Liquid Metal GS75 Stealth,

any one tried the liquid metal on a GS75 stealth ? any concerns or problems on the long run using LM on laptops ?!?... is it worth it ??...

thanks,

 

guys any one ?!?!

 

yes

 

can you share your results ?...

 

I haven't personally done it but here are some results from others;

https://www.reddit.com/r/MSILaptops/comments/bjatj8/gs75_2080_liquid_metal_results/

Disassembly;



Note the inverted motherboard. I would proceed with caution, as usual.

 

92 to 75°C in prime 95 small fft

Gpu doesn't benefit much from liquid metal. A good paste will do. I recommend new mastergel maker (flat headed) , it works better than kryonaut and has better longevity too.

 

GPU DOES benifit from LM. My GL63 1660Ti with LM and .05 12.8 wmk Odyysey pads instead of factory paste runs anything and everything maxed out and never breaks 86* CPU 76* GPU. And thats with LM only. No undervolting and no fan tweaking, LM and pads instead of paste.
Until people start making stencils for the heatsink copper buttons, that match EXACTLY the dimensions of the CPU-GPU mobo dies they are going to have problems.99.9 percent of failures with LM are because of sloppy application or outright laziness and taking a short cut.
LM will cut the temps more than any paste. LM the CPU but Not the GPU? Like do you think that cutting the GPU temps as much as you can does not have an direct effect on the CPU temps?
Everythings in a BOX with heatsinks and fans trying to get it out.A 2-3 degree difference is major to overall internal temperatures.To suggest to LM a CPU and not the GPU "because it doesnt help much" is an outright stupid statement.

 

Compared to a high end paste that lasts far longer and is only 2-3 C hotter, its not worth the effort to redistribute LM every 3-6 months just to maintain 2-3C cooler gpu temps than a high end paste.

And if you don't maintain it, you'll have higher temps than paste so whats the point?

You probably don't get it as the 1660ti has a far smaller die than a 2080 so you might not get the pooling issue in just a few months. In that case LM on the gpu might be worth it.

 

I did a fair amount of research before making the move to LM. I havent seen a trend to a re coat every 3-6 months and there are 5-10 million people out there doing it. I assume you are talking from personal experiences.Pooling issue? When your heatsink is removed and there is a wet "pooled" area surrounded by a dried out area? That pool is from surface tension "sucking" to the point of separation when you remove the sink .The "dried out" area is usually the alloyed area from the Copper-gallium interaction.I would say your "pooling"" problems are caused by an uneven die-heatsink button interface and failing to use a stencil for exact die contact patch with precise matching edges on your heatsink button.Some understanding of Surface Tension also helps.If the two coated surfaces contact exactly when mated, surface tension will maintain an even coating across an uneven interface (to a point). If the LM is just smeared onto the heatsink button, a bang or jar will cause the LM to seep "Down" and out onto the smeared heatsink, air enters the interface causing reactions and drying and you have to re coat. Vivid imaginations and sloppy workmanship are the main reasons for "Fails".
As far as "Not getting It". Give your head a shake! Die size has nothing to do with it A larger die size would only provide more surface area to transfer heat to the sink.
Why would I LM the CPU and NOT the GPU? The CPU "wouldnt" need maintainance ??? AGAIN! A 2-3 degree drop on the GPU DOES make an increase in the OVERALL cooling and thats what its all about isnt it?
If you have a high current 2080 card you should be the one most interested in that 3 degree drop!

 

It mainly depends on your skill set.I believe your rig is the new one with the mobo that has to be removed for upgrades.If so thats a con there. Secondly you might require a second coat of LM at a later date if you didnt get it quite right the first time so ergo twice the work. LM if applied properly should last longer than any of the pastes before a re do .If you have any doubt about your abilities, dont do it. If you do I can say that the Coollabratory Ultra is fairly easy to work with.Not quite as highly rated as the grizzly stuff by a minimal margin but more of a pasty consistency that does not run all over like those terrifying videos the sensationalists post,very stable.I stenciled my heatsink LM application and the sharp lines on the border after the tape was removed could not be moved by slapping the heatsink on the side of the table, So!
If you trust your abilities go for it.The biggest danger involved is when you lower the heatsink onto the mobo.If you dont have the die and heatsink contact patches aligned, things can go south quickly and you cant see it happening. Particularily when you are using the recommended Foam Gaskets for Just in Case Protection. (Check my Posts) .

