Thinkpad T61 GPU thermalpad

Hi there, Today I disassemble my T61 in order to clean the fan of the heatsink, I cleaned everything up (processor, northbridge and GPU) and applied new thermal grease, assemble all together again and checked the temps readings, and I found that it didn't help, I still get the same temps, 58*C - 64*C on the GPU core.

I have a question though, there are 3 main chips that have active cooling, the CPU, the GPU and the third one, but I'm not sure, the Northbridge?

I found that 1 of those 3 chips have a thermalpad (like hard paste) transferring the heat to the heatsink, my guess is that the thermalpad is on the GPU, so, it doesn't do a direct contact with the heatsink (like the CPU does), causing less dissipation.

According to TPFanControl the CPU is at 50*C idle, the BUS (Northbridge?) is at 37*C and the GPU is at 50*C.

I want to know if the GPU is indeed the one who has the thermalpad on it, and if I can remove the thermalpad and place a copper shime or a solid custom piece of copper onto the GPU contact area to decrease the GPU temps even more.

My dilema is that according to GPU-Z (0.3.8) the GPU while idle is at 60*C so I don't know if I should trust in the TPFanControl reading or the GPU-Z read.

Any thoughts?

Thanks for reading me.

 

The northbridge is a bigger chip than the GPU.

I would do the copper mod. The NVS 140m has the faulty G86 (or G84, don't remember which but they are both bad) core. In fact you can find out which chip is the GPU by reading what is written on them. The GPU should say "nvidia...G86..." or something. The other chip should be an intel chip (this is the northbridge, it will have an "i" written on it).

Anyway keep the GPU cool so that it doesn't fail.

 

GPU-Z is a nice little tool to identify what GPU core you have.

 

The 140M is the direct derivative of the 8400M GS, period.

 

Check the Hardware Maintenance Manual ( http://www-307.ibm.com/pc/support/site.wss/document.do?sitestyle=lenovo&lndocid=MIGR-67760).

Also, good idea- the G86 is a faulty chip and if it gets to hot- it's doomed to fail.

 

It has to do with heating cycles, and the packaging not conforming properly to the expansion and contraction of the GPU chip... If we all kept our 8600M GT laptops on all day, it won't fail for a longer time, that's why i use TPfancontrol to turn the fan down after gaming as soon as i exit, because the reduction in CPU clock speed while the fan continues on maximum RPM, cools the GPU extremely fast, like 65C to 50 in a matter of seconds because the fan continues to blow.

 

Haha, let me rephrase. It's a matter of chance, but taking care of keeping it cool over time is essential.

 

If the GPU is the smallest chip, so it has direct contact with the heatsink, leaving the northbridge with the thermalpad thing.

Though I'd like to know if would be good do the copper thing anyway?

G86M according to GPU-Z

Yep, G86M.

Thanks for the link, if broken though.

While I cleaned the fan and the heatsink and applied new thermal grease, the temperatures keep remaining the same on the GPU, so I guess that's the best that the heatsink can handle the heat generated from the GPU.

I dl TPFanControl, the stock fan settings are Smart, but that settings only put the fan on if the GPU reaches a temperature above a point (is my guess??) so I select whether BIOS or Manual (3) to keep the fan always on.

Is there any way to modify the pattern of Smart and change it to BIOS or Manual for good?

Thanks again all for your imput.

Happy Holidays

 

in addition if we only run XP and nothing else graphic intensive, then the GPU would be fine, but that seems to defeat the purpose of having a discrete gpu in the first place.