Linus: *tries to keep the door opening small so the moist air won't condense on the testbench* Also Linus: *BREATHES DIRECTLY ONTO THE TESTBENCH INSIDE THE CHAMBER*
My first thought for that problem was that they so easily could've just draped something (surely they have loads of moving blankets at LTT) over the door for Linus to duck under and mostly avoid all that drama. Wouldn't be a proper Linus/Alex duo video without a moment of total chaos though, gotta admit it makes for a better vid to just wing it
@@Piipolinoo oh my bad, guess I thought people are here for shenanigans. I must have missed LTT becoming a concise, no-nonsense consumer advice channel
As someone who lives in a tropical climate (the Yucatan peninsula in Mexico), I’m happy to see you guys test this, our electronics feels disposable and our phones are always hot. If I you want to play video games you need an AC for optimal performance. As an international viewer I’m excited to see what you will test next. I’m considering paying for floatplane.
> If I you want to play video games you need an AC for optimal performance. At such temperatures if you want to work you need AC for optimal performance. I had for some time a heating problem in summer(last floor with black roof did make +40-45C(with 70-90% humidity) in the room and it was a nightmare) and with AC i have comfortable +20C in the room all the time.
That glycol looked pre mixed. Since their was water in the loop you probably got no better than a 25/75 mixture. Their are charts for antifreeze to water mixture to temperature tolerances easily available online so you can properly mix the antifreeze for ur car. Just use antifreeze which is designed to known standards and specifications and will produce repeatable results and mix as necessary to not freeze at -40. For -40 you require 52% antifreeze to 48% water mixture to prevent freezing. Premix the solution, completely drain the loop and fill with solution.
Just buy 100% Ethylene glycol and mix it yourself. But I agree, just going from the viscosity of the antifreeze when he was pouring it, it was definitely not just ethylene glycol.
You're likely thermal shocking hardware if you're pulling from ambient to -40 and back again (~60c delta!). Ideally you should bring the chamber back up to ~50c to do a drying cycle before you come back down to ambient to negate the need to leave the system to dry out for a few days. There's a few fluids that you can use for "water" cooling that wont freeze at low temp (Engineered fluids EC110 is a good one - also dielectric). Source: I design military servers that operate in -40 to +55c environments and make a lot of use of a very similar climate chamber!
- leave at 50°C for a while - put in 10$ worth of dessicant, just a couple kgs of zeolithes from an aquarium petshop or sth - put in a 5$ humidity meter - cool down appropriately slowly while keeping an eye on the humidity meter
I love how there were no conclusions. Just went with it, ended it. Just because you know we only wanted for this to happen, not how or why. I love these cooling videos
@@BlissToby when they fix it with a compressor they will 100% do that... They just can't give conclusions with non controlled tests which i can understand
More content like this is important to me. What good are the labs, if we can't see these things in action? These die hard extreme tests are great fun to watch!
@@elbeetlebeasto impatient? They've been setting up a new building for well over a year now and are most likely running behind schedule. They have every right to rush out videos for the sake of content. I'm sorry but would you rather wait another year for a video like this to come out? Also keep in mind, what they do on camera is for entertainment purposes. Prototypes are funny to watch, any content creator with sense knows that a rushed prototype is funnier to watch then a fully fledged working project. In fact, they'll most likely have a follow up video where things go right. Basically what I'm saying is they make money off of TH-cam videos and a tight ship that's running as intended just isn't funny to watch. They have to have both the more informative videos and more entertaining ones. Also I'd recommend you check out the livestreams, specifically the WAN show which talks a bit more seriously about tech news and such if this kind of content "enrages" you so much.
