ML20211N641

From kanterella
Jump to navigation Jump to search
Insp Rept 99900401-201 on 990330-0701.Noncompliance Noted. Major Areas Inspected:Quality Assurance Program
ML20211N641
Person / Time
Issue date: 09/08/1999
From:
NRC (Affiliation Not Assigned)
To:
Shared Package
ML20211N628 List:
References
REF-QA-99900401 NUDOCS 9909130037
Download: ML20211N641 (8)
v • d • e


Text

l 4 U.S. NUCLEAR REGULATORY COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION

)

Report No: 99900401/1999201 Organization: ABB-Combustion Engineering, Inc.

Contact:

Ron Fitzgerald, Director Quality Assurance Nuclear Activity: Designs, manufactures and supplies various safety-related systems to NRC Licensees Dates: June 30 - July 1,1999 Inspectors: . Bill Rogers, Reactor Engineer Stephen Alexander, Reactor Engineer Approved by: Richard Correia, Chief Reliability and Maintenance Section Quality Assurance, Vendor Inspecticn, Maintenance and Allegations Branch Division of Inspection Program Management i

1 Enclosure 2 9909130037 990908 {

PDR GA999 EMVC-E '

99900401 PDR j

1 INSPECTION

SUMMARY

On June 30 - July 1,1999, the U.S. Nuclear Regulatory Commission (NRC) performed an inspection at the ABB-Combustion Engineering, Inc., (CE) facility in Windsor, Connecticut. Subsequent to the onsite portion of the inspection, the staff performed additional review of materials at the NRC office and performed the final inspection exit meeting during a telephone conference on August 5,1999.

The inspection was conducted to review selected portions of CE's quality assurance (QA) program, and its implementation, and the applicable programs and procedures used to design, manufacture and supply safety-related systems to NRC licensees.

Specifically, the inspectors reviewed CE's activities related to the CE qualification of Litton-Veam electrical connectors used in the Core Exit Therrnocouple system supplied to the Arizona Public Service Co. (APS), Palo Verde Nuclear Generating Station (Palo Verde) and other NRC licensees.

1 The inspection bases were:

- 10 CFR 50.49, " Environmental Quahfication of Electric Equipment important to Safety for Nuclear Power Plants"

. 10 CFR Part 21, " Reporting of Defects and Noncompliance."

2 STATUS OF PREVIOUS INSPECTION FINDINGS No previous finding were reviewed during this inspection.

3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 Backaround

! On October 24,1998, Palo Verde issued Licensee Event Report (LER)98-008 which indicated that the CE qualification reports used to document the environmental qualification of Litton-Veam electrical connectors for the Core Exit Thermocouples (CET) used in Combustion Engineering plants may not adequately demonstrate appropriate equipment qualification. The LER stated that the CE qualification testing, of a single Litton-Veam connector containing one thermocouple circuit, did not represent the installed configuration at Palo Verde which contained multiple Litton-Veam connectors wired with several circuits.

The LER indicated that the multiple connector, multiple circuit configuration was not accurately represented el ice, during a portion of the CE qualification test, moisture 2

had been present inside the Litton-Veam connector. The LER indicated that CE had not accounted for the possible system interactions caused by the moisture present in the Litton-Veam connectors which might occur with additional connectors and circuits.

The LER identified five general areas, all related to moisture intrusion, which APS had determined could potentially affect qualification: (1) system interaction related to multiple circuits, (2) system interaction related to multiple connectors, (3) maintaining the thermocouple at a single temperature during the qualification test, (4) the use of a supplemental Battery Effects Test to support qualification, and (5) the use of the Quality Safety Display Parameter Display System (OSPDS) to support qualification.

APS had determined that it would no longer use the original CE qualification reports to support the environmental qualification of the Litton-Veam connectors used for the CET system. APS had concluded that the use of an alternate Litton-Veam connector qustification report provided by the Litton-Veam company, combined with the action of replacing the Litton-Veam square section connector seal each time the connector was disconnected, demonstrated and maintained the environmental qualification of the Litton-Veam connectors used in the Palo Verde CET system.

