Solving Intel's Heat Problems..
Vložit
- čas přidán 14. 05. 2024
- Thanks to Seasonic for sponsoring this video! Buy Seasonic PRIME 1300W 80+ Platinum Power Supply
On Amazon: geni.us/ZjbHhax
On Newegg: geni.us/lUHu6Hc
Buy Intel Core i9-12900KF Desktop Processor
On Amazon: geni.us/xyfqgXm
On Best Buy: geni.us/Ca9UD
On Newegg: geni.us/FzO8n
Buy ASUS ROG Maximus Z690 Extreme (WiFi 6E) LGA 1700 (Intel 12th Gen) EATX Gaming Motherboard
On Amazon: geni.us/ssxF9Th
On Newegg: geni.us/e2mflf5
On BHPhoto: geni.us/dSocT9t
Purchases made through some store links may provide some compensation to Linus Media Group.
Discuss on the forum: linustechtips.com/topic/13935...
► GET MERCH: lttstore.com
► AFFILIATES, SPONSORS & REFERRALS: lmg.gg/sponsors
► PODCAST GEAR: lmg.gg/podcastgear
► SUPPORT US ON FLOATPLANE: www.floatplane.com/
FOLLOW US ELSEWHERE
---------------------------------------------------
Twitter: / linustech
Facebook: / linustech
Instagram: / linustech
TikTok: / linustech
Twitch: / linustech
MUSIC CREDIT
---------------------------------------------------
Intro: Laszlo - Supernova
Video Link: • [Electro] - Laszlo - S...
iTunes Download Link: itunes.apple.com/us/album/sup...
Artist Link: / laszlomusic
Outro: Approaching Nirvana - Sugar High
Video Link: • Sugar High - Approachi...
Listen on Spotify: spoti.fi/UxWkUw
Artist Link: / approachingnirvana
Intro animation by MBarek Abdelwassaa / mbarek_abdel
Monitor And Keyboard by vadimmihalkevich / CC BY 4.0 geni.us/PgGWp
Mechanical RGB Keyboard by BigBrotherECE / CC BY 4.0 geni.us/mj6pHk4
Mouse Gamer free Model By Oscar Creativo / CC BY 4.0 geni.us/Ps3XfE
CHAPTERS
---------------------------------------------------
0:00 Intro
1:10 Installation
4:20 Cooling Tech Tips
7:44 Fixing Block
13:00 Testing
15:50 Tuning
19:30 VT Curve
23:30 Disassembly - Věda a technologie
Next step: sub zero in a vacuum to eliminate moisture.
And you could call it "Space cooling" or something.
was just about to ask if they solved the condensation problem. reminds me of iron man 🤔
I mean just putting it in a Styrofoam box and drying the air inside the box first would work. No vacuum required. Hell, even a trash bag and some tape
would completely eliminate condensation. I'm sure they were well aware of this basic logic solution (it's probably the first thing anyone would think of) but of course then they'd have a 10 minute video and it wouldn't be nearly as profitable for them through youtube's system.
just seal it in a plastic storage container (provided sealed holes for the hoses, power, peripherals, etc) and also stick a little RV dehumidifier in it running as well
well, there's electrolytic capacitors, they wouldn't like to be in vacuum but you can put the board in transformer oil to stop the moisture build-up.
@@xtrariceplease excellent point sir. I guess a nitrogen chamber would be better then.
At least Sub-Zero-Nitro still sounds cool
At these temperatures, the condensation on long lines means a massive amount of power is wasted on condensation the water. You should really insulate the whole lines, don't underestimate the obsene amount of energy water takes for a phase change.
It is "only" 5 times as much energy as getting water from 0°C to boiling... who neds insulation when you can use bruteforce without thinking? :D
The issue is, is their is an extremely low load on the “Evap”. If they ran 2 more pcs in series. Than it wouldn’t feeze up like that. Basically if that condenser didn’t have an accumulator on the suction line that compressor would be dead.
The ice on the lines will insulate them and the more ice, the more insulation 😁 It's wasteful as it has to be kept cold and eventually a messy (drippy) equilibrium will be found therefore it is still better to insulate things where possible.
Yeahhh. I remember we learned this in physics, and then we saw the difference when we needed to calculate the energy for rubbing alchohol.
We all thought "wtf? It's just water right?"
Why does that waste energy? The tubes get cold, the water from the air frosts on it, and then what? That's pretty much it, right? You can only get so much frost on there.
linus has literally taught me - that so long as you get a midtier or higher computer part, you can almost certainly toss it around as if it were a childrens toy and the thing will still work to its full potential... Thanks linus for helping me overcome my fear of messing with expensive technology in fear of breaking them from the littlest of things when clearly, they can take a BEATING!
Another thing to possibly keep in mind, when you order pc parts shipping like ups and fedex have no idea what you’ve got, they are never as gentle with your hardware as you. Sure it’s still in the package, but…
True, and that's thanks to said fear. The things are already expensive, so they add in extra stuff to make sure it's not fragile since that'd mean both their and your money would be wasted. Heck, for some tech, the reason it's expensive is the protection not the specs.
and yet my brother yells at me for taking out a cpu cooler and cpu by myself..
