§.!QN



By

R. L. Tomlinson

AN-1527 August 1966

t ft N • : • • '· ·\ • , .._ ,"" _.., • • ..... ' •• ,,.; • ..~ ~ ; { ;· \a,( { ' " • •\ ,., ·•..\ ••• ' • ' .. i. ,

Approved by:~

R. H. Chesworth, Manager Engineering and Development Division

AEROJET 0 GENERAL NUCLEONICS A DIVISION OF AEROJET-GENERAL CORPORATION

iii

[. I

AN-152 7

. .


REACTOR PHYSICS TESTS*

by

R. L. Tomlinson·

ABSTRACT

_The Aeroj et-General Nucleonics Industr_ial Reactor (AGNIR)

achieved initial criticality on 9 July 1965. Following ini.tial

criticality, a series c:i.f neutronics and power calibration tests . .

were perfonned to characterize the reactor from both the physics

and thermodynamics sta.ndpoints; A description of. th.e conduct a.nd ·····. •' . . . .. -·--·

significant results of this test program is ·presented herein,

. . *P.ublished by Aerojet-General .Nucleonics, San Ramo~,. Calif.

V

~-...

.;.

AN-1527

CONTENTS

I. INTRODUCTION

II. SUMMARY AND CONCLUSIONS

III. CONTROL AND SAFETY ROD SCRAM TESTS

IV. FUEL LOADING

v. VI.

A. INITIAL FUEL LOADING

B. SECOND FUEL LOADING

ISOTHERMAL TEMPERATURE COEFFICIENT

REACTIVITY MEASUREMENTS

A.

B.

c. D.

E.

F.

G.

H.

TECHNIQUES EMPLOYED

CONTROL AND SAFETY ROD CALIBRATIONS

FUEL AND REFLECTOR ELEMENT REACTIVITY MEASUREMENTS

GLORY HOLE MFASUREMENTS

DUMMY ELEMENT IRRADIATION CAPSULE MF.ASUREMENTS

FLUX WIRE HOLDER MEASUREMENTS

THERMAL COLUMN MEASUREMENTS

LARGE COMPONENT IRRADIATION BOX

I. XENON POISON EFFECTS

VII. POWER CALIBRATIONS

VIII. POWER COEFFICIENT MEASUREMENT

IX. NEUTRON FLUX TRAVERSES

REFERENCES

vii

Page

1

2

5

5

5

8

8

9

9

9

14

14

17

18

18

20

20

20

24

27

31

Figure Number

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17'

18

19

20

Title

AGNIR Installation

AGNIR Core

AN-1527

FIGURES

AGNIR Core Loading Patterns

AGNIR Initial Approach to Criticality

AGNIR Isothermal Temperature Coefficient

AGNIR In-Hour Equation

AGNIR Shim Rod Reactivity Calibration

AGNIR Regulating Rod Reactivity Calibration

AGNIR Core Component Reactivity Worth as a Function of Core Position

Fuel Element Worth Versus Water in the AGNIR .Core

AGNIR Dummy Element Irradiation Capsule

Dry Irradiation Tube on Thermal Column

Large Component Irradiation Box

Xenon Poison Effects in AGNIR Versus Time

AGNIR Power Coefficient of Reactivity

AGNIR Pool Water Heating and Cooling Data (Without Heat Exchanger)

AGNIR Calorimetric Power Calibration

Axial Thermal Neutron Distribution in the AGNIR Core at a Power Level of 230 Watts

Radial Thermal Neutron Distribution in the AGNIR Core (230 Watts)

Radial Thermal Neutron Distribution in AGNIR Core and Thermal Column

viii

Page

3

4

6

7

10

11

12

13

15

16

17

19

21

22

23

25

26

28

29

30

AN-152 7



I. INTRODUCTION

The Aerojet-General Nucleonics Industrial Reactor (AGNIR) achieved i.n

itial criticality at San Ramon, California, on 9 July 1965. It was the

twentieth reactor to be built and operated at the San Ramon site. The previous

nineteen were AGN-201 and -211 research and training reactors which were sold

commercially to research and educational institutions in the U.S. and Europe.

One 20 watt(t) AGN-201 reactor was in use at San Ramon for over eight years

until the AGNIR became available for company research activities.

The AGNIR is a 250 kw(t) pool-type reactor; it is fueled with uranium

zirconium hydride; and is water-moderated and water-cooled. The reactor is

licensed for general purpose neutron irradiations and isotope production. The

open-core design and the 10-ft-diameter, 23-ft-deep water pool was designed to

simplify the installation of special purpose irradiation loops, Both wet and

dry irradiation facilities are provided within the reactor in addition to

special laboratory space for the setup of electronic gear adjacent to the

reactor. In-pool storage is provided for 21 irradiated fuel or dummy element

irradiation capsules.

The reactor is operated from a control console which permits the oper·

a tor to fully view all operations performed at the top of the reactor po~il..

The facility is served by a 3-ton bridge crane; a mechanically positioned,

large component irradiation box can be actuated from the top of the reactor

pool.

