ML20049J327
| ML20049J327 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 03/04/1982 |
| From: | ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY |
| To: | |
| Shared Package | |
| ML20049J325 | List: |
| References | |
| CEN-193(B)-NP, CEN-193(B)-NP-S01-NP, CEN-193(B)-NP-S1-NP, NUDOCS 8203150062 | |
| Download: ML20049J327 (18) | |
Text
I s
b CEN-193(B)-NP Supplement 1 -NP t
PARTIAL RESPONSE TO NRC QUESTIONS i
ONCEN-161(B)-P, IMPROVEMENTS TO FUEL EVALUATION MODEL i
March 4, 1982 COMBUSTION ENGINEERING, INC l
8203150062 820310
~
PDR ADOCK 05000317 P
1 LEGAL NOTICE e
This report was prepared as an account of work sponsored by Combustion Engineering, Inc. Neither Combustion Engineering nor any person acting on its behalf:
A.
Makes any warranty or representation, express or implied including the warranties of fitness for a particular purpose or merchantability, with respect to accuracy, completeness, or usefulness of the information contained in this report, or that the use of any information, apparatus, method, or process disclosed in this report may not infringe privately owned rights; or B.
Assumes any liabilities with respect to the use of, or for damages resulting from the use of, any information, apparatus, method or process disclosed in this report.
I e
~
l i
i
i Question 7.
Fission Gas Release Model Question A The significant grain size dependence of the fission gas release model does not appear to be supported by the gas. release data presented. Does Combustion j
Engineering have additional data to support this dependence?
Response A Figure 7A-1 shows the differences in the FATES 3 predictions and the experimentally measured fission gas release values plotted against the fuel grain size for each of the fuel rods included in the CEN-161(B) data base and for four additional fuel rods which are described below. Figure 7A-1 shows that the differences between predicted and measurcd releases are not a function of grain size. This confirms that the C-E fission gas release model correctly represents the grain size dependence.
Four additional gas release measurements from ramp-tested fuel rods with relatively large-grain fuel have become available to C-E since the submittal of CEN-161(B).[
.]
l
[
(
] This PWR fuel was loaded into rod segments which were preirradiated through three reactor cycles at Obrigheim in a program sponsored jointly by C-E and KWU. [
]
l E
]
7-1
h E
] Both rods were examined after the ramp and determined to be sound. Each rod was punctured following the post-ramp nondestructiveexamination,andfissiongaswascollectedandmeasured.[
]
E i
i l
]
I i
I g
f l
7-2 i
i m
ye,
=swpy*---
wyz,----w-e-
--w----
+-%
,,~~-- - - - < - - - -
w
-c,=
+-r we v
-+--,+- =-
e - = - - - --==~- --
+ ' * ~
I f
l 1
Question B What is the grain size of Combustion Engineering comercial fuel and how does the model compare with gas release data for grain sizes in the range of the comercial fuel?
Response B E
]
i l
l i
l 9
7-3 t
E Z
I S
N I
A R
G SU SR EV E
S A
m E
p 1
L E
E A
Z 7
R IS S
E N
R A
I U
G A
G D
R I
F E
G R
U SA E
M D
ETC I
D E
R P
t Y$dm u<0 OE2NmE I OY9 owe s
a
?o
I Question 9.
Fuel Swelling Model Question A.
Is the total swelling value the " unconstrained swelling rate", and is the solid r
fission-product value the " constrained swelling rate"?
Response A.
i The total swelling rate, which continues irrespective of fuel-clad contact conditions, is the same as the unconstrained swelling rate, 0.40% AV/V per 4000 MWD /MTU. The total swelling rate is composed of a solid fission product component of 0.24% AV/V per 4000 MWD /MTU and a gaseous fission product component of 0.16% AV/V per 4000 MWD /MTU. The constrained swelling rate is 0.16% AV/V per 4000 MWD /MTU. The magnitude of the constrained swelling rate is coincidentally the same as the gaseous fission product component of tha total swelling rate but each represents a different phenomena. Further description of how these swelling rates are applied is given in Response C.
l l
l l
l
/
9-1 s.
1 Question B.
How is fuel-cladding contact defined for the transition between unconstrained and constrained fuel swelling; e.g., is fuel relocation included in calculating contact pressure?
l Response B.
Fuel-cladding contact transition requires assumptions to be made for the gap closure and interface models. Power history is followed in FATES 3 by quasi-steady-state burnup intervals where conditions in the fuel rod are computed for f
the end of each interval. Fuel and clad temperatures, thermal expansion, and mechanical interference loads are all iterated upon to obtain a set of conditions which are consistent with gap conditions at the end of the transition interval. Clad creep is based on an average of the mechanical loads occurring during the transition interval. Unconstrained swelling is i
assumed for a transition interval going from open to closed; constrained swelling is assumed when going from closed to open. Relocation during a transition interval is assumed to at most close the fuel-clad gap and does not result in contact pressure.
l l
l i
\\
~
l l
9-2 i
I Question C.
i Once the fuel is constrained radially, are both the radial and axial swelling calculated using the constrained swelling rate? Are there data to justify the constrained value for both the radial and axial directions? We would expect that the radial direction is under more constraint than the axial direction.