I was checking for nice hi-fi headphones recently and was allowed to try out / play around with some 6000 euro headphones after asking nicely...
Was a tad scared at first but then I figured if I can break 6k headphones by using them normally and doing what could happen to anyone ... It's not my problem it's the headphones problem.
I'd say that should go for all consumer product in the high price range.
@@TH3C001 they fuckin do a 7 page muda on any expensive package
I remember being back in college in like 2005, seeing people on XtremeSystems running those phase change setups and really wanting one. I can't imagine how loud that would be running it 24/7.
I think you put it in a different room or something
@@kellywu4061 Put it outside like a real air conditioner
Hey! I was one of those guys :)
I ran a modified Prometeia (on Athlon XP's), which wasn"t actually that loud from what I can remember, but that is of course a retail product and not a huge compressor like an actual AC unit.
Mine had different gas put in to have it run even colder (Could be R404 but im not 100% sure)
idle temps were -58°C and with full load it still was somewhere around -40°C iirc
So much fun but also so many frustrating motherboard-replacement-moments :D
I remember being a child in 2005 and seeing that, lol. That was pretty cool stuff (pun unintended!).
@@serialced As a noob getting into more & more pc related info to learn more & more... -40c under load is breaking my brain hahaha. Extremely wild & impressive!!
An LTT classic. We've been missing this kind of stuff lately.
You know it's a classic when he drops one of the most expensive components before the 5 minute mark.
czcams.com/video/up2TjMuan6o/video.html
Yeah, this is the kind of stuff I subscribed for
Not just tons and tons of laptop, phone and TV reviews...
@@D3Vlicious ehh, I'm pretty sure he does it on purpose at this point.
@@whiteblack6865 Obviously not? 😂
"Seasonic, they make great power supplies and are willing to sponsor our nonsense" is what I expected him to say :)
suh-sonic
0:19 JUST LUBE IT UP
I'm glad Seasonic sponsored this madness. I've been missing this kind of content!
I see suggestions to immerse in mineral oil here; what about a less messy alternative:
A gas like Sulphur Hexaflouride is heavier than air and would displace any air around the mobo, mitigating moisture concerns without the mess of mineral oil. Basically just need to place the mobo in a bin and "pour" the Sulphur Hexaflouride onto it. Sulphur Hexaflouride is also an excellent electrical insulator, and is used to prevent arcing in high voltage systems.
That was very interesting
also it makes your voice funny (deeper)
Now this I kind of want to try...
Surprised its been a year and they haven't done this yet!
Argon would be cheaper, easier, and a lot less dangerous
Mechanical engineer here. You need to have your expansion valve on the cpu block if possible, more work will be done for pulling the heat out of the system in on go. And you won’t have your lines freezing up as bad.
It is, basically. Watch hhow they made the block, is a bit of a janky solution, but there should be expansion happening in the block.
came to post that...posted that, then read you posted that. That.
The expansion seems to be happening immediately after the compressor. The high side should be ...higher it seems.
@@MaxSeidel1 yeah I watched the first one and it about gave me a headache. Alex is pretty good at cad, but lacks the knowledge of thermodynamic systems. He made a pretty model that isn’t efficient enough and didn’t really consult the system, since the unit has a external Evaporator coil and it’s going to be closely mounted to the condenser/compressor unit there is probably already a expansion valve close to right off the compressor side and doing a partial sub cooling then he’s trying to do a final subcooling on the block he designed but it’s more than likely not efficient enough and causing issues with the work being done in the system
@@jacke8487 I agree. It looks like a more efficient system would be to have a chamber in which thermal expansion is taking place right above the cpu, where it would then remain in a neutral environment, but pipe the hot vapor up to a condensation chamber. The bottom pool of that collection chamber could have a contact plate to rapid cool the liquid before it can re-enter the expansion chamber.
That way, there's absolutely no condensation anywhere near the PCBs, and you can still introduce water at a temp below that of the ambient temp, so it may not be as potent of a cooling effect, but it might actually be able to be made into a smaller form factor component.
@@VincentAdultmanbearpig Thats how I made mine with a large shop fridge compressor running R22. Huge cpu eveporation chamber and the nozze where inside the cpu chamber. I made the same mistake and underestimated the need for surface area but I had the volume. So my pressure line never got cold like that. My friend who helped me build it built cooling systems for living, so it worked first try even though it was super janky. This was 20 year ago with all free junk parts with no fancy electonics. Just brute force with a lot of cooling fans and was on /off with the compressor. If I were to build another one I would just use 2 china copper block with multifins and solder thick walled pipe between. Doesn't get any cheaper than that.