1

AN-1527

The reactor core consists of zirconium-hydri_de fuel moderator elements

surrounded by graphite-filled reflector elements. The inherent safety of

this core design simplifies the procedure for obtaining licenses for experi

ments.

Reactor control is maintained by three boron carbide control rods. In

core irradiation facilities include a seven-element central exposure capa

bility; two 3-element exposure facilities; a glory hole; and nrultipurpose

dummy element irradiation capsules.

The facility consists of a high-bay metal-framed building 40 by 80 ft,

the low-bay portion of which is occupied by a general purpose laboratory; a

control room, and a change room. Six shielded pits are provided in the

facility for the storage of radioactive components, and a hot cell,which is

'licensed for 500 curie of Co-601 is also located with the reactor building. Non-

radioactive storage is provided above the low-bay area within access of the

three-ton bridge crane.

The initial physics tests on the reactor are reported herein. A cut

away drawing of the AGNIR installation is shown in Figure l; and a drawing of

the AGNIR core is shown in Figure 2. The facility was described earlier

(Ref. 1) as were the procedures used in performing the above-mentioned physics

tests (Ref. 2),

II. SUMMARY AND CONCLUSIONS

The initial criticality for the AGNIR was achieved with 63 altnninum-clad

TRIGA Mark I fuel elements. These fuel elements contained a total of 2265 gm

of U-235 in the form of uranium-zirconium hydrid~. The isothermal temperature

coefficient for the system was found to have an average value between 60 and

125°F of -0.15~/°F. All in-core void measurements indicated negative effects.

The power coefficient was measured to be -0.47¢/kw, resulting in a $1.17 in

itial reactivity deficit at 250 kw in addition to xenon, samarium, and fuel

burnup effects. The total worth of the control and safety rod system was

measured to be -$8.51. No data were obtained during these tests that measur

ably differ: from that presented in the AGNIR Hazards Summary Report (Ref. 1).

2

LARGE COMPONENT IRRADIATION BOX

FIGURE 1. AGNIR INSTALLATION

3

7 ELEMENT EXPOSURE FACILITY

3 ELEMENT EXPOSURE FACILITY

FIGURE 2. AGNIR CORE

4

DUMMY ELEMENT IRRADIATION SPACE

CONTROL INSTRUMENT CHAMBERS

AN-1527

III. CONTROL AND SAFETY ROD SCRAM TESTS

The experimental tests performed prior to the initial reactor critical

ity, and the four measurements subsequently performed as part of the quarterly

maintenance checks, indicate that the control rod drop times fall well within

the Technical Specifications of the reactor license. The Technical S.pecifi

cations state that the total rod drop time, including magnet separation time,

shall not exceed 600 milliseconds. For the three .control/safety rods the

magnet separation varied from 50 to 60 msec, while the total drop time varied

from 410 to 430 msec, A special relay rack panel was installed in the con

trol room to facilitate the easy measurement of the rod drop time with the aid

of a sweep oscilloscope and a series of microswitches. The panel also has

provisions for controlling a BF3

pulse counting assembly that was used for con

trol rod calibrations using the rod-drop technique, described in Section VI.

IV, FUEL LOADING

A. INITIAL FUEL LOADING

The initial fuel loading followed the reference procedures (Ref.2);

the fuel load consisted of 63 aluminum-clad TRIGA Mark I fuel elements and 23

graphite-filled reflector elements. The loading was purposely skewed toward

the control instrumentation to provide the maximum signal to the reactor in

strumentation during the initial critical experiment. The initial loading

configuration is shown in Figure 3A.

The nuclear instrumentation used during the initial critical ex

periment consisted of the normal four channels of reactor control instrumenta

tion; i.e., one BF3

pulse channel; one gamma-compensated ion chamber inter

mediate channel; and two uncompensated ion chambers used as power channels,

With no appreciable gamma background on the fuel, all four channels were on

scale providing useful infonnation. In addition, two additional BF3 pulse

channels and one uncompensated ion chamber were used during the critical ex

periment for a total of seven usable channels of nuclear instrumentation. A

plot of the multiplication versus fuel mass for the initial fuel loading

(Figure 4) reveals that the ion .chamber data proved more reliable than the

pulse counter data for this initial fuel loading. Cr.iticality was achieved

for the configuration within 35 grams of the value found in the initial

criticality calculation.

5

I 4u-r,,~-114; I

AGNIR CORE LOADING FOR CRITICAL+ 20,7~ EXCESS COLD CLEAN (SKEWED LOADING PATIERN) .

FIGURE 3A.

~ CONTROL RODS

SOURCE

FUEL ELEMENTS

9 GRAPHITE ELEMENTS

® GLORY HOLE

AGNIR CORE LOADING FOR $1.80 EXCESS COLD CLEAN (SYMMETRIC LOADING PATIERN)

FIGURE 3C.

FIGURE 3. AGNIR CORE LOADING PATTERNS

AGNIR CORE LOADING FOR S2.70 EXCESS COLD CLEAN (SKEWED LOADING PATTERN)

FIGURE 38.