This is a safety concern mainly with regard to further loss of void volume and, hence, higher gas pressure.
l Response C.
C-E agrees that the radial direction is under more constaaint than the axial direction, and that additional axial swelling will reduce internal void volume.
Once the fuel is constrained radially, both the radius and overall length changes of the U02 fuel pellet are determined by isotropic components of the constrainted swelling rate of 0.16% AV/V per 4000 MWD /MTV. Further reduction in internal void volume is accounted for through the component of swelling which is assumed to fill the fuel rod internal void volume as stated in CEN-161 (B). This component of swelling is the difference between the total swelling rate (described in Response A) and the constrained swelling rate.
Hence, the internal void volume is reduced by a fuel swelling rate of (total swelling rate - constrained swelling rate) + (axial component of l
constrained swelling rate), or (0.40 - 0.16) + 1/3 (0.16) % AV/V per l
4000 MWD /MTU = 0.293 % AV/V per 4000 MWD /MTV.
The measurements of internal void volume taken on C-E's fuel rods in Calvert Cliffs-1, have been very close to the values predicted by FATES 3.
A detailed comparison of predicted vs. measured values is give.' in CEN-161(B).
e 9-3
7 Question D.
i Fuel swelling is a temperature dependent phenomenon, and the value used in FATES-3 is apparently based on data from low temperature fuel. However, peak linear heat ratings will provide fuel temperatures very near the transition between low-and high-temperature fuel swelling. Are there data that show the j
applicability of this swelling rate at peak linear heat ratings?
Response D.
C-E recognizes that above some threshold temperature fuel swelling will vary with temperature. A temperature-independent swelling rate which is based on data from low-temperature irradiation is used in FATES 3, however, because of the effect of swelling on the important performance parameters in application.
In FATES 3, the two most important effects of the fuel swelling are (1) the effect on the calculation of fuel-pellet dimensions for computing gap conductance and (2) the effect on the calcu!ation of internal void volume and internal pressure of a fuel rod as a function of burnup and operating history.
Considering the first effect, the use of a swelling rate based on low-l temperature data might, if anything, result in a conservative prediction of l
gap closure which, in turn, will lead to lower calculated values of gap l
conductance. As for the second effect, it is recognized that fuel swelling is l
one of many considerations in a determination of the internal void volume within a fuel rod. The four most important variables which affect the changes in internal void volume within a fuel rod are:
(1) differential thermal l
expansion of fuel and cladding, (2) permanent external volume changes in the l
fuel (combination of densification and swelling), (3) permanent cladding j
dimensional changes (creep and growth) and (4) changes in open porosity content l
of the fuel.
The overall changes in void volume result from a super-position of the contribution from these factors, and it is often difficult to separate the contribution of the individual factors.
Speci fically, swelling ef fects on void volume are complex since the low-temperature and high temperature swelling 9-4 L
l' may act in an opposite manner. For example, the low-temperature swelling (dominated by solid fission products) generally reduces void volume by causing net external volume changes in the fuel and by reducing the open porosity i
through swelling accommodation. On the other hand, the high temperature swelling is dominated by coalescense and growth of gaseous fission products which eventually are interconnected to the open surface of the fuel and to the fuel-rod plenum when these gaseous products are released. This interconnected, open-porosity network provides additional space within a fuel rod for accommodating gaseous species, and therefore acts as a positive contributor to internal void volume within a fuel rod, much like the effect of a dish at the f'
end of a pelle'..
i The above considerations suggest that an accounting of high temperature swelling may not cause a significant reduction of the internal void volume within a fuel rod.
Proprietary data on fuel-rod void volumes obtained from several Over-Ramp rods (I) support the above view.
In Figure 9D-1, void volumes measured in several fuel rods after the power ramping are plotted against the corresponding ramp terminal peak powers.
Beginning-of-life (BOL) void volumes determined from pre-irradiation characterization data for these rodsarealsoindicatedinthisfigure.[
j
]
Post-ramp void volumes calculated by FATES 3 for the Over-Ramp rods are included l
in Figure 9D-2 together with the measured data.
It is apparent that FATES 3 provides a reasonable prediction of these data, and that the nature of the prediction does not change significantly as a function of heat rating. This behavior is consistent with the physical explanation provided above that the high temperature swelling may not cause a significant reduction in internal void volume within a PWR fuel rod.
In addition to the above comparisons, FATES 3 analyses that were performed for 9-5
the high-burnup RISO rods (2),(3) also provide an additional opportunity for comp,arison of measured and predicted internal void volumes of fuel rods irradiated at high powers.