Another solution to prevent condensation forming on the hardware is making a small sealed box for it and create a inert atmosphere in the chamber using nitrogen gas. A small flow of nitrogen gas works great for drying all kinds of stuff and is the go to method in chemistry labs to prevent water getting into your sample. Ask me for more info if you want to.
Nice.
_S C I E N C E!_
Ya I remember a company that sent me a device when i was into ln2 overclocking that was like a box i put the motherboard into that was similar to this idea, removed any air with the nitrogen gas evaporating from the ln2. I don't think there was enough of a market for them to ever succeed but was a cool idea. No condensation while using ln2 on the cpu
Or just pump it down with a vacuum pump and boil off all the moisture before they apply the cooling device. They could eliminate all the insulation.
@@ramosel yeah, also probably possible. But it brings a higher risk, since you are working with different pressures. The pipes need to be capable of dealing with that and you have quite a lot of seals in this setup which isn't optimal as well.
The fact that Seasonic is always sponsoring you guys for projects like this tells me that they appreciate how your craziness is a great demonstration of their quality.
The load isn't heavy enough for the compressor. I mean you'd most likely be better off calculating the actual btu's generated and buying a smaller 'reach-in' freezer compressor with an evap and cond coil to match. you'll get better performance out of the cooler and could potentially make this a more realistic venture. also you lose more of that risk of condensate as well. with a variable speed motor you could literally develop the first feasible refrigerated pc. Look for something low temp.
Theres also the possibility of submerging the pc in mineral oil and using a copper base plate, or refrigeration tubing coiled next to the board to super cool the mineral oil. which i believe would only crystalize at like -30 degrees Fahrenheit, so a low temp compressor would work wonders in a situation like that.
a mini-fridge compressor would be more then adequate to cool a pool of mineral oil that submerges an AIO radiator
Many many years ago back when I was making phase change system to overclock I submerged my motherboard in mineral oil so I didn't have to go through the hassle of insulating. That motherboard didn't last long, not sure if it was a fluke or if the mineral was the culprit. After that I just insulated it normally and it could run 24/7 with cpu core temps of -40c. Good times. After that I started making autocascades instead that do -100c and got more into that than the overclocking.
a "realistic venture" would be incredibly off-brand
I know that Jay once said that the EMI from his microphone was messing with the OC cards. I think the same thing happened when Brandon turned on the camera. I would set up an EMI shield room on the new building for this OC tests. Like engine dynos were we test engine runs from a separate room.
They should make the room dehumidified as well. Maybe double chambered.
Yeah, the cameras they use probably put out a lot of power, unlike consumer cameras, in heat alone. So I'd love to know what EMI stuff they spew out. (Plus it could just be wobbling as you walk along the floor/against the table :P ).
Just make a faraday cage for the test rig.
if you're going that far add an active UPS to the power chain. Shielded cables, the works.
Speaking of engine dynos: We had a problem with a big display showing a black screen for a few seconds at random intervals while running a car dyno. I guessed that the high currents of the recurporation created EMI in the HDMI-run or the TV electronics itself. Can you confirm this being plausible?
Linus drops motherboard - sees massive back-shield: "It's Linus-proof!"
Linus has screwdrivers. Board is NOT Linus-proof, lol.
It's Linus-resistant tho :)
@@sergiodiazesparza2584 this is a fair and valid argument. i give my respect.
HAHAH
It's clutzy Linus proof not handy tool Linus proof lol
When you're running computer hardware that far above spec, the electromagnetic emissions from a camera being on can legitimately cause instability.
If condensation is the primary concern with sub zero cooling, what if you did this kind of cooling setup in a mineral oil or other non conductive fluid bath?
The oil would probably solidify.
@@mehmeh1999 Does it really matter? I don't see that as a problem
@@mehmeh1999 Doesn't oil freeze at like -40 c? It has a huge thermal mass so it would take probably at least a full day running it full blast to get that cold
@@squidwardo7074 It's about localization of the issue. Yes it could take a day to raise the humidity of a room but it only takes seconds for it to form where it matters.
@@darekmistrz4364 Yes it would lower the efficiency of the oil if not completely alter its properties. If they were to try this they should probably use a 3M Novec derivative.
Has anyone tried putting the components in a chamber and flooding it with nitrogen? Moisture free! I bet that would be a cool video.
Seems Iike there would be some issues with maintenance and negative effects on the rest of the components of the board.
Maybe something not quite as cold?
A friend and I have done this when doing liquid nitrogen overclocking. It works very well
@@robertt9342 He didn't say liquid nitrogen, just pump in air temperature N2 like it's a glovebox or something
There are a lot of overclocking competitions using LN. The problem is that condensation starts immediately and extremely quick the moment you run out of nitrogen. Nitrogen is not air and it will run out unless you are the LN plant itself. Moisture free as long as you have constant supply of nitrogen which is not possible
@@catchnkill possible, if you would setup everything in a vacuum closed chamber.
I'd like to see this cooling rig used on the threadripper, just to see how the two CPUs compare when super cooled.
