Unstable Machines

  • Thread starter Thread starter Jethro
  • Start date Start date
J

Jethro

I am at my wit's end with these two machines. One is a Pentium 4 2.4G
on a ASUS P4B533-VM MOBO, the other an AMD Athlon 64 2.4G on a ASUS
K8V SE Deluxe MOBO. Both have 640G DDR. Both use the same mouse,
keyboard, and monitor through my KVM. Both are LAN'ed together via a
D-Link router. Both have 400W+ PSU's. Both have relatively new HDD's
as C (boot) drive.

I have had this problem before, and often I go several months before
it happens again, like now. In fact, it happens so often, that I have
to leave both towers open so as to make it easier for me to deal with
the problem.

I find that more often than not, when I boot-up either machine, it
either will not boot up at all or will boot up with a missing drive or
drives. Both machines are running with two HDDs and one DVD. If it
boots up and the missing drive is not the C boot drive, like say the
DVD drive, then I may not notice the missing drive right away, if at
all.

I have changed both IDE cabling to new ones, and it is still
happening. I have changed drive power connectors from the PSU's, by
adding Y-extenders.

Generally, I can jiggle (or re-seat) just the power connectors for the
problem drive(s), and the next re-boot will be correct, and stay that
way for a day or two.

I have not noticed loss of a drive or drives while running, but once
in a great while a machine will crash with a black screen or lose its
mouse/keyboard, so I guess it could be happening, so I can't be sure

I sure would like to try another course of action.

Since this happening on both of my side-by-side machines, I am
baffled.

Any ideas as to what I can try?
Thanks
Jethro
 
I am at my wit's end with these two machines. One is a Pentium 4 2.4G
on a ASUS P4B533-VM MOBO, the other an AMD Athlon 64 2.4G on a ASUS
K8V SE Deluxe MOBO. Both have 640G DDR. Both use the same mouse,
keyboard, and monitor through my KVM. Both are LAN'ed together via a
D-Link router. Both have 400W+ PSU's. Both have relatively new HDD's
as C (boot) drive.

What make model current on these PSU?


I have had this problem before, and often I go several months before
it happens again, like now. In fact, it happens so often, that I have
to leave both towers open so as to make it easier for me to deal with
the problem.

I find that more often than not, when I boot-up either machine, it
either will not boot up at all or will boot up with a missing drive or
drives.

What happens, were does it stop when it will not boot up at
all? Do these boards POST particularly fast, fast enough
that the HDDs might not have enough time to get up to speed?


Both machines are running with two HDDs and one DVD. If it
boots up and the missing drive is not the C boot drive, like say the
DVD drive, then I may not notice the missing drive right away, if at
all.

Check PSU voltages. Recheck bios settings. Check on bios
updates, though it is odd that it is happening to both
systems? Exactly the same situation on both? Exactly same
PSU on both?


I have changed both IDE cabling to new ones, and it is still
happening. I have changed drive power connectors from the PSU's, by
adding Y-extenders.

Generally, I can jiggle (or re-seat) just the power connectors for the
problem drive(s), and the next re-boot will be correct, and stay that
way for a day or two.

.... then haven't you found the problem?
Use a strong light to inspect the contacts on the drives and
power connectors. Use a fine tool to try to slightly bend
the power plug pins (on the PSU) slightly tighter. If it is
difficult to plug these plugs into (or remove from) the
drives, it is possible to crack a circuit board on them if
there was enough stress, or the board wasn't fastended to
the frame close enough to the power socket. Inspect the
circuit boards too as much as possible if it seems possible.


I have not noticed loss of a drive or drives while running, but once
in a great while a machine will crash with a black screen or lose its
mouse/keyboard, so I guess it could be happening, so I can't be sure

I sure would like to try another course of action.

Since this happening on both of my side-by-side machines, I am
baffled.


Is there anything else common to them perhaps running from
same UPS? I doubt that is the problem but I agree it is
odd, unless these are both using same PSU with some
particular problem (perhaps poor quality molex-clone plugs
based on what you'd written).
 
What make model current on these PSU?




