ML20236T074
| ML20236T074 | |
| Person / Time | |
|---|---|
| Site: | University of Michigan |
| Issue date: | 07/23/1998 |
| From: | Simpson P MICHIGAN, UNIV. OF, ANN ARBOR, MI |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| Shared Package | |
| ML20236T076 | List: |
| References | |
| NUDOCS 9807270417 | |
| Download: ML20236T074 (9) | |
Text
-_- _______ _________________ ______- ___ _ _ _ _ _ __
Th3 Univ:rsity cf Michigin Michig n Mem: rill Phoenix Prrject Office of the Director 2301 Bonisteel Boulevard Ann Arbor, Michigan 48109-2100 July 23,1998 Docket 50-2
. License R-28 United States Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555 Re: Non Routine Report - Preliminary Safety Analysis of Flow Blockage in the FNR Core
Background
At 1332 on November 12,1997 reactor operators shut down the Ford Nuclear Reactor (FNR) to remove a piece of a polyethylene irradiation container that was found to be panially obstmeting flow through fuel element number 255 located in position L65, see Figure 1. The debris was approximately 1 inch by 1/2 inch in size and roughly located on top of the fuel plates in the element as shown in the attached Figure 2. Once the reactor was shutdown the staff proceeded to remove the fuel element with the blockage. Before the element was completely lifted from the core, the debris floated free. The element was returned to the fully seated position. The polyethylene debris was recovered from t!~ cool and was subsequently disposed of. Element 255 (and the rest of the core) was visually inspected to confirm that there were no funher obstructions. The reactor was then retumed to 2 MW.
The reactor operator response to this event was in accordance with OP 103, Operation at Power with regard to debris on the core. OP 103 requires that the reactor be shut down and the debris be removed if more than 25% of a single channel or more than 10% of the total flow area of a fuel element is obstmeted. The debris discovered by the operators on November 12 obtructed approximately 40% of three adjacent channels and 7% of the total flow area. The guidance in OP 103 is based on a January 18,1974 memo to the Operations Staff from the Reactor Manager at the time, Roben Manin. A copy of the memo is included as Attachment 1.
There is no indication of any damage to fuel element 255. The detection of fission products in the primary coolant is our most sensitive method of ~ iing defective fuel. There has been no evidence of fission products in the weekly prima.y coolant analyses since the event occurred..
Subsequent to this event,in a meeting held June 25,1998, the Safety Review Committee questioned whether shutting down, removing the debris, and returning to power as per our operating procedures was an adequate response to this event. The question was raised as to h; \
whether the blockage could have caused localized boiling within the element.
The Safety Analysis Report for the FNR indirectly addresses the issue of primary flow ,Ol blockage. FNR Safety Limits are based on the fuel cladding in the hot channel of the most #f
/
9007270417'900723 h l
> PDR ADOCK 05000002 l g P DR g .- ,
^
2
. con)pxt core remaining below 2350F which is the boiling point of water at the level of the core. The analysis shows that the reactor can operate at 2 MW with the primary coolant flow as low as 164 gpm before the fuel clad reaches 2350F in the hot channel. In other words, the coolant slow could be reduced by " blockage" to only 16% of the nominal flow rate of 1050 gprn before the clad will reach the boiling point of water. This thought process was the basis of the Reactor Manager's 1974 memo, assuming uniform coolant flow through the core. This model does not take into account localized flow and degraded heat transfer that develops directly behind a flow obstruction, nor does it credit margin available in channels operating at lower power levels than the hot channel.