@@elbeetlebeasto Linus did explain exactly what they were doing wrong about temp management of the chamber, which means that this is just a for fun side project, rather than a serious take on the subject. It also means that they understand how they should do things but deliberately decide to ignore. As the other fellow said, they probably will have a follow-up where they make proper use of the equipment, and probably a third video once the compressor they're missing arrives to properly test humidity as well
@@elbeetlebeasto you're forgetting that labs will have its own website where the experts they hire will do proper testing. This is just a fun infotainment piece about the capabilities that they have now. So yes this is rushed and kinda bullshit but this was never supposed to be an in depth informative piece
Remember that the Lord Jesus Christ died on a cross for you because He loves you so much. He then rose up from the dead three days later The Ten Commandments are called the moral law, (most of us are lying thieving blasphemous adulterer at heart and deserve hell) you and I broke the law, Jesus paid the fine. That’s what happened on that cross. By believing that Jesus died on the cross and rose up from the dead 3 days later and not just confessing your sin, but also repenting of all sin you have done and putting all your trust in Him in prayer, He will grant you everlasting life as a free Gift.
Now that you mention it, i wonder if they still have access to, or have, the computer case that was effectively a giant passive cooler that they did a video on.
@@agalerex I remember seeing 12" wide enormous copper headsinks for desktop processors like 10 years ago, they were for applications that needed to be silent like acoustics labs, I remember you could run an i7 with one, just not at super high power levels
50 degrees Celsius stress test must be done by computer part manufacturers. Especially if they want to sell parts to Australia, India and other hot places.
I agree. My Asus laptop died while playing at 43°c, but here in summer the temperature can go up to 50°c with really high humidity. HP laptops here are famous for failing at temperatures over 35.
You can tell they have never used and enviro chamber before. You have to run a normalization cycle before you open the door! Bring the components back to the air temp of the lab before you open the door. It only adds a few minutes to the testing and you don't have to worry about condensation. Glad they figured it out at the end. Also, you can mix glycol 50/50 with water and it should still work fine for those testing temps.
Keep at it guys! It's possible. I've run my PC around -25°C. Left it in my semi during winter and wanted to see what it would do in Cinebench. 280mm corsair AIO, 5900x, strix x-570I. Can't remember the score but it's possible. Canadian winter overclocking is fun, free sub zero cooling lol.
Something you must remember your pc case (The Chamber) is actively cooling or heating the system inside to hold that 50c set point, we will not see a thermal run away because of that. The GPU running at 90% fan and 80c is getting cooled by the chamber’s 50c air.
A lot of components will get extremely angry once they go below -10-0C. Primarily PSUs and SSD HDDs can act up. There's also sometimes a "non-op" and "operating" difference. For example if you bring a PC that is running down to -40C, it usually is easier than doing a "cold soak" at -40C with it turned off and then attempting to boot. Source: have done a crap load of testing MIL-STD 810 at an aerospace company
Makes sense, I remember there being a similar issue with liquid nitrogen cooling they did in a video a while back, they had to get a load on the CPU before starting to really cool it or it'd go too cold and just stop working, I'd imagine similar issues in a lot of components in the PC, they just wouldn't have worried about it, cause... well... They weren't submerging the PC in the stuff lol.
Love from India ❤ During summers, the room temperature varies from 30°C in the night to about 35°C-36°C in the noon in India. Currently it's 4 in the morning and the ambient room temperature is 32°C. Testing at such temperatures is crucial for us who have tropical summers and kudos to the team for doing this. Thank you.
I am living in the Philippines. The highest room temp I got was 35c during "summer" march to may. GoW and Witcher 3 was about 73c on the cpu and gpu around 76c. Both did not reach 80c. But during cold months around nov-feb. Temps would go below 70 during the day and almost 60 during nights
Linus and Alex testing tech on the way to the North Pole would be a worthy sequel to the Top Gear Polar Special. (this video suggestion is on me today Linus)
Condensation should be only a problem if your PC is cooler then the ambient air and not if it is warmer. Reason being is the air has a humidity limit based on how warm it is. So the warmer it is the more water it can hold. If you have that warm air that has humidity come in contact with something much cooler the air will cool in response and if it cools enough to drop below the humidity limit of that air at that temperature then it will release water (Form Condensation) in response. If you take something warm and put it in a cold environment or warm something up in a cold environment that warming of the air makes it able to take on more water and instead will encourage the opposite of Condensation which is drying. In otherwords you don't need to worry about your PC forming condensation in extreme colds. However in this specific circumstance if you do freeze the air rapidly without releasing the moisture first you will get frost on everything in the testing chamber. However the warmer components in the PC will be the slowest place that frost forms. In summery though: -Warm air + Cold Stuff = Condensation. -Cold air + Warm Stuff = Drying or in least no condensation.