141.,-'-~-,14" I

o.

o •

. ... ·z 0. 0~ 1-

6 _. CL

I_. ::>

-~ . _.

<( u . o.

o.

··. ·8: . ' Cl.' .-u·

0.1

·O

A, F, G: PROPORTIONAL COUNTER . C, D, E: UNCOMPENSATED ION CHAMBER

B : COMPENSATED ION CHAMBER

2.2rg

0.2 0.4 0.6 0.8 l.O 1. 2 1.4 1.6 1.8 2~:o 2.2 2.4

FUEL MASS, u235 kg·

.FIGURE 4. AGNIR INITIAL APPROACH TO CRITICALITY

, ' .. ,: .·.\;~ •: ----

AN-152 7

The fuel loading proceeded until a total of 69 fuel elements and

23 graphite elements were loaded into the AGNIR core for a total cold, clean

excess of $2. 70 (Figure 3B), The initial control rod calibrations were per

fonned during this loading. The final fuel loading for this configuration

is shown in Figure 3B. Preliminary coEe component reactivity measurements

were made with this configuration.

B. SECOND FUEL LOADING

On completion of the preliminary physics tests the AGNIR core

loading was adjusted to a nearly symmetric pattern with the "glory hole," a

dry exposure tube, located at the geometric center of the reactor (Figure 3C),

With 71 TRIGA Mark I fuel element.s and 23 graphite elements, the reactor had

a cold, clean excess reactivity of $1.80. The control rod calibrations, core

component reactivity measurements, neutron flux traverses, and power calibra

tions were performed with this basic core configuration.

V. ISOTHERMAL TEMPERATURE COEFFICIENT

The water-filled pool, which serves as~ radiation shield and coolant

reservoir for the AGNIR, contains approximately 13,000 gallons of water, The

water tank was used as a low-grade calorimeter for two thermal measurements:

1) power calibration of the reactor (see Section VII), and 2) the isothermal

temperature coefficient and cooling characteristics of the reactor pool tank.

Nineteen 220-volt immersion heaters, with a total measured power rating

of 23.9 ±. 0.1 kw, were inserted into the reactor grid using the fuel element

positions, while water was continuously circulated through a purification loop

at the rate of R:i 6 gpm. While maintaining the reactor. critical, the water tem

perature was monitored at five locations within the reactor pool tank. During

the tests, the water temperature was varied from 69°· to 125°F. The control

rods were calibrated, using both period and rod-drop techniques (see Section

VI), prior to measuring the temperature coefficient. As the water temperature

of the pool tank is changed, the actual position of the poison section of the

control and safety rods differs from the control/safety rod position indica

tors, because of thermal expansion of the rods. The discrepancy is appreci

able since the submerged portion of the control/safety rods is 20.5 ± 0.5 ft

during normal reactor operation, the uncertainty in length being due to the

8

AN-152 7

allowable l ft variation in the water level within the reactor pool tank, The

average value of the isothermal temperature co.efficient from 60° to 130°F is

-0.15¢/°F (Figure 5), The insertion of controi .rod poison as a: function of

bulk water temperature due to the linear expansion o.f the aluminum hanger rods

accounts for the entire coefficient within. the acci:lr~cy of the experimental ·

measurements.

VI. REACTIVITY MEASUREMENTS

A. TECHNIQUES EMPLOYED

A series of reactivity measurements we~e pe'rformed,. using positive

period and rod-drop techniques, The circuitry Ufled in performing these· . \ ' ' .

measurements is similar to that used ai: ~ther· :reacto~. i~stallauons; however,

an AGN-developed neutronics code* was used in r~ducing the data front the rod

drop measurements.

· · The range of positive periods meastir~d; using thifl technique, ,• . •,.

varied between 100 and 10 seconds, which. corresporid~ ,.tci :excess reacti vitie~

from 10 .to 40 cents. The reference pro,cedures (R~f'. ·2.)··were' used during the

measurements. The In-Hour Equation used in the period measurements is plotted

in Figure. 6.

B. CONTROL AND SAFETY ROD CALIBRATIONS·

The la~ge reactivity worth of the AGNIR control ~nd safety rod

system ($8. 51 total) does not allow complete posit·i~e p~i;-iod calibration of ·

the ·sh~ and safety rods (appro~imately $3.80 and $3.75~· respectively), due to

the $3.00 license limitation and the safety. inte~lock o~ ~he safety rod. The ' . . . . . . . .

. safety rod is maintained in the "full "'-out" position ,du~ing ·reactor operation;

therefore, total worth measurements on all rods w~re 'obtained~. using rod;.drop

techniques. Intermediate points on the shi,;n and regulating rods were also ob-. ' ' ' .. ' ' . : . -

tained with rod-drop techniques and were ,found to ,be in: good ·agreement w:i. th· . .,. ' . ' . ,•

the period measurements. The calibr~tion, curves· obtai~ed u~in·g these tech

niques are presented in Figures 7 and 8 ..

In all reactivity measurements, .the accurat'e·determination of

criticality is very important. To assist in this' determinati~n, an expanded

scale was placed on the Channel 4 line~r power reco~d~r.: Using this recorder, ·*Internal Communication: T.P. Wilcox, DROP -· An IBM·Code to Solve for Reactor Power L~vels· After a Step Change in System ReactivHyJ AN.-COMP-134.

