ML20205Q192
| ML20205Q192 | |
| Person / Time | |
|---|---|
| Site: | General Atomics |
| Issue date: | 05/20/1986 |
| From: | GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER |
| To: | |
| Shared Package | |
| ML20205Q179 | List: |
| References | |
| NUDOCS 8605280205 | |
| Download: ML20205Q192 (38) | |
Text
,-
ATTACHMENT 1 SUGGESTED 'dORD CHANGES FOR R-67 TECHNICAL SPECIFICATIONS A New Section 5.2.3:
5.2 3 Standard low-enrichment fuel with high w/". uranium (a) Maximum uranium content in unieradiated fuel 30 wt 5 (b) Maximum uranium enrichment in unieradiated fuel 20 %
(c) Hydrogen-to*:1cconium ratio 1.5 to 1.65 K atoms to (in the ZrHn) 1.0 Zr atoms (d) Natural erbium content q _(homogeneous distribution) 0.4 to 1.8 wt %+
(e) Cladding material and 0.020" 304 or 304L stainless minimum thicknese steel, -Incoloy 800, or Hascaloy*X.
44,44,44,,,,,-+,,4,,44, ,a----4.--44 ..
+ Scaled according- to maximum Ud235 content such that a ccre containing
~
only these elements would require a critical loading' of A 50 fuel elements.
2 A Revised Section 5.3 2. -
5 3.2. Special fuel elements with high w/o uranium.
(a) Uranium' content in unirradiated fuel ->30.0 to 45.0 we %
(b) Maximum' uranium enrichment in unirradiated fuel 20 %
(c) Cladding material and thickness 0.020" 304 or 304L stainless steel,
. Incoloy 800 or Haste 11oy-X or 0.030" Zircaloy for 1.5" nominal diameter, or 0.015" to 0.020" stainless steel, Incoloy 800 or Hastelloy-X for 0.5-inen nominal diameter.
0605280205 860520 PDR ADOCK 05000163 p PDR P ACr t
At:achment 1 (conc.)
(d) Construction shall be suen that the calculated clad temperature scall not exceed 500*C at planned operating level.
(e) Special fuel elements with 0.5 inch nominal diameter scall be enclosed in a 0.020-inen metal shroud. The shroud, construcced feca one of the above listed clacding materials, shall have a nominal diasecer of 1.5 inches.
A Revised Section 5.5.1 5.5.1. Standard Fuel Elements (5.2) and Soecial Fuel Elements with Hign Enrienment (5 3.1).
(a) All fuel elements shall be inspected visually for damage or detericra-tion at leasc annually, and all uninstrumenced fuel elements shall be measured for length and bend at least annually.
(b) Uninstrumented fuel elements shall be measured for length and bend af ter each group of pcwer pulses as defined in 5.5.i(c) if:
(1) They are stancard elements (5.2.1) and are in a core suojected to reactivity insertions in excess of $3.00 of reactivity, or (2) They are standard FLIP. fuel elements (5.2.2) or standard LEU fuel elements (5.2.3) subjecced to pulsed reactivity insertions anc their measured temperatures exceed the following values:
Fuel Elemenc Lccation LEU (20-20) FLIP LEU (30 20)
In Compact Segmenc 680'C 680*C 600*C Adjacent to One Water-Filled Hole 580*C 500aC 325*C Adjacent to Two or More Adjacent WaterdFilled Hales 535'c 330*C 350*C (3) They are scecial fuel elemencs and ara in a core subjected to l power pulses wnich produced measured temperatures in excess of l 700ac, as measured by a scandard thermocouple.
? ACE 2 L
Attachment 1 (cont.)
(c) Same as existing wording.
(d) Same as existing wording.
A Revised Section 9.2.2.
9.2.2. The ' maximum allowable fuel temperature using a standard thermocouple shall be:
(a) 800*C for standard low-enrichment high-hydride fuel elements, (b) ~ 500*C for standard low-enrichment low-hydride fuel elements or aluminum
- clad fuel elements, (c) 920*. for special fuel elements, (d) Values selected for the applicable long lived fuel elements as follows:
Fuel Elements Location LEU (20420) FLIP LEU ( 30-20 )
In Compact.
Segment 780*C 780*C 690*C Adjacent to One Water-Filled Hole 670*C 575*C 490*C ,
Adjacent to Two or More Adjacent Water 4 Filled Holes 615'C 500*C 405'C A thermocouple element shall be placed in the hottest fuel element location next to each flux trap. A thermocouple element shall also be placed in the anticipated hottest fuel element location within the compact segment of the core unless the flux trap is located in the hottest section of the core. With no flux trap, the thermocouple element shall be located in the antici-pated hottest core position.
?tCE 3
9, O ATTACHMENT 2 TABLE OF CONTENTS Page Contents Nwaber
)
- 1. General, Brief Description of Long Lived TRIGA Fuels . . . . . . . . 3
- 2. Review of steady state and Pulsing Operations with FLIP Fuels . . . 4 3 safety Limit for TRIGA Fuel . ., . . . . . . . .... . . . . . . 5 4 Calculations and Experiments that Relate Measured Temperatures and Peak Temperatures in TRIGA Fuel .. . . ., . . . 6 4.1 Considerations of Temperature in TRIGA ELi? Fuel . . . . . . . 6 4.2 Considerations of Temperature in TRIGA LEU Fuels . . .. . . . 13 4.2.1 LEU (20-20) . . . . . . . . . . . . . .. . . . . . .. 13 4.2.2 LEU (30-20) . . . . . . . . . . . . ... . . . . . . - 13 4.3 Comparisons of Temperature for FLI?, LZO (2C-20) and LEU (30420) . . . . . . . . . . . . . . . . . ... . . . . . . 13
- 5. Qualification of TRIGA LEU Fuel . . . . . . . . . .. .. . . . . . 20 5.1 Brief Description of IRIGA LEU Fuel . . . . .... . . . .. . . 21 5.2 Basic Metallurgical Ccnsideration for TaICA LEir Fuel .. . . . 22 5.3 special LEU Fuel Test Program at TRICA Mar!( ?. .. .. . . . . 23 5.3.1 Pulse Tests with Three LEU Fuel . . . . . . . . . . . . 23 5.3.2 High Level Pulse Tests wie.h LEU Fuel . . . . . . .. . . . 24 5.4 High Burnup Tests with TRIGA LEU Fuel . . . . . . . . . . . . . 26 5.5 Operations with Commercial TRIGA LEU Fusi . .. . . . . . , . . 2S 5.6 Qualification of LEU for Use in standard TRIGA Reactor . . . . ,29
}
- 6. Prompt Negative Temperature Coefficient of Reactiv1?.y for Long Lived Fuel . . . . . . . . . . . . . . . . .. . . ... . . . 20 PAGE 1
- _s
Ar.tachment 2 Table of Contents (ConL } -
Page Centents Nuccer
- 7. Bases fcr Proposad Technical Specifteacions . . . . . . . . . . . . 31 7.1 Procedures for Detectining Hottest Fuel Element . . . , . . . 32 7.2 Delay Tins fet* Scram in Pulsed Operaticr. . . . . . . . . . . . 33 7.? Core Configuration for Long Lived Fuel Elementa . . . . . . . 33 i
- 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
? ACE 2 j
.- . . ~ .. -
_, .7 .
.e c ATfACHMENT 2 Discussion and Safety Conciceratienc For a Standard TRIGA Core Loaded wit.h LEU _ Fuel.
- 1. Generci Brief DeecrtDtiert of Long Lived TRIGAldelB.
FQr many years CA Technclogies has ccatinuen a fuel development program the nic of whii:n nas been to produce fuel suitad to the enanging neecs of its cas-t omrc. On e of thess goals, has been to produce fuci uith a censiderably longer life so that a ccre centhining these fuels pil.1 operate lenger and with lese f requent. fuel ' replacettent , To chts end ttte FL. fuel was developed using the license -peovialone for special fuels and special cores. Later, t,1cence Amend-citnt No. 29 elevated FLIP (ael co r,hc status of ctkrtdard fuel in 1978. It -
snould be pQint,td out that an bportant consideration in the design of the long lived TRIGA ftrelt is that ea6h thould be compatible a!.th all other TRIGA fuels.
This h0B meant that (no contributiccs *// each element to (1) the core reactiv-Aty t.nd (2) the peczpt n58ativ6 doaffl0 lect of ceactivity stiall be cotoparable.
Thus .3ach of the lortg 41ved fuel ylements is capable Of replacing other fuels
'in th& care on a cne-so-ort 9 taels. The design of the rh st or these, namely the ata%ars ELIf fGel (3 1/3 wt-5 U, 703 enriched, erbiwr poiten), arid the r6c9ac LEU RICA fuel (60 adq 10 wt-) C, 20% enriched, erstua poison) meets these crtterJE. Table 1 surr. crises several of the important t,&rameters fcf the celginal JcandsFd fuc'.. FrJP, I,E'1 (20-23) , acd LEU (30-20) reels. All fuel tyWE ic this ?lahle were designed 'c0 operac'a in TRIGA cores wich 80 fuel 61elnents (c4-
- megawatt r,d 100 fuel ej ements for 2 or 3 cegawatts.
M3ti 1. Coniparisen of Imoct' tant P3remeters for Several TRIGA Yuel Ty;;es.
- j. r- -
a x 10 8 l Type Wt f. U-Z35 U average CORE f U 5.r grime / element enN chgent over Lifetime i 23-700*C MWD (Ak/R'C)
= = - --
Original l 3.5 0.C 39 20 9.5 =t00 .
- _m FLIP 8.5 1.6 137 70 10.5 3500 LEU 20 0.5 99 20 10.5 1200
%; I w- _ m .--.
LEU 3C 0.92 162 20 8 3000
_ ~ _
LIU 45 1.3 232 1 20 5 4000 I -
PAcI 3
T~
Attacament 2 (Cont.)