Damn that rig ROI will sky rocket.
same
I think what's stopping them is just the sheer amount of money they stand to lose if the rig gets condensation on a Threadripper
Yessss. Do this Linus.
I just want to see this cooling setup used all the time, on everything. There's so much potential for something like this, I'm anxious to see what the evolution of it will be.
@14:35 - @15:00
This is so true. When you want it to work reliably all the time - stepping off the bleeding edge can do wonders for the stability to run long jobs.
"It's gonna run for years"
Also less than a minute later: *drops it*
Nothing is "Linus proof" only Linus resistant 🤣
"Linus-proof? Challenge accepted." - Linus Sebastian, probably
That's an important distinction
So Linus is like Bullets..Woah!
its just scripted stupidly, what else
My astronomy cameras use a sealed argon chamber on the cold side to prevent condensation, I run them at around -20. Building a large enough sealed chamber which allows for IO and power connections would be an interesting challenge.
Positive pressure would reduce the need for perfect seals. Thermal noise isn't really a problem for them so it doesn't have to be uniform, it just has to be cold.
I was thinking the same, at that point it might be easier to just prevent all water vapour from coming in than the insulation they had to do !
A lot of the older subzero stuff used to suspend everything in 3M Flourinert
This... THIS RIGHT HERE @Linus
@Sarh sstfu
3:18
"linus proof"
Every hardware need that
I'm trademarking that and I'm going to make millions
You guy's honestley never cease to amaze me. The problem with going too cold is components can become brittle and fail. I would probably try to reduce the tempurature to the 'Triple Point' for the best balance between, cooling, energy and life of the components.
Haven't seen Linus this excited in a while. It's really nice to watch.
Haven't seen Linus drop motherboards in a while. It's really nice to watch.
They probably script these "drops" now
If that's getting close to threadripper I'd love to see them do it on an actual threadripper
Thrradripper isn't aci that thermally limited to begin with.
That's what I was thinking
threadripper is too efficient for that to make sense i am afraid
@@SimonBauer7 any sources for that? I'm interested
Nah, fuck Threadshitter!
If you wanted to daily drive something like this , it would'nt be particulary expensive to get a "technical insulator" to insulate it properly :) Black stuff is called neoprene (closed cell foam) and is mostly used to insulate cooling pipes/pumps/ventilation :)
@7:33 "not great, not terrible" - this quote would be such a great reference in that moment 😆
I mean this is the only way to properly cool an Intel chip... So well done LMG
@Clik the link below 💋👇 xxxxx fuckin bot
You could also upgrade to amd
@@Jdomi amd is more of a downgrade, unless you consider worse performance as an upgrade
I got 2 phase change units that run much colder than this if U wanna buy 1 u can beat their #s
I hate bots like these... they're so annoying..
"Wow! It's Linus proof!" - I think it should be any hardware manufacturers QA testing to have that as the first thing they do.
You can''t simulate the Linus Drop Test. All hardware to be tested must be sent to him personally.
@@mal2ksc I'm waiting for Linus Drop Tips
@Thor Crowley "Linus resistant" would be more accurate
The metrics would be called the linu score.
Love the look given to Linus at 9:55 when he suggested cooling the GPU too 😂
17:45 so what your saying is you should do this to a theardripper and just max it
Electro magnetic radiation is a bech, especially when the test bench is not shielded.
Some overclocking guides even explain that devices like lavalier microphones cause the overclocks to fail.
So maybe it is the cameras fault afterall!
Even Jayztwocents talks about this in some of his overclocking videos.
Put the whole thing in a grounded faraday cage
most likely, when you oc it works the first time not second due to voltages etc.
@@wyattmattas8294 lol... Was just going to check the comment replies and mention Jay if no one else did, because the comment reminded me of what Jay said, but fortunately you mentioned him
I would recommend using some 3/8 soft copper (preferably 1/2” for the suction or 1/4 for the liquid, depending on your BTU load) if you want to run this long term. Those hoses will leak over time, the hoses actually breathe, and the gaskets get worn out. Flare some copper, put in a brass gasket to ensure seal, and nylog it up. You can flare it on to quick connect fittings if you want and just pop the lines if you need to move it, no need to recharge each time (over time you will).
Would lines like that be flexible enough to move around regularly? As I understand it, they want to have it mobile and easy to setup. I've never worked with soft copper lines (and also have no idea how thick inch measures of tubes are). But I've worked with Swagelok SS tubing and know that's easy to set up but not at all handy to move around all the time.
@@WouterVerbruggen fair point. They will stiffen up over time, or if you change the setup too much. But as long as you aren’t constantly undoing flares, and keep the setup the same, it should be fine. They have a new hose out that is pretty cheap, it’s some sort 3/16-1/4” ptfe with and aluminum layer then a shell. It stays really flexy, we constantly bend the display pieces back and forth (with no pressure in them), but the tooling and fittings are expensive. That would be ideal b/c theoretically it would last forever (or past the life of the equipment).
