spearjl%a1.missle.mrouter@AFSC-SSD.AF.MIL ("Capt Jon L. Spear") (07/07/89)
I'd appreciate any experience, suggestions, or comments from IBM users out
there. Have you seen or heard of this sort of problem? What did you do
about it?
I don't subscribe to these lists, so please reply directly to me at:
spearjl@afsc-sdx.af.mil (preferred address)
spearjl@arecibo.aero.org (backup address)
jspear@gryphon.com (usenet address)
If there is interest, I'll summarize the responses to the mailing lists.
The problem: On a project I'm involved with, we've had actual head crashes
(head disk interaction or HDI in IBM parlance), or excessive error rates
indicating impending head crash, in six 3380 DASD units.
IBM diagnosis: The proposed mechanism is that atmospheric amines react with
the 825 lubricant coating the platters to create a grease, raising the
surface and making the disk thicker. The buildup causes contact with the
head. Potential sources of amines include: descalers used in humidifiers,
plasticizers for paints, and curing agents for epoxy's. We've heard of a
couple sites around LA that report similar problems.
Action so far: IBM analyzed a crashed platter and found some amines there.
They also took an air sample, but no amines found. We're in the process of
getting an independent analysis of the air and surfaces near the problem
DASDs. Other DASDs in the same environment, some older and some newer, have
not shown any problems, but we want to ensure that they don't.
Thanks for your help,
Capt Jon Spear
Voice: AV 833-1402, (213)643-1402
Paper: SSD/CWCDR, PO Box 92960, LAAFB CA 90009LDW@USCMVSA.BITNET (Leonard D Woren) (07/12/89)
I thought that IBM had replaced all of those HDAs by now. It was my understanding that all HDAs built in some 6 month time frame a few years back (1983?) were going to fail as you described. IBM generally won't admit to this unless you already know and confront them with it. At my last company, when we finally got them to admit it, IBM brought in a loaner AD4/BD4 and a 3880, and we moved 8 volumes a day twice a week, for them to work on 8 HDAs a week. I think this went on for 6 weeks. Some of the HDAs were replaced, on others, just the air dams with deteriorating glue were replaced. Leonard D. Woren Senior MVS Systems Programmer <LDW@USCMVSA.BITNET> <LDW@MVSA.USC.EDU> University of Southern California Disclaimer: doesn't everyone know that people on the net speak only for themselves?
TYJ@SLACVM.BITNET (Ted Johnston) (07/12/89)
In-Reply-To: JXH -- 07/11/89 13:04 Because of plenum design it was only the back (I think) HDA that had to be replaced. We still have HDAs from that era that have had no problems. Replaced about 5 or 6 of them in 1983 1984 or somewhere back around there.