It should be emphasized that the eroium burnable poison performs at least two necessary functicns. First, it controls the reactivity contribution of each element so that a reasonaole minimum size core is produced wnile also being a 'ournacle poison that lengthens the core life. Second and perhaps even T.or f important, the erbium itself produces the large prompt negative tempera-ture coefficient of reactivity since the normal large cell component is essen-tially cbsent in these heavily loaced fuel elements. The success of this 1stter intent it cemonstrated in Table 1 wnere it can be seen that the prompt negativo coefficient, a, for fuel up to (30-20) is essencially unenanged even though quite different core lifetimes have been produced.
In the following we present a review of the tests and technical details that permitted FLIP fuel to be elevated from the category of special fuel to stsndard TRIGA fuel. The details of that earlier series of events is useful fu' more than simply an illustration of this process. More particularly, the LEU fuels (20-20 and 30-20) will be shown to share great similarities to FLIP fuel. In mhny areas of interest, the tests with FLIP fuel provide the impor-tant guidan0c for the consideration of standard LEU fuels. To round out the presentation, the tests conducted on LEU fuel at GA Technologies as special fuel and the exhaustive burnup tests conducted on TRIGA LEU fuel at the Oak Ridge L&borctcry Reactor (ORR) will be reviewed and set in perspective.
- 2. Review of Steady State and Pulsing Goeration with FLIP Fuel at GA Technologies.
The first tests (~1976) in the Fuel Lifetime Improvement Program (FLIP) were approacned cafefully, especially those that might involve variations in the large . prompt negative temperature coefficient of reactivity a. Conse-quantly a series of steady state and pulsing tests were conducted with 18 FLIP f uel elements in a standard core. After these very successful tests, a full core of FLI? (-100 elements) was installed in the Mark III TRIGA reactor (R*100, now decommissioned) and operated (-750 MWD) to about one fourth of the expected core life.
The FLIP core was then transferred as special fuel to the Mark F (R-67) rcactcr in 1973 when the Mark III (R-100) was decommissioned. After two years of add!tional but intermittent steady state operation, two series of pulse l tests ware conducted successfully. The first involved pulsing the full core
(=100 elements) to modest peak fuel temperatures. The second series of testa with a smaller core produced higher FLI? fuel temperatures. In these 1975 l
pulse teats on a core loaded with FLIP fuel, we demonstrated that the tempera-ture dep6ndent coefficient was much greater than required in Section 9.4.2 of l the Technical Specifications. The results showed that a reactivity insertion f
$2,50 wculd produce a maximum measured fuel temperature rise of less than 440*C comparea to cne allowed 700*C. In other words, the prompt negative temperature coefficient is considerably larger than necessary to assure safety of I operation. The first pulse tests (with a 100 fuel element core) consisted of 30 pultes with reactivity insertions up to $2.12. The peak measured fuel i temperature was 379*C. It may be noted that the fuel had about 1/4 of the l cxpected end-cf-life burnup and that all fuel passed the surveillance tests.
The second series of pulses with a small compact core (60 fuel elements plus 3 fuel follower centrol rods) involved 41 pulses with reactivity insertion up to PAGE 4
9 O Attachment 2 (Cont.)
$3.20. The peak measured fuel temperature was 523 C. All fuel continued to pass the surveillance tests.
3 Safety Limit for TRIGA Fuel The Safety Limit of 115C*C for all TRIGA fuel has been previously estab-lished and is governed only by the properties of the sicconium hydride alloy.
This limit has been tested successfully in an earlier pulsing program at GA Technologies. In that program special fuel elements were pulsed to produce measured fuel temperatures of 680-700*C. The calculated peak fuel temperature was approximately 1150*C. The measured temperature limit for standard low-enrichment high hydride fuel is 800*C which corresponds to a calculated peak fuel temperature of 1025'c which is well below a fuel temperature that produces a measurable pressure from hydrogen. As will be demonstrated below the limits placed on measured FLIP fuel temperatures operated with and without water-filled flux traps assures that the peak fuel temperature is no greater than 1025'C.
The following presents considerations of the fuel Safety Limit of 1150*C and possible uncertainty in measuring the fuel temperature. An extended series of high performance pulsing tests was performed on TRIGA fuel. One test in volved the use of several highly loaded fuel elements that were designed to lead the rest of the core in temperature. These tests involved hundreds of pulses and produced peak fuel temperatures that exceeded 1000*C and approached 1200*C. The measured pressure for the maximum pulses varied from 12 to 30 psi and was a trivial fraction of the equilibrium pressure that can be tens of at-mospheres. Although a nominal safety limit of 1150*C has subsequently been established for TRIGA fuel, it is clear that still higher temperatures have been teste.d satisfactorily and safely. The magnitude of the hydrogen pressure within the fuel cladding is governed mainly by the H/Zr ratio and the maximum fuel temperature. In summary the experimental testa performed with the above fuel are applicable to the expected performance of FLIP fuel and later to the LEU fuel.
The Limiting Safety System Settings and the appropriate reactivity inser-tions have been selected to limit the FLIP fuel to peak temperatures of 1025'C.
(See the discussion below that relates measured temperatures and peak fuel tem-peratures). The corresponding measured temperatures for FLIP fuel are 780*C, 575*c and 500*C, respectively. These settings provide a margin of safety of 125'C (even more, if one accepts the fact that pulsed temperatures considerably above 1150*C are still sare) . Two uncertainties in these temperatures have been considered:
(1) uncertainty in scram temperature trip point, and (2) the accuracy of determining the hottest temperature at the location of the hottest fuel.
The margin of 125'C is very much larger than the possible uncertainty in the temperature scram trigger point which is probably not larger than 10 to 20*C. In general, more than one thermocouple is set to provide a temperature scram. This could lead to an overall uncertainty which is probably smaller
? AGE 5
e ,
Attachment 2 (Cont.)
than 25'C. This procedure to select the hottest core location has been en-played dozens of times in this laboratory. Experience shows that rotation of the fuel element within the hottest ft-1 element position usually produces var-iations about the average high temperature of no more than s 15ac. Since an effort is made to position the element for maximum temperature reading, the chances of producing an error of more than sac are very small. The most prob-able error in scramming from the maximum fuel temperature is estimated to be less than 25'C which is small compared to the margin of safety.
4 Calculations and Exceriments that Relate Measured Temoeratures and Peak Temoeratures in TRIGA Fuel.
4.1 Considerations of Temperature in TRIGA FLIP Fuel.
A long series of efforts has been devoted to determining the magnitude of the peak fuel temperature in a pulse. In steady-state operation no similar problem exists because the central portion of the fuel is the hottest with the outer edges significantly cooler consistent with the thermodynamics of the equilibrium state. Thus, the maximum measured temperature determined in steady-state operation is essentially the hottest temperature in the fuel ele-ment. On the other hand, for pulsing operations which give rise to essentially adiabatic thermal conditions within a fuel element, the temperature distribu-tion in the fuel element is related directly to the U-shaped power generation, being higher at the outer edges of the fuel element. Consequently, the maximum temperature within the fuel element is higher than the temperature measured by a thermocouple near the center of the fuel element. This peak temperature is calculated from the applicable power peaking factors, the temperature depen-dence of the heat capacity of the U*Zr*Hx matrix, and the extensive heat trans-fer calculations. All of the applicable data have already been presented in the R*100 SAR (Docket 50-227) dated January 5, 1970. (see for example the peaking factors given on p. 3-57 and Table 3-7 on p. 3-58.) In Tables 2, 3, and 4 are presented the peaking facters used and the resulting peak fuel elea ment temperatures calculated for the list of measured temperatures.
The confidence in the peak-to-average power densities and corresponding temperatures thus calculated is given by a number of flux measurements per-formed over the years on related topics. By measuring the flux distribution over a cross section of a fuel element, one can determine the power density within a fuel element. This can then be compared with the calculated power densities. For instance, in an early test (1961) nine cobalt wires were in-serted along diameters of a TRIGA fuel element. These were positioned to give representative radial values of flux and to investigate the flux within the fuel element adjacent to regions with maximum water between fuel elements as well as minimum water between elements. Activation of the cobalt wires was accomplished by firing one pulse (about 28 MW-sec) on the Mark F. Each of the wires was then cut into about 25 roughly equal pieces and the activity of each segment was measured. A typical radial plot is given in Fig. 1. The
, butterfly" shaped flux distribution for the element in question is shown in Fig.
l 2 where the effect of larger and smaller amounts of water between adjacent ele-l ments is evident. A theoretical calculation of the expected power distribu-l tions across the various diameters shows good agreement with the experimental l results. In another test at a later time (1966), the peripheral activation of
Cr in the stainless-steel clad of one of the hignly loaded pulse test fuel l PACE 6 l .
Table 2. Temperatures Calculated for FLIP Fuel in Compact Core. -
~
- g E **
T(*C) (watt-sec/ce) (watt-sec/ce) (adiabatic) measured 800 2922 1615 530 600 900 3480 1924 600 680 1000 4080 2250 670 760 1025 4240 2344 690 780 1050 4400 2432 705 800 1100 4721 2610 740 840 1150 5050 2792 775 875 1200 5405 2988 810 915 A
- E =Ex 0.83 TC 1.43 x 1.05 ,
For this' calculation, we have A
P
-- = 1.43 where P is the peak power density, and P
E is the average power density in cell; P
= 0.83 where P 18 the power density at trie thermocouple location;andC P
(P/P) high T
= 1.05, which gives tha temperature correction of the peak-to-average ratio.
(P/E) 23*C
- TTC ~ . rm c e a cu at ns.
TC(adiabatic) x l
j
f Table 3 Temperatures Calculated for FLIP Fuel Next to Single Water-filled Hole.
b kC **
a (heat content) Cneat content) TC TC T('C) (watt-sec/cc) (watt-sec/ce) (adia5atic) measured 800 2922 1056 390 440 900 3480 1258 440 500 1000 .4080 1474 495 560 1025 4240 1532 508 575 1050 4400 1590 520 590 1100 4721 ',1706 550 620 1150- 5050 1825 575 650 1200 5405 1953 605 685 f
i
- ~
- 2.2 1.05 For this calculation, we have P- = 2.24 where P is the, peak power density, and E is the' average power density in cell;
= 0.85 where P is'the yower density at the thermocouple location, TC P
and A
(P/P) high T
= 1.05 which.gives the_ temperature correction of the (P/P) 23*C peak-to-average ratio.
)f
- T =T . f om A c e ca cu at ns.
-)
a .
Table 4 Temperatures Calculated for Flip Fuel Next to Three-Element Water-filled Hole.
Flip Fuel Next to Three-Element Water-filled Hole.
w "C **
. _ (heat content) (heat content) TC TC T (* C) (watt-sec/cc) (wat t-sec/cc) (adiabatic) measured 800 2922 872 335 380 900 3480 1038 380 430 1000 4080 1217 430 485 1025 4240 1265 443 500 1050 4400 1312 455 515 1100 4721 1408 480 545 1150 5050 1506 500 565 1200 5405 1612 530 600
- E TC "
- 2.6 1.05
~
For this calculation, we have P is the peak power density, and
. = 2.65 where P is the average power density in cell; C
= 0.83 where P is the power density at the thermocouple location; P
and
) = 1.05, which gives the temperatura correction of the (P/P) 23*C peak-to-average ratio.
- T ~ * '# * ^ " * * "" * ' "**
- _9_
)
s
- ~
,1I g 8
il '
j ,l l
._ 1 3 *
/ I
- 1 . .
+ D
- d r tn a e' wm o
5 1
2
's7 - e gf e l ! 6 1
T le 1
. - dou E F E
-, e
. /
g r 4 d t e c t
a a x x t -
- J-
=,
4 1
t u u n l l e F F
/ "
l E
m e
yf,m. ,
l e
u 2 F 1) m 9 m
A,
( C r n
,'sl--
e
,8 a
.. ~
i
- t
- n d e e
.. C r 0 u
_J 1 l s
_. e a
. u e i
- D -
F m
o M
e c
j r n
_ .f e u
m e l
- c F 8 n a f t o ,
s i n D o r l i
t a u i b d i a r R
,- i l 6
t i
s -
. D 1
l a
i d
a R
i
,r 4 .
, 1 g
. i j F 2
k,.
~
_ 0 s ,.oge J '
o x;w c ,- oG9m * " S3M o wn 4 s.- .
^yvoy8n 19 s
4 To Cora Center
.iFlux at edge of meat n Flux at r = 16 m
,\\ \'\ g\ g\ ,
.. ,
- Flux at r = 14 m
\ \ g' \ Y\
\' 9
\\ \ ,\g }%I O *\g g s ?'
\\04 s\ \
g\ \g g\ \
\'\ d i M
\ \\
el.5kg\\ \ g\ Yg\
g\s\ \ s . N- -
\ \ ,' \\ s\ \
\h \ g \' * \ \ g\ d \ [
\'$\'(n'g\\\'d' Y
\
s \ g\ \ l\ \\\\
l.h \ g\ '\'s\'h
\
g\ \ g sh \% g g \ . % '\ \' g\ g \ M\' \N\ g g\\\ \\' g\ g\ 'D[ MO,,lJ 1*3 f' '
c",' J'
\g\Y \\\\\ \ g\ h ' '~
e \ '
\\ gy g s
\\\g s' g\ : \ ' d ' \ ' ' \\ * \'
'f O \'b*'c t e "[Q\
[x -
d \\ \ N' s' ? ', [s I- \ s'
\ 3 \\ \ \ s' \ = \ \ g. %\ \ .\ \ \ d ' ' s s \' \ D \
( \\ Q N
\
\\ \ ,\
, gs ,g \
- g g\ \* h s \\ \'\ s\ \s
.\ A h\ g . [s g g% \ g' g\ g \
\
k* \
\\ \ Y(
%D", #&'9% s7 L's(;' $ i.z' 'h':6.+1k'n&n u sm. .. :s - y: , ':y.,.p + \ .\':::
N '1.'u e \'
's.$Q'? -
~'W '
sk\' .'.}'y'2%h 1:& % '
' ~"0.~M' '
gg.* .,Q\ g ,
. GQ \U\\N'3 w.
\\\\. .
- E ",
h 'h
\ *' *',-
, gc 37 q \h g
N' sh (Y, . \s s\ \ g
\(U N s ss\ h\ \ \s \ \ I \g\\ g \ s N s
/
,'g0. yN'A'\g\]'\
\g g \Y . g '\ '. h
($\'\\h~iU3^'N e#(O O' O
\ # 'N \\ g ) \\" ' \' \
\ '\
I s' *
'g \ g'
.\\
\ \\
\\ e.\ggs\ 'dg \ fd
\s \g\]s s
y \
\\,s -
\ s \" d ', N *.\ \\
\\ \A , ,\
\
' /h N 'D'Y'\ /d \" '\\Y Y \\\ \' N' s
)Q'f \'E /' ,[ $g'g\QN .Al *\Y '
- s .
\x\@'@g g\
\#'"
e'\l'$ #8gc' 'i
~ 'Nd
\
N' D \' \V (' k\ / 'i' gN'\\' kg\sM .Yh\\ } \ \dJ\'\ \ g' \ '- y ' ,\\\s N h \ \ \s"'
{.\\g\\ g\ y \\g \ g\ ' N' 'g 'g . 'gi U.i N \ \ -
s
\
I\\g
\
g
\'(^;;t."0.t':c M .
0
'c(ild ]\ ,0t:b%'\s '(?'Q Rt \' s\ij:'gQ "',':O s,9*d'
'N g)\ @
g 's \ g '. . \ \
g \\ \ '\ - \ g \
g [(s s O 'M '\\ \ g\ \\' \ \\ '. gsg,j ;. \ )d ...W
\g Q'
\\ .\ \' \ \ ,\ gg s\-
- 5'-\V ' \ .
\, hgg\>'j
, . , g,\s.3. -
Fig. 2. Relative flux in TRIGA fuel at position C-9 in Mark F (R-67) Reactor.
j
e e fv;;. -"* .d':.n.r : .., :r ..A. . .h : n; * -
c:nn-:4 :- T.i.+ .:.n :. [ :- ;:a. I::
- .t -
= - ny" :: Pr"" 1:: - " '
[= ~5 ly.2iU::@ ]i j p. -: - ?!?.~ :piili: :f ri!,
== >
..;; g
.n - [" " 3 .~. . ....
. l .n.: :;n: .- . i. t. n. =. . . ln. .. . .... . . :: =. ,
.5g .::. .".!iG- ~ Z ~ Iii : ..= n. i%-:i.;.hi.i. M i:-fi MI..I. Mr. .
bb hih W iNJ' ' li^:.*Mi!" : *5 5'I:I!EfbdNi'E 'f-i.Ii: . G E
[.@ 7 : . .-i u--iiii . :4.'-ti
- fk!iCi{Ub'~i[.Mi ;.{.ii,5f e Ed H..
p . . i.ii.
- =. ,
U
- ii..a
. ri.C. ..:..W.:nd.:
.. . . . . . %..% z:i.=.:.:id...d.i:
.. . i. '~:l.it::
- id.. .
a h__ - d i'- k g
- UU "ifk.b! O N !!?'- h bd2 -
35 =
Nh
- = 0 55
= n.
3 o NNM~i~-dNE ~ik.iiiM="*5!.:En b
5:N c c
n- : A :n=gu,;,,r"":"--- m,;=.n n . ::z , :,d u;;
33
.g =;;n ,.,, --: =-I n==.
L: c
~ n .N"-"t --":nr Q r4 4 [ T N~h5NINk E :NN
- < . - o h:.!!E.\i!N5'.i'. Ni$I!N M e
b,,
ibi .iNNN.5
,"4 _n.:=.:: :.;/ na. u.:a E!ITE -NI5'b _ 555hl
. --. r n.
'" a _ . . n'.: g .nn... =. :p. :: n: . . ---- g
.........f b:b $ .'_N Mg N N ! NE~ .IIN b;I:i.Mi 5'..$.a.=?E[s.h S.
- =P- _*"_- ..
. . . . . . . . . . . . _ . . . . . g
. -- u .i: n =.=.] n, d_:.n._.
- : . .-_ . . ._. ., . . .-. .. ..;:. . .=_ :::_ ~ ::. .=._a..;;; . . .n. t.;.;.:..
. =. . :: _ __ ..
=:.:.*= : .= : = .
= .:7:n. .:. ;;= .:=_. :. =.-
T
--=:-"- := r ,v.= .: = ---
v -:== r = :: c
.= ;,.;r - 4
- =_ :=._.
_ . .- 4.=. . . f. u . . - - - :
- r. b- .- g .:
.t =..=.-. :: _::u =
_ . . . . ._ h
...h--
I N
__...._..=_.1.._.:.._...=.....==_=..._ .:.:t=...-. . . . . . .u. .,u W
._.:__ _ . . .._=.=.....=.=u.=. - .
e "=. .n. . _.r. :.=_
.. ._ . r. t.. N.
g.. . . . . _ . _ . p r4 >
___._=...p=.=..=..n_s..u.\.....__..:= : ;;.=.. _ . . ,,.4 .f__=_._..__...=._p:.=.=._..
a.::. ... n. . m ye u gy
.._---;=.V_...;\==:-
- _ _.. ; L-- P_ %- u = r: n\7 :=. U ;,=_. .._.f.. - - g= r . ._,n :; ;,;
g g g g 4
y 4
- = .=_:.=_n == :n : .\ .= :: ;;.r *\: ==== =
- -.==- .=-: = j . u u.. _ . .v. . -; .:;; to
- s. ..:=*7 _ ._. Nr:--.=e. - g yy L_ T c
- =-*=--- *----= -- = : = : .:.=. . \=:: .T -t. ..=_:a . . ..t:= =r U .:= = _r_;=.= = = . .: . _= .=~. =- :: - <
c-e 3e .4.6 c ou
.._...____.==_=:==_._=.=_:..=y_n=_--_--
_A o :
.= = = = =_= . :. _._.=. > g,
-4 : =__ur .t .__...___.= =;;;-- *
=_..:=:: =.= :r m e
u- .
a Wc 'd
_.. .-=- : = . = _- . . - ."ub=_i:._==. r.= n==:?"E*~ =.~--.=..[J. @ @ l s
- * - " * - ~
- m - - := Z-N. _'lii" 'i- . '-4
.._ 4 M
-
- C
~
_._._,==.:.=..=._r.
._.==... . ._
{. ._._.... 5.:-==..:._=--.~-.=.--.- .. ..
- .c- .- : % .c y
- n; =..r.: =:.=. _:...,;g:.; ,., g
--"... ~.~_- _- u . .:.n.r . =-f:=1 = =Q.u he
.::_ . ..r.= . . . . .=_ .: =:. ..
-~~.u :.:
y
--*:= =_.*. .=.:.k_=. .. . _. .. x. . . ../. . ::.=.=_=. ~. _
. _ . .. _ . :. .:_:_=;.=_: 1. . .
0T
.. _. . . . . . _ g, g
.C_
" .:: .:.:= =: ;n g g=c . :::----- ==-j'*._.._..._..__.=.=_==.=::*.:=*===:
- ==.== u
_........\ _ P - "- -* :.
- q Q y:--- - -- ==== : = = = =:= - = = = . ou 2---- u -4
...__:..___.=..w_-__2..==r=_:=--=__=._=_._=.=._=.==-
_ . . _ nm w._.__._.. =
-- _-._ : x.: . __. . -==:.dm. ... =_ . ._ .,; . . .r.N..F:. ..
._:-. =. t- ;
.=. ..:..==.. .=:. = .~n-. . . . .c. _c. _ = . . .
- a. e4
_._...=.___=..::._-..=.:=.=___..._._=.==._=..=_L==....-
_-._.t._.-._._%.. _ .
. . . . . . . . . _ _ . . . =_.= = . . . = _ . . = . . .
o, g
= .;;=__- -- :- - --_ ~ ; n. : r a = ==n =.. . ..: g
--.-:._ r._--- : -- : =;._ :
u
.._._._..__._=...-._;. _ _ . . . ... u===__-:=_=_=
._..t.. . . . . . . _ . .. . . . .
. . . .. .... oxg
- _ . . _ _ _ _ _ . . . _ __ .. ._= .. ... _ _ _ , . . . . . _ . - . . _ . . . _ . . . _ .. . ,
n==
- - -r= c.-.._---.___=.=:2.------"-=:.u.--=:.==r.== - n33:,g
_ _._=u -----=....az. .:===. -, n<
ou r_*, = :
- == m == :==. .r a _ : u==n _ ==-r:t_ ,m, c
--- = n : =:; . . .. . . . == - = = :== = r " r n~ =/:u n- . === =nrna
- - - = r.=: =. r-==.:. = 7==a = . =.n : /.'.: . =;f.'::n =.:! = . = n-~~ - - - * --
4 0
== =L == : ==:2=-----_u====:== un -= n _ .. mW Cu
- =
- =_ . =_. .: =: == -=.- .. 2=. .=_._.::_ =_==_ . r=. _._=.. . f.=. . . f[.a. n. .=.=. . a. r. .:_p.n. .
.. =_. .u.. a _
aa oe
...p..........._=...
.=.=...-..:=c.l.:.._...:
._. - _ . . :.:=. :. . _. . . .. . :. . . =. r=. . n. -
n[...=. .;.rm. .:.;:::: ..._..... :::n. =: O g 55EEli55#iisi45;IW:
i EliIi N2!'i:~iM ! :E@i-Wi:i-iS5)."-- "E!!N .
~~
T ;i.M Ji
- i I-h /5 '" dji":!$;lE~ N - Ei"Iii M
. = . . . . . ..5
- =._..=. . :._ F=.
_ .: ::.:=. . ...'". ld '!53EN5:. . . . .
. . .:...:t.::. :=.m. . =. =... u=a. : .4' n
". . u s = ,. . .:. :::= =. = . . ._.::.
.._..t
. :i eo
. . . . . . . . . . . . . . ... ....a c= =. :: j u. ..= g .. c_u. t,. ... . .j . . = . :. . .. .=. y. .. q. ::. ., . . .: y. .. _:.
. . . . . . . . r. . . =. n :: =.. :.
. . . . . g siid,g. rS 'iP= Fi;i fin fi-4i--i:$-- ii i:!Ci i.:-! .g -
.]. : *. .n, a C;. :+ :meme- : ca.; :;..
=..!...n. p.r..a. .. @. .
Ci.ji:].*:.:.ii. . . E-r..^.. r
.i ~ ;i:d:f ,
. F.-+,.: .:t:: a c..- - :;V
.. ,. . , , , . . .y A3 TAT 30V 8AT38TMI k
Attacnment 2 (Cont.)
elements was carefully measured and compared with the computed flux distribu-tion around the periphery of the fuel element. Figure 3 presents these data and shows that the two are in reasonaole agreement within experimental errors.
The theoretical curve conservatively predicts a somewhat larger range of maxi-mum and minimum activity and, hence, power densities. Combining the results from the flux tests where the radial as well as the measured peripheral activie ties were compared with calculated power densities has given assurance that the calculated peak-to-average power densities used to evaluate the FLI? operat-ing parameters are valid.
4.2. Consideration of Temperature in TRIGA LEU Fuel.
The calculations that lead to the results in Tables 2, 3, and 4 for FLIP fuel were repeated for the two LEU Fuels , (20-20) and (30-20). The comparable results are presented below.
4.2.1 LEU (20-20) Fuel. Data are present in Tables 5, 6, and 7 for LEU (20-20) fuel operated in a compact core, and cores with a single or multiple water-filled hole (s) adjacent to the LEU fuel. From these Tables one finds the measured temperature for a peak fuel temperature of 1025'c to be: 780*C, 670*C, and 615*C respectively.
4.2.2 LEU (30-20) Fuel. Data are presented in Tables 8, 9, and 10 for LEU (30-20) fuel operated in a compact core, and cores with a single or multiple water-filled hole (s) adjacent to the LEU fuel. From these Tables one finds the measured temperature for a peak fuel temperature of 1025'c to be: 690*C, 490*C, and 405ac, respectively.
4.3 Comparisons of Measured Fuel Temoeratures in Several Long Lived TRIGA Fuels.
For ease of intercomparisons of the fuel temperatures, Table 11 is pre-sented in which direct comparisons are made for FLIP, LEU (20-20) and LEU (30-20) fuels. The measured fuel temperatures for the safety Limit of 1150*C and the lower value of 1025'c are listed from the accompanying Tables pre-sented above for a compact core, a core with one water-filled hole next to the fuel, and a core with more than one adjacent water-filled holes next to the fuel. It can be noted in the following results that the magnitude of, the temperature peaking increases with the U-235 loading (see Table 1 for the pertinent fuel loadings).
PAGE 13 l
x Table 5. Temperatures calculated for LEU (20-20) Fuel in Compact Core.
^ E
, (heat content) (heat c ntent) TC C T( C) (watt-sec/cc) (watt-sec/cc) (adiabatic) measured 800 2935 1618 530 600 900 3493 1926 600 680 1000 4093 2257 670 760 1025 4250 2343 690 780 1050 4410 2431 705 800 1100 4735 2611 740 840 1150 5070 2795 775 875 1200 5417 2987 810 915
^ 0.82 TC *
- 1 39 x 1.07 For this calculation, we have
= 1.39 where P is the peak power density, and P
P is the average power density in cell; TC = 0.82 where P TC is the. power density at the thermocouple P location; and h hT = 1.07, which gives the temperature correction of (P/P) 23 C the peak-to-average ratio.
PAGE 14
a .
Table 6., Temperatures Calculated for LEU (20-20) Fuel Next to Single
~
-Water-filled Hole.
^ E (heat content) (heat ntent) TC C T(OC) (watt-sec/cc) (watt-sec/cc) (adiabatic) measured 800 2935 1317 450 510 900 3493 1567 515 580 1000 4093 1836 580 655 1025 4250 1907 595 670 1050 4410 1978 610 690 1100 4735 2124 640 '
725 1150 5070 2274 670 755 1200 5417 2430 700 795 Ex
- ETC 1.75 x 1.07 For this calculation, we have
_ = 1.74 where ' is ,the peak power density, and P
P is the average power density in cell; TC = 0.84 where P TC is the power density at the thermocouple P location; and
= 1.07, which gives the temperature correction of (P/P) 23 C the peak-to-average ratio.
TC* TC(adiabatic) x 1.13 from TAC code calculations.
PAGE 15
Table 7. Temperatures Calculated for LEU (20-20) Fuel Next to Three-Element Water-filled hole.
^
E
, (heat content) (heat ntent) TC(
T( C) (watt-sec/cc) (watt-sec/cc) (adiabatic) measured 800 2935. 1178 420 475 900 3493 1402 475 535 1000 4093 1642 530 600 1025 4250 1705 545 615 1050 4410 1769 560 635 1100 4735 1900 590 665 1150 5070 2034 620 700 1200 5417 2173 650 735
^ 0.82 TC = E x 1.91 x 1.07 For this calculation, we have P
-- = 1.91 where P is the peak power density, and P
P is the average power density in celli
- TC = 0.82 where P TC is the power density at the thermocouple P location; and (P/ ) high T
= 1.07, which gives the temperature correction of (P/P) 23 C the peak-to-average ratio.
TC* TC(adiabatic) x . 3 kom TAC cme calculadons.
'PAGE 16
Table 8. -Temperatures Calculated for LEU (30-20) Fuel in Compact Core.
E TC **
, (heat content) (heat content) TC(o } TC(o T( C) (wa tt-sec/cc) (watt-sec/cc) (adiabatic) measured 800 2915 1374 470 530 900 3468 1635 530 600 1000 4062 1915 595 670 1025 4217 1988 610 690 1050 4374 2062 630 710 1100 4696 2214 660 780 1150 5029 2371 690 780 1200 5371 2532 725 820
^
0.76 TC = E x 1.55 x 1.04 For this calculation, we have P
-- = 1.55 where P is the peak power density, and P
P is the average power density in cell;
_TC = 0.76 where PTC is the power density at the thermocouple P location; and (P/P) high T
= 1.04, which gives the temperature correction of (P/F) 23 C the peak-to-average ratio.