@@WouterVerbruggen there’s a brand called Aeroquip that is pretty nice but pricy but there are others, I can’t remember their names
its not a long term setup and obviously you know copper is going to work harden as much as they will be messing with and it around.
@@Tlavite it will, but if it’s always the same layout it won’t matter, for the 20+ times they will set it up; alternatively they can leave it setup long term and not worry about leaks. It’s a win/win.
3:44
subtitles: (imitates fart sounds)
lmao
I used to run a modified Prometeia (ran -58°C idle) for years for extreme overclocking but also as my main PC. (Mainly used AMD Athlon XP's back then)
Didnt use dielectric grease but did use a lot of the provided black putty around the cpu area.
Never had an issue with CPU's dying, but I needed about 1 motherboard every 1 to 2 months.
This episode gave me so many flashbacks to those "fixing condensation" days XD
I worked in refrigeration for 7 years before moving into IT... I think you can improve your evaporator design greatly, but at the same time what you have created is pretty f*****g amazing and it's what I've dreamed of doing for some time now.
I would like to experiment with a vacuum chamber case also (but I can see this turning into a black hole of I/O seal design....), but getting a heat pump to run as a CPU/GPU cooler is something I would like in my personal rig.
You have done a really good job at janking up the control board, but this is something I feel may also be a huge time sink when trying to master it for every day use.
Wouldn´t go down the vacuum rabbit hole. At that point you might run into problems with stuff one usually doesn´t have to worry about like RAM temperatures, ssd heat and so on. Just seal the case,put desiccant in it, have some of the sub zero cooling exposed as a water trap (bottom of the case obvsly) and case-heatsink and have fans that circulate the air inside the case. That was always my dream, but money is a thing apparently...
why not just get a bunch of male>female cables for all of the IO ports, you can seal the whole thing, and not have to fuck around with io holes, it won't be super modular, if you use a motherboard/any add-in cards that doesn't have the exact number of ports you chose to have passthrough cables for you'll have some unused ports, but it's a simple solution
Yeah fair call on the other components. I could see how it could be possible to liquid cool components likes memory/SSD and run the GPU on the AC system, individual chips would be exposed to a vacuum. I mean that might not actually be an issue in theory as I'm pretty sure there are off the shelf chips exposed to the vacuum of space on current spacecraft, but I'm no lt sure how the chips on a gaming motherboard would like being in a vacuum.
I would like to disprove it by experiment rather than conversation though, that's for sure.
Thanks for the insights from the pro! Can you help us to find good documentary about how Lord Kelvin / Boyle tried to reach absolute 0K with evaporating liquids?
Use farts
I love to see the longer videos, it's great when you go into all the details
Agreed
I’m 100% cool with hour+ LTT videos
22:43 This man is passionate about what he does. That smile says everything 😂. Keep up the good work guys!
Massive respect to the man, who made the cooling block in the CNC and did the homework to make this possible (on that old 8700k) - xlnt job!
You wouldn't have problems with condensation if you cooled the whole room to -20C.
only if they didnt breathe :). Also, I wouldnt be happy being Seasonic recommended for stable OC in a vid full of BSODs :D
well they are Canadians, they are used to -20
That would make an interesting video.
Sounds like Canada
that is basicly how i chilled my waterloop to force my way into the 3D Mark Leaderboard :D
but is was only 7°C outside
3:05 Welcome back to Linus Drop Tips
frr 😂😂
I'd love to see a third round video with Splave involved using a hand picked cpu for best overclocking. Also the insulation used here seemed even better than the liquid nitrogen setup.
I owe my career in refridgeration to farting around with A/Cs and computers. lol. Glad to see you guys dig this hobby back out again.
I love when the Linus/Alex combo shows up to really get down and flex on something really awesome. This formula is by far my favorite, and in my opinion probably the best showcase of the legitimacy of "Linus Tech Tips", out of all the sorts of things you guys do (which I see all of). Alex and Linus rolling up their sleeves together is always an instant click.
I have always wondered why when overclockers are testing temps below dew point they dont just build a simple test chamber with a container/chamber of calcium chloride or even pass the air through a small evaporator to remove the water from the air. Seems extremely complicated to try and mitigate condensation than to just remove it before hand.
Why calcium Cloride?
Not sure where my message went but it is extremely hydroscopic and it is easily available from stores and a safe to use dessicant. Damprid is an example. In a few lab setups i have done, we pass/recirculate air through pellets of CaCl2 inside of an enclosure when we have to meet extremely low amounts of water vapor. In these cases we are measuring by absolute humidity instead of relative.
Use a walk-in freezer chiller running 410A, you should be able to get it to -85F with that. To stop all possibility of moisture, put the MB in a vapor tight case and use a deep vacuum pump. The blue return line [low side] should be twice the size of the red high-pressure line and there should be an expansion valve just before the refrigerant gets to the CPU cooler. Some newer split systems use variable rotary pumps and maybe you could fool the control unit to crank the compressor up or down as needed with a resistance modified thermocouple.
