I have a vintage '83 IBM XT that I use for modem type (TNC) communications.
The motherboard has been upgraded with 640K of memory.  It orignally
had space only for 256K.  Now it takes quite a while for the system to
check memory on boot.  Is there some way I can bypass the memory test?
I have tried hitting the Esc key, etc.

36 responses total.

There is no switch on the system board to disable the test, and no key
to press to bypass it.  One possible solution is to purchase a third party
BIOS that is compatible with your XT.  I know that they exist, but have
no experience with them - I assume they follow the "Press <ESC> to bypass
memory test" trend.

The "Press ESC to bypass memory test" feature didn't come out until
the AT clones, I believe.  I don't think the original IBM AT had it,
and the XT doesn't even display a memory count as it's testing.

No, I get a memory count as it's testing.  I don't want to buy a BIOS
chip for it, so I guess I'm stuck. 

        You can leave the computer on all the time.  It's about the best
solution there is.

yep.

I have a Seagate hard disk drive that I keep on all the time.  It won't
spin up without a jumpstart (turn the spindle). It works well otherwise.

How do you turn the spindle?  I rotate the whole thing quickly to have the
disk spin.  It doesn't happen often though (once a year?)

I have one of those type of drives too.  I use a screwdriver to get it
going.  An old Western Digital 40Mb IDE.

If it is an ST-1x8 you have to stick a screwdriver between the circuit
board and the sealed chamber of ther drive and turn the spindle.
The ST-2x8 has an "E" clip on the bottom of the spindle that can be
used to rotate the spindle.

(You could always try baking it in the oven)

I just leave it turned on.

The problem with the one in our zenith is that the armature that carries
the heads gets stuck in one place when you turn the power off.  Once you get 
it jiggled out of the PARK position it works fine.
I've taken it apart several times, and thought I had it fixed, but it's
stubborn.

You might try a different park utility? whose/which one are you using?

you misunderstand.  It does that whenever the power gets turned off
or goes away.  No park utility involved.

Then, I would ADD a park utility that works with full power applied,
your current park utility works as the Power Supply DIES and the PS
may be dying too fast now for the auto-park to work right upon
power-down!

Rubbish.

Perhaps, perhaps not. Thoeretical consideration - old PSs and old capacitors
and/or increased load on PS (since new) cause the voltage to die off
much faster. How much faster I don'T know. Anything that uses the
voltage die-off time is subject to that time slice and rate of decrease
of voltage. 
 
And there is another time element required to move the head to the
park position, a time which is variable. It is quite posible that the
head moves *near* the anticipated park position, either undershoot or
even overshoot. I woould suspect overshoot myself, initially, so that
on power up, the head is not wehre it's supposed to be, and needs
a mechanical nudge ("jiggled out of PARK position") to get going
again. 
  
The otehr consideration is that upoon power up, there is a small
"initiating pulse" to "kick" the head out of park, which is designed
into several varieties of electro-mechanical hardware. If the "kick pulse"
is absent or too small, head stays "stuck." 
 
Another problem I've run into, and solved, occurs when there is a
"bumper pad" at the extreme end of travel. The pad deterioriates over
time and, while acting as a bumper, also becomes sticky - just sticky
enough to require more of an "initiating kick" than is available. This

condition is most often associated with "arm overshoot," btw.
  
There are probably other situations which could cause the head to fail
to move, but the above are the ones I've run into, and solved, most
often, in all sorts of electro-mechanical actuating systems, large and
small.

A large problem (particularly with vintage ST-251 drives) is excess
lubricant on the spindle bearings.  After a while, the lubricant
oozes out of the bearings onto the platters, and collects in a lump
near the center of the platters.  The heat from an operating drive
softens the lubricant, and the heads then get stuck there while the
lubricant cools when the drive is turned off and the heads are parked.
Sometimes a rapid twist of the drive will free the heads; other times
the "sticktion" is more serious and you have to do something like
baking the drive in an oven to unstick the heads.  (Meg Geddes really
did that once, to get data off a couple of sticky drives...)