The FNR Technical Specifications require a Non-Routine Report be transmitted to the NRC in the event of " Discovery of any substantial errors in the transient or accident ,
analyses or in the methods used for such analyses, as described in the safety analysis or in I the bases for the technical specifications." The FNR Safety Analysis Report is not in enor. f However, this report is being filed because the SAR does not directly address the issue of I localized core blockage. l I
Safety Analysis l In response to this event, the FNR staff reviewed research related to fluid flow through partially blocked narrow channels. The review identified a recent report titled " Flow Blockage Analysis for the Advanced Neutron Source" by T.K. Stovall, et. al., published by Oak Ridge National Laboratory in January 1996. In this report a 3-D computational fluid dynamics model was developed and benchmarked against experimental data obtained from a mockup coolant channel for the Advanced Neutron Source (ANS) reactor. Tests were performed with 10% and 25% central and edge blockages. Results of the expenments and model show that a recirculation region develops dimetly behind the blockage. The coolant velocity is significantly diminished and back flow occurs in this zone. The recirculation region spans the width of the blockage and extends 3 to 6 times the length of the blockage into the channel. Beyond this region normal forward coolant flow is restored. The edge block resulted in the most restrictive case. The heat transfer coefficient in the recirculation region reached a minimum that was determined to be about 1/4 of that in the region of unobstmeted flow.
Although specific FNR core parameters are not identical to the ANS tests, the following conclusions are believed to be applicable to the FNR: 1) flow through a pardally blocked channel is approximately proportional to the fraction blocked, validating the basis of FNR procedure OP 103; 2) there is a recirculation region extending several times the length of the obstruction beyond the blockage that has substantially diminished but not zero convective heat transfer characteristics; and 3) full forward flow is restored in the channel beyond the recirculation region. This is significant because the location of degraded heat transfer within the affected channel occurs in the top several inches of the core, where axial power is much less than the peak axial power value and where the coolant temperature is lowest.
In order to bound the heat transfer consequences caused by the debris discovend on the FNR core in November, a simple and conservative 1-dimensional model of a blocked l
coolant channel was developed. The model simulated an infinite linear array of channels that were partially blocked for their full length. It was based on the core and flow
3 c characteristics used for the FNR Safety Analysis Repon. A complete description of the model and the results are included as Attachment 2.
The analysis determined that a block of up to 26% of the hot coolant channel (L37) in the most compact core (25 new elements) at the minimum flow rate (900 gpm) and maximum licensed power (2 MW) could be tolerated before the clad would reach the coolant boiling temperature. Analyzing the model for a typical core of 43 fuel elements and minimum flow showed a blockage of 34% could be tolerated in the hot channel. The model was analyzed a third time for the conditions that existed when the event occurred in November - panial obstruction of L65 in a 43 element core at 1150 gpm and 2 MW. Results indicated a 43%
blockage would cause the clad to reach the coolant boiling temperature. The blockage in November was estimated to be 40%.
Based on the above evaluation, it is concluded that the clad of element 255 that was partially obstructed in November 97 did not reach the boiling point of the coolant. This conclusion is further supported by the earlier visual observation of the fuel and lack of evidence of cladding failure in pool water analyses.
~ Remedial Actions to Prevent a Recurrence The root cause of this event was lack of oversight that allowed an abandoned sample encapsulated in a polyethylene bottle to be left in the reactor pool so long (about three years) that radiation and water exposure destroyed its integrity. A method will be developed to track quartz and polyethylene samples that are in direct contact with the reactor coolant.
This will assure they will be removed from the reactor pool in a timely manner.
Also, while it is impossible to assure that items will not be dropped in the pool, OP 103, Operation at Power will be modified to instruct the console operator to immediately shut down the reactor if a dropped item threatens to fall on the core.
Good housekeeping practices will be re-emphasized.
These actions shall be implemented within thirty days of this date.