9

Or----,------.------~--,-,--..-----.------.-----....---

-1

-3

-4

l! C: Q)

-5 u .. w C) Z. <( J: -6 u >-!:::: ;:: ti :-1 <( w 0::

)

-8

-9

-10

-11

-12 ...... ------.................................. _._ ........ .,..-__ ...,... __ .............. ,,___....,_ _______ ___. 60 70 80 • ·· ·• · 90 • · 1 do . · · 11 o • 120 130 140

\ PO()[ Wfa;TER fEMPEAA~UR~: °F .

FIGURE 5.•. AGNIR iS()lHERMAL:TEMPERATURE COEFFICIENT ",· -· , . ·. ·' . . .

10 .

I 4r.H

olo-

u 48

I

STAB

LE P

OS

ITIV

E R

EA

CTO

R P

ER

IOD

, se

cond

s

0

.,, N

G

) 0

C

;::o m °' .

;::o

)>

m~

G

) )>

o

()

z -f

;::o

<

z .=

i -< I

... I

0 °'

0 (!

) 0

C

:::; ....

;::o

"' m

0 C

)>

(X

) -f

0

0 z

0 0

(.

: 4od.-----.----,,------,------,,--r--------,----'-----r-------,.......,i------.----, . I

1380 I :

. ,

/ 360

I

I 340. I

I

i 320'

\

300 I

280

: 26d

i ! 240 u' .

~ .. ! 220 >! 6 !200

~: ~ '180

I

\ 160.

! 140 I

I 120

100

80

60

40

20

-- "'!'"..._ ....

', ' ' .,·.

. ' \ \

\

'

\ ·\.·

\ .

\

' ' ' ' I Q....._ __ ......., ___ _....._.._ __ ....._ ___ _.__....._..,...,..,___,........-_ _,...._. __ .._..._...__ _ __,

'· ·o 100 : 200 300 400 · . soo 600 700 800 900

· ROD POSITION.

FIGURE 7. AGNiR SHIM ROD REACTIVITY' CALIBRATION

12

lOOr------r----'T"----r----r----,.----.------.----.

90

. 80

70

60

"' .... C Q) u ..

>- 50 !:::: > ti <( ·w co:: 40

30

20

10

100 200 300 400 500 600 700 800

ROD POSITION

FIGURE 8. AGNIR REGULATING ROD REACTIViTY CALIBRATION

13

AN-152 7

with its full-range zero suppression and 100-to-l amplification, minute drifts

in power can be readily detected and exact criticality accurately determined.

C. FUEL AND REFLECTOR ELEMENT REACTIVITY MEASUREMENTS

The reactivity worth of the AGNIR fuel elements, dry glory hole,

and dummy element irradiation capsule were evaluated as a function of ring

position, starting at the geometric center of the reactor grid plate. The

fuel measurements and dry glory hole measurements both were made with refer

ence to water, The actual worth measurements are plotted as a function of

ring position in Figure 9.

It can be seen that a fuel element in the center of the loading,

ring A, is worth less than that in rings Band C, which are almost identical

in worth. Since the AGNIR grid structure is basically a TRIGA grid with a

few minor modifications, this effect is not what might be expected, LP.riori;,.

however, in most TRIGA reactors the central core position is occupied with a

pulse rod or a norunovable glory hole and are therefore not available for the

placement of fuel, The TRIGA grid is undermoderated in the center and over

moderated at the edge. Starting at the center of the core, the hydr~gen/

uranium (235) ratio increases approximately linearly with radius. Multigroup

neutron transport calculations (Ref. 3, 4) performed on the actual AGNIR core

loading indicate a definite depression of thermal neutron flux in the center

of the core with a fuel element in the central location. With the central

fuel position flooded with water, the standard loading pattern for TRIGA

reactors, a normal thermal neutron distribution across the core is obtained

(Figure 10), Therefore, the shape of the fuel element reactivity curves shown

in Figures 9 and 10 are wqat would be expected from an analytical basis for

the AGNIR with and without a fuel element in the central position.

Graphite reflector elements were evaluated in the F and G rings

of the AGNIR core, The average value for these elements were found to. be 9

cents for the F ring, and 4 cents and for the G ring.

D. GLORY HOLE MEASUREMENTS

A special dry glory hole is available for various radiation ex

periments, This glory hole is equipped with an internal shield plug that is

used to reduce the radiation streaming in the vicinity of the control rod

14

"' "E Q) u ..

>-!:::: ~ t:; <( w 0,,:

:·-.\:)

60

40

30

20

lO

/ FUEL VERSOS WA m

GLORY HOLE VERSUS WATER

' ' ' ' . . ·.· ... ',' /· CADMIUM IN CORE

. ' ' . . . . . . . ~ . . . ' .

