THANK YOU to everyone who watched this video, it really means a lot to me and i just wanted to say Tank U. make sure to subscribe and comment down below if you have anything you wanna see me do or check out.
@Xtreme Performance I fully/basically agree. I think your most important comment is "for best performance De-lid your CPU". I personally would apply liquid metal between IHS and a copper heat sink (be it water or air) only if I target enthusiast extreme overclocking out of 2 reasons: 1) The temperature down when applying liquid metal as shown here (above the lid) is only 1/10 compared to when applied on a delidded CPU (between CPU die and lid/nickel plated copper heat spreader, this means below the lid) 2) The longevity (usable time) is much more limited when applied on copper compared to nickel plated copper (gallium reacts much faster with copper than with nickel) Concerning application procedure I would say as follows: A) Mandatory is a protection against catastrophic short and corrosion (aluminum) damage by spilled liquid metal B) Both sides need to be wetted over the complete contact area, because liquid metal easily oxides on air with oxygen (that is the reason why it forms balls). And just one time movement when putting both surfaces together would wet both surfaces to may be only 10~20%. C) Ideal is to start the wetting on the smaller area and to realize minimum thickness/minimum danger of spillage then not to reapply liquid metal on the opposite larger area, but just use the remains on the tool used for the smaller area to apply a very thin wetting over the complete contact area. D) To increase the longevity (at slightly higher thermal resistance and strongly larger danger of spillage) you can intentionally target a thicker liquid metal layer/TIM, because the longevity is proportional to the amount of gallium/thickness of the liquid metal as the speed gallium disappears from the liquid metal into copper(/nickel) is basically constant at thicknesses of a few 10µm. The trigger point for the remaining liquid metal to become hard at room temperature is when the gallium percentage in the liquid metal falls below a certain percentage (Note: This video shows a small amount of hardened liquid metal built up already after 2 months, this means the hardening is not an abrupt process but happens continuously with steadily lower percentage of gallium in the liquid metal/over time).
You DO NOT apply liquid metal like you do thermal paste. You'll need to spread a thin layer out on both the head spreader and heatsink surface otherwise it'll spillover into your motherboard and fry it. I replaced my GPU's thermal compound with LM and the temps are great. Programs started crashing after a year though but a re-application/top-up fixed the issue. The surface was rough and slightly pitted, but wasn't detrimental to performance.
@Jake C you beat me to it, i saw straight away that he did not apply the LM correctly. how has this dude even got a channel, logic dictates that all surface area of the top of the cpu should have LM applied.
@@05C177I would not use nail polish as you are not guaranteed it will work, there is a special chemical in some nail polishes (acrylic) which is very effective at insulating electrical current. Not all nail polishes have this not even 30% as it's kinda expensive for such a cheap product. If you decide to go ahead and use nail polish you may have absolutely nothing protecting the SMD's making it completely pointless. Acrylic conformal coating is what you would ideally use to protect the SMD's and not the silicone type most use. They can both work but the acrylic version is significantly better at preventing a short due to it's electrical isolation properties. The silicone also does this but is not intended to be used on high current which is not something you want in this case. Conformal coatings mainly block electricity and provide a "air gap" between the current and sensitive traces/SMD's with the main benefit being it can withstand high heat these are the two traits that separate conformal coatings from common nail polish. Folks use silicone because it's easier to come by in a local store while acrylic is more difficult but odds are most will have to order it online so they should go with acrylic it just offers more benefits for this type of usage. Also if you are nervous still kapton tape can be your best friend even if you do/don't use a conformal coating it's main purpose is to prevent shorts in electrics where two boards may come in contact or the touch a piece of metal which could allow it to jump. Kapton tape can be about 10 times better in terms of electrical isolation than conformal coatings but conformal coatings offer a complete seal over the components while kapton tape only covers it. Kapton tape can also withstand about 300c which is far beyond what any graphics card will ever see even at it's hottest spot.
So it ate into the copper plate and now its heat conductivity is compromised. After watching this I'm never going to consider using liquid metal thermopaste.
3 things when using liquid metal. 1st. The flatness of the IHS is very important. this important for all thermal interfaces really but LM doesn't really like filling in big gaps. Its good to check with a straight edge, lapping might be necessary. Intel is infamous for their out of flat IHS's. AMD is better but you can still run into one. my 2700x was a dome, couldn't figure out why my temps were never that good. until i checked it on a piece of glass and it spun like a top..... 2nd. you need to spread the liquid-metal evenly on both surfaces to break the surface tension of the liquid metal. Its a wetting procedure. you'll use WAY less LM, save the applicator included by thermal grizzley; normal q-tips leave fuzz. 3rd i found when applying LM that i got the best "thermal" results when polishing both halfs of the thermal interface. I use Flitz. This removes the oxide coating that forms on the surface of the metals and allows the LM to "wet" the surface better. However it does allow for faster Ion transfer, it tends to make the LM form a hard alloy with the copper much quicker, it's not pitting or corrosion. This does not effect the thermal conductivity but can make your CPU get stuck to the copper cold plate..... also had that happen lol.
The raised bit you can feel on the heat block surface is probably the remaining tin and indium after the gallium migrated into the copper. While the discoloration doesn't appear to harm the thermal performance, the pits and crusty raised bits could worsen the contact between the surfaces. Lapping will fix that, but if your heat block has a convex surface, it will be flat after lapping. The Intel IHS is normally fairly convex, so that is something to consider if you do not wish to also lap the IHS. You're using a CPU that has relatively low performance heat paste between the CPU and IHS, and that's surely the thermal bottleneck, not the IHS to heat block surface. You'd probably see better results delidding and putting the Thermal Grizzly between the CPU die and IHS and using non-metal high performance thermal paste on the heat block. That seems to be the most popular use for Thermal Grizzly among hobbyists! If you want to keep using liquid metal on the heat block, the best way to do that, IMO, is to nickel plate the copper surface first. It sounds a lot more involved and difficult than it really is. I found a brush-on nickel plating kit online (search Caswell Plug N' Plate Nickel kit to get the one I have) for under $50. I wanted to improve the thermal performance of my gaming laptop (thermals on those can always use some help), but I'd tried the Thermal Grizzly with a laptop heat sink before, and while it worked great at first, it soon led to rising temps as the gallium migrated into the copper. I had the same raised crusty bits and pitting as your heat block. I did not want the same to happen to my new(ish) laptop's heat sink, and nickel plating was the most promising answer. I thought at first of finding a commercial plating company and seeing if they would do it, but the ones around all seem geared toward bulk jobs, not simple one-offs like mine. I would imagine the cost would also be high. The kit above was the perfect answer (FWIW, I have no connection to the maker of the above kit... I am just a computer hobbyist/enthusiast myself). I had expected there to be complications, things that were more tricky than they seemed, the kind of stuff that always seems to happen, but there was none of that. And now I have the ability to plate anything else I want... I used so little of the plating solution that the bottle still looks full. I've had the Thermal Grizzly on there for about 4 months, and the thermal performance of the laptop has not gotten any worse. That's not really that much time, but by this time my other laptop had already started getting hot with the copper heat sink and Thermal Grizzly.
there's little to gain using LM on top of the IHS. it's maybe 1C to 2C better than conventional thermal paste like Kryonaut, PK-3, or MX4. it's not worth the risk of damaging something. under the IHS is where LM shines. you can gain 15C to 20C under a de-lidded IHS. LM is even slightly better than solder.
The application is really bad...you should stop making videos like this, as it doesn’t help people less experienced hat would try to use LQM. Wrong application, the result is not accurate. Please , just stop...who did you see apply liquid metal like this?
@@mattnutt2738 most common is using a cotton swab and rubbing it all over the surface of not only the cpu but the cooler too. in the beginning the drop soesnt want to spread but with a little bit of preasure (so the LM heats up a bit) its spreads really good and becomes a thin shiny layer across the surface. make sure your cooler is made from copper or nickel as otherwise the gallium in the LM will destroy your cooler over time. when i bought my LM from thermal grizly there were some cotton swabs in the package.
I've been using liquid metal on a 6700k with a H115i as the cooler or at least a year now, no major temp differences between now and when I installed it. May be worth a check I guess.
Brake Clean works the best. And keep using those blue shop towels. They are super absorbent. And if you're worried about lint, you can wash and dry them with a load of laundry. They come out almost like microfiber! Also, put the LM on a piece of cardboard first, and the use the Q-Tip to pick it up. It can come out of that syringe very fast and unexpectedly. Apply a super thin layer to the block and the IHS, by way of "polishing" it on to the surface. Less is more in this case.