Sincerely, Philip A impson, Msistant Laboratory Manager for,-
Ronald F. Fleming, Director Michigan Memorial-Phoenix Project enc.: as stated 1
- cc: Theodore Michaels, USNRC Project Manager l
Thomas Burdick, USNRC Region III l
i
_ _ _ _ _ _ _ ___-______.__.___._.m___. -
- ' Revision 20: 030597 FUEL LOCATION DATA SHEET -
O FORD NUCLEAR REACTOR CORE LOADING
/
7,,G6 q@ t@ 7?4 tb f QN1 5h 6 7 8 9 10 w 15 w 16 di 17 z%
18 tsi 19 zal 20 h N1 7,,,,9 hh 1- Nb 25 26 27 28 29 30 g\ 305 SDk %% 1D 2[ '
35 36 37 38 39 40 3 60 c st\ pd C 32D 17% g 45 46 47 48 49 50 i
55
[\ \56 Ylb 57 58 59 bNO 60 7 1 DL 17h hk Nb 65 M 67 68 69 70 2h g 1 k TI -
75 76 77 78 79 80 C = ControlFuelElement H = Holder CR = Control Rod NS= Neutron Source E = Experiment P = Plug F = Standard Fuel Element SR = Shim Safety Rod FC = Fission Chamber Number: N3 Date: Il ND Number of FUEL MOVEMENT Entered by: MnEn 14mAJ
- v. CHE KLI TS fb!* Reviewed by: 3 -%
a q lI l
1 4
9 i
3,, 1
=
=
'95 =
= ryu
~
I
" ~ ^' '^ ~
i g
a _e -
t<i.1 s ?.
~ ,49 Q l
Ij ,
n' l
gp,# --
\ \
4 e
O
. - ,. m o o n .- . ,o m m 4
TH E UNIVERGITY OF MICHIG AN PHOENIX MEMORIAL LABORATORY To: FNR Operations Staff From: Robert D. Martin //
Reactor Manager Date: January 18, 1974 '
Subject:
Trash on Core A recent log book entry (January 17) by Talbert pointed out a deficiency in the Regulations Manual regarding any quantitative guidance to A crews re-garding objects on the core. The log book entry noted recalling numbers like blockage of 10% of element flow area and/or 25% of any one channel. These numbers happen to be quite adequate as guidelines which, if exceeded,' justify reactor shutdown to permit removal of A object.
Based on the thermal-hydraulic analysis performed for this reactor, when
' preparing for Tech Specs and other matters, it was concluded that for the most compact core we would ever consider operating, a total core flow rate of 800_gpm-would still provide adequate cooling of the core even when h reactor power '
was at or'near the scram level power. This means that a total core flow rate re-duction of 20% could be tolerated without developing boiling in the core. This 20% reduction applies as well then to individual element flow rates.
When one looks at the effect of blocking portions of a fuel element in the FNR on total flow rate, you find that the core has a virtually constant core
~
pressure. drop from flow ~. Thus, the amount of flow through any given element is dependent on the open area at the top of the element. Thus, a reduction of l 10% of h area of an element would reduce the element flow rate by that amount. - i Since we concluded ht a 20% reduction could be tolerated, the use of the 10% !
value as a guideline is reasonable yet still conservative.
When considering the amount of blockage tFat a given fuel channel can !
tolerate, it is easily shown that if coolant is available to one side of any given fuel plate, then the plate will be adequately cooled. Thus,' the selection of L
.o .'25% blockage of a single channel as a guideline is orbitrary but again conservative. !
r.
._ __-______._.____-__._.___._.b__..____
- . ,,y_ _ ___ _ ____ _- - _ _ - - _ _ _ _ _ - - _ - _ - - - - _ - _ _ - - _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ - - _ _
Clearly, however, the values of 10% and 25% are best used as guides since occurate determinations of "just-how-much" blockage, in quantitative terms, are not possible for on element in a core from o visuyl observation 25 feet away. However, visual observation con lead to o questitative judgement which, when comparad to these guides, con be the basis for a decision.
Other pertinent matters should be included in the judgement process such as:
- a. What appears to be the nature of the material causing the blockage ?
- b. If the object should move around because of flow could the situation worsen ?
- c. Does the material appear such that it might break in small pieces and get lodged within adjacent fuel channels ?
Depending on the results of these subjective judgements, a shutdown may well be justified for smaller element blockages than listed in the above guidelines.
In addition to this memo, a revision will be made in the Regulations Manual roflecting the formal adoption of these guidelines.
RDM:ll l
i l
_ - _ - _ . _ _ . _ _ _ _ __ . _ _ - . _______-_____-_-______________A