Thanks very much. Just applied Liquid Metal and thermal pasted the GPU. Made the world of difference to my X17 R2. Much quieter (fan), cooler to the touch and the frame rate is no longer stuttering. Boosted my cinebench multi core by 1000. Cheers 👍
На счёт предположения об окислении: 1. Разница состояния ЖМ между процессором и между медными пластинами не объясняется окислением. Т.к. хорошо видно что ЖМ на меди кристаллизуется! ЖМ - многокомпонентный сплав из Галлия, Индия и Олова (галлий 62 %, индий 25 %, олово 13 %). Оксиды компонентов ЖМ выглядят по-другому: Оксид индия имеет три модификации: Оксид индия(I) - черные кристаллы Оксид индия(II) - серый порошок Оксид индия(III) - светло-желтые кристаллы Оксид галлия(III) - белые кристаллы Оксид олова(II) - тёмно-синие, почти чёрные кристаллы Оксид олова (IV) - белые кристаллы 2. Кристаллизация ЖМ на меди происходит в результате взаимной диффузии меди и ЖМ. В меди образуется плёнка бронзы, а ЖМ приобретает состав, сходный с бессвинцовым припоем (Олово, Медь, Индий, Галлий) Так что, "застывание" ЖМ не связано с окислением кислородом воздуха. У меня аналогичный опыт застывания ЖМ между медным основанием кулера и крышкой процессора, в то время как при контакте никель-никель образуется такой же серый осадок, что и при контакте кремний-никель. Возможно, что в последнем случае действительно имеет место окисление Индия до Оксида индия(II), но тут встаёт вопрос - почему не ухудшается теплопроводность контакта кристалл-крышка процессора? Вероятно, всё-таки это не окисление, а осаждение того же самого индия или олова из ЖМ.
After 1, too much LM on the processor, and after 2 LM you need to apply it to the radiator and the processor because you can see how hard it sticks, so it will have air "holes" on the radiator
hey bro thanks a bunch and much love from Canada, Alberta i havent been inside a laptop for years now and all these latches and sequential screwing / unscrewing woulda broken me if not for a real G wingman! so I removed the old paste, element 31, which was plastered all over the motherboard like some child was learning to drywall in their sleep but then an earthquake ruined it all I then tossed out the CPU Barrier because the whole thing would need a holy baptism by God himself to clean it up but he doesnt take liking on alien stuff anyways :D so I applied Noctua NT-H1 (normal thermal paste) and put everything back to check it out, needless to say, it was bad, so it was time to disassemble it again and try liquid metal. I left the newly added thermal paste there, made a clearing in the center so the thermal paste would act like a frame / barrier for the liquid metal when it would compress down (this was important cuz we tossed out the old diaper of the CPU Barrier) So putting everything back together, ran some stress testing and its a huge difference but I think a new patch of Element 31 beats all cuz the GPU temps weren't as good as original (80c max) overall, might be better to just get Dell to put in their element 31 stuff but it will cost :/
Hey, great tutorial. I’m about to attempt this on my (same) laptop, but instead I’m going to use the Thermal Grizzly KryoSheet graphene thermal pad. Do you know the measurements of the CPU & GPU? Thanks!
Hi thank you for this great video. How often do you have to reapply it? I read that for lq it's necessary to change it 2-3 times for the first 6 months, it that right? Another question: when do you reapply it do you have to remove the old lq completely?
That is an insane amount of compound on the GPU. It's going to spread all over your motherboard. BTW some compounds may have conductivity of 3 to 8 w like arctic silver 5. That much could potentially damage your components.
Is this the only thing I need to do to get the thermals down? Or is there any undervolting or ocerclocking that I need to do? Not a computer expert so definitely in need of some help here.
I feel like that is too much. I don't say to criticize. I say to teach. If too much is applied it will leak or shoot out when you apply pressure and/or from gravity under certain circumstances. Liquid metal on the motherboard and other components will cause damage as it's conductive. The happy medium between too little and too much, can be tricky to find. Too much is way worse than too little in regards to lasting damage.
Bacause gpu‘s seldomly overheat. And also because of the transistors which are in the vicinity of the gpu. If LM would make contact with them, the mb could be damaged.
Not really. I'm just not happy that my idle temp is 60 deg. Celcius. And that's in air-cond' room with cooling pad on. The laptop is almost 2 years old so I think it's about time.
Quick google search states that the drivers are corrupt. Reinstall them and hopefully you should be good. If the LM did anything it wouldnt boot at all
Hello, could you tell me why this laptop makes a strange electronic noise ? It s the sound like an old big HDD working, or transistors noise, or maybe the fan.
It's called coil whine and it's because this laptop is only one out right now that pushes the 3080ti to 175 watts. Meaning you are getting the closest possible to a desktop feeling of the 3080
Hi, I'm about to LM my R2 too. Do LM ing your cpu will cause any after affect after a long use? Maybe after affect in 1-2 years? Because some people told me LM ing cpu is good but will slowly destroy the cpu in a long term.
I’m a trained it-technician, and i know that gallium (LM) is not the best in terms of being user friendly. But unfortunately it’s the only way to get the promised amount of performance from this laptop. Please be very careful and take your time when applying LM . Feel freee to ask questions if you’re struggling. !!
Liquid metal used for semiconductor TIM contains gallium, indium & tin. The issue is that when using bare copper & LM, the gallium will react and diffuse into the copper via the Rxn 2Ga + Cu -> Ga2Cu and this reaction speeds up with exposure to the high temperatures from the processor/SoC using the formula Rxn rate = Ae^(-Ea/RT) Pi([Ri]^ei). As more gallium diffuses into the copper, the larger tin & indium metal gets left behind and overtime builds up a solid uneven metal layer over the copper coldplate. This leads to air gaps between processor and cold plate which impedes proper transfer of thermal energy. The result, increase of core temps as time progresses. Hence, LM Should only be used on nickel plated copper. If you really want good temps without sacrificing longevity from LM, use Honeywell PTM-7950 pads. These pads are made of a phase changing material and has excellent thermal impedance compared to most non LM TIMs in the market. In fact, Lenovo uses a specially modified version of said pads on their legion series laptop! Hope this helps :)
@asf asfd I agree with you that a deep enough layer of CuGa will eventually reduce the diffusion Rxn to a negligible rate. However, during the initial stage of the Rxn the larger tin & indium metal will still be left behind and overtime builds up a solid uneven metal layer over the copper coldplate. This leads to air gaps between processor and cold plate which impedes proper transfer of thermal energy. I know that some people did not face this issue but in my experience , the tin & indium build up was so bad that core temps on my laptop were hitting 97-100dC.
@@VSN1001ell shit, Liquid metal or not to liquid metal lol. If this was that big a deal wouldn't Alienware not be using it ? I understand the chemistry 100% but I think you forgot to calculate magnetic induction which can change the thermal properties of metal including retention. Agreed yes gallium can "cold forge" essentiallly to untreated copper 100% I just don't think Alienware would be that stupid given that element 33 is also liquid metal hybrid , But then again a short call to anyone in customer service would change that opinion real quick Not the sharpest over there at all😂
@@jewishhella2481 Alienware element31 is a different case. They are aware of the issues of using pure LM. So they instead encapsulate the LM into a silicon based thermal paste to prevent direct LM to cold plate contact, which “solves” the issue. I say “solve” cause hotspots on die has been reported by multiple people due to uneven distribution of LM within the silicon paste. In the context of commercial production, the bigger issue is not the chemistry of LM with the cold plate but the application of it. For example, Asus LM applicator machine is so bad at its job that there’s countless reports of large potions of CPU/GPU dies being void of LM due to the coldplate and dies not being wetted enough with LM Also can you elaborate on how magnetic induction can changes the retain ability of LM? I’m actually quite curious of this phenomenon
@@humanoit7533 I also wondered this. CPU-barrier is present as you lower the heatsink assembly (22:39), but is no longer there when you start tightening the screws (22:40).
Does all X17-R2 came with that CPU mylar barrier? I have the x17-R2 and I plan to do LM soon. Core 3 and 5 are thermal throttlling with UV and TCC on at 17,400 Cinebench R23.
Long term it will oxidize. Considering the disassembly of this laptop, not a good idea lol. It will also oxidize faster the more you use it and longer you have it. I got to a point with mine where daily use, I needed to replace every 2 months to maintain temps. Afterwards it's impossible to completely clean without sanding.
Hey! I am new to these stuffs. Where do I get CPU barrier? I have an X15 R2 reaching 100°C quite easily. I am worried I might kill it if I apply liquid metal incorrectly. Lemme know.
Hey sir! 👋🏼 Most important rule is: take your time. If you follow my guide diligently you won’t have any problems. You can use thin electricians tape or high temp tape (Capton tape) With what exactly have you problems with?
@@humanoit7533 hey thanks for the quick reply. The problem with my laptop is that my CPU reaches a century quite easily even when casually gaming. And the problem with liquid metal is it's quite risky especially since I have never tried it before and it's quite an expensive laptop. Any bonus tip? Will Kryonaut do instead of Conductonaut as it's not electrically conductive?
And yeah, Dell says not to use alcohol wipes to clean off Element 31 as it can short circuit the CPU and damage it. Can you clarify about it too? Also something like an estimate of this would be amazing. Kryonaut Extreme vs Conductonaut vs Element 31
Hello my friend ! I strongly believe that the problem might not be the element 31 per se, but its application. I got about 5-10 C cooler temps with normal traditional tp
@@humanoit7533 yes you are 99% correct (I think it was Tom's HW) did some research on this and concluded that most of the time is poor application rather than the paste itself that's why is good practice to repaste, even Nvidia does this from time to time
hello I heard that Alienware would add element 31 to the CPU of X17R2 as a liquid metal to conduct heat, but I don't think I saw any sign of element 31 before you wiped it
I already do it, and I could say that it also work, you reduce de the temperature on idle around 10-15 °C , but on performance you could gain about 40% before thermal throttling, that's amazing, If you are experimented, be carefully and do it for sure...
@@John-lt8wvyes your computer will completely explode, strangle your mother murder your son bang your girlfriend and kill your neighbors cat! Use at your own risk!
if anyone see this video and planning to do as this video. you gonna have bad time with LM in the future. find someone know what they are doing. this guy didnt know how to use LM. this guy know nothing about surface tension of LM, and that amount of LM can fry your laptop if you bring it around. search for Gamer Nexus or Linus Tech Tips video for the right way to use LM. dont just follow someone that has 0 knowledge but making guide video. these people just spreading missinformation. for the sake of your own laptop, dont follow these kind of video
![[TUTORIAL] How To Apply Liquid Metal | Alienware X17 R2 i9-12900HK](/img/logo.svg){kind=logo}
