[comp.sys.ibm.pc.hardware] Runaway Motherboard

jdalin@devildog.att.com (Jimmy Dalin) (01/25/91)

I'm posting this on behalf of my brother, who regrettably does not have
access to the Net.  

He recently purchased a 386-20 motherboard at a computer show.  (Editorial
comment: I'm beginning to think that purchasing AYNTHING at a computer show
is a bad idea.)  It worked just fine for about 6 months.  Now, however, if
he runs a program that displays the time in hours:minutes:seconds format, he
can see that the seconds tick off about three times as fast as real clock
seconds.  To complicate the situation , the problem is sporadic.  Also, the
machine hangs up shortly after the symptoms appear.

Does anyone have any suggestions.  With my limited expertise I suggested
that perhaps the crystal has gone bad.  If so then 1) can he simply purchase
a new one, unplug the old one, and plug in the replacement, and 2) how can
he recognize the crystal on the motherboard?

Thanks in advance for your help.

Jim Dalin
Attmail: !jdalin
 

berger@iboga (Mike Berger) (01/30/91)

jdalin@devildog.att.com (Jimmy Dalin) writes:

>He recently purchased a 386-20 motherboard at a computer show.  (Editorial
>comment: I'm beginning to think that purchasing AYNTHING at a computer show
>is a bad idea.)  It worked just fine for about 6 months.  Now, however, if
>he runs a program that displays the time in hours:minutes:seconds format, he
>can see that the seconds tick off about three times as fast as real clock
>seconds.  To complicate the situation , the problem is sporadic.  Also, the
>machine hangs up shortly after the symptoms appear.
*----
Does he run Stargoose?  It has a known bug in that it leaves one of the
timer interrupts running at 3x normal speed.  I'd check for a software
cause first.
--
	Mike Berger
	Department of Statistics, University of Illinois
	AT&TNET     217-244-6067
	Internet    berger@atropa.stat.uiuc.edu

cd5340@mars.njit.edu (Charlap) (01/30/91)

The problem might also be in your clock battery.  Solid-state clocks
tend to speed up as the battery dies, until they stop all together.

Try replacing it.  A Ray-o-vac computer battery costs about $9 at 
trade shows.  Mine has gone about 2 years without a problem.  Before,
when I was using lithium watch cells, the clock would run away from
me all the time.

--- dave (cd5340@mars.njit.edu)

poffen@sj.ate.slb.com (Russ Poffenberger) (02/01/91)

In article <2169@njitgw.njit.edu> cd5340@mars.njit.edu (Charlap) writes:
>The problem might also be in your clock battery.  Solid-state clocks
>tend to speed up as the battery dies, until they stop all together.
>
>Try replacing it.  A Ray-o-vac computer battery costs about $9 at 
>trade shows.  Mine has gone about 2 years without a problem.  Before,
>when I was using lithium watch cells, the clock would run away from
>me all the time.
>

The CMOS real-time clock does NOT drive the system clock while the system is
running. It is read once at boot time, then the BIOS keeps the time. Also,
the CMOS clock is usually powered by the system power supply when on, not by
the battery (this is only while turned off).

Russ Poffenberger               DOMAIN: poffen@sj.ate.slb.com
Schlumberger Technologies       UUCP:   {uunet,decwrl,amdahl}!sjsca4!poffen
1601 Technology Drive		CIS:	72401,276
San Jose, Ca. 95110             (408)437-5254