In Table 90-1, the measured BOL and end-of-life (E0L) void volumes of the five RISO rods are shown together with the FATES 3 predictions. The table also includes information on the life-time peak, rod-averaged power and the rod-averaged burnups associated with these rods.
[
]no appreciable decreases in void volumes are observed as a result of high-power irradiations.
It is also evident that the FATES 3 predictions of E0L void volumes of these high-power and high-burnup rods matches measured values quite well.
In four out of the five cases, FATES 3 slightly under predicts void volume, which is a conservative input for calculation of internal pressure.
It should be noted that the fuel temperatures experienced by the RISO rods are much higher than the fuel temperatures which occur in the lead power rod of a current design PWR.
]
The above data ' comparisons demonstrates that the swelling model used in FATES 3 is applicable for calculation of void volume and internal pressure of fuel rods operated at the peak linear heat ratings of the lead rods examined in the licensing calculations. As mentioned before, use of this swelling rate is also conservative for gap conductance calculations.
The above considerations support the conclusion that the fuel swelling rate in FATES 3 is appropriate for its intended applications.
(1) T. E. Hollowell, P. Knudsen and H. Mogard, "The International Over-Ramp Project at Studsvik, "to be published in the Proceedings of AN0 Topical Meeting:
LWR Extended Burnup-Fuel Performance and Utilization, April 4-8,1982, Williamsburg, Virginia.
(2) P. Knudsen and C. Bagger, " Power Ramp and Fission Gas Performance of Fuel Pins M20-1B, M2-2B and T9-3B, "Riso National Laboratory (Denmark) Report Riso-M-2151, December 1978. Transmitted as enclosure to J. C. Voglewede (NRC) letter to J. C. Ennaco (C-E) dated June 17,1981.
9-6
I (3) C. Bagger, H. Carlsen and P. Knudsen, " Detailed of Design, Irradiation and Fission Gas Release for the Danish 002-Zr Irradiation Test 022", Riso National Laboratory (Denmark) Report Riso-M-2152, December 1978. Transmitted as enclosure to J. C. Voglewede (NRC) letter to J. C.
Ennaco, C-E) dated June 17, 1981, i
l i
l t
I I
i 9-7
,4aw g,.,,
c-Question E.
I What effect does the new (lower), as opposed to the old (higher), swelling rate
[
have on fuel rod internal pressure?
Response E.
The lower swelling rate would decrease fuel rod internal pressure due to less filling of the internal void volume. However, the lower swelling rate would g
also result in less gap closure, lower gap conductances, and, therefore, more fission gas release which would tend to increase internal pressure.
Since these are opposing effects the net effect will depend on power history, various fuel rod dimensions, etc.
However, as indicated by the detailed discussion given in CEN-161(B), the new (lower) swelling rate is appropriate for calculations of internci void volume and gap closure.
The appropriateness of this swelling rate is clearly demonstrated in the void volume comparisons of Figure 9-5 of CEN-161(B).
In addition, the applicability of this swelling rate at peak linear heat rates has also been demonstrated in Response D.
L l
i l
9 9-8
TABLE 9D-1 A Comparison of FATES 3 Predictions and Measured Values of Internal Void Volumes for Five Nonpressurized RISO Test Rods Test Series 022 022 013, 050 013, 055, 067 013. 055, 067 Rod Identity PA29-4 M2-2C M20-1B M2-28 T9-38 Li fetime-Peak, Rod Averaged Power, kw/ft 17.8 16.8 16.1 14.4 14.9 EOL Rod-Averaged Burnup, GWd/MTU 45.4 43.1 29.9 28.7 30.4 e
Measured Internal Void da Volumes, cm3 Beginning-of-Life 2.80 2.95 3.05 2.91 3.06 End-of-Life 2.90 2.93 2.78 3.07 3.08 FATES 3 Predicted Internal Void Volumes, cm3 Beginning-of-life 2.80 2.95 3.05 2.91 3.06 End-of-life 2.75 2.88 2.91 2.73 2.89
1
=
Figure 9D-1 POST RAMP VOID VOLUMES MEASURED IN OVER RAMP RODS PLOTTED e
AS A FUNCTION OF RAMP TERMINAL PEAK POWER f
n3o 2
9 8
2 S
O>
9 O>
s 42 b
E_
O O
CI:
e e
1 PEAK RAMP LHGR, KW/FT l
l l
9-10
1 Figure 9D 2 A COMPARISON OF FATES 3 PREDICTIONS AND MEASURED VALUES OF POST RAMP VOID VOLUMES OF OVER RAMP RODS i
U d
S 8
2 3
o l
c o>
a 42 b
e_
Qo ct:
e e
O PEAK RAMP LHGR, KW/FT 9-11 l