Sabertooth X58 random shutdowns
 

Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After turning
on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back. Booted
up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7 -960
and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

Sabertooth X58 random shutdowns
 

On 12/26/2013 3:21 PM, guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

And what was the CPU temperature? Perhaps it is time to do a little
housekeeping, such as cleaning out the fans and fan filters, the dust
and lint that might have gathered inside the case, renew the thermal
compound between the heatsink and the CPU. At the same time, refresh
the contacts of the peripheral cards and memory. Remember, the i7-960
is a "hot" CPU, even when not driven too hard.

GR

Sabertooth X58 random shutdowns
 

guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

Are you monitoring the CPU temperature ?

Both Intel and AMD processors have internal thermal protection.
The signal that comes out of the processor is THERMTRIP. If the
CPU thinks it is too hot, then the CPU asserts THERMTRIP and
the ATX power supply should shut off. THERMTRIP is a logic
term to the PS_ON# signal the ATX supply uses.

PS_ON# is an open-collector signal. You can ground PS_ON#,
by joining it to an adjacent COM (common or ground) signal,
and keep the ATX supply running. However, that would
also prevent THERMTRIP from protecting the CPU from overheat.
It would keep the computer running, but put your
CPU at risk if it really was overheating.

It could also be, that something is sending a "regular shutdown"
to the PS_ON#, and causing the fans to go off and the main
section of the ATX supply to stop. That would be harder to figure
out. As to why some software path might do that.

But THERMTRIP is an example of a hardware-only path. Nothing
gates it, and if the CPU thinks (even, erroneously) that
the CPU is too hot, then the power is going to go off.

1) Monitor CPU temperature.
2) If CPU seems too hot, remove heatsink/fan assembly, re-apply
fresh thermal paste, install heatsink/fan assembly again, and
retest the CPU temperature. Before starting that process, make
sure you have a tube of thermal paste handy.
3) When you think it's fixed, test with Prime95 torture test or that
Intel burn-in program. Those are programs that attempt to run the
CPU at 100%. The cooling system on a computer, should be
adequate to keep the CPU below critical temperature, even when
it does a length calculation. Sitting idle in the desktop and
not seeing it power off, is not a sufficient test. Use a good
"loading" program, to test that it is really fixed.

A loading program can also be used as a form of proof that
it is THERMTRIP that is doing it. In that, it might shut down
soon after starting the test program.

A range of Intel processors, Intel used low temperature solder,
to solder the top of the silicon die, to the metal lid on the processor.
For the latest processors, Intel has stopped doing that, and is back
to using thermally conductive filler between the top of the silicon
die and the metal processor lid. I think the previous soldering
scheme was a more reliable mechanism, in that there was less opportunity
for the solder to "leave the area". A recent picture of
a Haswell processor opened up, shows a grey material now
present as a conductor of heat.

If for any reason, the thermal path inside the CPU is interrupted,
that can be enough to cause the CPU to overheat. If any of the
Intel warranty remains on such a processor, you'd attempt an RMA
to get it replaced. The warranty terms on a boxed retail processor
with fan, may be different than a tray-container OEM one.
(I'd have to go and look that up, since processor replacement
doesn't happen all the often.) The warranty might be three years,
but you should go look that up.

It could also be something like the ATX supply overheating
internally. Power supplies have a thermistor attached to
an internal heatsink. If the supply thinks it has become too
hot, it may shut off. But in that case, it may require
toggling the switch on the back, to get the supply to
restart. So there are subtle hints like that, as to whether
the switch-off attempt is coming from the CPU/motherboard,
or from the ATX supply.

Paul

Sabertooth X58 random shutdowns
 

"Ghostrider" wrote in message
m...

On 12/26/2013 3:21 PM, guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

And what was the CPU temperature? Perhaps it is time to do a little
housekeeping, such as cleaning out the fans and fan filters, the dust
and lint that might have gathered inside the case, renew the thermal
compound between the heatsink and the CPU. At the same time, refresh
the contacts of the peripheral cards and memory. Remember, the i7-960
is a "hot" CPU, even when not driven too hard.

