Processor after thermal paste replacement - Heats up quickly under load

As in the subject, the problem is that the processor in the laptop quickly heats up to high temperatures after replacing the thermal paste. laptop it Samsung R540 model JA07PL . The processor is an Intel i3 380M 2x 2.53 GHz. I've had a laptop for about 10 years. Since then, the paste has never been replaced in this laptop, nor has anyone cared about it in any way. Before the replacement, the temperatures were fine, rarely exceeding 65 degrees under load, at rest it averaged 41-45 degrees and the temperature went up and down slightly, 1-2 degrees at a time. Now, after replacing the paste, the temperature at rest is ok in my opinion, because it is 45-49 degrees, but as soon as it starts to be loaded with anything, the temperature in a second can jump by 10 degrees at a time and often jumps to 70-80 degrees. It crashed a few times at 90. The paste is Arctic MX-2. Distributed throughout the entire processor core. The fan was all cleaned, had poor airflow, but it didn't change anything. When I opened the laptop again, I noticed two spots on the core under the grease-like paste, maybe I didn't degrease the surface too thoroughly, but after replacing this paste with a new one, checking if the core is well degreased on the mirror, the problem is the same. It heats up to 75-80 degrees under load and there is a lot to reach 90 and turn off. What could be the problem? There were no thermal pads. Windows 10 Pro x64

Ps. I would like to add that some time ago I replaced the paste on the Lenovo G7 laptop and the symptom is the same, the temperature constantly jumps to the red field, but there is no problem on my desktop computer. Temperatures without load barely reach 40 degrees, and under load they do not exceed 60 and the paste did not change anything. The only difference is when I applied them. Half a year on a stationary one, after a few months on Lenovo and now in Samsung but it seems to me that it should not matter, because the paste would not lose its properties so quickly.

Thanks in advance for your help!

Possible too much paste or incorrectly installed cooler.

sosarek wrote:

Too much paste possible

I put it on a pea and then spread it evenly.

sosarek wrote:

incorrectly installed cooler

I screwed the cooling "crosswise" it is even numbered how to screw it so there is no question of bad cooling assembly

There is also unsealing of the heatpipe, too much gap between the radiator and the fan ...

sosarek wrote:

There is also unsealing of the heatpipe, too much gap between the radiator and the fan ...

How can I check if it is working?

krzysztofer100 wrote:

sosarek wrote:

There is also unsealing of the heatpipe, too much gap between the radiator and the fan ...

How can I check if it is working?

The entire cooling system should heat up fairly evenly (have the same temperature). You can disassemble and, holding the radiator, immerse the part that adheres to the processor in hot water, a moment later it should have the same temperature over the entire surface.
You check the gap visually, if there is a large gap, cover it with Kapton tape.

7jacek7 wrote:

You check the gap visually

Where to look for a break? Where it doesn't get hot?

krzysztofer100 wrote:

Where to look for a break? Where it doesn't get hot?

I meant the gap between the heatsink and the fan. This can be remedied.

However, if the cooling system does not heat up evenly, it is suitable for replacement. There is no cure.

krzysztofer100 wrote:

The paste is Arctic MX-2. Distributed throughout the entire processor core.

Apply a small grain to the center of the processor core, do not spread. The cooling will tighten the paste as it should.

7jacek7 wrote:

However, if the cooling system does not heat up evenly

I put the beginning of the cooling where it adheres to the processor and the whole cooling heated up very quickly, so the cooling will be efficient here ... but where to look for the reason when everything is done properly?

Show me how it's mounted.

sosarek wrote:

Show me how it's mounted

Now I have the cooling dismantled but I haven't cleaned the paste from the core yet and the cooling pressure left such a trace


The ONLY thing I haven't mentioned so far is this thermopad thermal paste on the chipset, but I don't think it had such a huge impact, especially since before replacing the thermal paste everything worked great and never exceeded 60 degrees under load

The paste should only be on the cores and not the elements around because it can lead to a short circuit (if with silver).
If the thermocouple is intact, don't touch it.

sosarek wrote:

If the thermocouple is intact, don't touch it

I take back what I wrote, there is no thermopad here, only a paste that is already dried, but would it have such an effect on the processor?


Added after 16 [minutes]:

krzysztofer100 wrote:

only paste that is already dried

Everything cleaned and degreased on the mirror, I put everything together and in a moment I will send a screen of temperatures from the Speccy program

Dried paste is a poor conductor.

OCCT results from a laptop. The temperature jumps by more than 10 degrees, something is clearly wrong, the maximum temperature is 89 degrees, at 90 it turns off, fortunately the entire 5-minute test passed, I suspect that if the test lasted longer, it would not complete it due to the power being turned off at 90 degrees. The temperature at load jumps every 10 degrees, when it cools down it goes evenly after 1-max 2, but as soon as it starts to be loaded it jumps by 3-5-10 degrees, different.