CE had reviewed the APS LER, documented the existence of the LER and CE's conclusions in a November 4,1998, letter, and provided this letter to other applicable NRC licensees. The CE conclusion was that CE had demonstrated the qualification of the Litton-Veam connectors and that the connector seals were not required to be replaced each time the connector is disconnected.

3.2 General Scope of Inspection and Environmental Qualification Reauirements The purpose of the inspection was to determine whether CE had taken adequate actions to demonstrate and document the environmental qualification of the Litton-Veam connectors as installed in the application at Palo Verde and other NRC licensees.

The inspectors reviewed the environmental qualification documentation prepared by ABB/CENP (then CE Power Systems) (CE) for the CET system, including cables and connections, supplied by CE to APS for installation at Palo Verde as post-accident monitoring equipment (Category 1, Type C Variable) as prescribed by Regulatory Guide 1.97, " Instrumentation for Light-Water-Cooled Nuclear Power Plants to Assess Plant and Environs Conditions During and Following an Accident," Revision 2, dated December 1980 (RG 1.97) and required to be environmentally qualified by 10 CFR 50.49(b)(3).

The documentation reviewed included CE Nuclear Power Systems Report No.14273-PE 5800, " Core Exit Thermocouple - Mineral Insulated Cable Environmental Qualification Program for Arizona Nuclear Power Project Palo Verd'e Nuclear Generating Station," Revision 02, dated December 1,1987; Report No.14273-PE-5802," Income Instrument Assembly Environmental Qualification Program for Arizona Nuclear Power Project Palo Verde Nuclear Generating Station," Revision 01, dated July 30,1987; and related documents. The inspectors also interviewed cognizant CE 3

f,y=,

p l- staff, reviewed drawings 'of the tested equipment and examined samples of the tested equipment types.

L L - 3.3 ~ Review of Test Results For the General Effects of Moisture intrusion and Multiole .

Circuit Interaction L

a. Inspection Scope The inspectors performed a review to determine whether CE had adequately demon-stated that the apparent moisture intrusion into the Litton-Veam connectors observed during ~a portion of the CE environmental qualification test would not have an -

unacceptable, adverse effect on the performance of the CETs being qualified for the multiple circuit application at Palo Verde ,

- b. Observations and Findinas

The inspectors examined the actual chart recorder output from the CETs during the

[ CE environmental qualification test. The charts indicated that the signal from the -

CETs was steady and within the margin of error prescribed by item II.F.2 of NUREG-

~ 0737 as referenced in RG 1.97 (+/-22*F) with respect to the reference thermocouples throughout the CE environmental test with the exception of a few seconds of perturbation roughly corresponding to the peak pressure transient near the beginning of the loss-of-coolant-accident (LOCA) simulation.

1 The'CE environmental qualification report stated that the cable and connectors contained copper wires to simulate the presence of the ex-core leads for the rhodium detectors of the neutron flux, in-Core Instrumentation (ICl) system (a non-safety-

. related syr/em). These leads were open ended in the thermocouple oven, as they would be in ten installation, then passed through the LOCA chamber in the cables and the single Litton-Veam connector and were terminated outside the LOCA chamber at a terminal block set up to facilitate measurement of insulation resistance to ground.

l The CET element itself was grounded to its probe sheathing at the junction.

According to the report, periodically during the testing, CE measured the insulation

. resistance to ground of each of the ICI leads with all the others grounded. The

-insulation resistance eventually deteriorated to as low as 2500 ohms to ground,-

measured at 10 Vdc. Examination of samples of the stainless steel-clad and mineral-insulated, multi-conductor cable and the Litton-Veam connectors of the type used in

~

. the equipment qualification test and installed at Palo Verde, as well as the drawings used to build the test specimens, confirmed the description of the tested configuration

in the qualification report. Regardless of the low insulation resistance to ground, which was attributed principally to the moisture intrusion into the Litton-Veam connector, the CET output signal, as stated previously, was apparently unaffected.