After using Liquid metal on a few components I've noticed the problems started when its exposed to fresh air where oxidisation starts up and then I seen slight pitting on my X62 and no amount of cleaning would remove the stain, only a bit of sanding from 800 grit sand paper removed all marks to a mirror finish.
You know Liquid Metal absorbs into copper contrary to popular belief you have to apply it twice you removed the layer that would make the next application similar to thermal past because it creates a middle ground between the exposed copper and the bare Liquid Metal it makes the Liquid Metal act better as a thermal paste you are trying to use it for
Do what I'm doing. I'm using my Swiftech H240 X3 AIO to cool the 40mm×80mm copper water block attached to the hot side of two TECs, or Peltier Modules. Then on the cold side it too has a copper chiller block with temp controller. The chilled loop is completely independent of the AIO loop and will be chilling a super overclocked 8700K and OCd 1070Ti.. if I get the EK ASUS Turbo VGA water cooler. Either way I want to fine tune this for the next rig and scale it down as much as possible.. which I do from the get-go anyway. Stupid goal is to hit over 6Ghz on all 6 cores for over 24 hrs at, or near, ambient temps - trying to avoid the condensation.. although, the condensate could be used to chill the hot side's cooling system even further, if it decides to stick around. That being said it'll probably be more expedient for you since you got the hook from the show, I imagine. I just got my copper blocks today and ordered them 4 weeks ago. Aluminum works just fine and it's properties dissipate heat a lot faster than copper which tends to hold onto the thermals a bit longer.. .. okay maybe going with the aluminum block for the hot side.. if the thermoelectric loop inevitably runs too cool I can warm it up by running another block on the hot side to hopefully improve efficiency and performance. I would love to see your rendition of this project of mine. I tested one TEC1-1706 with two lithium 18350 batteries giving it 8.4V-8.6V out of its possible 12V and without a fan on a passive heatsink it reached -15° C
Even with that tiny drop , it goes a long way , i see the same thing in laptops but they never spread , so its safe to say it doesnt act like what normal thermal compound do
@@JoshuaG Yes they don't spread when pressed down with the cooler cold plate but they will move around if it's not tight or IHS and cold plate is not really flat enough since it's in liquid state not unless some of it started to hardened and formed a natural barrier for leak over time until it's fully hardened. I'm curious when it's fully hardened if your performance will degrade from day 1 good to hardened LM and will it thermal throttles?
@@RaimarLunardi it may not be, but it's still not going to stain it all that quickly, it'll 100% stain it yes, but not super quick as i said, and more often than not they are pure copper just for cooling efficiency, the NZXT and corsair stuff are generally 100% copper as are many of the custom waterblocks available
@@RaimarLunardi that still doesn't disprove my point, since, you know, I'm talking from experience here and not basing it off of it not staining pure copper quickly, but even down at 80% it doesn't stain quickly
What is stuck to the copper is likely gallium oxide ., its probably not chemically bonded, but to remove you will probably need an acidic solution like weak muriatic acid, or even just soldering flux. However its probably not worth the effort.
I have only applied LM to Nickel plated heatsinks. I don't want to deal with the reactive issues from copper. I can confirm that the heat sink on my EVGA 1080ti SC black is nickel plated.
Did you noticed the two drops of liquid metal on the side of the IHS that you forgot to clean up in the first place? They are there as you applied the new batch of liquid metal and could be a potential risk, if they drip down after you stand up the PC in its normal position. For now I hope everything is fine in your computer :-P
when i was editing the video i did see those and was like how did i miss that, i thought about taking everything apart and cleaning it up but to be honest i was to lazy so i decided i roll the dice and i haven't yet had any trouble and i move and jostle this system all over the place so i think i will be good "final last words"
I have serious doubts that the cooler cold plate is solid copper which would account for the reaction, it looks like copper coated aluminum solid copper will be copper colored all the way through.
The question I have, is that heatsink pure copper? Copper is quite expensive, so it wouldn't surprise me to find there are other metals in the mix. These rouge metals could cause unusual reactions. Manufacturers are not alway 100% honest and if they can save a few pennies, then they will.
I have had liquid metal between my 6850k and my fractal design S36 AIO liquid cooler since the summer of 2017. It is now December of 2018 and I have observed no decrease in performance of the plate or negative effects in temperature. I haven't opened it up since I applied the compound and I'm sure it looks like shit, but who cares about the aesthetics of a plate and a cpu you never see. If you're concerned about the stains on your cpu lid, turtle wax medium polishing compound can remove all of it (but, generally, the writting goes as well). I have personally tried this and it leaves the cpu looking shiny and new and other youtubers have demonstrated this.
exactly for some reason people think liquid metal was deigned to go between the die and IHS, which i wasn't it was designed to be a high performance thermal compound. Imagine if a company designed a product that required you do void the warranty on the vary product its was designed to be used with. Its probably because thats how most people see it used and due to a lack of research think thats that proper applications. I, like you have had no thermal issues no migration issues, obviously my copper is not stained but what does that matter is has no effect on its thermal conductivity and how often are you admiring your coolers copper plate.
@@MajorHardware Yes correct it only become mainstream to use it directly in die to ihs because no thermal paste have better performance than liquid metal. They can use regular high quality paste in between the die and IHS like Intel use toothpaste ahaha . But they will never reach the Liquid metal performance. That's why it become standard when you delid you need liquid metal. And liquid metal is advertised as thermal paste in liquid+metal form in between heatsinks. And since it's dangerous to delid a ryzen processor because it soldered I also plan it to use as paste replacement. Im doing more research about it before applying see how it performs and hits flaws. So far Nickle plated is best not copper but copper is fine but you need to re-apply after months just like this video in order to allow gallium to get absorbed by copper. It also enters Nickle but not as much as copper since Nickle have more resistant to it and it's not really noticeable and it's easier to remove in Nickle than copper except for some few gallium who got absorbed by Nickle well Nickle is the barrier coating of the copper. Good thing I have Nickle plated water block. There are few people who said to paint it let it cure (slightly absorbed or stick better in IHS and cold plate and re apply again but 1st coating must be thin really thin since your double coating it) So my understanding is, it's like painting the 1st coat will be the main and the 2nd finishing coat will be for protection or sacrificial coat. But in liquid metal the 1st coat will be sacrificial gallium loves to form a bond with metals so 1st coat then 2nd coat will be the main coat the will act as thermal paste and since there is already a 1st coating that is touching the unevenness or flatness of the heatsink the second one will be for proper contact or transfer of heat. I'm going to try this method but I will leave the case titled/Horizontally for 24hrs and leaving for the next coat and heatsink attachment I leave it horizontally for a week just for some curing in theory (since liquid might slide from cpu atleast for a week some of the gallium will bond with Nickle) And did some research how to properly apply safely #1 cutting a paper same size as heatsink, putting a tape in between IHS and paper and putting another bigger tape above the paper. #2 putting the waterblock or coldplate above the cpu, the paper will attach to the cold plate now remove it. #3 after removing use any type of tape and tape the corners, for much safer result put another layer covering slightly the side of exposed cold plate when you remove the paper (reducing the exact length and width of the cold plate to IHS contact using Liquid metal from the side to avoid potential leak). #4 clean with alcohol and apply liquid metal much thinner than recommended (let it cure) apply another layer. For cpu IHS apply a tape on the small corner of the IHS leaving very small allowance so that you can't paint it with liquid metal remove more like 90-95% coating and atleast 5% space on corners a really tiny space since unlike ps5 we don't have a protective seal and unlike Intel amd socket sucks nxt to the cpu is not the metal cover clip but the mobo IC and traces itself LOL. (Well I will consider of thermal paste to put tiny spread on the side of processor to block air going in and act as a barrier of the liquid metal since thermal paste is sticky and if you spread it flat it won't move to much and will stick with the coldplate heatsink and air will be also eliminated since paste fill up the gap better than liquid metal, well just a theory my choices are paste and tiny cut Thermal pad attached on the SIDE not top of ryzen iHS enough to make contact with coldplate heatsink to act as a seal protector at the same time block some air but I think there is easier) Finally attaching coldplate to cpu and let it cure for a week horizontally before tilting it back to normal
@@JoseSanchez-xj3xn ps5 is using this liquid metal and they cheap out the heatsink. And planned a sort of barrier and sealing the die preventing air or most air going inside. And only claimed they made their own Sony liquid metal that is different in others a long lasting liquid metal lol. As far as we know as long as it is touching Nickle and not exposed in the air it will stay in its liquid form for yearsssss example of this are delided Intel cpu specially when they glue the IHS again. Ps5 used cpu IHS design as its main heatsink design (basically a direct cooling but unlike GPU it is sealed like cpu) and inside there's a foam around the die not allowing the liquid to get out and if you remove the foam there is a insulator tape covering the IC. That's why when you open a PS5 the liquid metal is in its liquid state not hardened.
