LG Gram laptop overheating

[pguilaume]

I have an LG Gram laptop which is a few months old. It occasionally has a problem where it shuts down at random, more often if I am playing video.
I believe it's cooling related, but sensors don't show very high temperatures. I'm wondering if the fans are not being controlled properly, but am not sure where to look. Here is the output of sudo hwinfo --all | grep -i fan
Driver Status: thermal,fan are not active
Driver Activation Cmd: "modprobe thermal; modprobe fan"
Can it be that the fan is not running and causing processor to reach critical temperature, shutting down the computer?
I don't ever hear the fan running unless I pick up the laptop to listen, it's very quiet, and never seems to increase in speed before these shutdowns, so I think the fan isn't being properly controlled to compensate for increased heating.
I'm not sure where to look, but would appreciate any advice.
Please let me know if there are any specific tests I can run that will produce helpful information.
Here is some info from inxi -
CPU: Quad Core Intel Core i7-1065G7 (-MT MCP-) speed/min/max: 1300/400/3900 MHz Kernel: 5.4.0-88-generic x86_64 Up: 1h 11m
Mem: 3734.5/23697.0 MiB (15.8%) Storage: 1.14 TiB (35.8% used) Procs: 311 Shell: bash 5.0.17 inxi: 3.0.38

[GreyGeek]

Install lm-sensors from the repository. Then fetch the Thermal Monitor widget and install it. Configure it to read one or more of your cores.
Then you can see the temperature your core(s) are running at. I have an i7 CPU too,and its max temperature is 105C, so I set my max mark at 95C.
The Thermal Monitor can also be set to display temperatures for your SSDs, HDs, and GPU. But, KDE Neon 5.22 & 23 are not displaying CPU temps for some reason.

Since your machine is new I doubt that you have a thermal paste problem, but stranger things have happened.

[pguilaume]

I have installed the Thermal monitor widget like you suggested, also a fan speed monitor. I can also see the temperatures in ksysguard, and they are typically in the 50 - 70 degree range, as I type, the highest I am seeing is 47. I think I might be getting temperature spiking when I do more graphics heavy work, or even watching video. So far, most of the time I am getting these shutdowns, it's been while watching video.
What is troubling me is I can't find a way to tell if the fan is running, without lifting the laptop and listening carefully. I'd be happy to tolerate a noisier fan and shorter battery life if I knew it would keep temperature under control. It looks like the fan speed is regulated by UEFI, so I can't control it from the OS.
Can you suggest any way to achieve control of the fan? Or even to monitor it's status?
I think lm-sensors is supposed to show fan information, but I don't see it.
Here is the output of "sensors"

iwlwifi_1-virtual-0
Adapter: Virtual device
temp1: +30.0°C

CMB0-acpi-0
Adapter: ACPI interface
in0: 7.45 V

coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +51.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +50.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +50.0°C (high = +100.0°C, crit = +100.0°C)
Core 2: +52.0°C (high = +100.0°C, crit = +100.0°C)
Core 3: +49.0°C (high = +100.0°C, crit = +100.0°C)

acpitz-acpi-0
Adapter: ACPI interface
temp1: +32.0°C (crit = +119.0°C)

[Snowhog]

Unless the fan has a sensor you aren't going to be able to monitor its speed.

[pguilaume]

Would thermald affect the fan if it is controlled in the UEFI?
I have this as thermal-cpu-dev-order.xml in /etc/thermald -



<!--
Specifies the order of compensation to cool CPU only.
There is a default already implemented in the code, but
this file can be used to change order

The Following cooling device can present
-->

<CoolingDeviceOrder>
<!-- Specify Cooling device order -->
<CoolingDevice>rapl_controller</CoolingDevice>
<CoolingDevice>intel_pstate</CoolingDevice>
<CoolingDevice>intel_powerclamp</CoolingDevice>
<CoolingDevice>cpufreq</CoolingDevice>
<CoolingDevice>Processor</CoolingDevice>
</CoolingDeviceOrder>