'

---OWio-----..1;_ ____ ..J_ ____ ...1.-""""."""---'--'~--L---"""".""".-L---""""."""""""."""---'

A B C D E F .G

... CORE COMPONENT LOCATION_ BY RING·

FIGURE 9. AGNIR CORE COMPONENT REACTIVITY WORTH AS A FUNCTION OF CORE POSITION I • ' • '

15

VI ,._ C (I)

u .. >-!:: > t-u <( LI.J ~

~ ....... .......

I ~ '9

I ..... -: "l::j,,

180

160

140

120

100

60

40

20

0 A B

----

G WATER IN POSITION A-1

0 FUEL IN POSITION A-1

C D E

CORE RING POSITION

FIGURE ·10· •. FUEL ELEMENTWORTH VERSUS WATER IN THE AGNIR CORE 16

F

AN-1527

drives at the top of the AGNIR pool. The glory hole can be positioned in

selected locations in each of the seven rings of the reactor core. The

reactivity of water versus the voided glory hole for these locations are

shown in Figure 9.

E. DUMMY ELEMENT IRRADIATION CAPSULE MEASUREMENTS

To facilitate special short irradiations of a general nature, an

irradiation capsule was designed that could be adapted for a variety of

irradiations (Figure 11), It can be located in most fuel element positions

and has provisions for bringing instrumented tubes to the surface, A shielded

transfer cask is available for transporting the capsule within the AGNIR

building to the hot cell area for remote disassembly. The worth of the dummy

irradiation capsule versus water in the reactor core was found to be identical

in reactivity with the glory hole within the experimental error of the measure

ments.

SEAL

IRRADIATION VOLUME

FIGURE 11. AGNIR DUMMY ELEMENT IRRADIATION CAPSULE

17

AN-1527

F , FLUX WIRE HOLDER MEASUREMENTS

The AGNIR core was designed t o allow flux traverses to be readi ly

per f ormed. A total of 22 holes, 0.313 in. in diameter, penetrate the up per

and lower reactor grid plates to allow the use of flux measuring wires, or

0 . 25 - i n.-diameter (or smaller) neutron-sensitive chambers to be placed in the

reactor core. Special aluminum holders for flux measuring wires have bee n

fabr ica ted for use in these holes. The reactivity of these holders, with

respect to water, is so low as not t o be a consideration in any flux mea sur e

ments ; however, special cadmium tubing used with many flux wire measuremen t s

(0 .050-in. ID by 0.090-in. OD) has an appreciable effect on the reactivi t y o f

the AGNIR. Cadm i um tubes, with a l e ngth of 26 in. and a weight of 6.63 gm,

were inserted in t he flux wire hold ers and their r eactivity worth deter mi ned .

These da ta ar e plotted in Figure 9 as a function of core position.

G , THERMAL COLUMN MEASURE11ENT S

The AGNIR thermal column consi s ts of a large block of graphite

containing five rows of 1.5-in. diameter holes arranged at increasing radii

from t he core. The rows are placed 6 in. apart, and each row contains seven

irrad i ation positions (Figure 12). Flux wire holder positions are located

near the centerline of each row to facilitate performing neutron flux traverses

of the thermal column assembly. Four slotted beams, two on each side, are pro

vided to allow experiments to be attached directly to the thermal column. Ex

tensions of these beams allow experiments to be placed inunediately adjacent to

t he reactor core. The assembly is located adjacent to the reactor core on

ta pered pins and remotely bolted to the bottom of the reactor pool tank. In

stallation and removal of the whole assembly is accomplished with the facility

crane and remote handling tools on a routine basis. Figure 12 also shows a

6-in. dry irradiation tube in one of the rear positions of the thermal column.

When the thermal column was installed, the worth of the column was

measured to be less than one cent positive with respect to water, due to the

2-in. gap between the reactor core structure and the thermal column which ef

fectively separates the reactivity effects of the thermal column from that of

the reactor core. Neutron flux traverses performed in the thermal column are

described in Section IX.

18

FIGURE 12. DRY IRRADIATION TUBE ON THERMAL COLUMN

19

AN-1527

H. LARGE COMPONENT IRRADIATION BOX

The large component irradiation box (Figure 13) consists of an

aluminum box with an internal volume of 8 cu ft. The walls of the box are

r e latively thin to eliminate excessive parasitic neutron absorption. The

box is pressurized with CO2 to 0.5 psi above the water pressure with the aid

of a relief valve attached to the top of the box. The CO2 is supplied

through aluminum and plastic tubing from a supply at the top of the reactor

poo l . Another tube is available for bringing electrical leads to the top of

t be pool if required for any experiment. The box is weighted with lead to

el iminate buoyancy. The box is remotely installed and bolted to a movable

t a ble at the bottom of the AGNIR pool. Similarly, the movable table is re

mot e ly positioned on tapered locating pins and bolted to the bottom of the

AGNIR pool.

When the void box was installed, a reactivity loss of 9 cents

was measured for the voided box with respect to water. The box is designed

to handle the irradiation of components and subsystems up to 2 ft in diameter.

I. XENON POISON EFFECTS

Since AGNIR operates at an average thermal neutron flux of about 12 2 13 2

4 x 10 n/cm -sec and peaks at about 10 n/cm -sec at the center of the

r eactor core, the effects of xenon poisoning on the operation are appreciable.