Those Seagates make good pie filling. Mmmmm.

Forget the oven!  Just rotate the platter motor from the bottom to
break it free, then apply power and your off.  (That's part of what
I did to recove the data off meg's HD's.  The remainder of the procedure
was to allow the drives to spin over night in a small box so they would
come up to operating temp.)

Yes, and if you're really lucky you won't rip the heads off the actuator
arms when you do this, either.

Good point.  Guess I've been lucky.

   The sad thing about these drives is that there are so many of them
out there doing this!  Yucko.  I just gave some phone support to an
old friend whos ST-251 was doing just this.  The old rotate trick did
the job, at least this time.
   And no, the power supply can't be blamed for the sticktion problems.
The damn drives simply wern't made right.  Thanks, seagate, for losing
gigs of important data, all over the world.

I've never used a Seagate drive, so I don't know, but - can one get
at the bearings? Perhaps a smidgen of watch oil would be an efficient
repair. I know clocks, and proper lubrication will restore many to
health for many more years (caveat: *never* use 3-in-1, or any other
oil you can get from a hardware store).

I rock the box rotationally to move the platters.  Saves me disassembling
the box.

(Rane, where do you get "watch oil" and what brand/mixture/formula/
weight is best?)

Come to think of it, "clock oil" is probably better, for this heavier
mechanism. You get it from whales (a synthetic may be on the market now).
I have it, and if you can dissect the drive down to where the pivot
is accessible, I'd be glad to apply the requisite smidgen. I have 
smidgen applicators, also. I'm not just giving the same of a supplier
first, because there is some (slight) skill in oiling a clock (believe
it or not), which you can find in books on clock repair, or I could
convey (essentially - *very little*). That all said, one supplier is
Roland V. Tapp, PO Box 5510, Grants Pass, OR (503-471-0194). 1/2
ounce clock oil costs $2; an oiler costs the same. (Afterthought: clock
oil doesn't oxidize or become gummy, but it is made for low-shear
application - only one clock pivot rotates rapidly, and that is still
much slower than a drive pivot - I wonder what drive manufacturers use?)
,

   What is the clock oil that you mention above?  It isn't real whale
oil, is it?
   (Wanna give lessons in clock theory?)

(maybe enter a clockmakING item in the ing conference.)

(or a clockoilING item ;)

We're talking about hard-drive oiling, folks. Mechanical clocks are
obsolete, you know. I could tell you more about clock oil than you
want to know but, yes, some was made from whale oil. Say, do you
think there is enough interest to populate a clockoilING item? 
,

Yea, I like the old mechanical clocks!  I have a reproduction Regulator
School House clock that stays within a minute a week.  No big deal to
tweek the hands every week when I wind it.  I've oil it several times
over the years and oil sure makes a differance!

I've never seen a stuck Seagate, though I have seen stuck Fugi and Western
Digital drives.  I don't care much for Seagate. I have a little 2.5 inch
Seagate here that sounds like it has sand for spindle bearings!  I much
prefer the little Quantum, Conner, Toshiba and Integral.  the Integral
has a couple of 1.8" platters and measures 3" x 2" x .6" overall and holds
85 meg.  Cute!  They sound "nice" when they run.  Nice precision hum with
crisp, snapity sounding head drivers.

Seagate stopped making good reliable drives when they went to 3.5"

I don't know about that. The ST-251 and the ST4096 series were not very
reliable. 

I've never had any problems with many of the hundreds of ST4096s that
I installed.

I had one, and when the head moved, it shook my entire desk (I am not 
exaggerating). 

The ST4096 had a MTBF (mean time before failure) of only 20,000 hours or
40,000 hours (forgot which one). Compare that to today's drives' MTBF
of 100,000 to 500,000 hours (11 to 57 years). 

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss