ML19329D644

From kanterella
Jump to navigation Jump to search
Rept on Replacement Fuel Assembly 3A33.
ML19329D644
Person / Time
Site: Crystal River Duke Energy icon.png
Issue date: 03/26/1976
From:
FLORIDA POWER CORP.
To:
Shared Package
ML19329D642 List:
References
NUDOCS 8003160268
Download: ML19329D644 (8)


Text

a

(\ ,

'ss!

REPORT ON REPLACEMENT FUEL ASSEMBLY 3A33 1

March 26,1976 8 0 03160A gy

INTRODUCTION During the fuel receipt and unloading activities of November 8, 1975, fuel assembly 3A33 fell to the floor due to the failure of the new fuel handic.ig tool sling (License Event Report sent November 20, 1975). Because of the damage to the assembly, it was returned to Babcock and Wilcox's Commercial Nuclear Fuel Plant for repair / replacement. Upon ins

  • action by B & W, it was decided to replace the fuel assembly with an assembi of the type now being manufactured.

Replacement fuel assembly 3A33 is idential to the remaining 176 fuel assemblies in the first cycle core loading in Crystal River 3 except for minor mechanical, thermal, hydr'aulic, and nuclear aspects. The replacement assembly consists of 40 fuel rods removed from the original assembly and 168 replacement fuel rods.

This report quantifies the differences and demonstrates that the use of fuel

.- assembly 3A33 in the first cycle of Crystal River 3 is acceptabic.

MECHANICAL The only significant mechanical difference between the replacement fuel rods and the remaining fuel rods in the first cycle of Crystal River 3 is the internal spacers and fuel pellets. Spring spacers and Zircaloy tubular spacers have replaced the corrugated tube spacers and Zirconia ceramic spacers. The newer types of spacers are representative of current production and are in operation in several B & W reactors. The fuel in the new rods is slightly different in enrichment, density, and active length as shown in Table 1.

Of the fuel rods removed from the original assembly, many rods did not show evidence of unacceptable surface defects. These fuel rods were examined by X-ray and ultrasonic testing to confirm integrity of the fuel rod and internal

, spacers. The 56 fuel assemblies in Batch 1, including 3A33, win be discharged at the end of the f!1st cycle. Evaluations of the potential for clap creep collapse, fuel-clad interaction, fuel densification, and fuel swelling have been performed on each type of fuel rod in 3A33. The results of these evaluations conclude that the 3A33 fuel rods are within acceptable design limits for the first cycle operation at Crystal River 3.

~

THERMAL-HYDPAULIC Analyses of the thermal-hydraulic performance of the replace -nt fuel rods in fuel assembly 3A33 have been performed using the as-built fuel densities, fuel enrichments and fuel pellet diameters. A comparison of the results of these analyses to those shown in the Crystal River 3 Fue_ Densification Report

, BAW-1397(1) is shown in Table 2.

-- The results of the analyses show that, except for the engineering hot channel factor, all thermal-hydraulic performance parameters for the replacement fuel rods in fuel assembly 3A33 are no more restrictive than for the fuel rods in the limiting fuel assembly in the remaining 176 fuel assemblies. The engineering hot channel factor shown as =1.2% increase in the 3A33 replacement fuel rods is primarily due to the variations in fuel enrichment.

The effects of the higher engineering hot channel factor for fuel assem14y 3A33 is discussed in the nuclear section of this report.

I

NUCLEAR Analyses of the nuclear performance of fuel assembly 3A33 in the first cycle of Crystal River-3 have been performed assuming that 3/.33 is loaded into the unique core location K-9, as shown in Figure 1. T e loading of

-3A33 into cc-e location K-9 was selected on the basis of low power peaking

and the minimization of potential quadrant power tilt. The results of these analyses are summarized in Table 3. These results show that the =19% margins

, in radial and. total power peaking for core location K-9 are more than adequate to compensate for the =1.2% higher enginaering hot channel factor associated with fuel assembly 3A33. The axial-peak, and thus the total peak, will be reduced slightly in 3A33 because of the axial zone loading of fuel enrichments in the :3A33 replacement fuel rods as identified in Table 1.

Inasmuch as the limiting parameters for 3A33 do not vary more than a few percent between the constituent fuel densities or with remaining 176 fuel assemblies in cycle 1, the fuel rods in 3A33 need not be selectively positioned within the fuel assembly and may be randomly loaded. However, as a matter of good engineering practice, the replacement and original fuel rods have been symmetrically loaded within 3A33.

Evaluations have shown that the maximum ejected rod worth and maximum stuck rod worth are not increased by the use of fuel assembly 3A33 in core location K-9, and that no technical specification revisions'are required for Cycle 1 operation.

SAFETY ANALYSIS The differences in mechanical thermal, hydraulic, and nuclear parametcrs of fuel assembly'3A33 to those parameters assumed in the safety analysis have been evaluated for effects on the results of the safety analysis. Fuel assembly 3A33 has no discernable effect on the core parameters used in the safety

, analysis with the exception of power peaking. However, 3A33 will be selectively loaded into core location K-9 and, in that location, it will not present a more limiting power peaking condition than that previously analyzed. The input parameters to the safety analysis a e therefore unchanged by the use of fuel assembly-3A33 in core location K-9 during the first cycle operation of Crystal River'3. This evaluation has thus concluded that the results of the safety analysis as shown in the Crystal Mver 3 FSAR remain valid.

