Liquid metal TIM replacement question

[DaEnglishMuffin]

Hey all. I just bought a new laptop and apparently it benefits from LM TIM replacement. In order to make sure I didn't kill my new computer, I tested the process on my old Lenovo T430.

 

I applied conformal coating to the aluminum surrounding the copper on the heatsink

I applied conformal coating to the CPU and GPU processors around the die.

I applied just enough Thermal Grizzly Conductonaut to make the dies shiny and metallic. 

 

The results are included in the pictures. Can anyone better explain why the temperatures don't reflect what others have seen (10-20 degree drop). I think the cooling solution may be bottle-necking my results, but I wanted to see if anyone had any other ideas. With results like this, I don't really want to open my new laptop up.

 

Pre-LM Idle

 

Pre-LM Load

 

Post-LM Idle

 

Post-LM Load

 

Post-LM/Post-Stress Test Idle

 

 

 

 

 

 

[Amaranth]

I would repaste the laptop again since you shouldn't have that much difference between the cores unless you misapplied the liquid metal or have a warped, or otherwise damaged, heatsink. Also remember that you need to coat the underside of the heatsink itself for optimal performance since the liquid metal has a very high surface tension and doesn't like bridging the gap between the die and heatsink unless there is some liquid metal already there to attach to.

[AMDNeko]

 Laptops usually do not have enough mounting pressure to cause insane drops and the heatsink is still a limiting factor...some laptop hsf will actually make LM not work at all since its too low so it just spills out. If you do not want to ruin the new laptop you can use foam dam method and ensure mounting pressure is high enough before closing it up and calling it a day. In low pressure areas ICD and Gelid will be better suited than Kryo or any LM~

[DaEnglishMuffin]

So I followed the advice of Amaranth and did see an improvement. The temperatures took longer to rise, but unfortunately reached similar levels.

 

Post LM Idle #2

 

Followed by Post LM Load #2

 

So there was an improvement, but I think that PPCs-Kat may be on to something: the mounting pressure is now the limiting factor. Not ruling out my original conclusion either: the cooler/fan is now the cooling bottle-neck.

[DaEnglishMuffin]

Pangea2017, who is XMG? I would like to see the research and conclusion that you mentioned.

[DaEnglishMuffin]

Followed by Post LM/Stress-Test Idle