Hi,
I've run Intel Processor Diagnostic Tool and all tests successfully PASSED, temperatures were well within limits. I can send testresult if needed, however there was nothing unexpected there.
On the other hand I've ran few Prime95 torture loops with hwinfo software and temperatures logging enabled.
What I've found out thats thermal event shutdown
The system was hibernated due to a critical thermal event
ACPI Thermal Zone = \_SB.PCI0.LPCB.H_EC.TZ0
_HOT = 200K
always happened about 30 seconds after Memory temperature (NUC7i7DNHE sensor, not the one in the DDR4 modules) reached 75 C. Coincidence?
There was no PROCHOT event and no other thermal event or thermal alert when this happens in hwinfo's section for i7-8650U Performance Limit reason.
Maximum CPU Core/Package temperatures was 80-88 C range, so CPU temperature should really not be source for this ACPI Shutdown Thermal Event
Another observation:
When I run this command
wmic /namespace:\\root\wmi PATH MSAcpi_ThermalZoneTemperature get
on NUC7i7DNHEI get this result:
Active ActiveTripPoint ActiveTripPointCount CriticalTripPoint CurrentTemperature InstanceName PassiveTripPoint Reserved SamplingPeriod ThermalConstant1 ThermalConstant2 ThermalStamp
TRUE {0, 0, 0, 0, 0, 0, 0, 0, 0, 0} 0 0 0 ACPI\ThermalZone\TZ0__0 0 0 0 0 0 10
TRUE {3442, 3282, 3232, 3182, 3132, 0, 0, 0, 0, 0} 5 3922 3010 ACPI\ThermalZone\TZ00_0 0 0 0 0 0 9
TRUE {0, 0, 0, 0, 0, 0, 0, 0, 0, 0} 0 3922 3030 ACPI\ThermalZone\TZ01_0 0 0 0 0 0 9
However result from few other computers (other OEM brands with Intel CPUs)
wmic /namespace:\\root\wmi PATH MSAcpi_ThermalZoneTemperature get
Active ActiveTripPoint ActiveTripPointCount CriticalTripPoint CurrentTemperature InstanceName PassiveTripPoint Reserved SamplingPeriod ThermalConstant1 ThermalConstant2 ThermalStamp
TRUE {3442, 3282, 3232, 3182, 3132, 0, 0, 0, 0, 0} 5 3922 3010 ACPI\ThermalZone\TZ00_0 0 0 0 0 0 6
TRUE {0, 0, 0, 0, 0, 0, 0, 0, 0, 0} 0 3922 3030 ACPI\ThermalZone\TZ01_0 0 0 0 0 0 6
As you can see NUC7i7DNHE have the first ThermalZone added compared to the other computers
Active ActiveTripPoint ActiveTripPointCount CriticalTripPoint CurrentTemperature InstanceName PassiveTripPoint Reserved SamplingPeriod ThermalConstant1 ThermalConstant2 ThermalStamp
TRUE {0, 0, 0, 0, 0, 0, 0, 0, 0, 0} 0 0 0 ACPI\ThermalZone\TZ0__0 0 0 0 0 0 10
and this doesn't look like ACPI ThermalZone (THERMTRIP) for CPU
I think this ThermalZone is source for the Thermal Shutdown Event in Windows
I believe if CPU temperature would source for Thermal Event Shutdown then sensor would actually be TZ00 instead of reported TZ0.
So questions remaining to be answered:
1) What is \_SB.PCI0.LPCB.H_EC.TZ0 (aka ACPI\ThermalZone\TZ0__0)?
2) Why is reported _HOT temperature 200K instead of real temperature.
There are two possible scenarios:
A) Shutdown should be expected because there is really some other ACPI Shutdown Temperature threshold in BIOS based on other temperature sensor than CPU package.
- if that's the case can you please explain this thermal zone source and it's behavior (maximum allowed temperature/duration) for ACPI thermal shutdown
- can you please fix the reported temperatures for this thermal zone
B) Shutdown should not be expected, because there should be no other ACPI Shutdown Temperature threshold than THERMTRIP
- maybe in this case ACPI\ThermalZone\TZ0__0 should not be exposed to OS
Thanks for your help.