For comparison, the result of the same test on my pc
Here, the computer was clearly bored during the test compared to the temperature jumps on the laptop


Screenshot of OCCT after the test on a laptop
OCCT-Scree...134125.png Download (126.61 kB)

Possible damage to the temperature sensor in the CPU, e.g. by chipping the core.

miecho18 wrote:

Possible damage to the temperature sensor in the CPU, e.g. by chipping the core.

And the jumping voltage will also be caused by this?

krzysztofer100 wrote:

I would like to add that some time ago I changed the paste on the Lenovo G7 laptop and the symptom is the same, the temperature constantly jumps to the red field, but there is no problem on my desktop computer.

Don't compare a stationary to a laptop. In laptops, it is normal for CPU and GPU temperatures to be much higher.

krzysztofer100 wrote:

miecho18 wrote:

Possible damage to the temperature sensor in the CPU, e.g. by chipping the core.

And the jumping voltage will also be caused by this?

what voltage?

miecho18 wrote:

what voltage?

Please tell me what it is because I can't see it. The frequency and power consumed by the processor jumps due to throttling, but it is normal for the processor to overheat.

miecho18 wrote:

Please tell me what it is because I can't see it

That's what I meant about the POWER tab. I will note that before all this overheating I pulled out the bios battery because the laptop was not started for a long time and nothing happened after switching on. After that, I disassembled the entire laptop to clean everything and this problem began. I don't have the strength for it anymore, because I don't like to disassemble this laptop for the 5th time, because you have to disassemble the whole thing into the first parts

Try another processor.

Hi, what did you clean at first, old thermal paste?
When you reinstalled the cooler, I understand that you checked if the fan is spinning?

Piotr Partyka wrote:

Hi, what did you clean at first, old thermal paste?

Paper and acetone from nail polish remover

Piotr Partyka wrote:

When you reinstalled the cooler, I understand that you checked if the fan is spinning?

Yes, everything done in accordance with the art.

The idle processor has an average of 41-43 degrees if nothing is happening, but it is enough to switch between folders and there are jumps from 43 to 59 so in my opinion something is wrong, and if there is something in the load for longer, it quickly goes to over 60 degrees in an instant, the fan spins faster and you can feel it.

You know what would be better isopropanol, because nail polish remover often has some additional additives like vitamin F or some flower extracts ...

Piotr Partyka wrote:

You know what would be better is isopropanol

It would certainly be better, but when I have nothing else effective to clean the paste, I take something that is able to degrease well. For years, I have used it for occasional replacement of processors or other cleaning of components and there has never been a problem with temperatures. The hot water test of the entire cooling went rather as it was supposed to come out with efficient cooling, i.e. the top got hot within a second after being put into hot water. I handled the processor very gently with paper and a cotton swab around the core to clean the paste there. The thing is that it gets some weird temperature spikes when without load they are very correct. Before cleaning, when the laptop was used, the temperatures in the load never reached such temperatures, only with heavy use, e.g. a game, then you could feel it, but resting or browsing the internet was not a problem, now it is true, but not at higher temperatures ...

krzysztofer100 wrote:

Piotr Partyka wrote:

You know what would be better is isopropanol

It would certainly be better, but when I have nothing else effective to clean the paste, I take something that is able to degrease well. For years, I have used it for occasional replacement of processors or other cleaning of components and there has never been a problem with temperatures. The hot water test of the entire cooling went rather as it was supposed to come out with efficient cooling, i.e. the top got hot within a second after being put into hot water. I handled the processor very gently with paper and a cotton swab around the core to clean the paste there. The thing is that it gets some weird temperature spikes when without load they are very correct. Before cleaning, when the laptop was used, the temperatures in the load never reached such temperatures, only with heavy use, e.g. a game, then you could feel it, but resting or browsing the internet was not a problem, now it is true, but not at higher temperatures ...

I understand, but you know, sometimes these removers have some additional substances in the composition, a lot of it, some plant extracts. And I think that for cleaning the procek core, there should be an uncontaminated product.
He gave me the full stock of this remover.

Is it a normal phenomenon that without any load, in the POWER tab, and more specifically the Package in HWMonitor, there are such jumps from 1.25W to 25W? This happens every second with a slight temperature jump (3-4 degrees at a time). I am completely spreading my hands, which may be the cause of such behavior of the laptop, there has never been such a problem before, and above all, such jumps, earlier after 1-2 max temperature jumps and with a sudden load it jumps by more than 10 sometimes ...

krzysztofer100 wrote:

I'm already completely spreading my hands, which can be the reason for such behavior of the laptop

Have you tried another processor as I suggested?

miecho18 wrote:

Have you tried another processor as I suggested?

Currently, I do not have such a possibility or any other processor for this socket.