TC* TC(adiabatic) x . 3 from TAC code calculaMons.
PAGE 17
Table.9. Temperatures Calculated for LEU (30-20) Fuel Next to Single Water-filled Hole.
A
{
) }
(heat content) (heat ntent) TC(
T( C) (watt-sec/cc) (watt-sec/cc) (adiabatic) measured 800 2915 857 330 375 900 3468 1020 375 425 1000 4062 1195 420 475 1025 4217. 1240 435 490 1050 4374 1286 445 505 1100 4696 1381 470 530 1150 5029 1479 495 560 1200 5371 1580 520 590 0.78
- ETC - E x 2.55 x 1.04 For this calculation, we have P
-- = 2.55 where P is the peak power density, and P
P is the average power density in cell; TC = 0.78 where P TC is the power density at the thermocouple P location; and
= 1.04, which gives the temperature correction of (P/P) 23 C the peak-to-average ratio.
TC* TC(adiabatic) x 3 fr m AC eMe calculadons.
PAGE 18 i
e -
Table-10. Temperatures Calculated for LEU (30-20) Fuel Next to Three-Element Water-filled Hole.
A
{
} (
, (heat content) (heat ntent) TTC(
T( C) (watt-sec/cc) (watt-sec/cc) (adiabatic) measured 800 2915 670 270 305 900 3468 797 310 350
'1000 4062 933 350 395 1025 4217 969 360 405 1050 4374 1005 370 420 1100 4696 1079 390 440 1150 5029 1156 410 460 1200 5371 1234 430 485
^ 0.76
- ETC = E x 3.18 x 1.04 For this calculation, we have P
- = 3/18 where P is the peak power density, and 7
2 F is the average power density in cell; i TC = 0.76 where PTC is the power density at the thermocouple P location; and E
, = 1.04, which gives the temperature correction of (P/P) 23,C the peak-to-average ratio.
9 PAGE 19
_ - _ - _ . - _ - _ _ _ _ _ _ . . _ - _ _ _. ., ,~ - .
o Attachment 2 (Cont.)
Table 11. Comoarison of Measured Fuel Temoeratures for FLIP, LEU (20-20) and LEU (30-20) Fuel.
Peak Measured Temoeratures ('C) for 1150*C Maximum Fuel Temoerautre Core Configuration LEU (20-20) FLIP LEU (30420)
Compact 875 875 780 1 water-filled hole 755 650 560 22 waterdfilled holes 700 565 460 Peak Measured Temeeratures('C) for 1025'C.
Maximum Fuel Temoerature Compact 780 780 690 1 water-filled hole 670 575 490 22 water-filled holes 615 500 405
It was mentioned in Section 1.0 above that TRIGA LEU fuel with (20-20) and (30-20) uranium loadings have been designed to be one for-one replacements for TRIGA FLIP fuel. Noted in Section 3 0 above is also the fact that the Safety Limit for these LEU fuels is 1150*C as for other TRIGA fuel depending, as it does, only on the temperature characteristics for hydrogen in zirconium hydride. In view of this state of affairs, the qualification tests for TRIGA LEU have consisted of a four pronged program:
- 1. neutronic calculations;
- 11. metallurgical investigations of the fuel matrix; 111. operational tests at GA Technologies with selected LEU fuel elements; and iv. an extensive high burnup test at ORR of 19 LEU fuel pins and subse .
quent Hot Cell examination.
The guiding idea underlying the TRIGA LEU qualification program is the fact that neutronically TRIGA LEU fuel behaves the same as the FLIP fuel. Any varia ation in behavior will be mainly due to differences in the contained quantity of U-235 and, to a lesser extent, to differences in contained U-238. Conse-quently, the main areas of investigation have concentrated on performance chard.
acteristics related to the main difference: namely, the very much greater loading of uranium metal in the U4ZrHx matrix.
PAGE 20 j
- 4 ' .-
Attachment 2 (Cont.)
Comparisons of the neutronic calculations have already been presented above. In Taole 1, two important neutronic characteristics of the TRIGA LEU fuel: namely (1) the prompt negative temperature coefficient of reactivity and (2) the burnup lifetime, have been compared with those for FLIP fuel. The fuel temperature peaking factors and the resulting relations between peak and measured fuel temperatures have been presented for FLIP fuel (in Taoles 2, 3,
'and 4) and for LEU (20-20) and (30-20) fuel (in Tables 5 to 10). Comparisons of, these ' results show that the peak-to-average temperature factors depend essentially on the U4235 fuel loading (Tables 1 and 11).
5.1 Brief Description of TRIGA LEU Fuel.
The TRIGA LEU fuel is the same basic uranium-zirconium hydride matrix with erbium burnable poison as used successfully in the FLIP fuel. The basic TRIGA LEU fuel for reactors with power levels up to 3 megawatts comes in the standard format namely, fueled diameter, about 1.5 inches; fueled length, three 5dinch sections; top and bottom graphite end reflectors; same stainless steel, Hasteloy X or Incoloy 800 cladding. The main metallurgical difference between FLIP fuel and LEU fuel is the greater concentration of uranium metal as shown in Table 12. Although the LEU 30-20 TRIGA fuel contains a significantly greater mass of uranium compared to the FLIP fuel, it is important to note the significant metallurgical consideration that the uranium occupies but a tiny fraction of the volume, only 11.2%. Even for the LEU (45-20) fuel which will also be discussed below, the zirconium hydride still occupies in excess of 80%
of the volume. From other research reactor fuel development programs, it appears that significant metallurgical problems can set in when the uranium content approaches 50 volume percent.
Table 12. Comparison of Uranium Densities and Percent Volumes for Typical TRIGA Fuels.
Type wt-%4 Enrich grams U ,
% Volume per em 8 for U Standard 81/24 20 0.5049 2.6 FLIP 81/24 70 0.5049 2. 6' LEU 20
- 20 3 722 19.5 PAGE 21 1
Attachment 2 (Cont.)
5.2 Basic Metallurgical Considerations for TRIGA LEU Fuel.
This part of the LEU program included attention to the following features:
Fabricability Metallographic and microprobe tests Fission product release Physical property determination Hydrogen pressure Quench testing Additional details of this program can be found in Ref (a) and (b).
The verification of f abricability involved the production of about 100 sections of uranima zirconium-hydride fuel using the standard tecnniques of induction melting and furnace hydriding. No difficulties were experienced; none were anticipated since the volume percent of uranium metal was less than 20 percent in the 45 weight percent alloy. The metallographic and microprobe analyses indicated that the microstructures , phase distribution, and homoge, neity were well within normal and acceptable limits [Ref(a), Figure 4 and pertinent text].
The fission product release was retested using samples of TRIGA LEU fuel.
The results are consistent with those obtained with lower uranium concentra*
tions [Rer(a), Figure 5] except for one surprising but useful feature. This is the fact that the release of fission products from fuel with higher concentra-tions of uranium can be significantly reduced compared to the results for fuels with lower uranium concentrations. The explanation of this observation is related to the tendency for . micro-segregation of the contained uranium in fuel containing higher loading. See Chapter 5 in Ref (b). These tests confirm that the large retention of fission products still continues to be a major contribu-tor to the safety of the TRIGA fuel
- moderator system.
The physical properties of the LEU fuel matrix were examined experimen-tally. The thermal dif fusivity (a) is of considerable interest and shows little variation with uranium metal concentration as shown in Figure 6 of Rer(a). The thermal expansion coefficient was measured for (45-20) fuel. For an element with maximum power density, the maximum radial expansion would be about 0.6% for 45-20 fuel compared to 0.5% for standard fuel. Two other physical features were investigated (hydrogen pressure buildup and quench testing). Hydrogen pressure characteristics were d termined from close monitoring of the production hydriding process with LEU fuel [Ref (b)]. The results confirm that the equilibrium hydrogen dissociation pressure of the fuel depends only on the H/Zr ratio and the fuel temperature and is independent of the uranium content. The quench tests demonstrate the same resilience to quenching from 1200*C in water as found for earlier TRIGA fuel tests. Figures 8 and 9 in Reference (a) show photographs of the before and af ter quench cona dition of the fuel specimen. A culmination of the mechanical characterization PAGE 22 1 J
Attachment 2 (Cont.)
of the LEU fuel was the thermal cycling both in the laboratory and in the Mark F reacter to demonstrate dimensional stability. In these tests, samples were cycled successfully througn the uranium phase change temperatures (=680*C) over 100 times in the laboratory. In addition, three LEU TRIGA fuel (two with 1/2-inch 45-20 and one with 1 1/2 in 20-20) were cycled successfully through more than 2000 power increases above 1 megawatt and exposed to more than 40 reactor pulses. Figure 10 in Reference (a) provides a summary of the early part of this successful reactor test demonstration of mechanical stability.
All of these tests on physical properties indicate the completely satisfactory performance of the TRIGA LEU uranium zirconium hydride fuel.
5.3 Soecial LEU Fuel Test Program at TRIGA Mark F.
The experimental performance tests under the R-67 license for special fuel have included not only power excursion tests but also pulse tests. The power excursion tests were noted in Section 5.2 above and involved three examples of LEU fuel elements. These elements were placed in the B-ring core location while the reactor was cycled from low power to above 1 megawatt more than 2000 times. The peak measured temperature in these LEU fuel elements was 350aC for a power level of 1500 KW. All of the LEU fuel passed the surveillance tests. It is particulacly useful to note that even the half-inch diameter fuel pins sup-ported only a*. the top and bottom were remarkably resistant to bending or bowing. During these stability tests, the slignt but acceptable . bowing that was observed for the half-inch fuel pins was obviously in response to the flux gradient at the location of the fuel pin. By rotating the fuel pin 180 degrees in the same flux gradient it was possible to reduce the bending already incurred. Further experimentation is needed to demonstrate whether full control of. fuel element bending can be achieved by judicious rotation of fuel elements. However, it may be useful to report similar observation on bending with standard 1 1/2* inch diameter fuel. For the large diameter fuel, bending is usually experienced to a lesser degree; however, the bending experienced in the very high level pulsing tests on standard fuel in the 1960s appeared similarly to be controlled by rotating the fuel in its core location.