Just some thoughts.
9:49 "We're not risking anything too expensive here"
They say while handling a graphics card that gets sold for at least twice MSRP anywhere I could find it.
I wonder if they ever going to do a mineral oil cooled combined with this AC cooling unit. That would be a challenge.
Was thinking the same.
I had checked and commented on the last video. Mineral Oil you have to be careful, some have very "high" (most common is -15C) pour/freeze point. But 3M and a few others do make non-conductive immersion fluids especially for this in server/supercomputer applications, and they can go -60 to -100 C pour point. So they will stay liquid enough to circulate them at low temps.
Just for those interested: the artist eraser is used to lighten usually rather then erase completely and is used mainly where u need precision and of course want to eliminate smudging. Except if you stick and “drag” then it creates “equal” smudging.
Pardon the vague explination… cant explain it more clearly due to lack of correct terms
Is it the same as Blu-Tak?
(meaning "could you use blu-tak, the poster-hanging adhesive, as a cheap artist's eraser?)
(Blu-Tak is a brand name used in Australia. I don't know what company makes it overseas. It looks just like that, and you stick shit onto things.)
@@briannem.6787 I don’t know but I daresay it’s different, since it’s meant to be an adhesive
@@briannem.6787 Its similar to Blu-Tak, only an bit less adhesive. I know you can use Blu-Tak as a “press” eraser, however it does lose some of its adhesive qualities in cold weather and when wet. Which may cause problems
It's "silly putty." Same crap. Was a byproduct discovery from a man attempting to develop wallpaper glue.
We used to take it and copy comics out of newspapers with it until we got bored of it.
The formula has been slightly altered here and there over the decades, but it's essentially the same crap it always was.
Art supply companies just do the same thing every other company does. They see something that someone uses as an alternative use case in their purview of products and rebrand it as if it's exclusive to them.
2:17 “It’ll be a messy boi” 😂🤣😂🤣
Oh man... that bash on Seasonic had me crackin up! Hahaha
I once worked at an R&D dep where among other things we did a lot of thermal testing. What I learned there was that you really want to put the thermocouple inside the cooler block. The exact location depends on what is that you want to measure/model but typically I think it we drilled a tiny hole right above the chip, stuffed the wire in it and filled the hole.
linus: 'dielectric grease serves a very important purpose'
r/mk: 'stabilisers!'
linus: 'condensation barrier'
r/mk: cries tears of 205g0
Man I didn't understand one word that you wrote can you elaborate please oat milk. I love to learn
Yes
@@nutdealer3723 we use dielectric grease do lubricate our stabilisers on keyboards (its non-conductive and passive to the plastic)
the main reason to lube stabilisers is to reduce the rattle you hear on large keys such as the spacebar/backspace/enter/shift
krytox 205g0 is different type of lubricant we use for switches to reduce friction and make them sound less plasticy
@@nutdealer3723 Custom keyboard building joke. Stabilizers are a part of a keyboard switch that often needs to be lubed, best known lube is Krytox 205g0
@@nutdealer3723 it's mechanical keyboard terms. Dielectric grease is usually applied to the stabilizers for keys such as the enter and space keys. 205g0 is a kind of lube used for mech keeb switches.
May be you can prevent moisture by putting the test bench in a very small room or a case where you also put a dehumidifier.
If you have no relative humidity, no condensation. :)
yeah vacuum like C90 told 4 months ago, or transfo oil submerged, or why not both...the board under oil in a vacuum maybe you would need at some point to adjust the AC power on the watts you are adding in the cpu ...im sure the setup is limited by the remaining safeguards
11:14 Ah yes, the classic "Do as we say, not as we do" moment :)
1st
Have you never thought about skipping all the insulation-goo, and instead just build the board into a close-to-airtight box and flush it with dry air / nitrogen? Takes a bit of development at first, but it will make your live much easier the next hundred times you fool around with sub-zero cooling.
That'd be cool to see at their new building.
I mean don't they have the machinery now to do something like that?
Wonder if they could just do a superchilled mineral oil bath? LOL
@@Nevir202 That does not sound like "less of a mess" ;-)
@@ShieTar_ Was that a goal? Thought it was just to be cold and not short out? lol
The rubber eraser on the mother board hurt me as an artist. I have to keep all my art studs soooo far away from my computer to prevent any accidents, and you’re just slappin it on your parts!
increasing the dielectric constant from air 1 to DEgrease 2.9 will have some effect on the overall performance of the CPU as the (pin-pin) load capacitance will increase almost by 3 times. a small glass box with a filter and a compressor will dehumidify the air inside so it might solve the condensation problem
Or they could plumb a line from am air compressor that has been run through a dessicated or refrigerated air dryer.
If the GPU was cooled too could you just put the whole thing in a vacuum?
@@sperglet7503 yeah, but the chamber the chiller connects too would need to be airtight
"Linus-Proof" should become a new badge to qualify to PC parts. The test would be simple: let Linus handle the component for a couple of minutes until he drop it and test it.