if you have a contained volume of air and lower it past its dew point, it will condense. not directly on the parts, as you would get with cold equipment, but you'll get fog, and likely dew on everything. (evenly distributed condensation) if you were feeding it a flow of already cold air, any condensation would have already happened in the cooling process, so no further condensation would occur. Frost is simply condensation that freezes immediately.
@@nwimpney, yep is what I was saying that if it is trapped air cooling it will cause an even distribution of Condensation. But the warmer part of the computer will actually Condensat less technically.
I don't know if you have some laptops you were willing to destroy, but I feel like the cold test (even with moisture) could be a good one for laptops. It could kind of simulate taking your laptop from inside, transporting it in the cold, and then running it again. It's something I never really thought to worry about, but there were definitely times in college when I would walk to class in the cold and my laptop would be really cold and have some condensation on it. Now I'm wondering if that did any damage to it.
If you really want to know, you could see if your model of laptop has any MIL-STD-810 certification done on it. One of the parts of that spec has to do with thermal shock.
Ah my old laptop from college does show liquid damage signs and has for a while, despite having never spilled anything on it. I took that thing with me everywhere in my backpack though, through sun, wind, rain, dirt, and fog. Living in a relatively cool and humid forest area, I always assumed this was the case.
Yeah I really want to see it with a ROG Strix laptop since they run very hot. I'd also like to see some more gradual cooling so they can find the sweet spot between perfect performance, dropping performance, and total failure.
I remember trying to cool our server rack by pumping in outside winter air at around -10°C, but it seemed to warm up in the 5" pipe before getting there. Goal was to meet the old ATX recommendation of +40°C CPU temp max.
Thank you for actually testing in the warm configuration. The thing about tech reviews is that here in India we do not get a clear picture of what performance we could expect at home in normal living conditions and whether a chip is efficient enough to not throttle down to half its actual performance. I trust your reviews to be more inclusive than anyone else's. Hoping you can do this for most of the relevant gadgets ❤
Philippines here: A good case and a mix of Arctic F12 and P12 max fans should do it. I had overheating (above 80-85) or loud PC before too. Then I swapped all 5 case fans to Arctic F12 and P12 and since then I can sustain 70 degree on full Prime95 and 3D Furmark stress test on CPU and GPU. I was totally amazed how much a good airflow setup can make such a huge difference. Much more than in all the tests - since temps here are so much higher, than in the test setups of reviewer/youtuber! Went from 2 cheap (=shit, since Arctic is cheap too but sooo much better!) in front and 1 cheap upper rear and good but too loud rear out, to 3 F12 front in, 1 P12 rear top in (directly onto CPU cooler) and one P12 rear out. around 12-15 degree difference and almost complete silent, unless I am gaming. Of course water cooling would be ideal. But expensive here.
It was weird though. Idk about India, but here in Argentina we have a ton of humidity AND we do reach sometimes 40°C (or more) in certain places. My CPU at idle has been 55 or 60 tops, never reaching 75 or more doing nothing. Perhaps it's because mine is a ryzen 2600 and this was a high end Intel. Still, my case sucks ass (it's a generic badly designed one) and I have the stock cooler.
The suction line accumulator in your system linus is used as an expansion chamber in low temp systems where you run the chance of actually getting liquid refrigerant coming back from the evaporator. It uses the high volume of the accumulator to flash any liquid directly to a gas before it gets to the compressor
While you wait for an electric dehumidifier, you could use a kilo or 2 of heated Silica gel put inside the enclosure that is at room temperature. It will suck up the water before you freeze everything. It is a “desiccant,”
They’ve literally got a dehumidifier right there. The refrigeration system is exactly what a dehumidifier is. It naturally pulls moisture out of the air. They just don’t know what they’re talking about.
