I was wondering if liquid metal can be applied on the 1950x, like say Conductonaut?
In other words is the IHS or lid made of aluminum?
According to this Reddit thread, the Ryzen Threadripper lid is made of aluminium and you shouldn't use Conductonaut on it. : https://www.reddit.com/r/buildapc/comments/8779mc/do_not_use_liquid_metal_or_conductonaut_on_a/ .
Copied from above link Post question:
Google Photos image link: Nasty!
This is after 2 weeks of usage. I noticed my temps were getting worse and worse, and correctly suspecting the conductonaut TIM I broke it down. If you look closely you can see the pitting from the gallium eating into the IHS.
Warranty status: Voided.
To fix this, I used 000 steel wool to clean as much liquid metal as possible off the IHS and heat sink, and then used a rubbing compound to try and shine them both up.
After some elbow grease and lots of nervous sweat, I cleaned both surfaces with alcohol and applied Kryonaut instead. Much better temps!
The threadripper IHS's must contain a great deal of tin, which reacts to gallium. That or it's some kind of aluminum alloy, in which case in 2 more weeks my temps are going to get worse again as the corrosion continues.
I'm considering lapping my block and IHS. We'll see.
TLDR: Gallium + aluminum = Horrifying. Gallium + Tin = Also horrifying, but not as horrifying. Do not use liquid metal (or conductonaut) or any other gallium based thermal interface material on a threadripper IHS.
Edit: I've contacted AMD, and they say the IHS is made of aluminum.
EDIT: 04/09/2019 - This Reddit Thread is incorrect. A later Reddit thread and AMD Support varifies that Ryzen IHS is Nickel Plated Copper.
Thanks to another User that posted AMD response and the Reddit Thread that I found as posted below this comment.
Thanks for the correction and update.
This later Reddit Thread concerning using Liquid Metal seems to agree with your post. https://www.reddit.com/r/Amd/comments/8v03xf/ryzen_watercooling_and_liquid_metal/
From posted link:
The ryzen IHS is nickel plated copper so I think LM will be fine on it. Although you can't do too much research.
Nickel plated copper is perfect... The gallium shouldn't leach much out of the galinstan (alloy that is liquid metal) because nickel plated copper and galinstan have almost the same galvanic rating. There will be some leaching, but it will reach saturation at very low levels and be nothing like unplated copper, where it will actually discolor the IHS/heatsink and "dry out" the liquid metal after a year or so.
There are a bunch of scientific papers written on this that you can potentially pirate via scihub if you don't have access to a University network. About 20 years ago a ton of research was done on this, due to research grants from companies like IBM and Cray, who were looking for ways to better cool their high thermal density super computers.
If you have a nickel plated heatsink and a nickel plated IHS. With something like Thermal Grizzly Conductonaut, if applied properly, you normally don't even have to worry about reapplication due to all the gallium leaching.
btw, with a bare copper heatsink... If you don't sand/grind off the discoloration each time you have to reapply the liquid metal, the copper after a few reapplications will reach full gallium saturation, and no longer require reapplication due to "drying" the galinstan out.
I also received a response from thermal grizzly which is posted below
i have read on the internet and several persons stated it is made out of aluminium. If you have an official response from AMD saying it is made out of nickel plated cooper, you can use Conductonaut. Nickel plated cooper is the best-use case for Conductonaut.
Jannik Froehlich - Thermal Grizzly Support
Anandtech has a really good review of TIMs from liquid metal to toothpaste and denture cream. liquid metal is certainly the best, but the gap between it and the next is very slim. The real issue is the cooler not a degree or two provided by the TIM. I have seen images on the Internet of Ryzens exposed to liquid metal and it is not pretty. I think AMD has responded in this forum. I recommend users interested search for AMD comments here, read Anandtech's review and search for images of Ryzens with liquid metal exposure. Enjoy, John.
Hey there. I'm the guy from the original reddit thread. The pitting that Conductonaut LM created on my IHS, though you cannot see it very well in the picture, was easily sanded off this year when I decided to Lap my threadripper. After spending years with this CPU I am of the opinion now that I was unintentionally misled by AMD support on the material used for the IHS. I've heard from many other sources it is Nickel Plated Copper. My IHS hasn't turned brittle or cracked yet, which also indicates that it is not aluminum.
That being said, I can't vouch for the purity of the nickel used as the pitting caused by the LM had some depth to it, and was not sanded off too easily. You can see before and after on the bottom of my rebuild photo album: https://imgur.com/a/yy98XuF
I still do not recommend LM for Threadripper IHS. I still use LM for GPU direct Die cooling as it can't be beat, but the few c of gains from switching to Conductonaut from Kryonaut on a threadripper are not worth the potential maintenance down the line.