Pretty sure Steve from Gamers Nexus could include this test.
LOL @6:00 - I was one of those idiots on [H]ardforum, built a daily driver phase change setup. Opteron 165 cooled by a 1/4HP compressor, SO overkill.
Very very interesting and also congratulations for this huge record!!
For a 24/7 phase change cooling you just need a heating system to avoid the condensation in critical points... with my Vapochill I've used this system for several years without any condensation issues.
I also suggest to put the phase change system in a different room... you just need a hole in the wall in front of your case ;-)
That left handed complement to seasonic was the best. It was such an insulting compliment but yet gave them so much credit in one moment. It was beautiful, and well executed!
It's enjoyable watching him doing something he's genuinely into, and giving tips (OHH!) and teaching others about it.
Dope
Dope
Dope
Yep! When he said that there are some people daily driving chillers, he was one of them! That’s why he is so into it.
I haven't had a BSOD since... what, 2 years or so ago when I had a sudden every few days BSOD happening... after having done a last-chance update of Bios and such.... bam, no more issues. I had no idea what exactly fixed it, but going manually through Device manager and check for updates for everything there, as well as update my BIOS or such... it fixed it.
its crazy to see this a year later and seeing my 13900k thermal throttling 100c and still getting like 40k in cinebench with a 360 aio XD
How difficult would it be to do this in a Nitrogen only chamber? Or maybe even keep a steady stream of nitrogen blowing all over it so that there isn't an lot of water vapour to condense on the electronics?
czcams.com/video/kgnfvb2t-TQ/video.html
Either that or just a chilled insulated box. Should reduce the chances of frost if the inside of the box is closer to CPU/cooler temp then it is to room temp.
I also thought of this. Just put the whole board in a dry chamber
Exactly. This shouldn't be that hard to the LTT team.
near vacuum chamber might be easier 🤓
I actually ran a vapochill on my gaming rig for a couple of years, was insane when it worked but a nightmare when something needed reseating or upgrading. My Athlon 64 3700+ San Diego and later a core 2 duo 6600 really liked the cold when oced.
Oced and cooled like that i think the dual core will work perfectly fine this days
I got phase change unit in storage. Time to dust it and put back into use, it's able to do -60 degrees of C on evaporator head all day long.
For instance...the entire room/environment the system is in could be could to sub-zero. That should create a super dry atmosphere and eliminate any moisture which would allow formation of condensation. I have been in cold storage units for food where the interior of an entire (very large, 20,000 s.f.) building is below zero (Fahrenheit).
Given that this was lead by Linus I'm surprised it looked LESS scary than when Alex did it with the 8700K. I'm not surprised that he dropped the motherboard though.
Linus is not exactly new to this
Here is a super crazy idea. Install that super cooler, and just dip everything in a aquarium with mineral oil. That will prevent condensation :D. I would love to see what happens to the oil at -20 ºC.
It becometh mineral wax
Wait for rain
@@BlastinRope Even better :D
It freezes at -4 C
Is this feasible? What drawbacks does it have?
as an hvac tech... this has to be one of my favorite videos lol
I'd actually say it's kind of impressive how well a two gen old CPU is holding up all things considered.
Linus been on some nostalgia trip lately extreme overclocking everything, I'm loving it!!
This overclocking method pretty expensive, but worth it for servers if they can do it over windows 10 instead of windows 11. If they sold it on both eBay and Amazon, it could fetch people attention
ok, question:
Could you combine an "oil cooled pc" with a phase change one?
As in, could you have the mobo sitting in the oil, and then the phase change tubes hooked up the same way they would be here, but then not have to worry about the insulation?
Or at least, not having to insulate the board itself, just the tubes down to the surface of the oil.
I guess it would depend on the freezing point of the mineral oil.
Mineral oils good to at least -40-50C. You are gonna have to push dry ice, liquid nitrogen level to get there
Lol I literally just commented this same idea. It works, I just wanna see Linus do it.
Had this idea for years but never had the means to execute it. Would like to see someone try it
You wouldn't want to use oil for that. You'd want to use anhydrous alcohol. I'd also recommend a sealed enclosure with an air gap that has been purged with a non-condensing gas like nitrogen. I doubt they'd be able to run it cold enough to condense nitrogen with that system. Additionally, I'd personally be quite concerned about trying to use a fluid that is combustible and has a particularly low vapor pressure.
All you need is a sub ambient case with a seal that you can control the internal humidity of. I chose a 2 glass door refrigerator. This allows me to run a sub ambient coolant and have virtually 0 moisture issues Several pc's in the fridge all running of a manifold geothermal cooled water loop. gives me a steady 10 Celsius coolant temp. Now next project is the refrigerant cooled setup, going with full cover front and back mobo water plates which will circulate the geothermic chilled coolant. Then one of these phase blocks but I am considering a larger block that will serve as both the cpu cooler as well as the gpu cooler. like a big L shaped block. Don't have the machining tools here so will be using 3d printed patterns to cast copper parts from. Pretty sure I can cast a complex internal structure for the evap block no solder required.