What happens, were does it stop when it will not boot up at
all? Do these boards POST particularly fast, fast enough
that the HDDs might not have enough time to get up to speed?




Check PSU voltages. Recheck bios settings. Check on bios
updates, though it is odd that it is happening to both
systems? Exactly the same situation on both? Exactly same
PSU on both?

The PSUs are different brands.
... then haven't you found the problem?
Use a strong light to inspect the contacts on the drives and
power connectors. Use a fine tool to try to slightly bend
the power plug pins (on the PSU) slightly tighter. If it is
difficult to plug these plugs into (or remove from) the
drives, it is possible to crack a circuit board on them if
there was enough stress, or the board wasn't fastended to
the frame close enough to the power socket. Inspect the
circuit boards too as much as possible if it seems possible.





Is there anything else common to them perhaps running from
same UPS? I doubt that is the problem but I agree it is
odd, unless these are both using same PSU with some
particular problem (perhaps poor quality molex-clone plugs
based on what you'd written).

They are both plugged into the same surge protector - thence the same
wall socket. Might should by-pass the surge protector? I have a
whole-house protector outside anyway. You see, I was hit with a
destructive surge twice (a year apart) about four years ago - before
these machines.

Say - Do you think that leaving my cases open might hurt cooling by
the fans? After all, there is not much circulation going on. I have
measured temps though and they seem acceptable.

Jethro
 
What make model current on these PSU?




What happens, were does it stop when it will not boot up at
all? Do these boards POST particularly fast, fast enough
that the HDDs might not have enough time to get up to speed?

Boot-up fails when loss of the C boot HDD causes there not to be a
bootable drive (my other HDD is not bootable). When I check, the BIOS
shows either no C drive or no HDDs at all. After I jiggle the
connector, the BIOS detects the HDDs and boot-up works.
 
They are both plugged into the same surge protector - thence the same
wall socket. Might should by-pass the surge protector?

I doubt that's it, but might as well remove it from the
equation for the time being, doing so could only help to
isolate variables.
I have a
whole-house protector outside anyway. You see, I was hit with a
destructive surge twice (a year apart) about four years ago - before
these machines.

Say - Do you think that leaving my cases open might hurt cooling by
the fans? After all, there is not much circulation going on. I have
measured temps though and they seem acceptable.


If you think there isn't enough airflow, improve that.
Whether an open case helps can depend on the case, with a
very bad case it can help a little but with a good setup it
should have minimal, sometimes even detrimental effect. You
might open it and temporarily point a desk fan at it, temp
sensors aren't *always* right.
 
Boot-up fails when loss of the C boot HDD causes there not to be a
bootable drive (my other HDD is not bootable). When I check, the BIOS
shows either no C drive or no HDDs at all. After I jiggle the
connector, the BIOS detects the HDDs and boot-up works.

It seems reasonable then to assume you have a connector
problem. Drive or PSU, pick one and swap it out. If your
climate was particularly hostile I suppose the connectors
could be come corroded, but a visual inpsection should
reveal this.
 
I am at my wit's end with these two machines. One is a Pentium 4 2.4G
on a ASUS P4B533-VM MOBO, the other an AMD Athlon 64 2.4G on a ASUS
K8V SE Deluxe MOBO. Both have 640G DDR. Both use the same mouse,
keyboard, and monitor through my KVM. Both are LAN'ed together via a
D-Link router. Both have 400W+ PSU's. Both have relatively new HDD's
as C (boot) drive.

I have had this problem before, and often I go several months before
it happens again, like now. In fact, it happens so often, that I have
to leave both towers open so as to make it easier for me to deal with
the problem.

I find that more often than not, when I boot-up either machine, it
either will not boot up at all or will boot up with a missing drive or
drives. Both machines are running with two HDDs and one DVD. If it
boots up and the missing drive is not the C boot drive, like say the
DVD drive, then I may not notice the missing drive right away, if at
all.

I have changed both IDE cabling to new ones, and it is still
happening. I have changed drive power connectors from the PSU's, by
adding Y-extenders.