 

Ahhh! Dont know what happened here so I will fill the space.
To remove the MSI Blue Paste from the GPU use a wooden toothpick. Scrape the sides upward and work the paste into a ball on the top center. Then pick off the ball. The paste is rubbery and rolls into itself easily. Sounds intensive but after a minute you will see how quickly it goes. Used this method when I found that the paste could not be dissolved with Iso or Acetone.

 

from my testing "gt75 7820hk"
for stability and others --better thermal paste
,.,.when find any high end paste

and LM
works fine ,,.but this come after time
first rebuild not must works stable or long time "LM is absorbed in copper faster"
after 3x re-paste --possitible use LM --year stable "but you must have luck"

I only use LM ,,, but sometimes it is possible "when it's dry" - the problem with shutting down for 40C "difficult to determine what's wrong"
.,,.,when it works - 100W - 75C - no problem .D

Of course it wants to work with LM carefully
best is 3m tape // nail polish /// any isolation around - silicone or foam

for mobility --better paste
I'm still afraid that it will leak somewhere during transport and destroy the board "now only using foam"

using "tg conductonaut"


,.,.,.,.

 

https://www.liquidmetalalloy.com/about-us

 

I read somewhere that they only applied LM to the CPU and just regular paste to the GPU which may still negatively affect CPU thermals / performance to some extent. Someone (a Zephyrus G15 owner) also mentioned from his experience the stock LM application left much room for improvement. After repasting there's some sizable gains.

 

IWhat part dont you understand about OVERALL heat reduction. Half the clowns who post on here dont have the skills to open the case. If you take everything posted here as Oath, you are lost my friend."SOMEONE SAYS".yeah! Right!
This site should be mostly used as a reference of "What NOT to do".
eg: can you tell me whats the best "Paste" to use, or should I use the "pea size" method or "the cross" method of application.Ya SPREAD the damn stuff! I mean how do you put your Peanut Butter on your Toast FCS?? http://forum.notebookreview.com/threads/long-term-liquid-ultra-results.834821/page-3
Do it the way I have shown in the link.Nickel Plated and BOTH heatsinks.It will be done Once and Finished.There will be no reason to do again for the life of the laptop.

 

Pasting is overrated IMO. I switched to graphite pads after using them on my server for a couple of years with steady temps. Finally got ick of scraping / cleaning / reapplying various pastes over the years on the laptops.

https://www.amazon.com/dp/B08BL3SCWH

$10 and you can trim them to cover the BGA's and be able to do a couple of laptops off a single pad.

 

Whats that got to do with Liquid Metal? If your saying Graphite is best for a laptop you are truly FITH!
I wanna ask you. Why do you think the Big manufacturers of laptops went with Liquid Metal?
Stick to your desktops and servers Stranger!
Second question. Have you ever used Graphite on a laptop?
Can you show a picture or do I use my imagination?
There is NO compression in Graphite Pads which is required in laptops to make up variances in component to heatsink clearances and that problem is prevalent on ALL laptops.

 

I didn't say best. It's a nice in between of both worlds w/ less chance of frying something with LM. LM in a laptop isn't the best idea w/o tons of precaution due to laptops being moved around constantly vs a desktop being put in place for years. Movement & LM = issues

 

Using Graphite pads on laptops was shot down years ago. Its a non starter! PERIOD!
This 12" HP Elitebook Im using right now has been using CLU Liquid Metal for 4 years. Its been dropped on the floor busting the screen. Left on the porch in the rain and generally gets tossed around like a beer coaster.If done properly,LM is the ONLY way to go.Any failures are caused by peoples inability to pick their nose and scratch their axe at the same time, let alone being unable to open their case.
"Tons of precaution" Dreamer!

 

Whoa calm down man. I was referring to Asus' factory applied LM paste (and the specific example I brought up being Zephyrus G15 2021. I believe they came with TG conductonaut).