A test was run on the cold, clean reactor core to determine more specifically

the magnitude of these effects. The reactor was held at a constant power level

f 250 kw d 1 f 8S°F f 50 h d . h o an a constant poo~ temperature o or ours uring t e test.

Using the control rod calibrations, the effect of xenon poisoning as a function

of operating time was determined (Figure 14). Similarly, the poison worth of

xenon was measured following shutdown by making criticality determinations at

low power level and correcting for the power coefficient (Figure 15) and iso

thermal temperature coefficient effects (Figure 5). The results of these data

are also plotted in Figure 14.

VII. POWER CALIBRATIONS

An accurate method of reactor power level determination for pool-type

reactors is electrical heat substitution (Ref. S). The reactor pool tank

(containing 13,000 gallons of water) was determined to be a fair calorimeter

between 70 and 95°F. As previously discussed in Section rv, nineteen 220-volt

immersion heaters (with a total measured power rating of 23.9 ±. O.l_kw) were

2.0

FIGURE 13. LARGE COMPONENT IRRADIATION BOX

21

I 4!.l-b'--IIS2 I

150

140

130

120

110

"' 'i: 100 Q) 0

' 90 :I: I-~

0 ?;

80

r..) >- 70 "' !:: ;;::: ti 60 < w ~ 50

40

30

20

10

0 0 4 8 12 16 20 24 28

TIME, hours

XENON BUILDUP FOLLOWING COLD, CLEAN STARTUP

XENON DECAY FOLLOWING SHUTUP FROM EQUILIBRIUM OPERATION

32 36

FIGURE 14. XENO N POISON EFFECTS IN AGNIR VERSUS TIME

40 44 48

!-,

~ C Q) u

' V) V)

0 ..J

>!::: :2:: t; <( w 00:

120..-----.-----.,---'--..------..-----,---,.--.,.,...----,.----,-----,

40

30

20

10

25 50 75 100 125 150 175

. REACTOR POWER LEVEL, kw

FIGURE 15. AGNIR POWER COEFFICIENT OF REACTIVITY

23

200 250

AN-1527

inserted into the reactor grid, using the fuel element positions, while the

water was continuously circulated through a purification loop at~ 6 gpm.

Sirtce the inlet to the purification loop is near the top of the reactor pool

tank and the discharge nea_r the bottom, this flow slowly .stir s the pool

water, thereby reducing temperature stratification within the tank. The

water temperature was monitored at five locations within the reactor pool

tank. Plots of the heating and cooling characteristics of the pool water tank

are shown in Figures 16 and 17. Once the 23.9 kw electrical heating curve was

obtained, the reactor was flux-mapped at low power* and the corresponding read

ings of the Channels 3 and 4 ion chambers were made. The flux traverses were

made at an estimated power level _of 160 w, based on thermal flux integration

techniques. Using this estimated power as a ba·sois, a scale factor was applied

to the ion chamber readings corresponding to a power level of 23.9 kw. A 24-

hr nuclear heating run was performed at these ion chamber readings. The actual

pow~r level was found to be 34.32 ± 0.93 kw, based on the previously performed

electrical heating data; therefore, the initial flux mapping was performed at

230 watts instead of the estimated 160 watts (Figure 17).

Subsequent nuclear heating runs were performed at 200 kw and 250 kw

(Figure 16). Due to the increased heating rate over the initial calibration

runs, the heating curves at these power levels are practically linear and do

not show the effects of water evaporation that occurs at the lower heating 0 rates. At pool temperatures of 70 F, the water evaporation was found to be

about 0.5 gallons/hr; while at 125°F, the rate increased to 5 gallons/hr. The

cooling curve in Figure 16 clearly shows the operating limitations of the

AGNIR without its 250-kw heat exchanger.

VIII ~WER COEFFICIENT MEASUREMENT

The power coefficient (i.e., the reactivity loss as a function of

reactor power) was measured at constant water temperature without xenon in

the core. Using the control rod calibration curves, the reactor power was

increased in 25-kw steps above delayed critical and the reactivity loss de

termined. A plot of the data is shown in Figure 15. The measured average

reactivity coefficient was found to be approximately -0.47¢/kw, for a total

.._g_activity loss at 250 kw of $1.17. *Internal Communication: V .R. Forgue, Gold Wire Flux Mapping of AGNIR, AGN

Chem Tech Memo No, 887, October. 1965.

24

I'-.) 0,

14-t.(-~~-1212'

u. 0 .. w ~ :::, I-<( ~ w a.. ~ w I-~ w I-<(

3: _J

0 0 a..

250kw ~.:_:_~...:..:..:.........c:..:.:.:.c.:c..~~=-c:~-+-:-,-c-,-±:-;-=-:-:-ci-::-::=:c~-:-:-:::-:t-:-,-:-:--;t-:----:-c:--:--:-r::-:-:-=-:-:=:cc:-=l=:,-.77'/:-:=:-:-:-:-t:c-:-c-:--:-:-:-r=:c-:-:-7)":-:-:-:-:c-=,--=-_ I -- : --:.:: : : .

---- Nu CL EAR _..;c___:____:..:_:_:___;___::'.'!"""C..:.:..:..:..:~::_+-:.~:.:.c.:.:..:.:..:.=;=.:.:,:-=+-'~.:.;-:=-:-?-:-'--:-'-'~--'--:-:-'-'-'F-~:-:-:-==+-:-:-7:-:-:-:::-:-:-'-:-=~~:---:-c---:--t:-,::--,-:-:-: ::tc: ::-"'.:: fc:-:: :'_ !-:---,: : :.,.....:~.: L:--, -c-:-i_.

-! ·:,: --= . : :·:: !> ' ' t':: = :

1 Q01l----,!:_;____;;____;'---'c__;--'....:......;....:......;__:.___:_c:......;,.;----;...----'---1.:--'=-;,..-=-~-'-'-'----'----!--::--:---:----:---:-:-c--:--~:-:---::-c--;--,t . ' . . .. . • . . . . . . . . . . .

i ...:....;..._;__-_;_:_...c.:..:..:._:_.;__;_--"-'--'---'-,F-c-',..+-,---:-:--:-:---:-:-c~;:.; : .. ::C:: =::. (··. ·:: .. T : J: : i --~-'--.-!--.;:_: ,- ·-, = i -:Tc:=>:-::<=,,, : , -·:,==::: :::·., ·- ·· ... : .. = J: -::,, r,::,:

... :.:.···: .. ... .,

90

80Ul-:.:.__je----:---,;f!--~__;:-=-:1:...:...:,+;i_-'--':1.:.;__=1:..:.../.;.:1,...:.:::...:...=:~-'.:1+1'..:_\!,:..,::.:_:r;+-1~.:.:,:11-:-~:C.::::+-i~:,..:~:-'-=-~t'-':::.:Er::C.C,:-c'-,i~:CC.1E.1~:::-r~=-=:J~1if-'Cil'-':1;1"-'-t~~l~-':cir;:'-:::f.:.,,1t'-'-fi-'CC1:::c.:.~""f::*i·""i1:-:-:-:1:::.,..,;~-+:.;-:-:rC.:C:i:-=-;:c:-=1;.,...\-:-::1=.,...:r

7-t~-:-:):-:-:t:-:::~t--:-t:-:-1,-:-:~:-;-:-t::-::_~-:-~i:-:-:t:~ii1:-:-:[t:-:-:ti-:-r,.:l~-:-~~:-:-:\-:-:jti..,,.1:-::c-::);.,...~ft::t~-:-~t-:-1r:-:-:;;:ct1~-:-:-:,7.,;:-.::=

-· -;: • -:1 , : -~::: :!:;';i'!':l;:i:i;:::f ;~=;;~'l ~:;J~:f ;W~~i;'.;;: ;;~:,=;~::;'.;~f:lr ~'.';~: ~n:f ;';:;::::::=~~ •:'~': - -: = '.::r; •:,l~ '.: .,,.;: ,~~:: i:: :~'. ;!~;:~::: ~~r~m;;r~:;m'.~U ;~l:~;;r;~;;i:i~i;~;mm~~;~'.~~;~:,

601--_ .--__, __ : :f •; ,!::;::;:,!;::::=~: i:~~~~:;;~:;~c~:~i~~:t~,r~;;,;;~:~~J~'::: ~rn~~~~f :;;m~;t~c;;; ~:f •; :i~'::;: ! : : : -= ~;~~~~: :~: :: ;;=: :f:. • :[::;:t;::; [:~;:m;~<:i:;;~[~;: ;;[ =;~[~:::!'.~::;: ;• :!::_::~'. ::~;.[;;: -. _ i:::: ~:: · 0 20 40 60 80 100 · 120 140 160 180 200 220 240

HEATING TIME, hours

FIGURE 16. AGNIR POOL WATER HEATING AND COOLING DATA (WITHOUT HEAT EXCHANGER) . I

'; .I

u.. 0

... w

. O! :) I-<( O! w ·O..

-~

w I-O! w

"' I-

°' <(

3: --' 0 0 0...

'::ci· :: : ::: .; J/',: le 1 j.P',, :!'. :),;.J::;,/:;jc:'\'(:T:'=='=:Ti:1:(,: : (: :; -- -::',;:: !·': >J :::·:: ! --- :J:: ..... : i i . . . _L _____ i ____ _

70~;-1),>~: i : ! !~5.4·;~~i~k~1:: i i l : it : ; : __ ._ ; •- . '.- • : L--~------

65 '{::*r::':<; :: : .. i'.::'.;''.::kL::;;:=::=~~1::;:;.:-y,_;;:n-:~:~'.,:V:;~::=;;:''.'.i~:::i···:>.r:: ~- :';;.:~~ :·f ~:·I . ; ::::. ' . . ·j-:-: i ·:--: -- --0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50

HEATING TIME, hours

' i I

FIGURE 17~ I

AGNIR_ ~ALORIMEfRIC POWER CALIBRATION

.\ .