Please just use liquid metal on nickle coated products only or purely for delid & thermal paste for contact to copper will work well enough, that is why the Ihs doesn't tarnish. plus AIO'S aren't far off the price off a custom loop with a nickle coated cpu block with the option to add a gpu block
to be honest , im not sure why mercury and gallium isn't more widely used as thermal compound. its very good at conducting heat. they'll just have to be sure to "pre-alloy" it so its bound up with some metal already and thus shouldn't attack anything.
@@NeganLucilleForever I would hope your not eating your computer components but regardless. your not going to get mercury poisoning from metallic mercury especially if it's alloyed with something else. take tooth fillings for example, they have mercury in them.
@@logangraham2956 Actually eating mercury is (relatively) safe, but inhaling mercury fumes is death. You can make a case that almost everything contains mercury, but concentration is key. It stands to reason that a mercury-based thermo-paste will have a lot of mercury in it, which will evaporate as your processor heats up. Inhaling it will slowly poison you.
@@logangraham2956 Just read about mercury-containing amalgam used in dental fillings - it's deemed not safe and being phased out. So there you have it.
I can't find the update video to this. Did you ever check this after 5 years to see if it still worked ok? I'm stuck between conductornaut vs nt-h2 for a 13700k build.
Even light pitting is a bad sign. Light pitting is a start to full corrosion. Obviously over time , it's been months with no problem but I would invest in a new heat spreader for the cpu and just chalk the cooler up to a loss and keep at it! It's gonna take a long while to fully destroy it. But also keep in mind as it pits it will absolutely lose thermal conductivity
And also I don't care what gamers Nexus says. As it pits OVER THE REAL LONG TERM it will lose thermal conductivity, it just physically has to there is no way around it. But the time frames in play are so long that I bet you could use the components until your ready to upgrade before an issue arises
Long term those two metals reacting to each other will ruin that copper. Because of the heat it wont take to long. I use a laptop not sure it's worth it. I would love my cpu to run 10 degrees cooler though. For now I just use a much better paste than factory and it helped along with some undervolting on my msi gs65 stealth thin with gtx 1070 max-q. My gpu stays fairly cool unlike my heat prone cpu.
Is the liquid metal thermal compound largely based on Galiium? if so, that is why it eats the metal plate from the heat sink, I think that would not be a good idea to apply such kind of compund to COU cooling in the long run, where the gallium eats the metal plate would creat tiny air space/gap and negatively affect the themal conductivity.
yes gallium is the main thing in LM that likes to destroy things, Aluminum is a definite no go with LM but copper is said to be safe. But as you see its really stains the crap out of it. since that video my PC is still running that LM and i havent had any issues yet. temps are still good and system is still stable. However come september we will crack it open and see whats really going on in there.
@@MajorHardware I hope it will be still fine in coming sepetember, just worried if the Gallium eats the copper and other metal would form an alloy look like solid structure and forms tiny air gaps (the minor stains you have witnessed) within the structure, that may reduce the thermal conductivity since air is not well thermal coductive, but i dont think there will be serious damage to the overall temp performance given the copper thermal sink is still closely attach to the CPU cap.
The gallium in the liquid metal will react to copper but it's extremely insignificant and has no measurable effects in terms of cooling. For what it is worth it's not ideal to use between the IHS and the cooler can't really hurt I guess but it's generally frowned upon for what ever reason. Oh yeah the gallium will absorb into copper tho but it does not degrade the copper like it does with aluminum what you see left is caused but oxidation but once the copper has absorbed all the gallium it can (not much really) it will no longer cause drying like this. If for some reason you decide you want to remove the oxide metal polish works pretty well but I would strongly suggest not using sand paper (some folks think it's a good idea) even car polish (compound meant for scratch removal) can also remove it but the process isn't all that quick (but you will not be at it for a hour) about 5 mins depending on the brand. Just avoid cleaner waxes which is suppose to be a hybrid of the two, they either have micro abrasives (which is what you want will not effect the look of the copper block) and chemical which should be avoid as it will have no effect on the liquid metal.
good info thank you when i take this thing back apart i will be trying to get the copper back to its pre LM days so all ur recommendations are noted for sure
Eeeeh, the current will simply cut off if you do that, it won't work in any way, thats a terrible comparison. It will never work. Liquid metal act as a link between the cpu and the heatsink micropocketed surface.
Gallium foprm CuGa2 alloy with copper with good conductivity it does not harm at all the performance also by seeing that bloob looks like you dont cover whole IHS, also u need apply thin layer on the cooler plate as well other wise gallium wont stick.
i have been using LM for about a year now under and on top of my IHS.... i am using a rocket cool IHS. but the only thing that i got on my system is the staining and yes a bit of pitting. but everything looks good. i will say that when i went to add new LM it was way easier to add it to the copper. also for ppl that are just getting into LM from what i have found if you are adding it onto bare copper let it sit or about a week go back in and add just the tinniest amount more. this will be smaller then you think. but i say to do that as i have found that i get better temps and coverage after i let it sit as it seams that letting the LM stain the copper made it way easier to get it to cover right. but also good vid
I’d say it’s not worth the risk of having my entire laptop (which is aluminum btw) fall to pieces if any gallium gets on it. And besides, I hear it destroys copper too- only slower.
I'm just about to upgrade my PC, but I've had LM as my TIM for 3 years (without changing for fear of what lies beneath. However, I have been running it like that with a Swiftech H240x on a 2500k @4.5GHz the entire time and performance has not degraded. Max temperatures I see are 63 degrees on the package (I use HW Monitor to watch for any issues). I'll be interested to see what it looks like when I do separate, as I plan to re-use the H240x with a 9900k...
from what i have heard/seen/read your copper cold plate will look like poop but it should work just a good as it did when it was clean, i would be interested to see what yours looks like as well.
Hey @@MajorHardware I replaced my chip a couple of months back, and grabbed an image of my cold plate (shows how badly I applied it too). If you're interested in seeing any, let me know. Nothing out of the ordinary, some slight pitting in one area, but nothing too bad. Currently keeping my 3900X cool and calm, this time with MX-4. Though I will be switching it out for an EK block at some ponit.
Why would you risk using liquid metal like that where it's not contained (like under the IHS in a delid)? Like everything else, it's subject to gravity and any liquid metal that's not clamped between two perfectly flat surfaces will move unless your PC is horizontally mounted (which it isn't). If even a tiny droplet of the liquid metal gets onto your motherboard you have 2 issues: 1. It's conductive so you will likely have an electrical short - RIP motherboard 2. It can react badly with the solder used since every board uses different solder compounds and some might react a lot worse than others. The reason most use liquid metal is to remove the crappy TIM under the IHS on intel CPUs, NOT as a replacement for non-conductive thermal paste (Thermal Grizzly. Arctic Silver etc) between the IHS and cooler heatsink. You would get the same results and be far safer using a regular non-conductive paste for your application. In this case the risks far outweigh 1-2 degrees difference (if that)
thank you for the comment my man, however i have had none of the problems you have stated and although gravity is not just a good idea its the law i have had no LM leak out onto anything and everything is working great as of now. however if things change down the line i will make sure to let you know because using LM as a thermal compound is only for SCIENCE as you can get comparable performance from a good thermal compound. Also LM was not designed to be used for delidding as no company would design a product that required you to void and potentially destroy your CPU, however give Intel's hate of solder (until recent) that was what it was mostly used for, LM was designed to be a thermal compound and does a good job, however with this product there are risks and everyone knows that myself included and i have choose to see what happens so hit that subscribe button and come along for the ride.
@@MajorHardware haha! Of course! Did you know that Coollaboratory make a liquid metal that's a copper/silicon mix that's not electrically conductive and designed for high temps? It's heat dissipating properties are better than most thermal paste and not the risk of regular liquid metal. I hope nothing goes wrong, that would be sad as it's a nice build. Happy holidays
@@MajorHardware Definition of RTFM in English: RTFM ABBREVIATION Computing vulgar slang Read the fucking manual (used especially in email in reply to a question whose answer is patently obvious). -oxford English dictionary en.oxforddictionaries.com/definition/rtfm
you can use nail polish remover to remove the liquid metal, alcohol wont quite do the trick, also as others have said you applied it improperly but thank you for learning for all of us :)
as for colors such as MOB and GPU i tried to use only neutral colors like black and steel silver. because im a big RGB fan (yes im that guy) i wanted to stay away from any permanent colors.
Im a Noob i just now figured out you are probably talking about the color of my RGB, at that time my GPU and background RGB color was neon green, the fans, ram, case edges were like a dark purple, i cant remember the color code but i was trying to go for a color close to what you see given off by a black light and thank you i did like that color scheme.
This is why I don't use liquid metal. I use Gelid Solutions GC-Extreme Thermal Compound TC-GC-03-A. My I9 7900X (Skylake) is running at 4.4MHz on all cores and runs about 30 C without load and about 60 C full load.
Nice video. Curious though it looked like the liquid metal wasn't cleaned off completely from the AIO pump head. If it was cleaned - then was that grey area where the liquid metal had been?
i put LM on my asus strixx 2080 ti for a few hours today and it stained the cooler already, only brought 5 degree better temps as well, so not worth it tbh
Has anyone tried using a metal polish for copper alloys instead of alcohol ? this might help with the staining and pitting as a polish will remove the surface layer of the copper and might smooth it out again
it also changes the appearance of nickel coated parts with more time, used it on my NH-D15 and is made the base of that and the IHS darker after about 5 months
Great video and I've just subscribed. I am however, sticking to Arctic Silver, as I've never seen any kind of pitting whatsoever with the hundreds of PCs I've built.
Rufinoman Isn't that the beauty of watching other people's results? The only reason we know the negative results of LM is because others risk their equipment and show the aftermath. I agree with you as well on sticking with more standard thermal pastes. I've never had a reason or need to change from Arctic MX-4.
AIO Liquid cooled a I5 3570k (H60 model) using traditional Thermal paste in 2012 when I built a customer's PC. Then in 2019 I recently acquired the customer's PC, and I just sold that CPU off. When I removed the AIO the CPU looks good as new no pitting no marks you could not even tell if it was ever used. Then I look at that AIO and CPU you got...... I think I will stick with traditional Thermal Paste.
Can you apply thermal paste after using LM for the first time. Today I used thermal grizzly conductinaut first time on my z73. Immediately realized mistake as it started spilling out. I have already ordered the grizzly thermal paste. Can I apply paste within a day or two application of liquid metal
Peace of advice if you have expensive equipment.. Just use thermal paste.. Liquid metal performs okay, but after a 1 year on my machine. I had to replace my CPU cooler due it. Not worth it.
do you think it is better for me to apply liquid metal in my 5800x? I have a noctua d15 heatsink with obviously noctua thermal paste and while I play I always remain around 70/75 degrees c.
In NO WAY did you apply that correctly. Its to be evenly distributed across the entire IHS. It is not paste. It needs to be thin. You will see a difference if applied correctly. And please tell people what product you use. Liquid metal is vague AF. All this pulling shit open to look is all for views. This stuff is Gallium. You can research all you need to know from scientists and real life testing without all the drama. The pitting you are getting is not a chemical reaction between the metals but a reaction to contaminants left during the application. Cotton swabs leave resume and cotton behind.
Bro i subbed you def deserve it. i hope you make it one day!! some tips : try to relax more on your body language, get bettet thumbnails and a better profilname / and picture 😬, the name and the pictures are just not flashing my eyes if u know what im saying. They dont really wanna make me click on your videos or channel, even tho your content deserves to be around 15k-50k or even more views.
thank you, i will do my best to improve, i appreciate any and all feedback, you cant improve if you dont know what you are doing wrong, THANKS for subbing hope to see you around.
@@MajorHardware Plan to liquid metal my 2k rtx2070 laptop, but feeling kinda unsure. Laptop runs hot and high temperatures accelerate this process. Bud good thing is that after reapplying, the process slows down due to the fact that copper becomes plated with gallium and further adding more gallium with each consecutive application reacts less and less. One of the concerns is that when eventually dried, the dry liquid metal powder may scatter when the laptop is carried around and short some circuits.
for something that new i dont know if your going to see any big benefits to using LM as a thermal paste, personally if it was me i wouldnt risk killing that laptop. I did make a video where i LM my old alienware laptop but that laptop was unusable so i had nothing to loose also it was really old anyhow. being that old the thermal paste was rock hard as i show in the video. that laptop is still running strong no signs of anything going wrong. in my opinion you wont notice any cooler temps liquid meteling or re-pasting that new laptop as the thermal compound in it from the factory is probably in great shape. to me the risk out-ways the benefits on a new machine.
yeah, the only problem is that stock paste that asus uses is a complete junk, paired with mediocre cooling renders lots of thermal throttling and fan noise
THANK YOU to everyone who watched this video, it really means a lot to me and i just wanted to say Tank U. make sure to subscribe and comment down below if you have anything you wanna see me do or check out.
Which are the blue paper towels you used? I really like them :P
dude, you forgot a tiny drop on a cpu heatspreader, on bottom left corner... clean it up before your pc burns itself down
Good job keep on doing the good work! :)
We need an update on this
@Xtreme Performance I fully/basically agree.
I think your most important comment is "for best performance De-lid your CPU".
I personally would apply liquid metal between IHS and a copper heat sink (be it water or air) only if I target enthusiast extreme overclocking out of 2 reasons:
1) The temperature down when applying liquid metal as shown here (above the lid) is only 1/10 compared to when applied on a delidded CPU (between CPU die and lid/nickel plated copper heat spreader, this means below the lid)
2) The longevity (usable time) is much more limited when applied on copper compared to nickel plated copper (gallium reacts much faster with copper than with nickel)
Concerning application procedure I would say as follows:
A) Mandatory is a protection against catastrophic short and corrosion (aluminum) damage by spilled liquid metal
B) Both sides need to be wetted over the complete contact area, because liquid metal easily oxides on air with oxygen (that is the reason why it forms balls). And just one time movement when putting both surfaces together would wet both surfaces to may be only 10~20%.
C) Ideal is to start the wetting on the smaller area and to realize minimum thickness/minimum danger of spillage then not to reapply liquid metal on the opposite larger area, but just use the remains on the tool used for the smaller area to apply a very thin wetting over the complete contact area.
D) To increase the longevity (at slightly higher thermal resistance and strongly larger danger of spillage) you can intentionally target a thicker liquid metal layer/TIM, because the longevity is proportional to the amount of gallium/thickness of the liquid metal as the speed gallium disappears from the liquid metal into copper(/nickel) is basically constant at thicknesses of a few 10µm. The trigger point for the remaining liquid metal to become hard at room temperature is when the gallium percentage in the liquid metal falls below a certain percentage (Note: This video shows a small amount of hardened liquid metal built up already after 2 months, this means the hardening is not an abrupt process but happens continuously with steadily lower percentage of gallium in the liquid metal/over time).
You DO NOT apply liquid metal like you do thermal paste. You'll need to spread a thin layer out on both the head spreader and heatsink surface otherwise it'll spillover into your motherboard and fry it.
I replaced my GPU's thermal compound with LM and the temps are great. Programs started crashing after a year though but a re-application/top-up fixed the issue. The surface was rough and slightly pitted, but wasn't detrimental to performance.
Ive did this too, alittle bit stain on the cooler auf my gtx 1080 after about 3 months of use but performance was good.
heat resistant nail polish helps this in the delliding process no? The TIM overflow
@Jake C
you beat me to it, i saw straight away that he did not apply the LM correctly. how has this dude even got a channel, logic dictates that all surface area of the top of the cpu should have LM applied.
@@05C177I would not use nail polish as you are not guaranteed it will work, there is a special chemical in some nail polishes (acrylic) which is very effective at insulating electrical current. Not all nail polishes have this not even 30% as it's kinda expensive for such a cheap product. If you decide to go ahead and use nail polish you may have absolutely nothing protecting the SMD's making it completely pointless.
Acrylic conformal coating is what you would ideally use to protect the SMD's and not the silicone type most use. They can both work but the acrylic version is significantly better at preventing a short due to it's electrical isolation properties. The silicone also does this but is not intended to be used on high current which is not something you want in this case. Conformal coatings mainly block electricity and provide a "air gap" between the current and sensitive traces/SMD's with the main benefit being it can withstand high heat these are the two traits that separate conformal coatings from common nail polish.
Folks use silicone because it's easier to come by in a local store while acrylic is more difficult but odds are most will have to order it online so they should go with acrylic it just offers more benefits for this type of usage. Also if you are nervous still kapton tape can be your best friend even if you do/don't use a conformal coating it's main purpose is to prevent shorts in electrics where two boards may come in contact or the touch a piece of metal which could allow it to jump. Kapton tape can be about 10 times better in terms of electrical isolation than conformal coatings but conformal coatings offer a complete seal over the components while kapton tape only covers it.
Kapton tape can also withstand about 300c which is far beyond what any graphics card will ever see even at it's hottest spot.
agree
So it ate into the copper plate and now its heat conductivity is compromised.
After watching this I'm never going to consider using liquid metal thermopaste.
It didnt eat into it. It combined with it. Youre supposed to add another drop after the first month. It will eat into aluminum, not copper or nickel.
3 things when using liquid metal.
1st. The flatness of the IHS is very important. this important for all thermal interfaces really but LM doesn't really like filling in big gaps. Its good to check with a straight edge, lapping might be necessary. Intel is infamous for their out of flat IHS's. AMD is better but you can still run into one. my 2700x was a dome, couldn't figure out why my temps were never that good. until i checked it on a piece of glass and it spun like a top.....
2nd. you need to spread the liquid-metal evenly on both surfaces to break the surface tension of the liquid metal. Its a wetting procedure. you'll use WAY less LM, save the applicator included by thermal grizzley; normal q-tips leave fuzz.
3rd i found when applying LM that i got the best "thermal" results when polishing both halfs of the thermal interface. I use Flitz. This removes the oxide coating that forms on the surface of the metals and allows the LM to "wet" the surface better. However it does allow for faster Ion transfer, it tends to make the LM form a hard alloy with the copper much quicker, it's not pitting or corrosion. This does not effect the thermal conductivity but can make your CPU get stuck to the copper cold plate..... also had that happen lol.
Great information thank you for the post
Honestly I am shocked you did not destroy your cpu and motherboard with the way you went about things.
The raised bit you can feel on the heat block surface is probably the remaining tin and indium after the gallium migrated into the copper. While the discoloration doesn't appear to harm the thermal performance, the pits and crusty raised bits could worsen the contact between the surfaces. Lapping will fix that, but if your heat block has a convex surface, it will be flat after lapping. The Intel IHS is normally fairly convex, so that is something to consider if you do not wish to also lap the IHS.
You're using a CPU that has relatively low performance heat paste between the CPU and IHS, and that's surely the thermal bottleneck, not the IHS to heat block surface. You'd probably see better results delidding and putting the Thermal Grizzly between the CPU die and IHS and using non-metal high performance thermal paste on the heat block. That seems to be the most popular use for Thermal Grizzly among hobbyists!
If you want to keep using liquid metal on the heat block, the best way to do that, IMO, is to nickel plate the copper surface first.
It sounds a lot more involved and difficult than it really is. I found a brush-on nickel plating kit online (search Caswell Plug N' Plate Nickel kit to get the one I have) for under $50. I wanted to improve the thermal performance of my gaming laptop (thermals on those can always use some help), but I'd tried the Thermal Grizzly with a laptop heat sink before, and while it worked great at first, it soon led to rising temps as the gallium migrated into the copper. I had the same raised crusty bits and pitting as your heat block.
I did not want the same to happen to my new(ish) laptop's heat sink, and nickel plating was the most promising answer. I thought at first of finding a commercial plating company and seeing if they would do it, but the ones around all seem geared toward bulk jobs, not simple one-offs like mine. I would imagine the cost would also be high.
The kit above was the perfect answer (FWIW, I have no connection to the maker of the above kit... I am just a computer hobbyist/enthusiast myself). I had expected there to be complications, things that were more tricky than they seemed, the kind of stuff that always seems to happen, but there was none of that. And now I have the ability to plate anything else I want... I used so little of the plating solution that the bottle still looks full.
I've had the Thermal Grizzly on there for about 4 months, and the thermal performance of the laptop has not gotten any worse. That's not really that much time, but by this time my other laptop had already started getting hot with the copper heat sink and Thermal Grizzly.
Probably really late, but how is the laptop holding up after 2 years
there's little to gain using LM on top of the IHS. it's maybe 1C to 2C better than conventional thermal paste like Kryonaut, PK-3, or MX4. it's not worth the risk of damaging something. under the IHS is where LM shines. you can gain 15C to 20C under a de-lidded IHS. LM is even slightly better than solder.
The application is really bad...you should stop making videos like this, as it doesn’t help people less experienced hat would try to use LQM.
Wrong application, the result is not accurate.
Please , just stop...who did you see apply liquid metal like this?
They liked it without reading it xD
Curious, how do you apply it then?
When ever I have seen liquid metal applied, it has been this way.
Matt Nutt You spread a thin layer over the entire ihs, He made a semi spread out blob on the cooler. So the application was inconsistent.
@@mattnutt2738 most common is using a cotton swab and rubbing it all over the surface of not only the cpu but the cooler too. in the beginning the drop soesnt want to spread but with a little bit of preasure (so the LM heats up a bit) its spreads really good and becomes a thin shiny layer across the surface. make sure your cooler is made from copper or nickel as otherwise the gallium in the LM will destroy your cooler over time. when i bought my LM from thermal grizly there were some cotton swabs in the package.
I've been using liquid metal on a 6700k with a H115i as the cooler or at least a year now, no major temp differences between now and when I installed it. May be worth a check I guess.
as long as you dont take it apart its fine. the metal sets over time but as it is metal the conductivity stays the same.
@@Raruschmaru so even it hardened there will be no real big difference correct as long as you don't break the bond correct?
@@gamertechlive1780 I guess so.
The staining on the copper can be removed with hydrocloric acid, but copper/gallium alloy is perfectly stable and does not need to be removed.
Brake Clean works the best. And keep using those blue shop towels. They are super absorbent. And if you're worried about lint, you can wash and dry them with a load of laundry. They come out almost like microfiber! Also, put the LM on a piece of cardboard first, and the use the Q-Tip to pick it up. It can come out of that syringe very fast and unexpectedly. Apply a super thin layer to the block and the IHS, by way of "polishing" it on to the surface. Less is more in this case.
After using Liquid metal on a few components I've noticed the problems started when its exposed to fresh air where oxidisation starts up and then I seen slight pitting on my X62 and no amount of cleaning would remove the stain, only a bit of sanding from 800 grit sand paper removed all marks to a mirror finish.
You know Liquid Metal absorbs into copper contrary to popular belief you have to apply it twice you removed the layer that would make the next application similar to thermal past because it creates a middle ground between the exposed copper and the bare Liquid Metal it makes the Liquid Metal act better as a thermal paste you are trying to use it for
Use nickel plated coolers. They are much better than exposed copper.
Unfortunately I haven't seen any "AIO" with an Nickel plated "block"..
Do You know of any ??
Best regards
@@mrdr9534 The Be Quiet! Silent Loop has a nickel plated block. There might be others but this has one for sure.
5:00 is this copper plated aluminium contact plate?
That is exactly what I was thinking... Copper wouldn't be that color... it would be darker or greenish
Do what I'm doing. I'm using my Swiftech H240 X3 AIO to cool the 40mm×80mm copper water block attached to the hot side of two TECs, or Peltier Modules. Then on the cold side it too has a copper chiller block with temp controller. The chilled loop is completely independent of the AIO loop and will be chilling a super overclocked 8700K and OCd 1070Ti.. if I get the EK ASUS Turbo VGA water cooler. Either way I want to fine tune this for the next rig and scale it down as much as possible.. which I do from the get-go anyway.
Stupid goal is to hit over 6Ghz on all 6 cores for over 24 hrs at, or near, ambient temps - trying to avoid the condensation.. although, the condensate could be used to chill the hot side's cooling system even further, if it decides to stick around.
That being said it'll probably be more expedient for you since you got the hook from the show, I imagine. I just got my copper blocks today and ordered them 4 weeks ago. Aluminum works just fine and it's properties dissipate heat a lot faster than copper which tends to hold onto the thermals a bit longer..
.. okay maybe going with the aluminum block for the hot side.. if the thermoelectric loop inevitably runs too cool I can warm it up by running another block on the hot side to hopefully improve efficiency and performance.
I would love to see your rendition of this project of mine.
I tested one TEC1-1706 with two lithium 18350 batteries giving it 8.4V-8.6V out of its possible 12V and without a fan on a passive heatsink it reached -15° C
ITS TO MUCH LIQUID METAL ON THIS TOOOOOOOOOOOOOO MUUUUUUCHHHHH
yup like 3x too much
Even with that tiny drop , it goes a long way , i see the same thing in laptops but they never spread , so its safe to say it doesnt act like what normal thermal compound do
not really
@@JoshuaG Yes they don't spread when pressed down with the cooler cold plate but they will move around if it's not tight or IHS and cold plate is not really flat enough since it's in liquid state not unless some of it started to hardened and formed a natural barrier for leak over time until it's fully hardened.
I'm curious when it's fully hardened if your performance will degrade from day 1 good to hardened LM and will it thermal throttles?
U missed 2 drops at the side of the processor.
I noticed that too. So stressful to watch him miss those two drops.
major mistake :)
This guy looks like american version of pewdiepie
that's not really a correct application of LM, and LM won't really stain copper all that quickly, but it will stain it either way, but that's fine
Copper used on heatsinks may not be pure copper...
@@RaimarLunardi it may not be, but it's still not going to stain it all that quickly, it'll 100% stain it yes, but not super quick as i said, and more often than not they are pure copper just for cooling efficiency, the NZXT and corsair stuff are generally 100% copper as are many of the custom waterblocks available
@@blazingmatty123 even on heatpipes they are at best 97% copper...
@@RaimarLunardi that still doesn't disprove my point, since, you know, I'm talking from experience here and not basing it off of it not staining pure copper quickly, but even down at 80% it doesn't stain quickly
try polishing with very fine grain emery cloth.that should take out the pits. then finish with metal polish.
Not smart. Youll have fine metal powder everywhere afterwards. Nice way to short and fry you components.
@@Chris-yy7qc if your very sloppy yes.
Did you come back with more info about this running with time?
What is stuck to the copper is likely gallium oxide ., its probably not chemically bonded, but to remove you will probably need an acidic solution like weak muriatic acid, or even just soldering flux. However its probably not worth the effort.
I have only applied LM to Nickel plated heatsinks. I don't want to deal with the reactive issues from copper. I can confirm that the heat sink on my EVGA 1080ti SC black is nickel plated.
Did you noticed the two drops of liquid metal on the side of the IHS that you forgot to clean up in the first place? They are there as you applied the new batch of liquid metal and could be a potential risk, if they drip down after you stand up the PC in its normal position. For now I hope everything is fine in your computer :-P
when i was editing the video i did see those and was like how did i miss that, i thought about taking everything apart and cleaning it up but to be honest i was to lazy so i decided i roll the dice and i haven't yet had any trouble and i move and jostle this system all over the place so i think i will be good "final last words"
Hehe, good luck 👍 ☺ 😄
I have serious doubts that the cooler cold plate is solid copper which would account for the reaction, it looks like copper coated aluminum solid copper will be copper colored all the way through.
Not supposed to be on copper
So you didn't do even a second of research before putting this all over your system? Well done.
The question I have, is that heatsink pure copper? Copper is quite expensive, so it wouldn't surprise me to find there are other metals in the mix. These rouge metals could cause unusual reactions.
Manufacturers are not alway 100% honest and if they can save a few pennies, then they will.
heat sinks are expensive too, lol
now you know why
There were two drops on the edge of the processor. I hope you cleaned it
Bro messing with everyone's PTSD in this Video.
I have had liquid metal between my 6850k and my fractal design S36 AIO liquid cooler since the summer of 2017. It is now December of 2018 and I have observed no decrease in performance of the plate or negative effects in temperature. I haven't opened it up since I applied the compound and I'm sure it looks like shit, but who cares about the aesthetics of a plate and a cpu you never see.
If you're concerned about the stains on your cpu lid, turtle wax medium polishing compound can remove all of it (but, generally, the writting goes as well). I have personally tried this and it leaves the cpu looking shiny and new and other youtubers have demonstrated this.
exactly for some reason people think liquid metal was deigned to go between the die and IHS, which i wasn't it was designed to be a high performance thermal compound. Imagine if a company designed a product that required you do void the warranty on the vary product its was designed to be used with. Its probably because thats how most people see it used and due to a lack of research think thats that proper applications. I, like you have had no thermal issues no migration issues, obviously my copper is not stained but what does that matter is has no effect on its thermal conductivity and how often are you admiring your coolers copper plate.
@@MajorHardware That's also a good point. At the end of this day that was it's original intention.
@@MajorHardware Yes correct it only become mainstream to use it directly in die to ihs because no thermal paste have better performance than liquid metal.
They can use regular high quality paste in between the die and IHS like Intel use toothpaste ahaha . But they will never reach the Liquid metal performance. That's why it become standard when you delid you need liquid metal.
And liquid metal is advertised as thermal paste in liquid+metal form in between heatsinks.
And since it's dangerous to delid a ryzen processor because it soldered I also plan it to use as paste replacement.
Im doing more research about it before applying see how it performs and hits flaws.
So far Nickle plated is best not copper but copper is fine but you need to re-apply after months just like this video in order to allow gallium to get absorbed by copper.
It also enters Nickle but not as much as copper since Nickle have more resistant to it and it's not really noticeable and it's easier to remove in Nickle than copper except for some few gallium who got absorbed by Nickle well Nickle is the barrier coating of the copper.
Good thing I have Nickle plated water block.
There are few people who said to paint it let it cure (slightly absorbed or stick better in IHS and cold plate and re apply again but 1st coating must be thin really thin since your double coating it)
So my understanding is, it's like painting the 1st coat will be the main and the 2nd finishing coat will be for protection or sacrificial coat.
But in liquid metal the 1st coat will be sacrificial gallium loves to form a bond with metals so 1st coat then 2nd coat will be the main coat the will act as thermal paste and since there is already a 1st coating that is touching the unevenness or flatness of the heatsink the second one will be for proper contact or transfer of heat.
I'm going to try this method but I will leave the case titled/Horizontally for 24hrs and leaving for the next coat and heatsink attachment I leave it horizontally for a week just for some curing in theory (since liquid might slide from cpu atleast for a week some of the gallium will bond with Nickle)
And did some research how to properly apply safely
#1 cutting a paper same size as heatsink, putting a tape in between IHS and paper and putting another bigger tape above the paper.
#2 putting the waterblock or coldplate above the cpu, the paper will attach to the cold plate now remove it.
#3 after removing use any type of tape and tape the corners, for much safer result put another layer covering slightly the side of exposed cold plate when you remove the paper (reducing the exact length and width of the cold plate to IHS contact using Liquid metal from the side to avoid potential leak).
#4 clean with alcohol and apply liquid metal much thinner than recommended (let it cure) apply another layer.
For cpu IHS apply a tape on the small corner of the IHS leaving very small allowance so that you can't paint it with liquid metal remove more like 90-95% coating and atleast 5% space on corners a really tiny space since unlike ps5 we don't have a protective seal and unlike Intel amd socket sucks nxt to the cpu is not the metal cover clip but the mobo IC and traces itself LOL.
(Well I will consider of thermal paste to put tiny spread on the side of processor to block air going in and act as a barrier of the liquid metal since thermal paste is sticky and if you spread it flat it won't move to much and will stick with the coldplate heatsink and air will be also eliminated since paste fill up the gap better than liquid metal, well just a theory my choices are paste and tiny cut Thermal pad attached on the SIDE not top of ryzen iHS enough to make contact with coldplate heatsink to act as a seal protector at the same time block some air but I think there is easier)
Finally attaching coldplate to cpu and let it cure for a week horizontally before tilting it back to normal
@@JoseSanchez-xj3xn ps5 is using this liquid metal and they cheap out the heatsink.
And planned a sort of barrier and sealing the die preventing air or most air going inside.
And only claimed they made their own Sony liquid metal that is different in others a long lasting liquid metal lol.
As far as we know as long as it is touching Nickle and not exposed in the air it will stay in its liquid form for yearsssss example of this are delided Intel cpu specially when they glue the IHS again.
Ps5 used cpu IHS design as its main heatsink design (basically a direct cooling but unlike GPU it is sealed like cpu) and inside there's a foam around the die not allowing the liquid to get out and if you remove the foam there is a insulator tape covering the IC. That's why when you open a PS5 the liquid metal is in its liquid state not hardened.
I'm looking forward to see the results.
Please just use liquid metal on nickle coated products only or purely for delid & thermal paste for contact to copper will work well enough, that is why the Ihs doesn't tarnish. plus AIO'S aren't far off the price off a custom loop with a nickle coated cpu block with the option to add a gpu block
it wont hurt copper
Nickel plated surfaces are your best bet for Liquid Metal, although I wouldn’t recommend it for much, direct die, but not on a GPU in most cases
to be honest , im not sure why mercury and gallium isn't more widely used as thermal compound.
its very good at conducting heat.
they'll just have to be sure to "pre-alloy" it so its bound up with some metal already and thus shouldn't attack anything.
mercury? you serious? ever heard of mercury poisoning?
@@NeganLucilleForever I would hope your not eating your computer components but regardless.
your not going to get mercury poisoning from metallic mercury especially if it's alloyed with something else.
take tooth fillings for example, they have mercury in them.
@@logangraham2956
Actually eating mercury is (relatively) safe, but inhaling mercury fumes is death.
You can make a case that almost everything contains mercury, but concentration is key. It stands to reason that a mercury-based thermo-paste will have a lot of mercury in it, which will evaporate as your processor heats up. Inhaling it will slowly poison you.
@@logangraham2956 Just read about mercury-containing amalgam used in dental fillings - it's deemed not safe and being phased out. So there you have it.
I can't find the update video to this. Did you ever check this after 5 years to see if it still worked ok? I'm stuck between conductornaut vs nt-h2 for a 13700k build.
I still have LM under the IHS, 4770K (+2 years now) Still going strong.
Even light pitting is a bad sign. Light pitting is a start to full corrosion. Obviously over time , it's been months with no problem but I would invest in a new heat spreader for the cpu and just chalk the cooler up to a loss and keep at it! It's gonna take a long while to fully destroy it. But also keep in mind as it pits it will absolutely lose thermal conductivity
And also I don't care what gamers Nexus says. As it pits OVER THE REAL LONG TERM it will lose thermal conductivity, it just physically has to there is no way around it. But the time frames in play are so long that I bet you could use the components until your ready to upgrade before an issue arises
Long term those two metals reacting to each other will ruin that copper. Because of the heat it wont take to long. I use a laptop not sure it's worth it. I would love my cpu to run 10 degrees cooler though. For now I just use a much better paste than factory and it helped along with some undervolting on my msi gs65 stealth thin with gtx 1070 max-q. My gpu stays fairly cool unlike my heat prone cpu.
You dont need LM On Direct Die Cooling! Just Use a good thermal paste!
schwartz that's what I'm doing for now. I used artic paste this time, It's ok. I like grizzle better though.
my copper block doesn't look like that after using 6+ months of liquid metal?
Different alloy, more in the inert side id say.
Copper =//= copper
You're supposed to use something like a ear bud to spread the liquid metal all around it
Your supposed to cover the entire die with this stuff. FAIL.
You can clean the surface of that aio with 0000 steel wool. It is very fine and cleans well.
Is the liquid metal thermal compound largely based on Galiium? if so, that is why it eats the metal plate from the heat sink, I think that would not be a good idea to apply such kind of compund to COU cooling in the long run, where the gallium eats the metal plate would creat tiny air space/gap and negatively affect the themal conductivity.
yes gallium is the main thing in LM that likes to destroy things, Aluminum is a definite no go with LM but copper is said to be safe. But as you see its really stains the crap out of it. since that video my PC is still running that LM and i havent had any issues yet. temps are still good and system is still stable. However come september we will crack it open and see whats really going on in there.
@@MajorHardware I hope it will be still fine in coming sepetember, just worried if the Gallium eats the copper and other metal would form an alloy look like solid structure and forms tiny air gaps (the minor stains you have witnessed) within the structure, that may reduce the thermal conductivity since air is not well thermal coductive, but i dont think there will be serious damage to the overall temp performance given the copper thermal sink is still closely attach to the CPU cap.
I sure hope since the publishing of this video you've learned how to properly apply a liquid metal thermal interface.
i did not
Sand it off would be my solution
Nothin like a good ol boy delivering me tech tips with the familiar ring of down home.
Did you use conductonaut? This looks really clean for copper on nickel.
The gallium in the liquid metal will react to copper but it's extremely insignificant and has no measurable effects in terms of cooling. For what it is worth it's not ideal to use between the IHS and the cooler can't really hurt I guess but it's generally frowned upon for what ever reason.
Oh yeah the gallium will absorb into copper tho but it does not degrade the copper like it does with aluminum what you see left is caused but oxidation but once the copper has absorbed all the gallium it can (not much really) it will no longer cause drying like this. If for some reason you decide you want to remove the oxide metal polish works pretty well but I would strongly suggest not using sand paper (some folks think it's a good idea) even car polish (compound meant for scratch removal) can also remove it but the process isn't all that quick (but you will not be at it for a hour) about 5 mins depending on the brand.
Just avoid cleaner waxes which is suppose to be a hybrid of the two, they either have micro abrasives (which is what you want will not effect the look of the copper block) and chemical which should be avoid as it will have no effect on the liquid metal.
good info thank you when i take this thing back apart i will be trying to get the copper back to its pre LM days so all ur recommendations are noted for sure
I have a feeling that using liquid metal is like using a Toaster to hold your bath water warm. It works and sometimes it Kills.
Eeeeh, the current will simply cut off if you do that, it won't work in any way, thats a terrible comparison. It will never work. Liquid metal act as a link between the cpu and the heatsink micropocketed surface.
Gallium foprm CuGa2 alloy with copper with good conductivity it does not harm at all the performance also by seeing that bloob looks like you dont cover whole IHS, also u need apply thin layer on the cooler plate as well other wise gallium wont stick.
i have been using LM for about a year now under and on top of my IHS.... i am using a rocket cool IHS. but the only thing that i got on my system is the staining and yes a bit of pitting. but everything looks good. i will say that when i went to add new LM it was way easier to add it to the copper. also for ppl that are just getting into LM from what i have found if you are adding it onto bare copper let it sit or about a week go back in and add just the tinniest amount more. this will be smaller then you think. but i say to do that as i have found that i get better temps and coverage after i let it sit as it seams that letting the LM stain the copper made it way easier to get it to cover right. but also good vid
Why did t you spread the metal on the surfaces, you are no supposed to put a drop down and squish it with the aio.
I’d say it’s not worth the risk of having my entire laptop (which is aluminum btw) fall to pieces if any gallium gets on it. And besides, I hear it destroys copper too- only slower.
Liquid Metal doesn’t »dry out«, it’s literally just a gallium alloy that’s liquid at room temperature.
I'm just about to upgrade my PC, but I've had LM as my TIM for 3 years (without changing for fear of what lies beneath. However, I have been running it like that with a Swiftech H240x on a 2500k @4.5GHz the entire time and performance has not degraded. Max temperatures I see are 63 degrees on the package (I use HW Monitor to watch for any issues). I'll be interested to see what it looks like when I do separate, as I plan to re-use the H240x with a 9900k...
from what i have heard/seen/read your copper cold plate will look like poop but it should work just a good as it did when it was clean, i would be interested to see what yours looks like as well.
@@MajorHardware Yeah, hopefully no damage to the plate, but I'll reply back to this once it's in bits *nervous laughter ensues*...
Hey @@MajorHardware I replaced my chip a couple of months back, and grabbed an image of my cold plate (shows how badly I applied it too). If you're interested in seeing any, let me know. Nothing out of the ordinary, some slight pitting in one area, but nothing too bad. Currently keeping my 3900X cool and calm, this time with MX-4. Though I will be switching it out for an EK block at some ponit.
Why would you risk using liquid metal like that where it's not contained (like under the IHS in a delid)?
Like everything else, it's subject to gravity and any liquid metal that's not clamped between two perfectly flat surfaces will move unless your PC is horizontally mounted (which it isn't).
If even a tiny droplet of the liquid metal gets onto your motherboard you have 2 issues:
1. It's conductive so you will likely have an electrical short - RIP motherboard
2. It can react badly with the solder used since every board uses different solder compounds and some might react a lot worse than others.
The reason most use liquid metal is to remove the crappy TIM under the IHS on intel CPUs, NOT as a replacement for non-conductive thermal paste (Thermal Grizzly. Arctic Silver etc) between the IHS and cooler heatsink.
You would get the same results and be far safer using a regular non-conductive paste for your application.
In this case the risks far outweigh 1-2 degrees difference (if that)
thank you for the comment my man, however i have had none of the problems you have stated and although gravity is not just a good idea its the law i have had no LM leak out onto anything and everything is working great as of now. however if things change down the line i will make sure to let you know because using LM as a thermal compound is only for SCIENCE as you can get comparable performance from a good thermal compound. Also LM was not designed to be used for delidding as no company would design a product that required you to void and potentially destroy your CPU, however give Intel's hate of solder (until recent) that was what it was mostly used for, LM was designed to be a thermal compound and does a good job, however with this product there are risks and everyone knows that myself included and i have choose to see what happens so hit that subscribe button and come along for the ride.
@@MajorHardware haha! Of course! Did you know that Coollaboratory make a liquid metal that's a copper/silicon mix that's not electrically conductive and designed for high temps?
It's heat dissipating properties are better than most thermal paste and not the risk of regular liquid metal.
I hope nothing goes wrong, that would be sad as it's a nice build.
Happy holidays
I did not know this, i will look into that stuff and maybe pick some up if my computer doesn't go blue smoke on me 😁. HAPPY HOLIDAYS my good sir.