If the power supply drive connector is bad, all you're doing is moving
the problem a little further away from the drive.
Generally, I can jiggle (or re-seat) just the power connectors for the
problem drive(s), and the next re-boot will be correct, and stay that
way for a day or two.

I have not noticed loss of a drive or drives while running, but once
in a great while a machine will crash with a black screen or lose its
mouse/keyboard, so I guess it could be happening, so I can't be sure

I sure would like to try another course of action.

Since this happening on both of my side-by-side machines, I am
baffled.

Any ideas as to what I can try?

To check the quality of the power connections at the drive, grasp the
wire(s) to a contact in the power connector and push and pull the
wire(s) in and out. If you don't feel any or much physical resistance,
then the connection may be poor or intermittent. Do this for the four
contacts.
To tighten the contacts in the power connector, use a .093 contact
extractor to remove the problem contact, and use pliers to squeeze the
split socket contact.
<http://www.action-electronics.com/molex.htm#Tool>
W-HT-2054 STD For .093 Pins
 
If the power supply drive connector is bad, all you're doing is moving
the problem a little further away from the drive.

Oh I recognized that. I just had to try something and I thought a new
cable end-connector might help, since it would be tighter. But you
are right of course.
To check the quality of the power connections at the drive, grasp the
wire(s) to a contact in the power connector and push and pull the
wire(s) in and out. If you don't feel any or much physical resistance,
then the connection may be poor or intermittent. Do this for the four
contacts.
To tighten the contacts in the power connector, use a .093 contact
extractor to remove the problem contact, and use pliers to squeeze the
split socket contact.
<http://www.action-electronics.com/molex.htm#Tool>
W-HT-2054 STD For .093 Pins

Interesting site. Thanks - I'll look this over in detail to see if I
want to spend the money for the tools etc.

Thanks

Jethro
 
I am still having the instability problem wherein I have to jiggle
power connector sometimes to get my machines to boot.

I re-configured my AMD machine so that right now I have:

Primary Master - Seagate Barracuda 120GB ST31200026A
Primary Slave - TSST CORP DVDW TS
Secondary Master - WDC WD1600JB-00GVC0
Secondary Slave - MAXTOR 60GB 6Y060L0

This is the way I really want to run this machine.
I have activated all fans so that now I have two case fans running
plus the CPU and PSU fans. I have enclosed the tower.

I am still having the instability, and now I even find that the third
HDD I added is showing problems. When I first boot, BIOS identifies
the Maxtor as a 'CALYPSO', and after I reach the WXP desktop, and look
at MyComputer, I find that neither HDD on the secondary IDE is
detected by XP, If I immediately shut down/restart - the BIOS
correctly identifies all the HDDs and MyComputer shows all HDDs.
Go figure!

Jethro
 
Jethro said:
I am still having the instability problem wherein I have to jiggle
power connector sometimes to get my machines to boot.

I re-configured my AMD machine so that right now I have:

Primary Master - Seagate Barracuda 120GB ST31200026A
Primary Slave - TSST CORP DVDW TS
Secondary Master - WDC WD1600JB-00GVC0
Secondary Slave - MAXTOR 60GB 6Y060L0

This is the way I really want to run this machine.
I have activated all fans so that now I have two case fans running
plus the CPU and PSU fans. I have enclosed the tower.

I am still having the instability, and now I even find that the third
HDD I added is showing problems. When I first boot, BIOS identifies
the Maxtor as a 'CALYPSO', and after I reach the WXP desktop, and look
at MyComputer, I find that neither HDD on the secondary IDE is
detected by XP, If I immediately shut down/restart - the BIOS
correctly identifies all the HDDs and MyComputer shows all HDDs.
Go figure!

Maybe the power supply is a bit under powered and you have a
problem spinning up the drives initially, but the restart done from
within Win sees them because by then they have all spun up fine.
 
Maybe the power supply is a bit under powered and you have a
problem spinning up the drives initially, but the restart done from
within Win sees them because by then they have all spun up fine.


May be. I have ordered a better PSU and should have it in a few days.
I hope that will help.

Thanks

Jethro
 
Back
Top