L . The inspectors also performed a review to determine the effect, if any, of the l interaction between the CET signal and any electrical signal that may be present on l any of on the ICI wires. The inspectors determined the following: (1) the ICI rhodium L detectors are only used to provide very low amplitude (below the mV range) pulses for 4

L

an indication of power range neutron flux distribution during reactor operation, (2) the ICI signal drops off to an undetectable level after reactor shutdown as neutron flux decays into the source range, and almost immediately with the boration of the coolant, and (3) no detectable neutron flux signal would be expected from the rhodium detector circuits, in the Litton-Veam connectors, when the CETs must perform their post design basis accident safety function. In addition, the inspectors determined that any effect of the presence of the ICI wires would have been indicated during the qualification test because representative wires were installed the tested cable and connectors.

c. Conclusion

The inspectors concluded that CE had adequately configured the test specimen to represent the multiple circuits present in the Palo Verde installation. The '

documentation of the test results indicated that the connector functioned adequately, with the moisture present during a portion of the environmental qualification test, with respect to multiple circuits. The inspectors did not identify a concern in this area.

3.4 Review of Test Results Related to Multiole Connector Interaction

a. Inspection scope The inspectors performed a review to determine whether CE had adequately demon-stated that the moisture intrusion into the Litton-Veam connectors observed during a portion of the CE environmental qualification test would not have an unacceptable, adverse effect on the performance of the CETs being qualified for the multiple connector application at Palo Verde.
b. Observations and Findinas The CE qualification reports for the CET equipment at Palo Verde did not address the potential effects of multiple connectors. The tested equipment included a single 1 l

Litton-Veam connector exposed to the harsh environment of the LOCA simulation.

The Palo Verde qualification report did not contain an analysis to reconcile the q difference between the tested configuration and the configuration of the equipment l installed at Palo Verde which could contain os many as five Litton-Veam connectors l between the containment penetration and the isactor vessel head. The inspectors  ;

determined that CE had not configured the test er performed an analysis to l demonstrate that there would not be an interaction between multiple connectors, due to moisture discovered during a portion of . a environmental qualification test, and therefore that CE had not demonstrated that a singb connector would be representative of a multiple connector appMation.

j The inspectors reviewed a qualification report prepared by Southern California Edison (SCE) for the CE vessel head instrumentation connection system (including the CETs) installed at its San Onofre Nuclear Generating Station (SONGS), SCE EQ Document l Package SONGS Unit No. 2 & 3 M38382, Revision 5, dated September 6,1996. The 5

qualification report contained or referenced several CE environmental qualification documents that had also been used in support of the qualification of the CET cable and connector system at Palo Verde. The SCE document also contained analyses intended to address the limitations of those CE documents.

The inspectors determined, based on a limited review, that the SCE analysis may have had technical relevance to the issues of multiple connectors with the presence of moisture in the connector. However, CE had not prepared a review of the SCE analysis that was applicable to Palo Verde or included the SCE analysis in the Palo l Verde qualification package. ,

1

c. Conclusion I The inspectors concluded that CE had not performed suitable qualification testing of a i representative sample or performed an analysis to demonstrate that the single connector configuration, with moisture present during a portion of the environmental qualification test, would be representative of the installed multiple connector configuration at Palo Verde. This was identified as an example in Nonconformance 99900401/1999201-01. )

3.5 Review of Test Results Related to Maintainina the Thermocouple at a Sinale Temoerature Durina the Qualification Test

a. Inspection scope The inspectors performed the review to determine whether CE had adequately ,

demonstrated that maintaining the thermocouple at a single temperature during the CE environmental qualification test was representative of the installed configuration at Palo Verde considering the effect of the presence of moisture in the connector during a portion of the qualification test.

b. Observations and Findinas The inspectors reviewed the CE environmental qualification reports to determine whether CE had adequately demonstrated that testing the CET system with the temperature of the oven that contained the thermocouple element (external to the LOCA chamber) being held at a constant 400*F, was representative or predictive of the performance of the CET over the entire temperature range of 200'F to 2300*F through which the instrument is supposed to be qualified as prescribed by Table 2 of RG-1.97, considering the potential effects of the moisture intrusion.