4 years later I'm here. Can you tell me the brand and style number of the shirt you're wearing? I would love to buy one for myself :D
Is it necessary to apply liquid metal to the copper and the processor? Or should it only be applied to the processor
You did not apply that compound correctly at all. RTFM dude.
what does RTFM mean if you dont mind?
@@MajorHardware read the fucking manual, meant in the middle nicest possible way.
lol no worries im not easily offended I honestly never heard of that acronym.
@@MajorHardware
Definition of RTFM in English:
RTFM
ABBREVIATION
Computing
vulgar slang
Read the fucking manual (used especially in email in reply to a question whose answer is patently obvious).
-oxford English dictionary
en.oxforddictionaries.com/definition/rtfm
Let me guess you’re here after the PS5 Teardown
duuude howd you know! I do possibly think there might be problems with this down the road on the ps5
Liquid metal is such a meme.
My advice as a SysAdmin: Don't use it at all. Just use thermal paste. You're not going to miss that 3 degree difference.
What about the 3 drops the where in the corner on the cpu? You left it there , if they run down your motherboard will bet it bad man
you can use nail polish remover to remove the liquid metal, alcohol wont quite do the trick, also as others have said you applied it improperly but thank you for learning for all of us :)
What brand of liquid metal thermal paste did you use?
I had liquid metal on my CPU for only two days and it corroded the copper water block pretty significantly
Gallium doesn't corrode copper. It does bond with and destroys aluminium though. I'll let you figure out what went wrong.
Coollaboratory liquid extreme is the latest product claims not to dry out but has lower performance. hope to see long term test.
Random question, what colors are you using in your case? Looks fantastic.
as for colors such as MOB and GPU i tried to use only neutral colors like black and steel silver. because im a big RGB fan (yes im that guy) i wanted to stay away from any permanent colors.
Im a Noob i just now figured out you are probably talking about the color of my RGB, at that time my GPU and background RGB color was neon green, the fans, ram, case edges were like a dark purple, i cant remember the color code but i was trying to go for a color close to what you see given off by a black light and thank you i did like that color scheme.
im sticking to thermal paste liquid metal looks to crazy for me
How is the current status on this? Couldnt find the follow-up video 😅
Just got the exact same thing.. can i use normal thermal pastd afterwards?
This is why I don't use liquid metal. I use Gelid Solutions GC-Extreme Thermal Compound TC-GC-03-A. My I9 7900X (Skylake) is running at 4.4MHz on all cores and runs about 30 C without load and about 60 C full load.
Yeah I'm not going to use liquid metal on my components until they figure out how to use something other than Gallium on it.
i can understand that, i picked up some Kryonaut to test repasting a GPU rather than liquid metal so stay tune!
Nice video. Curious though it looked like the liquid metal wasn't cleaned off completely from the AIO pump head. If it was cleaned - then was that grey area where the liquid metal had been?
its copper gallium alloy, the only way to get it off is to lap the contact plate.
i put LM on my asus strixx 2080 ti for a few hours today and it stained the cooler already, only brought 5 degree better temps as well, so not worth it tbh
Has anyone tried using a metal polish for copper alloys instead of alcohol ? this might help with the staining and pitting as a polish will remove the surface layer of the copper and might smooth it out again
Cool rig dude love the attitude and vibe i'll sub.
Love the profile pick!!
pic
How do you remove the liquid metal like you just did ? Using what chemical or liquid ?
isopropyl alcohol is what i used
Put it on the q-tip outside the case over a towel. Everything else is just asking for trouble.
Y i would only use it with nickel coated parts.
it also changes the appearance of nickel coated parts with more time, used it on my NH-D15 and is made the base of that and the IHS darker after about 5 months
@@pikerdm7466 any pictures of it
Do you think that nickel plated on all coolers and water block that would protect the liquid metal from eating into copper
yes, it doesn't eat the copper away tho it just forms an alloy with it which looks crappy but doesn't hurt performance at least it hasn't for me yet.
@@MajorHardware it mental how liquid metal comes alive making alumilum metel becomes brittle
When we will find best liquid component for processor. Until now we still using normal liquid
just get a vinegar salt rub and gently brush it with a toothbrush. should fix that spot fast. or get a copper corrosion cleaner.
put gallium in a titanium padlock and leave for about 45 minutes and after a few knocks on the padlock
Your water block is not 100% copper ?
3 year with liquid metal and no probleme.
We have a 0.01 mm copper oxidation.
Its nothink crazy
Look at that under a microscope, I guarantee you wont use it again. Nickel plated copper is the only safe material for LM that I'm aware of.
Great content looking through the backlog, subbed and keen to see what comes. Hope to see u get more subs in the future
Great video and I've just subscribed.
I am however, sticking to Arctic Silver, as I've never seen any kind of pitting whatsoever with the hundreds of PCs I've built.
yes i would do the same no big benefit to using LM in place of high quality thermal paste.
Rufinoman
Isn't that the beauty of watching other people's results? The only reason we know the negative results of LM is because others risk their equipment and show the aftermath. I agree with you as well on sticking with more standard thermal pastes. I've never had a reason or need to change from Arctic MX-4.
Liquid metal more useful under the heat spreader rather than above it
AIO Liquid cooled a I5 3570k (H60 model) using traditional Thermal paste in 2012 when I built a customer's PC. Then in 2019 I recently acquired the customer's PC, and I just sold that CPU off. When I removed the AIO the CPU looks good as new no pitting no marks you could not even tell if it was ever used. Then I look at that AIO and CPU you got...... I think I will stick with traditional Thermal Paste.
Can you apply thermal paste after using LM for the first time. Today I used thermal grizzly conductinaut first time on my z73. Immediately realized mistake as it started spilling out. I have already ordered the grizzly thermal paste. Can I apply paste within a day or two application of liquid metal
My OCD was super triggered on that application.
All of these products clearly state NOT to use it on COPPER
Peace of advice if you have expensive equipment.. Just use thermal paste.. Liquid metal performs okay, but after a 1 year on my machine. I had to replace my CPU cooler due it. Not worth it.
do you think it is better for me to apply liquid metal in my 5800x? I have a noctua d15 heatsink with obviously noctua thermal paste and while I play I always remain around 70/75 degrees c.
In NO WAY did you apply that correctly. Its to be evenly distributed across the entire IHS. It is not paste. It needs to be thin. You will see a difference if applied correctly. And please tell people what product you use. Liquid metal is vague AF. All this pulling shit open to look is all for views. This stuff is Gallium. You can research all you need to know from scientists and real life testing without all the drama. The pitting you are getting is not a chemical reaction between the metals but a reaction to contaminants left during the application. Cotton swabs leave resume and cotton behind.
LOL this dude is pissed ^
@@MajorHardware Was he wrong though?
Interesting, so it's eating its way through both devices. Wonder how long before it ruins them?
so far we are still good but i guess we will wait and see.
@@MajorHardware Right on. I wonder do they all eat at or cooler's and CPU's?
do you mean all brands of LM
@@MajorHardware Jess, jess I do.
Bro i subbed you def deserve it. i hope you make it one day!! some tips : try to relax more on your body language, get bettet thumbnails and a better profilname / and picture 😬, the name and the pictures are just not flashing my eyes if u know what im saying. They dont really wanna make me click on your videos or channel, even tho your content deserves to be around 15k-50k or even more views.
thank you, i will do my best to improve, i appreciate any and all feedback, you cant improve if you dont know what you are doing wrong, THANKS for subbing hope to see you around.
Really waiting for that additional review and check after 6-12 months!
Oh it will come, as of now my system is still running strong no sings of anything going wrong as of now
@@MajorHardware Plan to liquid metal my 2k rtx2070 laptop, but feeling kinda unsure. Laptop runs hot and high temperatures accelerate this process. Bud good thing is that after reapplying, the process slows down due to the fact that copper becomes plated with gallium and further adding more gallium with each consecutive application reacts less and less. One of the concerns is that when eventually dried, the dry liquid metal powder may scatter when the laptop is carried around and short some circuits.
for something that new i dont know if your going to see any big benefits to using LM as a thermal paste, personally if it was me i wouldnt risk killing that laptop. I did make a video where i LM my old alienware laptop but that laptop was unusable so i had nothing to loose also it was really old anyhow. being that old the thermal paste was rock hard as i show in the video. that laptop is still running strong no signs of anything going wrong.
in my opinion you wont notice any cooler temps liquid meteling or re-pasting that new laptop as the thermal compound in it from the factory is probably in great shape. to me the risk out-ways the benefits on a new machine.
yeah, the only problem is that stock paste that asus uses is a complete junk, paired with mediocre cooling renders lots of thermal throttling and fan noise