Heat monitor/histograph

[Industrial One]

Lookin' for an application that sensors and logs CPU/Mobo heat 24/7,
display as a histograph where I can conveniently browse for any spike
that lasted for a couple/ten minutes or essentially caused the system
to shutdown due to heat risk quota.

Also, preferably in addition to it, logs of CPU usage as well (like
what the task manager does only more professionally.)

Thanks in advance. (Room is frozen like the goddamn arctic and still
the system shuts down from overheating, but my PC possibly came with
some humidity-sensor if there even is such thing that autohibernated
the system when it detected a risk of a system short-out.) I need this
app to confirm this is all, since I nap while my comp does the
encoding I leave overnight in the cold room, thus am not around to see
why it shuts down.)

 

[Paul]

Lookin' for an application that sensors and logs CPU/Mobo heat 24/7,
display as a histograph where I can conveniently browse for any spike
that lasted for a couple/ten minutes or essentially caused the system
to shutdown due to heat risk quota.

Also, preferably in addition to it, logs of CPU usage as well (like
what the task manager does only more professionally.)

Thanks in advance. (Room is frozen like the goddamn arctic and still
the system shuts down from overheating, but my PC possibly came with
some humidity-sensor if there even is such thing that autohibernated
the system when it detected a risk of a system short-out.) I need this
app to confirm this is all, since I nap while my comp does the
encoding I leave overnight in the cold room, thus am not around to see
why it shuts down.)

Run the Prime95 Torture Test. It starts a thread per core (multithreaded).
It will also heat the processor up pretty good. You don't have to
"Join GIMPS" to use it. The default memory settings should also result
in a significant amount of system memory being tested.

http://majorgeeks.com/Prime95_d4363.html

Temperatures can be recorded with Speedfan from almico.com .

Both modern Intel and AMD processors are equipped with THERMTRIP.
If all other measures fail to control CPU temperature, the
THERMTRIP signal can be wired from the processor to the
power supply controls. The supply will be turned off without
warning, if THERMTRIP is triggered. On an Intel processor, the
trip point is perhaps 20C above the throttle temperature. So
first there would be throttling, and if it is still getting
hot, then finally it turns off the power.

By using Prime95, you can test for thermal problems immediately.
Ten minutes should be enough to get some idea where the
temperature is going.

Power supplies can also turn themselves off, if an internal
fault is detected. That would include an overheat of a
part of the power supply. The heatsinks sometimes have a
thermistor tied to them, to detect when the heatsink
has become too hot. That can help cover the case, of a
PSU fan failure.

Paul