GR

Ghostrider,

I keep the computer pretty clean, and that was the first thing I had looked
at.
There was not any buildup, so I was trying to figure what hardware might be
over heating.
I had cleaned and reset the CPU / heatsink which is a Corsair H70 liquid
cooler with 2 fans.
Nothing seems overly hot, and the last temperature monitoring I had done was
well within range.
Exact temps escape me now!
I am thinking it is the CPU or mother board. I had it boot to safe mode with
networking and it ran for about 1-2
hours before I shut it down yesterday. Didn't know where to look to pinpoint
any specific codes or messages.
Just ran it in safe mode again and it shut down in 5 minutes...

Thanks for your help, I will probably reset the CPU again and see what
happens.
Guy

Sabertooth X58 random shutdowns
 

"Paul" wrote in message ...

guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7 -960
and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

Are you monitoring the CPU temperature ?

Both Intel and AMD processors have internal thermal protection.
The signal that comes out of the processor is THERMTRIP. If the
CPU thinks it is too hot, then the CPU asserts THERMTRIP and
the ATX power supply should shut off. THERMTRIP is a logic
term to the PS_ON# signal the ATX supply uses.

PS_ON# is an open-collector signal. You can ground PS_ON#,
by joining it to an adjacent COM (common or ground) signal,
and keep the ATX supply running. However, that would
also prevent THERMTRIP from protecting the CPU from overheat.
It would keep the computer running, but put your
CPU at risk if it really was overheating.

It could also be, that something is sending a "regular shutdown"
to the PS_ON#, and causing the fans to go off and the main
section of the ATX supply to stop. That would be harder to figure
out. As to why some software path might do that.

But THERMTRIP is an example of a hardware-only path. Nothing
gates it, and if the CPU thinks (even, erroneously) that
the CPU is too hot, then the power is going to go off.

1) Monitor CPU temperature.
2) If CPU seems too hot, remove heatsink/fan assembly, re-apply
fresh thermal paste, install heatsink/fan assembly again, and
retest the CPU temperature. Before starting that process, make
sure you have a tube of thermal paste handy.
3) When you think it's fixed, test with Prime95 torture test or that
Intel burn-in program. Those are programs that attempt to run the
CPU at 100%. The cooling system on a computer, should be
adequate to keep the CPU below critical temperature, even when
it does a length calculation. Sitting idle in the desktop and
not seeing it power off, is not a sufficient test. Use a good
"loading" program, to test that it is really fixed.

A loading program can also be used as a form of proof that
it is THERMTRIP that is doing it. In that, it might shut down
soon after starting the test program.

A range of Intel processors, Intel used low temperature solder,
to solder the top of the silicon die, to the metal lid on the processor.
For the latest processors, Intel has stopped doing that, and is back
to using thermally conductive filler between the top of the silicon
die and the metal processor lid. I think the previous soldering
scheme was a more reliable mechanism, in that there was less opportunity
for the solder to "leave the area". A recent picture of
a Haswell processor opened up, shows a grey material now
present as a conductor of heat.

If for any reason, the thermal path inside the CPU is interrupted,
that can be enough to cause the CPU to overheat. If any of the
Intel warranty remains on such a processor, you'd attempt an RMA
to get it replaced. The warranty terms on a boxed retail processor
with fan, may be different than a tray-container OEM one.
(I'd have to go and look that up, since processor replacement
doesn't happen all the often.) The warranty might be three years,
but you should go look that up.

It could also be something like the ATX supply overheating
internally. Power supplies have a thermistor attached to
an internal heatsink. If the supply thinks it has become too
hot, it may shut off. But in that case, it may require
toggling the switch on the back, to get the supply to
restart. So there are subtle hints like that, as to whether
the switch-off attempt is coming from the CPU/motherboard,
or from the ATX supply.

Paul

Paul, please see previous message. I don't remember exact temps, but they
were always well within range when I was monitoring them.
I guess it could be the power supply as it is probably 6 years old, and an
Antec II 550. It might be the most cost effective item to start replacing!
It has been used thru 3-4 separate builds over those years. I assumed if it
was the PS it would just quit altogether. It is strange that it would
work for 2-5 hours and sometimes shut down in a few minutes after windows
started.

Sabertooth X58 random shutdowns
 

On 12/27/2013 7:29 AM, guy wrote:


"Ghostrider" wrote in message
m...