Additional tests of the LEU fuel involved two series of pulse tests: (1) 3 LEU fuel pulsed with reactivity insertions up to $2.57, and (2) 1 LEU fuel (45420) pulsed with reactivity insertions up to $3.50.
531. Pulse Tests with Three LEU Fuel.
Three LEU fuel elements were located in the B-ring for a series of 16 pulses. The driver core for this test contained FLIP fuel in a compact array.
The FLIP fuel had undergone about one fourth burnup (=750 MWD). The three elements were:
LEU-1 1/2-inch diam 45-20 LEU 42TC 1/241nch diam 45d20 LEU-3TC 1 1/2-inen diam 20-20 The pulse series commenced with small reactivity insertions and continued with increasing insertions up to $2 57. The reactor performance as well as that for PAGE 23 b
r ,
Attachment 2 (Cont.)
the three LEU elements was monitored as usual. This means that prompt reactor period, peak reactor power and selected fuel temperatures were determined or measured experimentally. The peak measured LEU fuel temperature was 415'C, a value lower than the peak measured temperatures (about 450*C) for the FLI?
driver core.
This series of pulses demonstrated two desirable features of the LEU fuel.
The shutdown coefficient of the core was essentially unchanged by the insertion in the 3 ring location of three LEU fuel elements. As related above (Section 1.0) the peak temperature rise of the core to shut the reactor down through action of the negative temperature coefficient is still only about 450*C, much less than the 700*C permitted for special fuel. Secondly, the LEU as well as the FLIP driver fuel all passed the surveillance tests.
5.3.2 High Level Pulse Tests with LEU Fuel.
A single highly instrumented LEU fuel element (45-20) was located in the B" ring of the same compact FLIP core described above for a second series of pulses that culminated in a reactivity insertion of $3.65. This LEU fuel element was constructed especially for the test and included (1) a pressure transducer to measure the gas pressure produced in a pulse, and (2) four embedded thermocouples in place of the usual three. Instead of the usual three thermocouples located one at the center axially with the two others positioned 1-inch above and below the centerline, these four were all located at the core centerline, each about 0 3 inches from the center of the fuel element at 90 degree intervals around the fuel. Thirty pulses were performed starting with an insertion of $1.25 and ending with $3 65. The pertinent data are presented for typical pulses in Table 13 In reviewing these data it helps to recognize that the LEU element being tested has a (45-20) fuel alloy. From the data in Table 1 it can be seen that this U-235 loading of 282 grans in the element is much greater than the 137 grams in a normal FLIP fuel element. As such, the LEU fuel was expected to lead the core in peak fuel temperature, a beneficial result since one goal of the test was to observe performance of LEU fuel at or near the safety limit of 1150*C.
One of the technical problems in the series of pulses was the pulse clip-ping experienced for the largest pulses. Only one central transient rod was used. To obtain a single, long stroke insertion of the desired magnitude can, and frequently does, result in pulse clipping caused when the transient rod travel is not completed sufficiently early in the pulse development. The proba lem was overcome in these pulse tests by a variety of techniques, one of the best being to initiate the pulse from a subcritical condition.
Prior to the pulse test, calculations were performed to predict (1) the expected LEU fuel temperatures, and (2) the temperatures to be expected in the FLIP driver core. The conditions to produce a peak fuel temperature of 1150*C in the (45-20) LEU fuel would result in a peak measured LEU fuel temperature of 7504800*C. The peak measured FLIP fuel temperature would be 600"650*C. From the data in Table 13, we see that it was not possible to insert enough reactiv4 ity to produce the desired fuel temperatures. However, the LEU (45-20) peak measured fuel temperature of about 700*C was only 50-100*C below that required PAGE 24
)
Attachment 2 (Cont.)
to produce 1150*C in the LEU element. Thus the LEU fuel performance was tested for fuel temperatures in the range 1050-1100*C. The corresponding measured peak fuel temperature in the FLIP fuel was about 540'C, well below the tempera-tures permitted for routine operation.
Table 13 also presents representative data on the measured pressure rise.
As in all previous tests, the very first pulse (even with very low reactivity insertion) results in a modest, measurable pressure pulse. In this case, the pressure rise was 8.5 psi. As usual, the very next pulse gives a small, barely measurable pressure pulse. Even for the largest subsequent reactivity inserd tion, the pressure pulse was only 1.8 psi. The pressure rise during the first pulse is interpreted as due to the original air filling within the fuel clad.
The temperature rise during the first pulse is sufficient to cause chemical reaction between the fueled alloy and the air filling (producing either nitrides or oxides on the fuel surface). All subsequent pressure pulses are due presumably to the residual gases within the clad. Not until pulsed fuel temperatures are in excess of 1150*C has hydrogen release been observed as was the case in the very high level pulse tests noted in Section 3 above. In that former case, the maximum hydrogen pressure was about 35 psi. Subsequently, in that earlier test a short steady state run rehydrided the fuel alloy and reduced the measured pressure to essentially zero. For TRIGA fuel temperatures at 1150*C and below (in this case 1050-1100*C) no evidence has been found in the present or earlier tests for measurable hydrogen release. The performance of LEU (45420) in this regard is the same as for all previous TRIGA fuels tested and is found to be satisfactory.
Tab.9 13 Summary of Experimental Data for High Level Pulse Tests on LEU (45420).
LEU T 'C (Ak/8 ) P(MW) 48- ES* --(---)* meas ( AP 1 2 3 4 FLIP pai (3)
I 1.25 71.7 200 200 240 277 185 8.5 2 1.50 294 282 283 275 287 263 -0 3 1.75 595 322 370 310 347 31 2 1.5 7 2.50 2374 525 557 392 532 448 2.0 11 3 00 3799 625 663 441 626 513 2.0 20 3.34 4330 673 695 471 665 541 1.60 29 3.65 4399 685 696 476 678 540 1.80 The highly loaded (45420) LEU fuel element was examined carefully through-out the series of pulses that ended with the production of 105041100*C fuel PAGE 25
Attachment 2 (Cont.)
temperat ures . All surveillance examinations were passed satisf actorily. Al-though the peak fuel temperature did not reach the postulated safety limit of 1150*C, completely satisfactory performance was observed at a sufficiently high temperature (50 to 100*C below ll50*C) to provide assurance that the LEU fuel can perform in pulsed TRIGA operation analogous to that with FLIP fuel. This conclusion is consistent with the fact noted earlier that (1) the safety limit of 1150*C relates only to the performance charact. eristics of zirconium hydride and (2) LEU TRIGA fuel elements are predominantly :!rconium hydride even for the higher loaded alloy such as the 45 weight percent fuel tested here. It may be further noted that the FLIP driver fuel elements having suostantial burnup (51/4) reached peak measured fuel temperatures of about 540*C and successfully underwent these pulse tests.
5.4 High Burnuo Tests with TRIGA LEU Fuel.
In most TRIGA reactor applications with reactor power levels up to 3 mega-watts, the long lived fuel does not reach its ultimate burnup capability.
Typical 1 or 2 megawatt reactors may achieve only 25-35 percent burnup of the I fuel and that only at the termination of a lifetime core. To gain experience with much higher levels of burnup, accelerated burnup tests were performed over a 5-year period in the Oak Ridge Reactor (ORR). Starting in 1979 a burnup test of 19 half 41nch fuel pins was conducted in the ORR faciltiy. TRIGA fuel alloys of three types (20-20, 30-20, and 45-20) were tested. By the end of the program, a burnup as high as 66% in (45420) LEU fuel was achieved. Significant data have been presented during the progress of these tests [ References (a) and (c)] and in a final report on the qualification of TRIGA LEU fuel [ Reference (d)]. ,
Two unfortunate events occurred during the ORR tests involving the inleak4 age of water with the subsequent rupture of the clad. These events did however provide some beneficial results in that early Hot Cell examination of those two fuel elements provided excellent data points on the status of the fu21 matrix af ter large but not terminal burnup. The examination of the ruptured clad:
confirmed that these were ductile ruptures and not associated with fuel-clad interactions. These unfortunate mishaps did perhaps provide the early impetus to examine more closely the metallurgy of the clad used (Incoloy-800) and the clad
- fuel interface. On this point, there is no evidence contrary to the cond clusion that the clad itself and the clad-fuel interaction are completely ac" ceptable [ Reference (c), Figures 13,14, 15, 16, 17 and associated text].
The mechanical rigidity and stability of the LEU fuel even af ter high burn-up is demonstrated. The stability of the half
- inch diameter LEU fuel even after pulsing has been noted in Section 5 3 above. The TRIGA LEU fuel elements withstood the rigors of the high burnup ORR test remarkably well. Photographs of three elements are shown in Figure 4, Reference (c). It may be noted that all the test elements were repeatedly removed from, and replaced in, the 16 element irradiation cluster during surveillance examinations performed during the ORR test program.
PAGE 26
Attachment 2 (Cont.)
Post ~ irradiation-examination of the TRIGA LEU fuel from the ORR tests were performed for three LEU fuel elements re=oved prior to completion of the test to full burnup and for selected fuel elements that successfully completed the burnup test. Several of the examinations were performed to evaluate perfor mance related to:
- effect of fission products on clad
- fuel matrix porosity.
The above items of interest are treated in considerable detail in References (b) and (d).