[pguilaume]

I still haven't made any progress with this. Interestingly, I ran Windows yesterday after one of these random shutdowns. (It's dual boot). The fan was quiet while running windows just like in Kubuntu, but when I shut down the computer, the fan ran at a speed that was fast enough for me to hear it for several seconds while shutting down. I take this as proof that there isn't a hardware problem and the fan does work.
The thermal monitor widget is showing temperatures within normal range, but the temps do spike when doing things that I assume are taxing the GPU.
For example I had a schematic diagram open in Okular and zoomed in to see a detail. While watching the thermal monitor widget, if I pan around in the zoomed document, the temperature was hitting 90 degrees plus, and it did eventually cause a shutdown.
My feeling here is that whatever controls the fan isn't doing a good job.

I can see 9 cooling devices in /sys/devices/virtual/thermal/
Eight of the nine cooling devices show type as "Processor", and the last one shows as "intel_powerclamp"
The files in the cooling_device0 to cooling_device8 folders have the following contents (cooling_device0 shown) -
-rw-r--r-- 1 root root 4096 Oct 26 09:33 cur_state
lrwxrwxrwx 1 root root 0 Oct 26 09:06 device -> ../../../LNXSYSTM:00/LNXSYBUS:00/LNXCPU:00
-r--r--r-- 1 root root 4096 Oct 26 04:36 max_state
drwxr-xr-x 2 root root 0 Oct 26 09:06 power
drwxr-xr-x 2 root root 0 Oct 26 09:06 stats
lrwxrwxrwx 1 root root 0 Oct 26 04:36 subsystem -> ../../../../class/thermal
-r--r--r-- 1 root root 4096 Oct 26 04:36 type
-rw-r--r-- 1 root root 4096 Oct 26 04:36 uevent

If I cat the cur_state file, the result shows 0, which I take to mean the device is not active.
Max state is 10
in the stats folder, I can see a summary of running time for these devices and it looks like they have never run, ie. their state has always been recorded as 0.

Any ideas on what else I can try, or a way to take control of the fan?

[Snowhog]

Old school troubleshooting.

IF your issue is an overheating problem, raise the laptop off of the surface it sits on. It doesn't have to be much. Even a 1/2 inch would do. The intent here is to provide better air space beneath the laptop, and without blocking, or blocking as little as possible, the vents on the bottom. This alone allows more air to circulate beneath and through the laptop. Run it as usual, doing the things you normally do for as long as you normally do. Do the shutdowns still occur?

Given it's a laptop, several internal issues could be contributing to your problem. Dirt and dust, especially around the exhaust vent(s). It doesn't take much dust to affect thermal cooling; to increase overheating. Also, if there are any thermal heatsinks (CPU/GPU), they could be loose. Thermal paste doesn't last forever.

And, as this laptop is only a few months old, is it still under warranty? Have to contacted the manufacturer about the problem?

[pguilaume]

Warranty is no longer an option. As far as I can see, everything looks pretty clean in the vents and fan ducting. I haven't checked or changed thermal paste. The temps do stay lower if the laptop is elevated off the table so the extra air space does help. I just think that the fan should run and run faster if temperatures even approach critical levels. The computer definitely should not get to the stage where it shuts down at critical temperature if the fan can run and cool it down. I know the fan can run because it did when I ran windows recently. It just seems that the OS isn't configured to take control of the fan and spin it up when it needs to. It's strange that this has only been happening in the last month or so. I am wondering if something changed with an update. I see when I run dmesg that the cpu temperatures rise above normal every so often, and the cpu is then throttled to control heat and temps come down again within a second or so. It appears that this is the default cooling method used and it doesn't seem to affect performance.
I guess this is controlled by thermald?
If that is the case, then maybe the configuration can be changed to allow the fan to kick in earlier, but I'm not sure how to do that.