ML20069G958

Final Deficiency Rept Re Cracking of 3/4-inch Carbon Steel Pipe.Initially Reported on 820430.Manufacturer Installed Eddy Current Tester to Aid in QC of Pipes.Faulty Sections of Pipe Replaced

ML20069G958
Person / Time
Site:	Fermi
Issue date:	09/24/1982
From:	Wells D DETROIT EDISON CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
10CFR-050.55E, 10CFR-50.55E, EF2-59395, EF2-59395.55E, NUDOCS 8210190332
Download: ML20069G958 (3)

Text

= 1 1

\

Detroit Eclison R$5552-September 24, 1982 EF2-59395 Mr. James G. Keppler, Regional Administrator Region III

$p '

U. S. Nuclear Regulatory Ccn1 mission 799 Roosevelt Road Glen Ellyn, Illinois 60137

Subject:

Final Report of 10CFR50.55(e) on 3/4" Carbon Steel Pipe Cracking (#64)

Dear Mr. Keppler:

The final report on the cracking of 3/4" Carbon Steel pipe has been prepared to provide you with the final status of this problem.

The problem was originally reported to Mr. J. Konklin of NRC Region III by Project Quality Assurance's Mr. E. L. Thmpson, Acting Supervisor-Construction Quality Assurance, on April 30, 1982.

As previously reported, this item deals with a length of 3/4" Carbon Steel pipe that cracked during a flushing operation even though the Certified Mill Test Report showed an acceptable hydrostatic test pressure of 2,500 psi.

The manufacturer, Quanex Corporation, Gulf States Tube Division, was contacted and responded via a letter dated March 19, 1982, that the flaw appeared to be an I.D. rib caused by a score or pit in the extrusion mandrel. When the re-draw shell was extruded, this rib could crack in the subsequent cold drawing operation. The manufacturer stated that the crack would be very difficult to see in a visual inspection and, presumably due to human error, was missed on the hydrostatic test. Quanex's letter stated that an eddy current tester was being installed in the mill to aid in the Quality Control system and they were hoping to have the test in operation by mid-year. The tester will help detect flaws that could elude detection on hydrostatic tests or visual inspec-tion.

The hardness tester was being repaired and testing on the cracked pipe would be delayed until the next week. Test results would be forwarded. Quanex stated that a section taken through the flawed area indicated the metal is uniform in structure, therefore, no foreign steel, such as tool steel, was drawn into the tube wall.

1 Since our last report to the NRC dated August 27, 1982, we have received a letter frcxn the Quanex Corporation dated August 26, 1982 addressing our addi-tional inquiries. In this latest response, Quanex stated that their procedure is to inspect extrusion tooling after usage to determine if such tools should l

l

~

,0CT.14 882

i. 8210190332 820924 r PDR ADOCK 05000341 1 (U - _PDR 4u7

.g Mr. James G. Keppler, Regional Administrator September 24, 1982 Page Two EF2-59395 be repaired or scrapped. - Records of particular tools are not maintained, but it can be assumed that the mandrel in question has been either repaired or scrapped.

To our question, "Did you report the tube in question to the NRC under 10CFR Part 21?" Quanex responded that they did not report the tube, since they interpret 10CFR Part 21 as not applicable to ccmmrcial grade products and the pipe in questicn was furnished as ccmtercial grade ASIM/ASME SA 106 Grade B pipe. They went on to state they do not furnish material specified to ccmply to 10CFR 21.

Quanex also stated that the micro-hardness test in .the cracked or split area read 192 KNOOP @ 500 G load and a reading of 165 KNOOP away frcm the c 'fected area. This hardness difference would be due to more cold work caused /

excess material frcm the I.D. rib of the extruded shell.

In regard to the eddy current tester, Quanex stated that work has proceeded on the installation of the tester, however, they are still 60 to 90 days away frcm ccupletion. In the interim, they are insisting that all quality control perscnnel be more observant to preclude repetition.

As far as the Enrico Fermi 2 job site is concerned, this 3/4" schedule 80 SA-106.

Grade B Fipe was originally ordered as Quality Invel I ASME Section III Class B by Detroit Edison on April 27, 1981 frcm Marnon/ Keystone of lemont, Illinois. '

A total of 200 ft. was received on site with acceptable Certified Material Test Reports frcm the manufacturer, Quanex Corporation, Gulf States Tube Divisicn traceable to the Heat Number JB 6311.

The section of pipe that was found with the crack, has been cut out and replaced with acceptable material. Detroit Edison Engineering has taken the position to use-as-is all other pipe provided by this manufacturer and installed in other locations since it will be field hydrostatic tested. Any other test failures will be docunented as required on separate Nonconformance Reports.

Therefore, this deviation is no longer considered to be a condition that would create a substantial safety hazard and no further Corrective Action is required.

If you have questions concerning this matter, please contact Mr. G.M. Trahey, Assistant Director-Project Quality Assurance.

Very truly yours, n

/y"

• l Il -

J DAW /DF/cp n y- c= v ,-

.. p

. ( -.

Mr. . James G. Keppler, Regional Administrator September 14, 1982 Page.Ihree EF2-59395 cc: Mr. Richard DeYoung, Director Office of Inspection and Enforcement Division of Reactor Inspection Programs Washingtcn, D.C. 20555 Mr. Bruce Little, Senior Resident Inspector U.S. Nuclear Regulatory Ccmnission 6450 North Dixie Highway Newport, Michigan 48166

• .- . , - ,