On 12/26/2013 3:21 PM, guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

And what was the CPU temperature? Perhaps it is time to do a little
housekeeping, such as cleaning out the fans and fan filters, the dust
and lint that might have gathered inside the case, renew the thermal
compound between the heatsink and the CPU. At the same time, refresh
the contacts of the peripheral cards and memory. Remember, the i7-960
is a "hot" CPU, even when not driven too hard.

GR

Ghostrider,

I keep the computer pretty clean, and that was the first thing I had
looked at.
There was not any buildup, so I was trying to figure what hardware might
be over heating.
I had cleaned and reset the CPU / heatsink which is a Corsair H70 liquid
cooler with 2 fans.
Nothing seems overly hot, and the last temperature monitoring I had done
was well within range.
Exact temps escape me now!
I am thinking it is the CPU or mother board. I had it boot to safe mode
with networking and it ran for about 1-2
hours before I shut it down yesterday. Didn't know where to look to
pinpoint any specific codes or messages.
Just ran it in safe mode again and it shut down in 5 minutes...

Thanks for your help, I will probably reset the CPU again and see what
happens.
Guy

Guy, I presume that you are not over-clocking. But even if you were,
a 550-Watt PSU is inadequate for this ASUS motherboard and CPU. As a
custom builder, I usually use a minimum of a 750-Watt PSU for an X58
motherboard with the Intel I7-950+ family of CPU's. Paul might have
the right idea of issues with the [old] PSU.

GR

Sabertooth X58 random shutdowns
 

"Ghostrider" wrote in message
m...

On 12/27/2013 7:29 AM, guy wrote:


"Ghostrider" wrote in message
m...

On 12/26/2013 3:21 PM, guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

And what was the CPU temperature? Perhaps it is time to do a little
housekeeping, such as cleaning out the fans and fan filters, the dust
and lint that might have gathered inside the case, renew the thermal
compound between the heatsink and the CPU. At the same time, refresh
the contacts of the peripheral cards and memory. Remember, the i7-960
is a "hot" CPU, even when not driven too hard.

GR

Ghostrider,

I keep the computer pretty clean, and that was the first thing I had
looked at.
There was not any buildup, so I was trying to figure what hardware might
be over heating.
I had cleaned and reset the CPU / heatsink which is a Corsair H70 liquid
cooler with 2 fans.
Nothing seems overly hot, and the last temperature monitoring I had done
was well within range.
Exact temps escape me now!
I am thinking it is the CPU or mother board. I had it boot to safe mode
with networking and it ran for about 1-2
hours before I shut it down yesterday. Didn't know where to look to
pinpoint any specific codes or messages.
Just ran it in safe mode again and it shut down in 5 minutes...

Thanks for your help, I will probably reset the CPU again and see what
happens.
Guy

Guy, I presume that you are not over-clocking. But even if you were,
a 550-Watt PSU is inadequate for this ASUS motherboard and CPU. As a
custom builder, I usually use a minimum of a 750-Watt PSU for an X58
motherboard with the Intel I7-950+ family of CPU's. Paul might have
the right idea of issues with the [old] PSU.

GR

I replaced the power supply with a new 750W, one and everything is working
great!
I ran prime 95 for almost 2 hours and monitored the temps which never
exceeded 70C.
Looks like the old PSU was starting/did fail and the info from you and Paul
helped
me to get everything resolved. Many thanks to both of you, and I hope you
have a
Happy New Year!!

Guy

Sabertooth X58 random shutdowns
 

Guy wrote:


"Ghostrider" wrote in message
m...

On 12/27/2013 7:29 AM, guy wrote:


"Ghostrider" wrote in message
m...

On 12/26/2013 3:21 PM, guy wrote:
Just recently my computer started to shutdown randomly. Sometimes after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

And what was the CPU temperature? Perhaps it is time to do a little
housekeeping, such as cleaning out the fans and fan filters, the dust
and lint that might have gathered inside the case, renew the thermal
compound between the heatsink and the CPU. At the same time, refresh
the contacts of the peripheral cards and memory. Remember, the i7-960
is a "hot" CPU, even when not driven too hard.

GR

Ghostrider,

I keep the computer pretty clean, and that was the first thing I had
looked at.
There was not any buildup, so I was trying to figure what hardware might
be over heating.
I had cleaned and reset the CPU / heatsink which is a Corsair H70 liquid
cooler with 2 fans.
Nothing seems overly hot, and the last temperature monitoring I had done
was well within range.
Exact temps escape me now!
I am thinking it is the CPU or mother board. I had it boot to safe mode
with networking and it ran for about 1-2
hours before I shut it down yesterday. Didn't know where to look to
pinpoint any specific codes or messages.
Just ran it in safe mode again and it shut down in 5 minutes...

Thanks for your help, I will probably reset the CPU again and see what
happens.
Guy

Guy, I presume that you are not over-clocking. But even if you were,
a 550-Watt PSU is inadequate for this ASUS motherboard and CPU. As a
custom builder, I usually use a minimum of a 750-Watt PSU for an X58
motherboard with the Intel I7-950+ family of CPU's. Paul might have
the right idea of issues with the [old] PSU.

GR

I replaced the power supply with a new 750W, one and everything is
working great!
I ran prime 95 for almost 2 hours and monitored the temps which never
exceeded 70C.
Looks like the old PSU was starting/did fail and the info from you and
Paul helped
me to get everything resolved. Many thanks to both of you, and I hope
you have a
Happy New Year!!

Guy

Good diagnostic work. The poster does all the work, all
we do is throw out the suggestions :-)

Paul

Sabertooth X58 random shutdowns
 

On 12/27/2013 3:58 PM, Paul wrote:
Guy wrote:


"Ghostrider" wrote in message
m...

On 12/27/2013 7:29 AM, guy wrote:


"Ghostrider" wrote in message
m...

On 12/26/2013 3:21 PM, guy wrote:
Just recently my computer started to shutdown randomly. Sometimes
after
several hours and sometimes after a couple of minutes.
After several weeks of this I didn't use it for a week or two. After
turning on again, the monitor said no video. Today I replaced the
video card and same message. Replaced the monitor and got video back.
Booted up and everything came up for about 5-10 min.
and then shut down again. The unit is 16 months old, with an Intel i7
-960 and 3 sticks of 2G dominator memory. I don't know what to
do next..
Appreciate any help to resolve this issue.

Guy

And what was the CPU temperature? Perhaps it is time to do a little
housekeeping, such as cleaning out the fans and fan filters, the dust
and lint that might have gathered inside the case, renew the thermal
compound between the heatsink and the CPU. At the same time, refresh
the contacts of the peripheral cards and memory. Remember, the i7-960
is a "hot" CPU, even when not driven too hard.

GR

Ghostrider,

I keep the computer pretty clean, and that was the first thing I had
looked at.
There was not any buildup, so I was trying to figure what hardware might
be over heating.
I had cleaned and reset the CPU / heatsink which is a Corsair H70 liquid
cooler with 2 fans.
Nothing seems overly hot, and the last temperature monitoring I had done
was well within range.
Exact temps escape me now!
I am thinking it is the CPU or mother board. I had it boot to safe mode
with networking and it ran for about 1-2
hours before I shut it down yesterday. Didn't know where to look to
pinpoint any specific codes or messages.
Just ran it in safe mode again and it shut down in 5 minutes...

Thanks for your help, I will probably reset the CPU again and see what
happens.
Guy

Guy, I presume that you are not over-clocking. But even if you were,
a 550-Watt PSU is inadequate for this ASUS motherboard and CPU. As a
custom builder, I usually use a minimum of a 750-Watt PSU for an X58
motherboard with the Intel I7-950+ family of CPU's. Paul might have
the right idea of issues with the [old] PSU.

GR

I replaced the power supply with a new 750W, one and everything is
working great!
I ran prime 95 for almost 2 hours and monitored the temps which never
exceeded 70C.
Looks like the old PSU was starting/did fail and the info from you and
Paul helped
me to get everything resolved. Many thanks to both of you, and I hope
you have a
Happy New Year!!

Guy

Good diagnostic work. The poster does all the work, all
we do is throw out the suggestions :-)

Paul

Concur. Happy New Year!

GR