CPU heat sink detachment

[Avtandil]

Hello everyone, I am new here. Since the CPU overheats, it’s necessary to replace the thermal paste and the fan on the side of the heat sink, which does not seem to work. Disassembling the heat sink seems a little problematic. I tried to unscrew the two hex nuts, but they spin freely and do not detach from the support. How should I proceed for disassembly?
Pc HP Pavilion 590-a0001nl Motherboard: HP name: KensingtonP, SSID: 843E

[ATTACH type=“full” size=“600x800”]14281[/ATTACH]

[PeterOz]

Is this your computer?
Go to 24 minutes in video and watch

[MEDIA=youtube]2U_gz5-7j8o[/MEDIA]

[Avtandil]

Is this your computer?
Go to 24 minutes in video and watch

[MEDIA=youtube]2U_gz5-7j8o[/MEDIA]

Yes, it is. Thank you very much for your support! ???

[PeterOz]

Did that video help with your problem?

[Bruce]

I don’t think your picture is of the processor, looks more like the southbridge heatsink.
Pan out and give us a more overall shot thanks.

Any that hex nut - Sweet Mother of Jesus - who did that?
Looks as dodgy as f@ck!
Would love to see a shot of that from the other side of the motherboard.

Look as though the normal pin broke off and this MacGyver hack has been used to get the heatsink to stay on the mobo.
I’m guessing the hex head is spinning because on the other side of the board is probably a nut.

[Bruce]

@Avtandil - any news?

[Avtandil]

Did that video help with your problem?

I don’t think your picture is of the processor, looks more like the southbridge heatsink.
Pan out and give us a more overall shot thanks.

Any that hex nut - Sweet Mother of Jesus - who did that?
Looks as dodgy as f@ck!
Would love to see a shot of that from the other side of the motherboard.

Look as though the normal pin broke off and this MacGyver hack has been used to get the heatsink to stay on the mobo.
I’m guessing the hex head is spinning because on the other side of the board is probably a nut.

Sorry for the late reply, I hadn’t been around for a while.
From a general examination, the motherboard looks like not containing any other components similar to the processor. This PCs series is equipped with push-pin heath sinks instead of screwed-in heath sinks for CPU cooling. To detach them, it’s necessary to remove the motherboard, then tighten the pins on the back by nose pliers.

[Avtandil]

Did that video help with your problem?

Sorry for the late reply, I hadn’t been around for a while.
Yes, it did. My unit is equipped with a push-pin heat sink. I had to remove the motherboard entirely as shown in the video (min. 24), then I could detach the heat sink tightening the pins on the back by nose pliers and replace the thermal paste. CPU cooling is improved. Now I’m waiting for the new fan’s delivery to replace it.

[Bruce]

@Avtandil - what’s the latest?

[Avtandil]

@Avtandil - what’s the latest?

I’ve mounted a replacement fan for the CPU heat sink, but it doesn’t spin either. In BIOS-Thermal, the CPU fan speed reading is not available. A visual check confirms that it doesn’t spin. Monitoring program reports that only the case fan is working and doesn’t recognize any other. The CPU fan appears to be electrically connected and recognized by the system (unplugging its connector, BIOS after the POST reports disconnection and stops the boot), but it doesn’t spin. I tried a BIOS update, but the installed version is the most recent one. CPU cooling was improved after previous thermal paste replacement, but it’s not enough jet.
So, the system requires a fan to be mounted but doesn’t seem to want to run it.

https://pchelpforum.net/attachments/ventola-e-dissipatore-cpu-montati-su-scheda-madre-03-webp.14457/?hash=ba4dcb4e7d24070064102cfb9eccf151

[ATTACH type=“full” alt=“1 (2).webp”]14458[/ATTACH]

[Bruce]

Does that motherboard have another fan header you could plug the fan into to double check?
Or do you have another PC you could test it on?

[phillpower2]

Your CPU fan is only three pin and so is not able to control/report fan speeds, it works off voltages and the only way to increase or decrease the fans RPMs would be by likewise adjusting the voltages, too low a voltage would not be able to spin the fan so no RPMs would register.

Your MBs system fan header is four pin and so should report the RPMs of any four pin fan that is connected to it.