The migration of hydrogen in the presence of a thermal gradient is well known. See Section 2.5 in Ref (b). In steady state operation, thermodynamics of the cooling process assures that the edge of the fuel is substantially cooler than the fuel centerline. This thermocline induces long term migration of the hydrogen to the outer edges of the fuel resulting in lower hydride in the fuel center with higner hydride near the outer edges. Two major opera-tional consequences of this hydrogen might arise. First, massive hydrogen migration from the central region could reduce the hydride to H/Zent. In this case, the phase change with large resulting volume change at T >550*C would involve serious consequences for mechanical stability of the fuel. Second, the resulting increase in the H/Zr ratio at the outer fuel boundary can result in high fuel matrix pressures during a large pulse, leading to possible excessive fuel swelling, because of the hydrogen pressure
- temperature relationship. The latter becomes important mainly in pulsing heavily burned up fuel to very hign t emperatures . The excellent mechanical stability of the fuel involved in the long term burnup testa [Ref (c), Fig. 4] assures that the observed hydrogen migration is sufficiently limited even for the large burnups examined. The hydrogen distribution will actually reach an equilibrium condition for a given temperature distribution on a time scale much shorter than the burnup life of the fuel. These equilibrium hydrogen distributions have been reached in the FLIP fuel. )
The distribution of uranium before and af ter high burnup is of interest
\
since excessive micro 4 segregation could lead to local overheated regicns within the fuel matrix. The distribution of uranium with increasing fuel loadings has been studied in detail using microprobe images (See Chapter 5 in Ref (b)]. For 8 1/2 wt*% uranium loading the distribution is very uniform with uranius parti-cle size less than 1 micron. For 30 wt*% uranium, the distribution is still essentially uniform but with a tendency to concentrate at the ZrH denCrite bo undaries . The tendency to segregate increases noticeably in the (t5420) fuel but the maximwn thickress of the uranium concentrations ranges only up to 5-microns. (It may be noted that in Section 5.2 above, the somewhat larger dimensions of the uranium particles in the (45-20) fuel were considered to te reason for reduced fission product release.] Even a 5-micron dimension is att11 a small size for uranium concentration and is too small to create any problems with localized heating of the fuel matrtx. At the relatively low temperature (5600*C) experienced in the long term burnup tests in the ORR PAGE 27 m
Attacament 2 (Cont.)
reactor, no further concentration of the uranium can be expected since this temperature is considerably lower than the original melt tenperature and the long term hydriding temperatures in the neignborhood of 800'C. Inde'ed, the physical appearance of the fuel cladding and the excellent =echanical stability of all the TRIGA LEU fuel in the ORR tests both confirm tne absence of problems related to overheating due to excessive uranium segregation or redistribution.
The benign interaction of fission products with the clad has been noted earlier. Figures 13 and 14 of Reference (c) demonstrate the well known fact that the maximum range of recolling fission products is very limited. Metal-lurgical examination of the fuel-clad interraco confirms that no change occurs in the crystalline structure of the cladding aaterial (even at full burnup) as a result of the assault by recoiling fission products.
Microvoids or porosity of the fuel matrix has never been a pecblem in TRIGA fuel with the lower uranium loadings (8 1/2 wt*%) and the modest fuel burnup )
for the stancard TRIGA reactor with power levels 5 2 MW. With the higher fuel loadings in the TRIGA LEU fuel and especially for much higner burnup, the potential for increased porosity was considered. Areas around crystallite Interfaces and lattice imperfections are active as pore sitas which can attract fission product gases and possibly hydrogen. Reference (d) contains informa-tion from post irradiation examination Cf several sections of highly burned up LSU fael from the ORR tests. These showed the expected presence of micropores.
However, the micrographs confirmed the smallness of these micropcres wnich is also consistent' with the observed, very limited growth of the fuel diameters during the burnup tests. Nothing in the results indicate any basis for concern acout porosity in very nigh burned up LEU fuel [60% burnup of (45-20) fuel].
Consequently, no question whatever exists for the suitability of (20-20) and (30-20) LEU fuel for TRIGA reactors operated in the 2 to 3 MW or lower power range.
5.5. Ooerations with Commercial TRIGA LEU Fuel.
In addition to the tests set forth above on TRIGA LEU fuels, several exar*
. pies can be listed where since 1980 TRIG l. LEU fuel has been shipped for use at foreign TRIGA installations. TRIGA LEU fuel has been available commercially as a fully warranted product since 1978. In Table 14, we list the location, date and number of LEU fuels delivered. In the Taiwan application, the LEU fuel in a fourdrod cluster was used as a direct replacement for a portion of the MTR fuel bundles. For the TRIGA reactors in Thailand, Yugoslavia and Malaysia, the LEU fuel elements are replacements for existing standard TRIGA fuel (8 1/2 '
wt-%, 20% enrichment). The two reactors in the Philippines and Bangladesh are to be loaded with a full core of LEU fuel during the latter part of 1986. The entirely satisfactory operational performance of the LEU fuel as replacements for MTR fuel and standard TRIGA fuel provides substantial additional experience for TRIGA LEU fuel.
PAGE 23
~ .
Attachment 2 (Cont.)
Table 14 Commercial Applications of TRIGA LSU Fuel.
Location Reactor Number of LEU (First LEU Elements Delivered i Shipment Date) (wt%-enrichment) i Taiwan GE Reactor 44 (20-20)
(1980 Thailand GE conversion 30 (20-20)
(1980)
Malaysia TRIGA Mark II 11_(20-20)
(1982)
Yugoslavia TRIGA Mark II 10 (20-20)
(1983)
Philippines GE Conversion 100 (20-20)
(1986)
Bangladesh TRIGA Mark II 100 (20-20)
(1986) 5.6 Qualification of LEU for Use in Standard TRIGA Reactor.
The above set of discussions summarizes the highly satisfactory state of the understanding of TRIGA LEU fuel performance. TRIGA LEU fuel (45-20) is qualified for use in steady state reactors especially with power levels of 5 megawatt and above. The ORR tests confirm all aspects of its steady state performance. The qualification studies herein presented are primarily aimed at the domestic TRIGA reactor user who has facilities licensed to pulse and to operate in steady state at power levels up to 2 or 3 megawatts. Part of this presentation has been devoted to detailing the fact that the LEU long lived fuel (especially 20-20 and 30-20) follows the same pattern of behavior as the well proven FLIP fuel. The test program both at GA Technologies and at ORR has concentrated on all those areas of consideration where possible differences from FLIP fuel may arise. Every test performed in the qualification program confirmed that LEU (20-20 and 30-20) TRIGA fuel behaves at least as well as FLIP fuel. As expected, the pulsing tests that produced peak fuel temperatures near the Safety Limit were uneventful. The measured gas pressure was very low
(=2 psi) for even the highest pulsed fuel temperatures. The power cycles and pulsed operation with 1/24 and 1 1/241nch diameter fuel confirmed their mechan
- ical stability. The high burnup tests of 19 halfminch diameter LEU fuel elea ments in ORR confirmed not only the mechanical stability of the fuel elements but also provided comforting information that the metallurgy of the LEU alloy (even up to 45-20) is completely acceptable for use in routine operations of TRIGA reactors. It is also enlightening to point out that the fission densi-ties tested in the ORR irradiation were nearly 3 times those wnten will result when equilibrium burnup is reached in 30420 fuel and nearly 6 times those which will be reached in the 20-20 fuel if used in the TRIGA Mark F reactor, or any other reactor with 1 1/2 inch diameter fuel.
PAGE 29
Attachment 2 (Cent.)
The expected performance of LEU fuel in pulsed operation af ter sub4 stantial Durnup can be judged to be satisfactory from the observed splencid performance of similar FLIP fuel. As reported herein, a core of FLIP fuel witn at least one fourth turnup has been pulsed repeatedly to measured temperatures that reached up to 5AO'c (Sections 2., 5 3.1, and 5 3 2). The FLIP fuel subse-quently passed all surveillance tests. Since the physical phenomencn involved here is related directly to the zirconium hydride and since the volume percent of zLeconium hycride in the LEU fuels is more than 80%, we believe that the pulsed performance of partially turned LEU fuel will also be satisfactory. In addition, as the LEU authorized in this licence application becomes a standard fuel, it will be committed to long term steady state usage in the Mark F -(R867) reactor and its characteristics and stability will be monitored as it traces the same burnup history that FLIP fuel has now and by then will already have ccepleted. It is important to note in this regard, that there has never been a rapid, fuel related failure of TRIGA fuel with the release of fission products
.in the Mark F reacter. The failures have resulted from mechanical distortions (swelling and/or bending) which occurreJ over long periods of time or with repeated very high power pulses but have never resulted in clad failure or release of fission products.
6.0 Prompt Negntive Tem erature Coefficient of Reactivity for Long Lived fuel.
A major ccaponent of the inherent cafety of the TRIGA reactor system has always teen the large prompt negative coefficient of reactivity, r. It was a characteristic cf the origins 1 TRICA fuel (8 1/2 wtw1, 20% enriche'i) .nd due in large measure to the cell and inhomogenieties first pointed out by F. Dyson.
This so-callsd Dyson Effect is due to the relatively long range of the thermal neutron (-2 cm) in standard TRIGA fuel. These thermal neutrons sample the en-vironment both in the water outside tne fuel and inside the uranium-zirconium hydride fuel element.
The first long lived TRIGA fuel was the FLIP fuel. Because of tne heavy +
loading of U"235 plus er4tum poisco, the range of Enersal neutrens is reduced to a few millimeters. The thermal neutrons can no longer sample the environ-ment inside and outside tnis new fuel type and the Dysoc Effect is thus severely reduced. To r.aintain the lar5e value of o, a specially selected burn-able poison erbium was enomen, one that has a broad, large resonance at about 0.5 eV. The uniform distribution of this burnacle poisen throughout the fuel element assures that all the thermal neutrons (wnich have short ranges) will experience the enhanced absception within the arbium rescnance for neutrons with a hardened energy spectra. The success of tne resonacco in erbium to maintain the large value for a ~
for long lived TRICA fuel is indicated ty the values in Table 1.
The newer long lived TRIGA fuel is the LEU family of LEU fuel elements.