AN-1527

IX. NEUTRON FLUX TRAVERSES

The thermal neutron flux mapping of the AGNIR core and thermal column

was performed using 0.010 in. diameter gold wires inserted into the AGNIR

flux wire holders and irradiated in the AGNIR. core and thermal column, using

the 28 flux wire positions (22 in the core and 6 in the thermal column - see

Section VI-F). Epicadmium measurements were made, using 0.050 in. ID by

0.090 in. OD cadmium tubes with the 0.010 in. gold wires running axially down

the tubes. The 26Pin. long gold wires were cut into 2-in. increments and

wrapped around a dowel to form an 0.32 in. diameter ring and counted on a

scintillation counter that had been previously standardized with 0.002 in.

by 0.50 in. diameter gold foils by reference to a National Bureau of

Standards calibrated neutron flux. In regions of the AGNIR core where the

neutron flux changes rapidly with position, 1/2-in. long wires were used.

The 0.010 in. gold wire and the standard 0,002 in., 0.5-in. diameter gold

foils were cross-calibrated, using the AGN 201M reactor (basic AGN-201

reactor modified for 20-watt operation). A typical axial thermal flux plot

performed in the Band G rings of the AGNIR is shown in Figure 18, The B

ring represents the flux plot at the center of the core, while the G ring

represents the flux distribution in the reflector region. Figure 19 shows

the thermal neutron distribution radially across the AGNIR core at the center

line of the fuel, Figure 20 shows the thermal neutron distribution radia,lly

across. the centerline of the AGNIR. core and thermal column. The details of

the thermal neutron flux measurements.were documented earlier*.

*Internal Communications: V.R. Forgue, Gold Wire Flux Mapping of AGNIR, AGN Chem Tech Memo No, 887, October 1965.

V.R. Forgue, Flux Traverse in AGNIR Thermal Column~ AGN Chem Tech Memo No, 952, March 1966.

27

u (l) Vl I

N E ~

C: .. X ::::> __.

1 X 10 l Q ~----,----__.;.--,----.,..-~-....;..,r--.....----,----"T"""'---W

u..

Z 5x 109 0 0,:: I::> w z __. <(

~ 0,:: w :c I-

0 4 8 12 16 20 24 28 DISTANCE BELOW REACTOR TOP GRID PLATE, inches

FIGURE 18. AXIAL THERMAL NEUTRON DISTRIBUTION IN THE AGNIR CORE AT A POWER LEVEL OF 230WATTS

28

·-·--------

(_

'N: .EI ~.

! C . .. ·[ X'

3, U.. I

l X 1010

z;. o· 5xl09 ~-1-:::, W.'

'I Z; !

...I' <( '

~ ~ w :::c I-'

REGULATING ROD POSITIO~

(f OF FUEL

11 10 9 8 "7 6 5 4 3 2 0 . l 2 3 ·. 4 5 6 7 8 9 l O 11

. iijjsi'Xr,rtt FROM CORE CENTERLINr~~ ,fncliies';I. . . .• . '' ,. ... \ . ' .... ' .

. . FiG.URE 19 •. RADIAL THE,RMAL NEUTRON DISTRIBUTION IN THE AGNIR--CQRE '(230 'WATTS) l • • • • • • • •

.- .. 2-9.

GRAPHITE REFLECTED

WATER REFLECTED

1012.__·~1--~-·-R_E_A_._cr_o_R~~~·-·-11.-i:=-~~---T-·H_E_R_M_A_L_c_o~Lu_M~N----~--~M

u . Q)

"' I N

E ~

C .. X ::, ....I LL 1011 z 0 O! I-::, w z . ....I <( ~ O! w :c I-

. 9 ld ._._ ___ ..._ _______ __._ __ ___,....___._ ...................... __ ...... __ ....... ________ ....._ __ .... . · 15 10 5. 0 . 5· 10 15 • 20 25 30 35 40

DISTANCE FROM REACTOR CORE CENTERLINE, inches .

FIGURE 20. RADIA~ TH~RMAL NEUTRON DISTRIBUTION IN AGNIR CORE AND THERMAL COLUMN .. 30

AN-1527

REFERENCES

1. R. L. Newacheck, tl al., Aerojet-General Nucleonics Industrial Reactor,

Hazards Summary Report, AN-1193, Aerojet-General Nucleonics, San Ramon,

California, September 1964

2. R. L. Tomlinson, Aerojet-General Nucleonics Industrial Reactor (AGNIR)

Critical Experiments and Power Calibration, AN-1406, Aerojet-General

Nucleonics, San Ramon, California, April 1965

3. L. D. Connolly, Los Alamos Group-Averaged Cross Sections, LAMS-2941,

IASL, July 1963

4. B. G. Carlson, e. E. Lee, and W. J, Worlton, The DSN and TDC Transport

Codes, LAMS-2436, LASL, October 1959

5, R, L. Tomlinson, SNAP Shield Test Experiment Reactor Physics Tests,

NAA-SR-7368, Atomics International, Canoga Park, Calif,, July 1962

31

engineering and development division

Documents