Which Thermal Paste to buy and apply (Traditional and Liquid Metal)

Intel cannot actually. Their chips are way too power hungry lol and ARM were designed from scratch without the hole backward compatibility mess that x86 has to deal with. Imo intel got caught off guard with all these mobile chip stuff. Hopefully they can come up with something that can kick the A series chips from Apple so Android phones can benefit from it.


Sent from my iPhone using Tapatalk

 

please keep this on topic.

 

Sorry.

So back on topic. The test mentions the super thick paste are not good for low pressure HS then why does IC Diamond do so good on laptops? I got the 24 carat at home and I tested it with my fingers and yes, when I squeeze it I cannot make it spread like other pastes. It has almost playdoh consistency.


Sent from my iPhone using Tapatalk

 

The exact opposite is true. Either you misunderstood or the person that said this is mistaken. Thicker paste is best for low contact pressure or (heaven forbid, but it happens) no contact situations.

 

Delete

The paste ain't the problem!!

 

Is IC Diamond worth paying extra for in a laptop?

If IC Diamond isn't good for laptops, why do some very reliable resellers offer it as an upgrade. HID and Gentech both seem to think it lowers Temps.

Sent from my SM-G955U using Tapatalk

 

IC diamond lowers temps and it's good for heatsinks with poor contact. However it can scratch dies, and there have been bad batches floating around as well, but it's good for not degrading and for lasting a long time. IC diamond is always better than stock paste. Besides crappy jobs done at Alienware, some stock pastes are actually pretty decent. The stuff MSI uses actually works okay and lasts a long time without degrading, but it's not suitable for high overclocks.

IC both 7 and 24 are outperformed pretty significantly by Kryonaut now, however. Many people are swearing by Kryonaut now, although some people have seen it degrade after a short time (which it should not be doing). It's best to make sure you have a completely firm and airtight, solid mount however (a very VERY VERY light and careful lapping with very fine sandpaper can help a bit here, but do NOT lap like you're working on a desktop heatsink, because laptops have very low compression pressure, and removing any material besides nanometers may hurt cooling. If you screw the screws partway in a criss cross pattern before tightning all the way, and have a perfectly mounted heatsink, performance should last a long time.

Liquid metal is the best to use on laptops but if you have NEVER repasted before, don't try it. use Kryonaut until you learn how to repaste easily and efficiently. LM require a lot of preparation (for soldered CPU's with cooper heatsinks, tape/nail polish around resistors on the slug around the BGA, soft compressible foam that does NOT touch the main slug (you don't want it touching the LM!) but acts as a "dam" around it to stop any from moving and getting on your PCB, if you plan on using your laptop as a laptop and traveling all around with it, but the results are worth it. Once you've seen the temps you get, your jaws drop and you can't go back to regular paste.

 

That's a lot of good information. Thank you. The company I'll probably order from only offers IC Diamond and Coollaboratory.

Should I go with the liquid metal? If I do go with the liquid metal, would they put that on the GPU as well?

Also keep in mind, I don't really like tinkering with my laptop for the first 2 years. I would like to go with whatever paste is going to last at least 2 years. If I do need a repaste, I'll likely take it to Micro Center and have them do it.

Sent from my SM-G955U using Tapatalk

 

Well i was referring to Grizzly Conductonaut.
That's what I'm using.
Some have said that newer batches of Coollaboratory Liquid Ultra are not as good as the older stuff. Namely their newer stuff is more liquidly than the older stuff which was closer to a wet paste, in which case you're better off going with the Grizzly.

Put Kryonaut on the GPU's. Liquid metal shows almost no improvement on laptop GPUs because of the combination of high surface area (compared to BGA CPU's) combined with lack of heatsink size and pressure, meaning the heatsink is usually already going to be working at close to maximum efficiency with just Kryonaut.

Put LM on the CPU's (copper OR nickel plated heatsink only with Liquid Metal, NEVER aluminum).

 

@Mr. Fox @TBoneSan @Ashtrix @Phoenix @ajc9988 @Falkentyne @Vasudev @Mobius 1 @Johnksss@iBUYPOWER @hmscott @Donald@HIDevolution @bloodhawk @Arrrrbol @aaronne @Prema @Jon Webb @Coolane @ole!!! @bennyg @cj_miranda23 @ssj92 @D2 Ultima @Stress Tech @leftsenseless @Georgel ++++++

This is the Thermal paste KING!! Old news, I know... But Run and buy. Dump Liquid metal if you use it, into the dustbin and re-paste with the best yoo can get Why go for a lower quality than Mx-4?


Edit. Hope you all understand my harsh ironic comments in this post regarding Mx-4. FYI It's written in an ironic style. Arctic Mx-4 thermal paste is a no go for use on hardware in laptops!! Even a bad choice if you think use it on your desktops Cpu or Gpu. This not very good Thermal paste is popular due the creamy aka low viscosity for a very easy application. But as I pointed out... A really bad choice!!

 

@Papusan What did I just see?

Tested on Geforce 8800 GT
Data Source: PC Games, Hardware, Germany August 2010, page no.20-21

.....um...........uh...........
Where's my code for the blue shaking face?

 

Wait a second, Phobya Nanogrease Extreme needs to be applied with their spatula? This means that the typical X or blon method don't work wth it? O_O

 

For best performance, use spread method by spatula. Same is true for TG Kryonaut, Cooler master makergel nano. For IC Diamond 7 you must use a blob or pea size or X method because the paste itself is very thick.

 

I agree with Brother @Vasudev on this. Spatula is the best way. Application is superior and there is less waste of the product as well. I actually started spreading IC Diamond as well.

 

Good luck to you. I haven't seen that much thick paste in my life.

 

Wait a second, so using a Spatula won't create isues like.... air bubles?

Also, I wanted to ask about that Phobya Grease thingy, it is safe to use for absolute noobs? Like ABSOLUTE NOOBS WHO NEVER EVER EVER APPLIED A PASTE BEFORE?

Also, I would love to know how much I should apply. And how it should be properly applied on P775? I imagine that the screws must each be screwed only a very little at a time and such.

 

No problems with air bubbles here.

Yes, Phobya Nanogrease Extreme is totally safe. It is not electrically conductive.

Lay down a line of paste on the edge of the die and pull it across the die like spreading peanut butter on bread. If it is not enough to reach the other side, lay down another line of paste on the opposite edge and pull it toward the paste you already spread. If you have any leftover paste on the spatula, wipe it off on the center area of the heat sink contact surface. Then clean off the spatula and put it away for later use.

 

Yes its safe unlike LM which is conductive. Start with a small blob on gpu and cpu, start spreading it horizontally just once with a little force. Don't worry if you didn't spread at the edges it will be spread evenly when HSF is screwed in.

 

Interesting material, that recently caught my eye is this one. Its high-end material used as thermal paste. Iam not quite sure about the price. The catch is that it is changing a lot so it differs from transaction to transaction. Its mostly about 60 dollars per syringe so maybe too expensive to use at home. Not it is mostly used in industrial area. The properties are quite impressive though.

 

Cooler master Maker Gel nano, phobya nano grease, ICD 7 use same things as you posted in the link. 60$ too much expensive.

 

Perhaps my newest thread I just finished posting tonight may be of interest to some of the people here in this thread:

http://forum.notebookreview.com/thr...cientific-opinions-on-heatsinks-pipes.811453/

Have any of you messed with Indium Corporation products before? Their (super expensive by the box, but not by the piece) metallic heat springs boast an 86 W/mK temperature conductance. (That's +13W/mK better than TG Conductonaut's 73W/mK.)

I've been reading up on a lot of Indium Corps really interesting products and technologies recently. They have tons of data, blog posts, videos, technical documents, etc to go with all their products and they sell to little guys, albeit not exacly the cheapest.

HEAT SPRINGS: http://www.indium.com/thermal-interface-materials/heat-spring/

HEAT SPRING BLOG VIDEO: http://www.indium.com/blog/solder-redefined-part-4-baseplate-to-heat-sink.php

EDIT:

Also, why doesn't anybody consider using Fujipoly XR-m thermal pads that are rated at 17 W/mK instead of thermal greases or pastes? By the numbers Fujipoly XR-m pads are even better than GELID Extreme 8.5 W/mK and TG Kryonaut 12.5 W/mK which are some of the most highly regarded non-conductive TIMs.

 

Maybe because it's pads!!

 

This is why I am asking these questions. My direct experience is limited and I only have the numbers presented to me to go off of...

So my takeaway from that chart is that although the pads are technically rated at a higher thermal conductivity, their thermal resistance is also higher.

 

Thermal pads do not fill in microscopic imperfections on the contact surface, as thermal compound does.

 

Thermal pads will never be as thin as thermal paste. The right amount of paste will only fill up where metal against metal ain’t possible.

 

Yes, this much I am familiar with. On a given surface area I can't imagine the microscopic imperfections amount to very much of a percentage of the total area though which is why I was curious if a higher rated thermal pad could overcome a given (lower rated) TIM. My takeaway based on what I have seen from others and what I know and the graphs and charts I have seen so far seems to boil down to BLT and how thin the TIM can be made to fill the void between die and heat spreader plate.

 

I think the person is asking why can't you use a thermal pad, that has a higher w/MK rating, on a CPU.
It's not about the microscopic imperfections. It has to do with what item you are using it on.

A CPU is a VERY high wattage device that uses a LOT of heat. I'm not technically minded here, but a CPU will overheat to 100C instantly if there is no heatsink on it.
Thermal pads are designed to be used on power mosfets, chokes (VRM system) and RAM chips (or SSD cooling), where the amount of power and heat these devices use are MUCH MUCH MUCH lower than a CPU or GPU. A CPU can use 45W of power. A mosfet does NOT use this type of power, nor does a BGA RAM chip. They generate heat, but much more slowly, so thermal pads will work very well here. Someone else with an explanation more precise than mine, who actually knows what they are talking about, will have to fill in the rest here.

 

If you use thermal pads instead of thermal paste, you will loose everything of aviable contact point between HS and IHS. The thermal pads will functions like meat/salad in a sandwich.

 

take a look also on this one (sometimes it melt same as CLU and must be scratched away)
*** MSI 16L13 (Eurocom Tornado F5)/EVOC 16L-G-1080 15.6" Owner's Lounge ***
20W/mk
Indigo Xtreme™ Metallic Thermal Interface Solution

 

Post some before and after CPU/GPU temps after you used Indium extreme.

 

sorry was a long time ago and I must dig into hdd archive (sk775 and nehalem cpu on desktop) but I looking to buy again this month to test on Tornado F5 7700K + Bitspower lid (and share result here obviously, but from what I remember it act like every other LM compound but more safer)

Edit! end of life from EK, they purchased license and ended it some year later
https://www.ekwb.com/shop/ek-tim-indigo-xs-intel-115x
new formula 40W/mk
post edit:
I used the CLU metal pad, not the indigo, so I talked about reusability if not thermal stressed too much

 

Well conductonaut and CLU advertise 73m/WK and 40-50 W/mK.

 

This is interesting stuff, keep us posted.

 

I really love this one from the website https://bestthermalpaste.com/ @Mr. Fox @Donald@HIDevolution

 

You might wanna update your best thermal paste list. Cooler Master maker Gel Nano is the best one you can get from CM.
The list already has IC diamond, Notcua NH1, Conductonaut, gelid extreme.

 

Changed every month...RIDICULOUS!!

 

That sounds fishy to me, I don't think I've seen any paste that requires replacement that often.

 

Let me clarify. By RIDICULOUS I meant that is a ridiculous comment by the reviewer. We used IC Diamond for years before switching to Grizzly Conductonaut. When applied correctly, it lasts for years.

 

We all know ICD can work wonders when HSF is warped and only downside is the die surface gets etched for some reason.
@Phoenix said ICD was better than TGK for him.

 

Yes, we think IC Diamond is better than TGK as well. However we think Gelid GC Extreme beats them both.

 

Did you read that web?
It's like a little kid wrote it.
Typos everywhere.
Wrong words and spellings everywhere.
Terrible english.
I honestly don't even know if it's just a collection of web feedback put onto someone's 'review' page in an attempt to look good or popular or something, but it was cringe worthy. People should at least BUY the paste and test it themselves and learn how to use and balance heatsinks and not blame the paste.

 

I'd say a good review/comparison should help people buy one time instead of having to test out many on their own. That said, this page was not a good review/comparison.

 

Then again 3M won't make crippled product in the first place.
I don't have personal experience on TGK and Gelid extreme. I'm happy with CM MakerGel Nano. Its like ICD but it has the consistency of TGK and spreads easily. Its not thick too.

 

According to most reviews, GeLid extreme seems to provide best results in regards to other non conductive thermal pastes, and has a good price to boot.

For best possible results in terms of cooling, Conductonaut thermal paste seems to be the one, affording another 5-8 degrees Celsius reduction vs GeLid extreme.
However, conductonaut is also corrosive, so you need to be careful applying it to the cpu (usually what works is to cover the surrounding area of the die with a thermal tape to prevent thermal paste from accidentally spilling over and damaging the components).

 

I think you meant conductive instead of corrosive. It only eats Aluminium and not copper.

 

Yes. Thanks for correcting me.
And 'only eats aluminium' is not exactly a reassuring statement for a thermal paste. :-D

 

It does embed itself and "soak" into copper over time though with minor pitting/oxidation. It can be sufficiently addressed between repastes with a little elbow grease however.

SEE SPECIFICALLY:: "Dangers of Liquid Metal - Reactivity with Copper Heatsinks":

http://forum.notebookreview.com/thr...-works-why-it-fails-and-how-to-use-it.809332/

 

Just wanted to add, just like what many others have said, NEVER EVER BUY *Cough Splutter* *grey toothpaste* MX-4 FOR USE ON A LAPTOP

I never knew of this thread before I bought MX-4. That transaction was essentially $10 down the toilet.

I did make a thread about it before. On stock thermal paste, AIDA 64 stress test caused the CPU to become a maximum of 90C. On the first day of repasting, temperatures dropped to 75-80C max.

I've repasted with MX-4 at least 5 times now. The last time I did it was two weeks ago and I've simply given up. Now, temperatures shoot up to 97C within a minute, triggering BD PROCHOT and thermal throttling. That's worse than the 1 year old stock paste I had before. In fact, I've had 4 laptops in about 6 years, and not one of them have ever given me thermal throttling.

Well then again, Acer did decide to stuff a 30-40W dGPU in here with a 15W CPU (which I recently forced to run at 23-25W) cooled by ONE single SHARED heatpipe with ONE fan. And to think I've seen chinese knockoff laptops with the exact same CPU cooled by two heatpipes, it indeed is a poor effort by Acer.

Honestly, my old Acer laptop (15W CPU with 30W dGPU) never peaks above 70C during AIDA 64 Stress Test. For my current Aspire V Nitro, whilst the dedicated Geforce 945M is much more powerful and adds more heat to the heatsink, I'd only consider a reasonable temperature to be at most 80C during max load. Believe me, it does happen on the first day of an MX-4 repaste.

Maybe adding another heatpipe (Which I plan on doing) and using Liquid Metal would reduce max load temperatures to 70C? I'm not planning to use liquid metal any time soon but I'm hoping that heatsink mod will help thermal dissipation.

 

I used to repaste every 2-3 weeks until judal57 told me to look for better paste, In fact I did and it has been 6 months w/o any degradation in temps. Consistent as always. Instead of LM look for phobya or Gelid extreme or kryonaut for best results.

 

I'd probably skip liquid metal on a laptop too. It's fine on a delidded desktop, but you benefit from a more viscous compound in a laptop.