17:00 because the thermometer is inside the CPU, on the hot end of the thermal gradient. It's closer to the heat source than it is to the frost.
I wish I could've seen how they got that thing painted. The paint job is gorgeous.
I love that they kept the sound pads up from when they tested them like a year ago
Missing out on the best way to do this. Use a liquid cooling system and replace the radiator with a reservoir with the AC hooked to a coil of copper tubing submerged in the coolant to super cool it. Would be sweet. Would also alleviate the condensation issue
This so satisfied my nerd inside for this whole video My soul was at peace
I'm going to take a guess at why it showed thermal throttling during the test. The temp probe is using the delta r (change of resistance over period of time) as a variable to calculate the temperature. The software calculating it can only calculate natural numbers, including 0 (N0). When the board starts measuring the delta r it starts at a negative C temp and acknowleges it at 0°C so that the difference in temp (delta T) might very well be 100°C but -30°C+100°C=70°C. That is what's causing the weird meassurment, at least that's what I am assuming is happening here.
Remember kids (and obviously ASUS in this case), always work with Kelvin and then convert to C or F for the consumer.
"We're just nice people."
-Seasonic
is still my favorite
This has turned into a very legit looking extreme overclocking platform, I love it. Brings back memories of people posting pictures of their phase change cooled Pentium 4s on message boards.
Have you ever had a submerged subzero setup?
Would that take care of the condensation issue?
Dropping the mobo, CPU, and ram in a few inches of mineral oil may have been easier than all that disassembly and insulating just to keep air off it. Just use a PCIE extension ribbon for the GPU. This also has the benefit of cooling the VRMs and RAM. Cleanup with alcohol and a blow nozzle.
i think another commenter said that mineral oil freezes at -4°C, which would be bad
Also the mineral oil won't magical repell the water. Since water is more dense it would displace the oil ant could lead to a short
@@mrdoode1991 you would only have water if air got to the system to form condensation. With the system submerged and lines insulated where would the condensation come from?
@@os4b4 why would that matter?
@@Noah-lj2sg I don't understand how frozen mineral oil would affect anything. Also, if it is staying frozen .. you have a lot more possible headroom for over voltage. I just don't understand the premise of this point.
You guys definitely need to aim for the highest possible score using the threadripper and removing the cap for better cooling and go all out with ac it would be amazing to see the highest score
30 seconds in and I'm like 😭😨😱😩 experiencing a whole range of emotions I didn't expect to feel
You also need dielectric grease. It's messy but stops all air (therefore frost) getting past the tape gaps.
I remember back in the day, early 2000, that I was amazed by the Asetek VapoChill in my local computer store.
It would be interesting to see the preformance on a modern setup if you could find one in working condition.
I still have one , Hasn't been used for about 15 years (condition unknown ) . Would be happy to donate it, I'm in Australia though
@@jamezxh could always hook it up to YOUR system and see how far you can crank her up before she explodes
Why not do another collab with Steve for this? I feel like you'd really make some progress together with this this time with the improved chiller and insulation.
I think collabs just depend on steve being close by for something else
Border controls and travel difficulties probably make that not terribly attractive right now.
@@mal2ksc They could livestream or zoom if possible. Time is an issue though. Esp Linus with his house moving and whatnot.
Saline wrap that to keep the moisture out… it helps a bit!
You should try to drop the whole asembly in how candle wax and let it dry. Ofc you have to put masking on the parts that still need to be plugged at the las moment. In terme of moisture insulation, that should be perfect. No need to add tons of foam 😁
I’ve managed to overclock and under volt my 12900k and reduce heat simply. Go in BIOS, set my vcore to 1.23v, set P cores to 5.1 all core, and 4.1 on E cores. Ends up pulling 250 watt power draw, and 74c package temps when doing R23. Using Galahad 360👍
whats your superpi 1m and 32m score?
just running it stock instead of those stupid auto-OC values would already be more than enough.
@@velo1337 Took me a bit to look superpi up, as that is a new benchmark for me. I ran it and i believe the "score" of superpi is on the lowest amount of time it takes to run 32million prime numbers of Pi. My time on it was 0h 06m 00.430s PI value output. Hope that helps.
The look Alex gives Linus at 10 minutes about having the chiller work on a GPU killed me, like he seemed so not about it but then realized it would be dope too XD
Cool project, I have an idea for your next iteration … encapsulate the whole thing in a see thru plexiglass box and put it into a vacuum then maybe you don't need to worry about condensation!
You really know to tackle a challenge and win. In the 1970s, Chevy produced a car named the VEGA. It had a feeble four-cylinder engine. Owners would attempt to stuff a small block V-8 under the hood. Some actually succeeded, but the VEGA needed many modifications. This proves that any DIY project, like yours, may require a massive infusion of Time, Effort, and Money.