ML20203A197
| ML20203A197 | |
| Person / Time | |
|---|---|
| Site: | Arkansas Nuclear  |
| Issue date: | 08/09/1984 |
| From: | ARKANSAS POWER & LIGHT CO. |
| To: | |
| Shared Package | |
| ML19292F311 | List:
|
| References | |
| NUDOCS 8604160252 | |
| Download: ML20203A197 (4) | |
Text
r, tt
.~
REPORT OF OTSG MECHANICAL SLEEVE 40% WALL REDUCTION TEST i
August 9, 1984
\\
8604160252 860401 PDR ADOCK 05000313 P
PDR l
i E ----
g
- g q
j i".'
9
.~
OTSG MECHANICAL SLEEVE 40% WALL REDUCTION TEST OBJECTIVE The objective of this test was to determine whether mechanical sleeves installed in the free span of OTSG tubes which have as much as 40%
I through-wall degradations have a reasonable chance of satisfying the qualification program requirements for strength and leak tightness.
METHOD This was a simplified test which used 0TSG tube specimens which had, been machined to 60% of.the original wall thickness to. conservatively represent tubing with 40% through-wall intergranular attack (IGA). Two high yield (54,700 psi) tubes were each sleeved by rolling and the assemblies were subjected to incremental axial loads while the joint slippage and leakage under 1600 psi internal pressure were measured.
RESULTS Ultimate failure of specimen QD-1 occurred i
Specimen QD-2 did not
~'
fail at was not pulled to failure.
l The joint slippage or displac'ement is plotted in Figure 1 from data in Aapendix A.
The average unpressurized displacement is plotted as well as the two individual specimen displacements. Note that any displace-ments measured as negative were attributed to measurement error, and they are all treated as zero displacements in this figure. Specimen 00-1 is considered to be representative of the expected slippage of this joint, while specimen Q0-2 is abnormal,Ae to tooling failure l
during expansion.
F l o The. joint leakage is plotted in Figure 2 from data in Appendix A.
Both individual'and average leakages are included due to the unusual leakage of specimen QD-2. Specimen QD-1 had very low leakage under nomal operating loads, and it fs considered to be more. representative of the expected leakage of this joint.
DISCUSSION AND CONCLUSIONS Specimen QD-2 was rolled in an abnormal manner,'and tfie resulting
~
slippage and leakage of this specimen are not representat.ive of a normal rolled joint. Specimen QD-1 represents the expected perfor-mance of this joint.
Thistestisquiteconservativebecausecomplekeremovalof40%ofthe tube wall eliminates'any contribution of the undegraded portion of a tube which has suffered 40% through-wall IGA.' A rolled joint in IGA tubing is expected to be significantly stronger ~and more leak-tight than these specimens.
High yield tube material was selected for this test in order to offset some of the conservatism of the machined tube.
In addition the speci-mens were rolled The resulting strengtih was found'to be more than required which indicates that average rolled joints in average yield tubes with average IGA will probably have adequate strength for both normal operating and accident loads. -Specimen QD-2 showed very high displacements at low axial loads and they are not considered to be representat.ive. The l
slippage of specimen QD-1 is more typical of rolled free span joints.
With regard to leakage, the use of a high yield tube and the hard roll both tend to make tighter joints, which helps to offset the conservatism of the very thin tube wall. The leakage of QD-2 was unusually high at normal operating conditions, decreased, and then increased as expected at accident loads.
If QD-2 leakage is' disregarded, QD-1 indicates a
, maximum leakage.
7 -
j 3
- =f 4..
s*
Technical Specification limit. At accident condi-tions the leakage would increase to Ithe Technical Specifi-cation limit.
The a'bnormal roll of specimen QD-2 was due to failure of a tooling adapter.
An expansion tool
' was adapted for use in these short specimens by inserting temporary spacer sleeves between the specimen and the stop collar of the tool. During
- expansion one spacer sleeve telescaped inside another so that the stop collar was ineffective and let the rolls walk about 2" too deep and into the clamped area of the supporting fixture. When thi; was discovered, that expansion was rerolled to the normal stop and the upper expansion was rolled as us,ual.
The result was an obround upper roll with creases where the edge of the clamp block dug into about 70% of the circumference. The resulting strength and leakage of this specimen are quite irregular, b.ut they are reported for information.
As.a general conclusion, the strength and leakage of these conservative free-span rolled joints indicates that mechanical sleeves can be success-fully installed in thin-wall tubes.
If metallurgical examination shows no unacceptable damage to an expanded IGA tube, then a ful-1 qualification program should be undertaken to verify the performance of mechanical sleeves in IGA tubing.
O e
6 A.
1