The CE qualification reports for the CET equipment at Palo Verde did not address the effect of subjecting the thermocouple to its full temperature range during the qualification test. The testing was performed while maintaining the thermocouple at 400 F for which, at this temperature, the thermocouple would produce approximately 8 mV. Neither the environmental qualification test, nor the qualification report, accounted for any potential differences which could be experienced if the 6

ir--

~

l thermocouple was heated throughout its operating range and produced correspondingly higher voltages (at 2300*F the thermocouple would produce approximately 50.8 mV) with consideration of the effects of the presence of moisture in the Litton-Veam connector.

i The inspectors determined, based on a limited review, that the SCE analysis j (previously discussed in Section 3.4) may have had technical relevance to the issues of the system response to the full range of CET output voltage with the presence of  !

moisture in the connector. However, as stated previously, CE had not prepared a review of the SCE analysis that was applicable to Palo Verde nor included the SCE analysis in the Palo Verde qualification package.

c. Conclusion

The inspectors concluded that CE had not performed suitable qualification testing of a representative sample or performed an analysis to demonstrate that performing the I environmental qualification test with the thermocouple maintained at a single temperature, with moisture present in the Litton-Veam connector during a portion of ,

the qualification test, would be representative of the installed configuration at Palo l Verde. This was identified as an example in Nonconformance 99900401/1999201-01.

3.6 Review of the Batterv Effect Test

a. Inspection Scope The inspectors performed a review to determine whether CE used the CE Test Report No. CE NSPD-230P, Supplement 1-P, (Battery Effects Test) to support the qualification of the CET system, including the Litton-Veam connectors, and whether demineralized water was representative of the composition of the moisture that would be present during a LOCA in the Palo Verde containment and could penetrate the degraded connector seals.
b. Observations and Findinas Review of the Battery Effects Test report revealed that demineralized water and a solution of Boric Acid consistent with containment chemical spray was used during the Battery Effects Test. Further, according to the report, the two halves of the tested connector were mated while submerged in the test solutions to ensure complete penetration of the solution into the space between the connector faces and among the conductor pins. The test was intended to assess the potential severity of the possible electro-chemical effects of this moisture under the worst-case conditions, but there j 1

was no indication that the environmental qualification depended on the results of this test, nor that CE had intended to take credit for it to support the environmental qualification.

7 l

l l

n 4 '.

c. ~ Conclusion The inspectors concluded that CE had performed a supplemental test to assess the potential effects of a worst case situation but there was no indication that the test was required for the original environmental qualification nor was there indication that the supplemental test invalidated any portion of the original qualification. The inspectors did not identify a concern in this area.

3.7 Review of the QSPDS Inout Sianal Processina Alaorithm

a. Insoection Scope .

The inspectors reviewed the documentation relating to the input signal processing algorithm for the quality safety parameter display system (QSPDS) to determine whether it could be expected to accommodate signal errors resulting from the moisture intrusion into the Litton-Veam connectors which was observed during a portion of the environmental qualification test.

b. Observations and Findinas The inspectors determined that there was a potential for increased uncertainties in the CET signal resulting from moisture intrusion into the CET cable connectors. The error had been determined to be able to exceed the specified tolerance for the CET signal on individual channels on a random basis. However, review of the associated documents and interviews with cognizant CE staff indicated that the QSPDS signal i processing algorithm would be capable of rejecting erroneous signals, intermittent or sustained, from the various CETs such that the value of core exit temperature eventually displayed at any given time would be expected to represent a valid temperature. The inspectors determined that the QSPDS at Palo Verde is used for monitoring various safety-related parameters, but is not the primary instrumentation to be used by operators in executing the emergency operating procedures.

c.' Conclusions The inspectors concluded, on the basis of the review of the CET signal conditioning I l

algorithm used by the QSPDS, that there was reasonable assurance that the QSPDS could accommodate the errors that might be introduced into the CET output signals by moisture intrusion into the CET cable connectors during a LOCA and still determine and display a valid core exit temperature value. The inspectors did not identify a j concern in this area. l i

.4 .. PERSONS CONTACTED l Ron Fitzgerald, Director Quality Assurance Joseph Burger, Supervisor Field Services Ted Bernard, Senior Project Manager ,

Mike Linden, Project Manager Virgil Paggen, Licensing Engineer Ed Sirica, Consulting Engineer Reactor Equipment 8

Retrieved from "https://kanterella.com/w/index.php?title=ML20211N641&oldid=3247904"