Cooler Master apparently absolutely screwed up. I only hope that the costs for the fix Framework is working on will be upheld by Cooler Master, who is responsible for this mess, and not by Framework.
6 Likes
I wonder if someone could try removing the stock copper shim and connecting the 0.8mm shim with liquid metal to bith the die and the vapor chamber. Could be cool to see what the cooling system is capable of.
i wouldn’t try this, as the Spillage Risk is alot higher. I will again try to solder the 0,8mm Shim. i need to get my Gear dialed in 
1 Like
I agree 101%. If Cooler Master wants to have a long-term partnership with FW they need to get this issue worked out and clean up the mess they created.
2 Likes
Thank you @Qyygle for the good explanation of what is wrong.
My understanding therefore is that this is the problem part:
Nice experiment!
I wonder if it will help if you try to press down the shim while heating it so that it will reduce the amount of solder.
@PSierra117
Nice Catch 22 there.
Solder the shim and the heat pipes come off. Solder the heat pipes and the shim comes off. 
Maybe use lower temp solder on the shim? Would need extremely tight temperature control though cause the melting temp of different solders aren’t that different.
Probably not worth it.
@Alex_Uta
The Heatsink fell apart in the Oven, i wanted to preheat the whole assembly.
When i Heat the Shim with Low Temp solder (138C) it is melting really fast, but as soon as it touches the Vaporchamber its solidifying, as the Vaporchamber and Heatpipes dissipate the Heat.
Thats why i wanted to preheat the assembly but my oven whas to uneven.
I do not have a good controlled temperoven, where i would set the Temp to 140C and i have no heat resistant Clamps to hold the OG Parts together. Loose Loose on my experiments 
Edit: i regeated the already soldered assembly and pressed on the shim every now and then until i saw solder beeing pressed out under there.
Now it looks alot more uniform. Tomorrow i will try the assembly installed on my Mainboard.
5 Likes
That looks quite nice.
I’m quite curious about how it will perform.
Thanks for testing new ideas 
Replying to this thread, I’ve been putting up with this for quite a while now. It’s nice to see a fix is being worked on. But I wish there was a time-frame.
1 Like
I overheard a conversation a while ago on a repair shop while getting my phone repaired. Gaming laptops with liquid metal can have the liquid metal “creep” sideways if you put the laptop vertically like carrying it in your backpack. This will cause uneven thermals. The solution is reapply the LM or put it vertically the opposite direction. Could this be one of the reasons of this problem?
I think it’s probably part of the problem but there is nothing most people can do about it. How can you guess in which direction to store your laptop without first separating the cpu from the heat pipes?
I had the exact same issue with an asus laptop that we are seeing with the FW16 today. Liquid metal just seems to age and degrade rapidly so it might be slightly better than paste when it is first applied but it doesn’t last and is soon much worse.
Laptops are also more exposed to this issue than desktops because they move a lot and are often stored in a different position each day.
Until Framework came along is was just part of the business model: laptops and gaming laptops in particular are made so that you buy one and you really want to replace it with a new one 2 or 3 years later.
It would be really nice if Framework is able to change that … but at the same time I fear that Framework is being held to a higher standard than other companies and it’s going to be difficult and costly to make their business model work.
3 Likes
Hi everyone,
Thank you for your huge work ! Seems that my batch 2 needs some help. Stop reading this thread in august, I just took time to read every post I missed.
Testing with cinebench, no more than 13871.
I just opened a support case.
Again, thank you
1 Like
I have had some time to test my modified heatsink with shim a PTM sandwich.
I am basically getting consistent 16400+ scores in CB, cores are at 4400-4500Mhz (Ryzen 9) and core temps are at 88~100c. The cores have a fluctuating ~5-10c delta over all cores between cores during the benchmark. So it looks like al cores get enough thermal headroom to push to the limit.
1 Like
A question regarding the shim.
What would happen if there was no solder there?
I.e. Heatsink → shim → ptm → cpu.
Surely the pressure applied between the heatsink and cpu would be enough to hold it in place? Maybe just a dot of solder each edge to stop it moving sideways.
Then, why not remove the shim, and just replace it with thicker ptm ?
Note: I am no expert, so these are more questions than suggestions. I am just thinking that removing the solder for the shim might make it cheaper to manufacture.
Also, PTM7950 vs PTM7958. What is the difference, Lenovo use PTM7958.
My guess is maybe PTM7958 is easier to use on a production line.
The thermal transfer between the heatsink and shim would be terrible, same as if you omitted the thermal paste between the chip and the shim.
PTM is liquid during operation so the max thickness has some pretty hard limits not to mention the thermal conductivity of the ptm itself probably being a lot (like orders of magnitude) worse than copper.
Thermal conductivity is measured in W/mK which tells you the thermal transfer is a function of temperature difference and distance, the bigger the distance the more it hurts so to speak.
1 Like
My machine since I got it 3 months ago has been sitting flat and never put in a bag or carried yet. I still have the same thermal issue like others. It is a design flaw and Cooler Master needs to step up and work with FW to get us a fix.
Opening a support case is good it just provides more evidence (with each case opened) that the issue is a fundamental design flaw. Unfortunately, you will go through the RMA process and get another board with the same issue. I had opened a RMA as well and requested FW to put my RMA on hold until a validated/official fix is announced. Of course you are welcome to go through with your RMA but I am just providing info based on my experience. Thanks!
Yeah I see your previous post saying that you put on hold your RMA, so I expect to be able to ask the same. Having a motherboard with the same issue is useless and framework is a small company, I don’t want to “charge” them for a second motherboard for no reason. I prefer to wait for a proper solution.
2 Likes