-FUIL LOADING PROCEDURES Initial Fuel Load Procedure, FP-201, states that (a) prior to the move that will insert each fuel assembly into the core, the Operator will insure that the fuel assembly serial number and orientation are correct and (b) prior to lowering the fuel assembly into'the core, the Operator will vc-ify that the core position is correct. The fuel assembly move is then signed off by both the Operator and an independent checker. After the loading of all the

, fuel, a complete verification of the core will be performed by two independent

. persons. This series of checks insures that each fuel assembly will be in its prescribed location.

2 e

START UP PHYSICS TESTING In addition to al- .he above, the previously planned start up physics testing will verity the values of parameters af fecting the results of the safety analysis, thus verifying the validity of the Crystal River 3 FSAR Safety Analysis.

SUFDIARY lt is concluded from the evaluation discussed in this report that the use of replacement fuel assembly 3A33 in core location K-9 for the first cycle of operation of Crystal River 3 does not alter the FSAR analyses and will not result in:

1. An increased probability of occurrence of any accident previously analyzed, or
2. An increase in the consequences of any accident previously anal.yzed, or
3. An increased probability of malfunction of any equipment important to safety previously analyzed, or
4. An increase in the consequences of the malfunction of any equipment important to safety previously analyzed, or
5. The creation of the possibility of an accident of a different type than previously analyzed,
6. The creation of the possibility of a malfunction of a different type than previously analyzed, or
7. A reduction of the margin of safety in the basis of any technical specification.

References (1)

BAW-1398, " Crystal River Unit 3 Fuel Densification Report", B & W, October 1, 1973.

4 O

g 4

e

W TABLE 1' COMPARISON OF FUEL PARAMETERS NO.

FUEL ASSEMBLY FUEL RODS ENRICHMENT %TD STACK LENGTH, IN.

W/% U 235 3A33 h40 1.93 92.5 144 168 -[1.98 95.35 23-1/2 (Upper Zone) 1.94 90.9 95-3/4 (Central Zone) > 142-3/4 1.98 95.35 23-1/2 (Lower Zone)

Remaining Assemblias 208 1.93 92.5 144

' in Batch 1

~_ .%- w-e 4

e e

4

%v-- -wns tm p--eye,w+

TABLE 2 COMPARISON OF THERMAL-HYDRAULIC PARAMETERS 168 REPLACEMENT 40 ORIGINAL FUEL RODS FUEL RODS IN IN 3A33 add FUEL RODS FUEL ASSEMBLY IN REMAINING 176 FUEL THERMAL-HYDRAULIC CRITERA 3A33 ASSEMBLIES

1. Linear Heat Rate Limit Based on Central Fuel Melting, KW/Ft.

For Fuel Density:

a. 95.35% TD 21.46 -

~~

b. 90.9% TD 19.96 -

_c. 92.5% TD -

19.7

2. Average Linear Heat Rate, KW/Ft. 5.765 5.771
3. Average Fuel Temperatures (Stored Energy), F
a. At Average Linear Heat Rate:

(1) 95.35% TD 1285 -

(2) 90.9% TD 1327 -

(3) 92.5% TD -

1335

b. At 18 KW/Ft.

(1) 95.35% TD 2840 -

(2) 90.9% TD 3066 -

. (3) 92.5% TD- -

3110

4. Engineering Hot Channel Factor 1.026 1.014
5. DNBR Penalty Due to Fuel Densification,%* 1.9 2.9
  • These values are included for completeness; however, direct comparison should not be,made because the power spike models differ in the two calculations. The power spike model used for the replacement fuel rods ecploys the currently accepted statistical base (Fg = 0.5 and Fk is a linear function), whereas the power spike model used for the original fuel rods employs the original statistical base (Fg = l'.0 and Fk is a Gaussian distribution).

- . . TABLE 3 COMPARISON OF NUCLEAR PEAKING FACTORS CORE POWER PEAK LOCATION K-9 CORE MAXIMUM MARGIN %

  • Radial 1.34 1.60 19.4 Total 2.51 2.99 19.1
  • Margin = Peak, Core Maximum - 1_. 100% *

, _ Peak, Location K-9 _

i I

1 i .

l 6-4 .y _i a

FUEL TRANSFER

. . 'NAL e

0 O O c

O O O O o

O O O O O E

Q Q Q 'Q Q Q r O O O O O O O I

~

o O O O O O O H

O O O O O O O

!O O O W O' O

'- O O O Ol O O O

" O O O O O O N

O O O O O o

O, O O O P

R Q Q Q 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 O con 1 Rot Roo tocAtioN

[-LOCATIONOFFUELASSEMBtY3A33 gmmg m]D TY ~

f o J A\ o JL 1 lii m FIGURE 1 CORE LOCATION FOR FUEL ASSEMBLY 3A33 O

4 e

.