Also link to the post I was mentioning: https://www.reddit.com/r/GamingLapt...us_zephyrus_g15_5900hs3070_factory_lm_review/

 

Sorry there mate! On heavy medication for the big "C" and am subject to rapid mood swings. As far as I know I am the only one Nickel Plating his heat sinks so you dont get the copper absorbing the LM and drying the interface. Check the link here and note the picture #5 showing just how well that copper sucks up the LM. I got a sore wrist grinding that down to bare copper for Nickel Plating.

. http://forum.notebookreview.com/threads/long-term-liquid-ultra-results.834821/page-3

 

I personally use only copper and works well 1-2 years "is my best results"
,.it is only necessary to repeat this soon for the first application "some rebuild works few days only !!"

and only cleaning ,..,do not grind the cooler or minimal
best results - nickel plating cooper for LM

//
and know where your limit is! "" know if it is possible to target this limit but not damage the device ""
EC shutdown limit "this not mean visible heat problem !"

I think everything that can be monitored in system is not entirely true ! "dry zones on LM is problem ,.,extreme risk for dead HW"
and at MSI I think all the data is a bit tweaked "you not see real data"

no limit - possible scale up vcore "above safe numbers" or hidden temperatures problems

edit
it is a pity that it is not possible to log the reason for shutdown on MSI "probarly from EC"
and what this chip really sees !

 

I've used all 3 in my GS75 and I'll say this. Graphite Pads only work well if your heatsink is lapped to a mirror finish. Paste works moderately well on the cpu and pretty good on the gpu while LM is undoubtedly the best for the cpu but it results in hotspots on the gpu after a few months even after the heatsink has finished alloying with the gallium in LM, itll just naturally pool in some areas of the gpu due to its cohesive nature. Besides, theres no thermal benefit to having LM on the gpu vs a high quality paste so why bother?

 

There is no such thing as "Pooling". There will be "Dry Areas" due to absorption into the Copper leaving "Wet" areas "OBSERVED" as Pooling. Nickel Plating removes this problem. If properly plated, the procedure is done ONCE for the life of the laptop.3+ years on a HP Elitebook with absolutely NO CHANGES!

 

The LM no longer being absorbed into the copper as i mentioned before, the copper has already alloyed fully. The LM amount doesn't change and when i spread it around it can still cover the whole die well but after 3-6 months the hotspot problems with resurface.

On the cpu side this isnt a problem but for the gpu it is an actual problem once you get to larger dies like the RTX 2070/2080 or RTX 3070/3080.

 

AGAIN! Nickel Plating is the ONLY and FINAL solution. I have Four rigs all LM on Nickel.Oldest 3+years.I had heatsink off 3+ year Elitebook for cleaning Radiator fins last week for the second time and there is no change from the day it was done.Carefully removed and cleaned fins, re installed and done! No need to re wet! If you have Stenciled an exact match of the die contact area on the NICKEL PLATED GPU/CPU Heatsink with Sharp,Matching edges there should be no problems no matter Die size or less than perfect alignment.The "Gallium Oxide" skin formed along the sharp edging will keep contaminating air out of the die/heatsink Interface. If you dont understand the physical properties of a material (LM) your working with, maybe you shouldnt be using it?? Looking at the extremely poor workmanship in the posted photos,there is no wondering why people cant get simple things to work.
eg: mixing "Gorilla Snot and Thermal Pads"????
People cant seem to find the time to get some Modeling Clay, put on the heatsink DRY and then measure compressed thickness.The clearance between heatsink and components can vary "EVEN ON SAME MODELS" and "Exact same Machines"
Got an email from a guy I sold a Nickel/LM GL63-9SDK a while back. Said "You sure got the cooling figured out. Runs cooler than my Desktop".
BTW: GL63 takes .05 pads to replace the FACTORY Gorilla Snot, which is removed by rolling into balls with a toothpick and then picking those balls off.It will not dissolve with iso or acetone and will only make a mess.
FYI: Once the thermals are transferred thru the Nickel plate to the copper button and heat pipes, that heat is retained in the copper by the Nickel plate reflecting the heat back into the copper heatsink and NOT radiating it back out the bottom onto the motherboard and CPU/GPU areas. The Black Heat Paint keeps it in the copper on the other side.
On newer rigs, MSI GT75 onward, where they "cheaped out" by not painting the Heatsinks and piping?? Paint em! with Black automotive "Header Paint" Check your thermals before and after. Surprise!