[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference ssag::ask_ssag

Title:Ask the Storage Architecture Group
Notice:Check out our web page at http://www-starch.shr.dec.com
Moderator:SSAG::TERZAN
Created:Wed Oct 15 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:6756
Total number of notes:25276

4871.0. "RA73 ECM High Failure Rate" by ODIXIE::PADDOCK () Mon Oct 23 1995 10:14

T.RTitleUserPersonal
Name
DateLines
4871.1Blitz for HSC K.sdi for Paulse State parity errors.JOBURG::HARRISTue Oct 24 1995 00:395
4871.2BAHTAT::MILLINGTONPTue Oct 24 1995 10:177
4871.3RA73 ECM Capacitor C2 reversed on etch rev D, CS REV E05BOOTM1::CHANDLERCNo ProblemTue Oct 24 1995 12:1755
4871.4Pulse or State parity errors blitzSSDEVO::THOMPSONPaul Thompson, Colorado SpringsTue Oct 24 1995 12:2777
4871.5C2 is a repair centre ECO BOOTM1::CHANDLERCNo ProblemWed Oct 25 1995 08:2910
4871.6SSDEVO::KOWALLStorageWorks Engineering SupportWed Oct 25 1995 09:438
4871.7For what it's worth...TOPTEN::KELLERWed Oct 25 1995 15:2215
4871.8What was the ECO number for the Capacitor problem GIDDAY::LUSTENHOUWER$Set success=work_as_team/permMon Sep 09 1996 00:348
4871.9TINCUP::RIDGEI don't know where I'm at, but I'm nearly there!Mon Sep 09 1996 11:5247
4871.10Is this problem capacitor-related?ACISS2::MARESyou get what you settle forThu Feb 06 1997 12:5243
    Attached here is a customer description of an RA73 failure situation
    they are experiencing.  Does it sound like it is related to the
    reversed capacitor scenario described in this string???
    
    Any/all help appreciated,
    
    Randy
    
    ------
    
	We have two DEC 7000 Model 630 (3 CPU's) AXP Computers at the site.  
Each computer has eight RA73 disks (@2.0 GB) and two RA72 disks (@1.0 GB).  
While the computers were being staged here in the USA (before shipment to 
Hong Kong in May 1995), three of the RA73's failed; these were replaced by 
DEC USA under warranty.  Since they've been powered up and running in the 
field (only since May 1996), four of the RA73's have failed; in fact, a fifth 
one is now failing.  This failure rate seems excessively high to us.

	It seems that the "ECM Board" of the disk drive is most prone to 
failure, not the mechanical drive itself.  (Hmmm, that's certainly counter-
intuitive.)  DEC Hong Kong went to site, replaced the ECM board on a few of 
these, and shipped one of the bad ones back to us.  It's a DEC P/N 5421396.  
On the same side of the ECM board as the power input connector, there is a 
small toroidal transformer (L2).  This transformer appears to be burned.

	Our electronics guys, while admittedly not familiar with DEC circuit 
boards, surmise that this transformer is being used in a "chopper" circuit to 
provide a different voltage on the board.  They don't believe that this 
transformer caused the failure, but rather that this is a symptom of the real 
failure, whatever it is.  It doesn't seem that fluctuations of the AC power 
source, whether present or not, would cause such a problem on this board.  I 
believe that the disks are fed DC from the Alpha power supply.  (Is that 
correct?)  What do your technical support people say about the relative 
reliability of this drive?  What is your observed MTBF for this drive?

	In summary, we can't escape the nagging suspicion that there's a 
design flaw in this drive, which is causing these high failure rates that we 
are seeing.  We are pglanning to go to the site during the week of 17 FEB 97 
to replace some of the disks, but we'd like an answer from DEC well before 
then.  


    
4871.11yes. (better response taken offline)SUBSYS::VIDIOT::PATENAUDEAsk your boss for ARRAY's...Thu Feb 06 1997 16:210