For the LEU (20-20) and LEU (30420), the values of a are comparable to that for FLIP (see Table 1). The mechanism for maintaining the magnitude of this inpor*
tant component of reactor safety is exactly the same as that in FLIP fuel. 1 However, the precise quantity of erbium to accenplish the desired effect PAGE 30
Attachment 2 (Cor.t.)
differs depending upon the desired performance characteristic fee t4e fuel (such as core lifetime). As already noted in Table 12 for LEU (20-20) and LEU (30-20) fuels, the :lrconium hydride occupies 89 percent er more of the fuel volume so the mecnanica (thermal neutron scattering by irconium hycride) to harden the thermal neutron spectra remains essentially the same as in the FLIP fuel. As a result of these considerations the mechanium fce, and the magnitude of, a is essentially the same as for the qualified FLI? ruel.
Table 15 has been prepared to exhibit the components oi" 2 for tne various TRIGA fuels. The value of a is temperature dependent for the long lived fuels
[ FLIP, L2U (20-20) and LEU (30-20)] out is nearly constant for the standard TRIGA fuel. For comparison purposes, the temperature deoendent values of a were evaluated by averaging over a fuel temperatyre operating ratige of 23 to 700*C.
Table 15. Components of Prompt Negative Temperat' ire Coefficient of Reactivity (a = b ,f- ) x 10+s Standard FLIP, Steel Clad LEU (20-20) LEU (30-20)
TRIGA Fuel Water Reflected Water Reflected Water Reflected (23-700*C) (23-700*C) (23-700*C)
Cell 5.5 58% 8.5 81% 7.0 67% 4.5 46%
Doppler 2.0 21% 1.5 14% 3.0 29% 35 441 Leakage 2.0 21 % 0.5 55 0.5 5% 0.0 --*
9.5 10.5 10.5 8.0
- 7. Bases for Proposed Technical Specification.
The information presented below is a discussion of, and basis for, the Technical Specification limits set forth in Section 9.2.2(d), (e), and (f)
(ATTACHMENT 1) for temperatures measured by thermocouples located in standard FLIP and standard LEU (20420) and LEU (30-20) fuel elements. The procedures used for administering this Technical Specification are also presented below.
The temperature limits established in Sections 9.2.2(d), (e), and (f) are taken from Tacle 11 and are to be interpreted as follows. In a compact core composed of FLIP fuel elements, the maximum allowable measured fuel temperature PACE 31
Attachment 2 (Cont.)
sna11 De 780*C. Similarly, for LEU (20-20) and LEU (30-20) the limiting mea-suced fuel temperature snall te 780*C and 690*C. If there is a single water-filled hole anywhere within the core, then the maximum allowable measured fuel temperature for any FLIP tiement adjacent to this flux trap shall be limited to 575'C. All otner FLIP fuel elements in this core not adjacent to the water" filled region shall have their temperatures limited to 780*C since the peak-to-average power generation for tnese elements is the same as for a compact core.
For LEU (20*20) and LEU (30-20) fuel, the corresponding peak measured fuel temperature for fuel adjacent to a single water-filled hole shall be 670*C and 490*C , respectively. All other LEU fuel in this core shall have their temperatures limited to 780*C (20-20) or 690*C (30-20). If there is a flux trap composed of two or more adjacent water 4 filled holes, the maximum allowable measured temperature (cr any FLIP element adjacent to the water-filled region shall be 500*C. All other FLIP fuel elements in this core not adjacent to this water-filled region shall have their temperatures limited to 780*C. For a flux trap ccaposed of two ce more adjacent water-filled holes, the corresponding maximum allowable measured temperature for any LEU (20420) or LEU (30-20) fuel element adjacent to the va er-filled region shall be 615'C and 405'C, respec-tively. All other LEU fuel elements in the core not adjacent to this water-filled region shall have cheir temperatures limited to 780*C (20*20) and 690*C (30420).
In the event of a temperature scram when a water
- filled region is present within the ccre, it is not clear a priori whether the scram will be caused by the element in One comptet region or the one adjacent to the flux trap. It will depend heavily upon wntre the water-filled region is located within the ccee. For the purposes described herein, a water-filled hole in the outer-most fuel ring or a larger w3ter-filled region in the two outermost fuel rings will not be considered a flux trap for the purposes of establishing maximum permissible fuel temperatures.
7.1 Procedure for Determining Hottest Fuel Element.
The hattest fuel element shall be located experimentally by inserting an instrumented fuel element successively into a number of likely core locations, operating tha reactor L: a given steady 4 state power with the transient rod (if present) withdrawn and the other rods banked, and measuring the fuel element temperature. The power level shall be high enougn to produce a fuel tempera-ture measuraoly above ambient, but the power need not necessarily be the licensed maximum value. The instrumented fuel element will then be placed in that location which produces the highest fuel temperature thus determined. The position in the case of the hottest fuel element thus determined in steady state is unchanged when the reactor is pulsed because the end-of-pulse control rod configurations are very similar to that used for the steady state determi-nations described above. For those instrumented fuel elements which have their thermocouple located near the vertical centerline but at a point along the radius less than one4 half the radius measured from the center, the fuel element may have to be rotated about its vertical axis in order to ascertain the maxi-mum fuel temperature. This is because the portion of the instrumented fuel element nearest to any water
- filled flux trap will have higher power density and, hence, higher temperatures. If the thermocouple were to be located very near to the central axis of the fuel element, rotation of the fuel element about its vertical axis would not be expected to produce a measurable change in the indicated temperature.
PAGE 32
ee a Attachment 2 (Cont.)
l The transient operations to be conducted with a partial or full FLIP or LEU core will require careful attention to the permissible reactivity insertions.
l This attention is not different in any degree from that which has been required in all the previous operations both with standard and special reactor cores.
Before a pulse program starts for an untried core configuration, an estimate is made of the inserted reactivity to be required to produce the maximum allowed fuel temperature. This estimate is based on related calculations and/or past measurements on a similar core. Regardless of the magnitude of this estimated reactivity insertion for the untried configuration, the pulsing program starts with very small pulses (about $1 50) and increases the pulse size by small in-crements (perhaps 25 cents). The measured reactor performance characteristics are recorded and analyzed to assure that maximum allowable fuel temperatures are not exceeded. Using this conservative approach to pulsing previously untried core configurations together with the procedure outlined above for initially determining the location of the hottest fuel element has assured safe operation in all the R-67 reactor operations.
7.2 Delay Time for Scram in Pulsed Operation.
As has been established in all of the pulsing history for the family of TRIGA reactors, the effectiveness of the temperature scram is limited to reducM.
tion of the " tail" of the pulse. The reason is that the thermal inertia of the thermocouple results in producing the peak measured temperature reading after i the pulse has been completed. Thus, if the measured temperature should exceed the permissible limit and result in a scram, the scram would be ineffective in reducing the pulse size since the pulse would have already occurred. However, this scram would be effective in dropping the control rods if the automatic post-pulse scram had not already done so. In this case, the earlier ti=e for dropping the control rods would result in some reduction in the tail of the pulse and would reduce the total integrated power by reducing the tall of the l power transient.
73 Core Configuration with Long Lived Fuel Elements.
As implied in the discussion given in Section 7.1, the core configurations of importance are generic rather than specific to the exact core location.
With specific reference to Sections 9.2.2(d), (e), and (f), the consideration of long lived fuel, water
- filled regions anywhere in the core are of importance because of the power peaking induced in an adjacent fuel element. Calculations and measurements have demonstrated that care must be given to these fuel elements.
The sater in a single position significantly softens the neutron spectrum compared to a compact core and leads to increased power production, but only in that portion of each long lived fuel element adjacent to this water region.
Within the long lived fuel element the mean free path of thermal neutrons is always only a few millimeters. This limits the power peaking closely to the i region next to the flux trap. Since a single water
- filled hole is not quite sufficient to provide full thermalization of the neutron spectrum, a larger PAGE 33 i
Attachment 2 (Cont.)
water-filled region produces more thermalization and increased power peaking in adjacent long lived fuel. Three adjacent close packed water-filled fuel-element positions provide somewhat more water than necessary for optimum ther-malisation. Therefore, the power peaking evaluated for the optimum thermaliza4 tion is arbitrarily attributed to the presence of two adjacent waterHrilled holes and is incorporated in that fashion in the Technical Specification. As discussed above, it is required to employ at least one instrumented FLIP or LEU fuel element at the hottest location adjacent to each water-filled region with its scram set according to the Technical Specification. It is also required to install at least one instrumented FLIP or LEU fuel element at the hottest location within the compact zone with its scram set at 780*C [ FLIP or (20-20)]
or 690*C (30-20). In each of these cases, the instrumented fuel element will have a Ud235 loading equivalent to the highest loaded non4 instrumented elements in the region.
Reg,ardless of the location within the core of any water 4 filled flux traps, all fuel is protected by the considerations that will have led to the location of the instrumented fuel elements and to the scram settings noted above. In particular, if the flux trap is located far from the center of the core, the centrally located fuel in the compact zone would reacn its maximum permissible temperature limit while the temperature of the instrumented long lived element adjacent to the flux trap would be well below its limit. It is impossible to anticipate the exact location and number of water-filled regions that will be required for future in-core experiments. However, the reactor and the public are protected because the fuel at each of these important locations is fully protected by the procedures specified above.
4 PAGE 34
a* o Attachment 2 (Cont.)
- 8. References The references listed in the cover letter as enclosures are repeated here for the convenience of the reader.
Reference (a) " Current Status of Uranium Zirconium-Hydride Lcw Enriched Uranium Fuel Design and Development," by R.H. Chesworth, September 1982.
Reference (b) "The U-ZrH Its Properties and Use in TRIGA Fuel,"
M.T. SimnaN, Alloy: February 1980.
Reference (c) " Final Results of Qualification Testing of TRIGA Fuel in the Oak Ridge Research Reactor Including Post-Irradiation Examination,"
by R.H. Chesworth and G.B. West, October 1985~.
Reference (d) " Post Irradiation Examination and Evaluation of TRIGA LEU Fuel Irradiated in the Oak Ridge Research Reactor," by M.T. Simnad, G.B. West, May 1986.
PAGE 35 i