ML20084T726
| ML20084T726 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 03/06/1973 |
| From: | Sewell R CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | Oleary J US ATOMIC ENERGY COMMISSION (AEC) |
| Shared Package | |
| ML20084T717 | List: |
| References | |
| NUDOCS 8306240075 | |
| Download: ML20084T726 (4) | |
Text
I..
m U
\\'
x=::x 4
g3
/'
Y CORSum8IS 4[0h g/.
!li g
POW 8r
/g N
Company 2
3
.W u;f{ [3)P{'
S di
, Jackson. MicNgan 49201. Area Code 517 708-0550 General Offe..
12 We %c A
.u 6
e a
March 6, 1973 Mr. John F. O' Leary, Director Re: Docket No 50-255 Directorate of Licensing License No DPR-20 US Atomic Energy Commission Washington, DC 20545
Dear Mr. O' Leary:
Our letter of January 29, 1973 provided preliminary information with respect to the steam generator tube leakage that was first detected on January 15, 15773 This letter will provide additional information with respect to investigations conducted and repairs performed.
Indications of primary to secondary leakage in the "A" steam generator were observed on January 15, 1973 The leakage was subsequently confirmed and, when it became apparent that the Technical Specifications might be exceeded, the unit was removed from service and cooled down for inspection.
When the "A" steam generator was refilled (it had been drained prior to reducing the primary pressure to atmospheric to avoid any inleak-age of secondary water into the primary coolant), it became apparent that three tubes had complete penetrations of the tube wall.
)
Combustion Engineering investigative teams began an eddy-current i f testing (ECT) program in conjunction with personnel of Zetec Inc to deter-
{ mine if additional tubes had wall thinning. Eddy-current testing of tubes from the inside surface is a testing technique developed over the past ten years and reliably proven to detect the presence of tube wall defects.
Zetec Inc is considered to be the leader in this field, having field ex-JV perience with numerous steam generator field problems (including Mihama)[.
7 Electronic signals generated by variations in magnetic permeability of the tube wall are compared to similar signals measured from artifically pro-duced defects in heat transfer tubing identical to that in the Palisades steam generators. The signals are interpreted by highly skilled and ex-perienced technicians as to the depth of wall thickness penetration. The i
ability of this test to indicate defects of the wastage type has been demonstrated at Mihama, f
r
(
)
I 8306240075 730723 PDRADOCK05000g S
i_
l
'Y ' ~..
(
2 Mr. John F. O' Leary Docket No 50-255, License No DPR-20 March 6,1973
)
Recults of the ECT of the "A" steam generator indicated a very well-defined area of the steam generator where indications of tube wall reductions were present. Tubes tested in the first eleven rows (" rows" of tubes run parallel to the divider plats ~with Row 1 nearest the divider plate) of the "A" steam generator had varying degrees of tube wall reduc-tiony No tubes outside of this area 3ad any ECT indications of tube wall thfnning with the exception of a minor indication in one tube in Row 12.
The ECT investigation was extended to the "B" steam generator.
. Similar results were found in the first elevenyws except the number of affected tubes was significantly less and the degree of wall thinning was q much less severe. In all,1,014 tubes were inspected by ECT techniques in the "K"Meam generatbr an'd 929 tubes in the '"B" steam generator.
One of the leaking tubes in the "A" unit and a nonleaking tube with an indication of significant wall thinning were examined by a radio-graphic technique especially developed by Combustion Engineering. This technique is used to characterize the nature of the ECT indication or tube penetration to assist in evaluating the cause of the attack. Results of the radiographic examination reveal a relatively broad area of attack of varying _ depth and irregular shape'.~ ~(Ratographs show the attack area f in the leaking tube to be approximately 2"_long~and to extend over a 90
/
included angle centered at the top of the tube.) The radiographs taken at Palisades and those taken earlier at Mihama both show a general wastage attack of the nature seen on defect tubes removed from the Mihama steam generators and examined metallographically.
The defects and indications in_both steam generators all appeared in the 180-degree bend portion of short bend radius U-tubes, within a section of the bend which is enclosed by divider htrips on
~
either side (ie, the divider strips form a " wall" on either side of a line of tubes). A tight configuration exists only at the junction of three divider strips in a "Y" shape in the immediate vicinity of the first eleven rows of tubes. The divider strips separate with expanding tube row geometry outside of tube Row 11.
It is surmised that the tight compartmentation in the first eleven tube rows leads to a steam blanketing condition whereby impurities in the steam gen-erator water are concentrated.
The apparent cause of wall thinning is accelerated localized corrosion of Inconel-600. This attack is characterized by general tube wastage in a specif G a (no intergranular corrosion). It is similar to that previously identified on Incosil660 Tubes removed from the Mihama steam generators. Briefly described, the most likely mechanism is con-centration of free caustic (a fault chemistry condition identified in Q
Palisades steam generators on different occasions, most of which were near the time of the initial plant start-up) at a pseudostagnant steam water interface. Alternate wetting and drying, due to the tight com-partmentation, result in a tube _ degradation /healin'g cycle promoting a localized wastage region.
o.
- . p..
o o
3
'Mr. John F. O' Leary i
-Docket No 50-255, License No DPR-20 March 6,1973 Combustion Engineering is plugging all tubes in the_first eleven rows of.eachJteam_ generator and the one tube having an indicatToETn~Ifow 12 in the "A" steam generator. This very conservative approach to the problem terminates heat-transfer from these tubes and removes the concen-trating mechanism for any contaminant which may be present in the steam generator water. It is believed that this precaution, combined with good water chemistry control, will completely arrest the corrosion mechanism.
Extension of the affected area or the emergence of new areas which may be affected as plant power is increased to new higher levels is not anticipated. Beyond Row 11, the tube support strips are separated, which provides " ventilation" of the tube rows; therefore, a steam blanketing condition is not expected to occur. Combustion Engineering considers the abrupt halt in eddy-current indications at Row 11 to be primarily a func-tion of geometry and that, consequently, the extent of defects will be unaffected by power level.
Tube plugging operations were performed using Inconel plugs de-signed and manufactured by Combustion Engineering and seal welding pro-cedures developed and qualified by Combustion Engineering. This method uses a single-pass, no filler metal process of maximum reliability and integrity. All welders were qualified and certified by Combustion Engi-neering welding supervisors and worked under the direction of Combustion Engineering erection supervisors and Combustion Engineering nuclear compo-nents department engineers. Combustion Engineering Quality Control in-spectors performed weld inspections.
The plugging of the tubes in Rows _l-11 reduces the primary coolant mass flow rate _by_appr.oximately 2.1%. Therefore, the minimum reactor yessel_ mass flow. rate.at.2200 MW.and i800 psia will be 134.6,
~
millionIVhour.-Thiscomparestoadesignflowrateof125million
~
lb/ hour. This amount of flow reduction will slightly_ reduce the core thermal margin. Our January 3, 1973 Request for Change to Interim Special Technical Specifications requesting authorization to operate the Palisades' Plant at 2200 MWt.at reduced primary pressure assuming fuel densification-existed, documented that design criteria as-specified in the Palisades FSAR are met. As noted in the January 3,1973 submittal, the overpower -
margin to a DNBR of 13 is 125% of rated reactor power for the conditions reported in the submittal as compared to 122% in the FSAR. With the 2.1%-
flow reduction as compared to the flow assumed in the submittal, the minimum overpower margin will still be well above 122%. It is ' concluded that the design criteria as specified in the FSAR will be met.
In order to increase the effectiveness of the steam generator.
water chemistry control program, we are planning modifications to im-prove the sampling and chemical feed systems. These modifications include provisions for continuous monitoring of phosphate concentra-tions and pH. Equipment is on order to accomplish these modifications; the longest lead time' item is scheduled for delivery in about ten weeks. Much of this equipment can be installed while the plant is
.in service and will be installed as soon as possible after delivery.
y
. [1"[.
4
+ Mr. John F. O' Leary Docket No 50-255, License No DPR-20 March 6,1973 Data obtained during this repair are-still being evaluated.
When.this evaluation is completed, we will report the results of the evaluation and describe long-term surveillance programs if any are
. deemed necessary.
Yours very truly, Ralph B. Sewell (Signed)
RBS/ map Ralph B. Sewell Nuclear Licensing Administrator CC: BHGrier, USAEC l
t l-
- -. -, _.. -. _ ~.. -,.
-