Liquid metal thermal paste

[rds1220]

Has anyone here ever used liquid metal type thermal paste like Coollab Ultra? I've been reading up about it and it seems to give better cooling performance over regular thermal paste. If you have used this type of thermal paste how was it?

 

[rds1220]

No one at all has ever experimented with this type of thermal paste.

 

[madsweeney]

As much as I enjoy playing with mercury, I have not tried that paste.

 

[rds1220]

That's just a myth. The metal they use is gallium not mercury or lead as is often said.

 

[madsweeney]

I know, just a joke man, it is the internet.

Haven't tried the stuff personally, sounded not worth the hassle over Artic Silver. The stuff is conductive and corrodes aluminum and copper, not exactly something I'd want to put in my computer.

 

[JackNaylorPE]

All ya ever wanted to to know about TIM's can be found in this 80 way comparison

http://benchmarkreviews.com/index.php?option=com_content&task=view&id=150&Itemid=62&limit=1&limitstart=12

here's the top 5 (A+ rating)

Tuniq TX-3 (0) Aluminum Oxide Moderate 37.65°C A+
Gelid GC-Extreme (0) Aluminum Oxide Low / Thin 37.65°C A+
Thermaltake Grease A2150 (4) Polysynthetic Silver Low / Thin 37.65°C A+
Arctic Silver 5 Polysynthetic Thermal Compound (4) Polysynthetic Silver Low / Thin 37.55°C A+
Shin-Etsu MicroSi G751 (0) Aluminum Oxide Moderate 37.55°C A+

Here's how the Liquid Metal stuff came out ....26th place

CooLaboratory Liquid Metal Pad (0) Liquid Metal Liquid 38.40°C B

http://benchmarkreviews.com/index.php?option=com_content&task=view&id=150&Itemid=62&limit=1&limitstart=1

Some Thermal Interface Materials are made of liquid metal alloys or gallium. These TIM products are rare, and often times corrode metals they are not meant to come in contact with. At the time of this article, Benchmark Reviews has tested only one liquid-metal product (from CooLaboratory).

You'll note that AS5 and Shin Etsu tied for best thermal performance but there's an asterisk regarding the curing time for AS5 that's further explained here....bold part is off AS5 website:

http://benchmarkreviews.com/index.php?option=com_content&task=view&id=138&Itemid=62&limit=1&limitstart=5

Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity.

So by my estimation of this statement it would take almost a year of normal use to properly cure the AS5 compound, or almost nine days of continuous power cycles to meet their recommendation.

 

[rds1220]

I've never heard that it corrodes copper. From everything I've read aluminum will corrode but copper was fine.

 

[rds1220]

Jack that's pretty bad. From what I have read it was said that liquid metal conducts heat better, so much for that.

 

[AdrianMan]

I currently use Coolaboratory Liquidmetal compound and I believe it delievers as promised. Back in 2009? I used this compound to "weld" my then new Q6600 to a Swiftech Apogee waterblock. I still use that PC to this day. Now, the soon pride owner of a i7 5820k with probably EK EVO X99 waterblick, I have to decide if my Coolab Metal compound is better than Gelid Extreme, the new TH choice in the matter. More soon.

 

[4Ryan6]

This is what liquid metal will actually do to aluminum!

Before injecting the rest of the contents of that tube shown into the aluminum channel to leave overnight you can already see what it had done to the aluminum ram sinks in about approximately 8 hours. After seeing what it did to the ram sinks my curiosity was up so a 2nd test was to put it into a piece of aluminum channel. The rubber you see in the aluminum channel is to hold the liquid metal in that test area.



After over night It seems the moisture from the morning dew even accelerated its weakening to literal disintegration levels.



Do not use that product on anything that has aluminum content!

http://www.tomshardware.com/forum/274824-29-thermal-compound-roundup-2011

Scroll down to the Warning:

It will not corrode copper or the CPU heat spreader, but it is very difficult to remove with possible warranty loss from defacing the CPUs heat spreader, I had to lap the heat spreader on the CPU that was used for the testing.

That should be enough information for you to decide what to use.

Edit: Obviously since I conducted a thermal compound roundup, I have used a lot of thermal compounds.

I am presently using the Geilid GC Extreme and so far it is an excellent performer.

 

[AdrianMan]

LOL! Well, very interesting experiment with amazing results. But the product clearly states it will corrode aluminium, and with that in mind, I think we're all safe. As I was saying, my old Q6600, on the PC i'm using to write this right now, it's using this compound, and for what its worth, I believe it performs amazingly well. And I just found this benchmark from TH,

http://www.tomshardware.com/reviews/thermal-paste-performance-benchmark,3616-17.html

that clearly shows that liquid metal is the top performer, beating GC Extreme with a whole 1.5 degree. So, the half syringe of this wonderful compound I still own will be put to good use and will save me a few bucks too.

 

[4Ryan6]

@AdrianMan

There's no doubt liquid metal is the top performer with a fresh application of it however after I had completed the Thermal Roundup I linked above and saw my AMDs heat spreader was already lost, I reapplied a fresh application on the test machine and decided to just do the lapping later on.

I ran that machine running the same OC the tests were run with for 3 months, and then decided to go ahead and pull the heat sink and lap it, and the CPUs heat spreader.

But before I pulled it, I reset the rooms ambient testing parameters and reran the original tests.

The liquid metals cooling performance actually dropped by 3c, and when the heat sink was removed there was evidence of the liquid metal drying out in the center of the heat spreader where the highest concentration of the CPUs heat output was concentrated.

So from my own actual experience the cooling performance of the liquid metal will degrade over time and it will begin drying out under high heat loads.

You probably should check out what's going on under your heat sink.

Edit: Supplying some pictures of your discoveries would be appreciated.

 

[AdrianMan]

I don't really care, my Q6600 still going perfectly, equal cooling between 4 cores with only a 6 degrees increase from idle to 100%. (38º idle, 44º C full), with 27 ambient temp or so. No reason to mess around with it. And with over 7 years running, I believe the situation is pretty much stable. I have no doubt that between applications, a thermal compound might give different results. In my case, I have lapped ICH and water block base to mirror finish, as it seams this kind of compound is better spread thin at high pressure. It makes sense.

 

[synphul]

People get too concerned over the thermal pastes. That link JackNaylorPE posted (along with similar found around the web) shows how little difference there is among most common thermal compounds. From a+ to b+ rated compounds there's less than 1c difference. That in itself is close enough to consider margin of error. Mounting variations of the same cooler through several attempts to attach it will likely result in similar or larger temp differences.