[COLOR=rgb(44, 130, 201)]Edit to add: You could do a quick test of the CPU fan by connecting it to the four pin header, if you do, just power up to see if the fan spins then turn off the power again, we don`t want to fry the CPU.[/COLOR]

[Avtandil]

Does that motherboard have another fan header you could plug the fan into to double check?
Or do you have another PC you could test it on?

There are no other headers, but when I selectively disconnect the yellow wire or the black wire of the CPU fan, the system notifies the anomaly during the POST and stops the startup, so the device is recognized.
I arranged an extension cord to position the fan outside the PC while keeping it connected and being able to continuously visually monitor its operation. I discovered that, as soon as the PC is started, the fan starts working for a few seconds, then stops. This procedure confirms to the system that the fan can spin correctly when required. Maybe the fan starts working only when the CPU temperature rises above a certain limit? The fan that was mounted before had dust deposited uniformly on the blades, so at certain times it used to start working. Also, before replacing the thermal paste, when the temperature even rose to 95 °C I sometimes heard an increase in the fan noise; maybe the second fan started working?
I performed a stress test on the CPU from the site

https://silver.urih.com

to check if the fan starts working when the CPU gets very hot. But even after 35 minutes of testing with an average temperature (after replacing the thermal paste) of 80 °C, and peaks of 84 it remains inactive. Maybe higher values are required? But shouldn’t the fan always run, possibly varying its speed?
When the CPU (Intel®Pentium® J J5005) is running at medium load the temperature is around 70, shouldn’t the optimal value be lower?
Here are some tests for this CPU model. The tests were run for three configuration - fanless, with two 50x50 fans (bottom, blowing up through the case, case on stands to provide air intake) and third with PC with fans outside the window for much much cooler air.

[ATTACH type=“full” alt=“Opera Snapshot_2024-09-22_165644_rkblog.dev.webp”]14467[/ATTACH]

[ATTACH type=“full” alt=“Opera Snapshot_2024-09-22_170410_rkblog.dev.webp”]14468[/ATTACH]

The best solution would be to be able to set the fan activation point to lower temperatures.

[phillpower2]

There are no other headers,

See attachment;

[Avtandil]

Your CPU fan is only three pin and so is not able to control/report fan speeds, it works off voltages and the only way to increase or decrease the fans RPMs would be by likewise adjusting the voltages, too low a voltage would not be able to spin the fan so no RPMs would register.

Your MBs system fan header is four pin and so should report the RPMs of any four pin fan that is connected to it.

[COLOR=rgb(44, 130, 201)]Edit to add: You could do a quick test of the CPU fan by connecting it to the four pin header, if you do, just power up to see if the fan spins then turn off the power again, we don`t want to fry the CPU.
[/COLOR]

[COLOR=rgb(44, 130, 201)]
The connectors seem to have different shapes. I would prefer not to risk damage as I am not an expert.
I arranged an extension cord to position the fan outside the PC while keeping it connected and being able to continuously visually monitor its operation. I found that, as soon as the PC is started, the fan starts working for a few seconds, then stops and keeps inactive while the pc is running. This procedure confirms to the system that the fan can spin correctly when required. Maybe the fan starts working only when the CPU temperature rises above a certain limit? The fan that was mounted before had dust deposited uniformly on the blades, so at certain times it used to start working. Also, before replacing the thermal paste, when the temperature even rose to 95 °C I sometimes heard an increase in the fan noise; maybe the CPU fan started working?
I performed a stress test on the CPU to check if the fan starts working when the CPU gets very hot. But after 35 minutes of testing with an average temperature (after replacing the thermal paste) of 80 °C, and peaks of 84 it remains inactive. Maybe higher values are required?
When the CPU (Intel®Pentium® J J5005) is running at medium load the temperature is around 70, shouldn’t the optimal value be lower or it’s good?
Here are some tests for this CPU model. The tests were run for three configuration - fanless, with two 50x50 fans (bottom, blowing up through the case, case on stands to provide air intake) and third with PC with fans outside the window for much much cooler air.

[ATTACH type=“full” alt=“Opera Snapshot_2024-09-22_165644_rkblog.dev.webp”]14469[/ATTACH]

[ATTACH type=“full” alt=“Opera Snapshot_2024-09-22_170410_rkblog.dev.webp”]14470[/ATTACH][/color]