Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!ucla-cs!elroy.jpl.nasa.gov!sdd.hp.com!wuarchive!uunet!munnari.oz.au!metro!utsmips!amdahl!james
From: james@uts.EDU.AU (J Boswell)
Newsgroups: comp.sys.mac.hardware
Subject: Re: New SE/30 problem
Message-ID: <976@utsmips.csd.uts.EDU.AU>
Date: 7 Jun 91 09:36:36 GMT
References: <676094904.5@blkcat.FidoNet>
Sender: news@utsmips.csd.uts.EDU.AU
Distribution: comp
Lines: 40
Nntp-Posting-Host: amdahl.ccsd.uts.edu.au
Organisation: University of Technology, Sydney, Australia

Bill.Whitacre@p916.f421.n109.z1.FidoNet.Org (Bill Whitacre) writes:

>I sure could use some help with a perplexing problem.

>We have just got in 10 brand new SE/30's at work for use as equipment controllers.  Since these machines are going overseas we've learned to open up everything and burn it in for awhile before shipping it out the door.  In the process of doing so we've noted some disturbing behavior.

>These are SE/30's that've been upgraded to 5MB of RAM by Falcon (the sole GSA vendor for Apple h/ware!) and internal 40MB hard drives.

                             ^^^^^^^^^^^^^^ what brand etc etc

>Turning on the machines out of the box we got the question mark'd disk instead of booting from the internal HD.  After booting from a floppy and having it recognize the HD it would boot from it upon restart.  

There is a strong possibility that the hard drive is not "ready" when the SCSI
bus resets. I have had similar problems with an external drive taking too long
and the internal drive booting when the external is the selected start up
device.

If this is the case and your HD is "missing the SCSI bus" then there are some
ways around it if the mob who upgraded them cops out. Two possible H/W mods
do effectively the same thing. i) mod the logic board so the reset pulse
is longer (not recommended) or ii) put a delay on the 5 V line to the logic
board so that the 12V is slightly ahead of it, giving the drive time to spin up.
This one is not hard, a small inline inductor will work O.K.

The third option is installing more memory thereby delaying the time when the
Mac checks the SCSI bus to see who's available.

The best option is of course to get the upgrader to supply a HD that is faster
from power on to ready, usually the main factor is time to reach correct
spindle speed.

As for the other problems you mentioned check mem speed is fast enough and
compatability of s/w. I would never judge a machines performance on crashes
from shareware / PD stuff.

James

Of course I could be totally off the planet and it would'nt be the first time,
but it's worth checking out, esp. if the machine boots ok after
Restart/Reset/Shutdown etc etc. good luck