At these temps you might actually need to consider contraction and expansion of materials. Maybe thats why the temps were so high for the direct die cooled pc, because of the cooler not making good contact anymore?
If they cooled everything down and then applied the cooler, I bet they would get waaay better temps. (also probably not using frozen liquid would help)
It does not make any sort of difference in regards to high temperatures. Which is only natural. Normal thermal pastes are only affected by very low negative temperatures, when they freeze. I doubt that’s the case. Anyway, there are special thermal pastes, which can whitstand extremely low temperatures when LN2 overclocking.
@@ron200088 Oh that's is very helpful thank you. I was worried about playing games on max during the summer, I worried that my thermal paste would dry out quickly or something like that
That refrigeration setup is called cascade cooling, I service these chambers for an aerospace company in Florida. It's only 1 stage but it's two circuits, circuit 1 evaporator cools circuit 2 condenser, circuit 2 evaporator cools the chamber.
I have always wanted this video. Seeing a PC operating in a really cold room or a really hot one. It surprises me it took this long for you guys to do it. Great video and thanks for making it! ❤
I think you have the humidity and temp thing backwards. The total water the air can hold goes up with the temp, so by heating it first, you likely had more total water even though it was low relative humidity, since that's relative to max it can hold.
That's what I came here to say too. Heating the chamber increases the capacity of the air to hold moisture, so the RH% goes down, but the absolute amount of moisture in the air remains the same (unless you introduced more when you opened it). This is why humidity is an issue and your AC removes a ton of it in the summer when it's running (because lowering the temperature raises the humidity), and also why indoor air gets so dry in cold climates in the winter (heating the indoor air doesn't remove moisture, but warmer air holds more moisture, so RH% goes down).
One of them few times I've tried to visualize a psychrometric chart in my head... When heated, relative humidity would lower, but the absolutely humidity/humidity ratio stays the same assuming the test chamber is a closed volume with only thermal energy crossing the boundary. If anything, precooling as low as possible and literally toweling up the condensate, then raising to a desired cold temperature could work.
You were really close to the job of the accumulator. Instead of moisture in the system that would turn to acid, the accumulator stores and traps liquid refrigerant that hasn't turned to a low pressure gas before entering the compressor turning it into a high pressure gas. Getting liquid in your compressor is the number 1 reason a compressor fails, so this is a way to prevent that especially in bigger systems. But it is cool lately you've been kinda doing some research on how air conditioning works on some of your chillers. Keep up the great work.
you all need to run that test with a non heat pipe air cooler, something like the slightly more beefy intel stock coolers the ones that are triple height and have a big slug of copper in the middle, that way there are no heat pipes to freeze over, and no water or pump to seize up with the cold either, that would be the best bet for running at -40 with it running without issues. pretty sure i saw one of those intel coolers I'm talking about on a video you previously did, might have been comparing intel coolers over the years or something. I know the coolers I'm talking about were meant for workstations, I got one years ago bundled with my i7 970 and it actually performed well enough that the CPU was usable in Aussie summers, though was noisy and I ended up upgrading it pretty quick
When I would go to the desert it would often get into the 115 to 120s so I think it's awesome to see stuff like this. I've had more than one phone stop entirely cause of the heat
I wonder if it would of worked if the CPU cooler was swapped with an aluminum or copper heatsink with no pipes, and the paste swapped with a graphite pad.
Sometimes the gas for the heater would empty before it could be refilled again. Playing while there was no heat on my PC was actually nice. The ambient low temperature prevented my PC from overheating. It was an old PC but it was my first PC. I would play games outside it's specs on low graphics with low framerates. The heat generated by my PC could warm up my hands when they got clammy. I swore it also performed better. It regularly reached 40-57 on hot days even in Canada. Like you guys said poor ventilation in both the room and case would cause the internals to heat up to a surprisingly high temperature. One day it was shutting down whenever it would reach a certain temperature. This corrupted the BIOS. The poor thermal paste hadn't been changed in over 10 years. The PC was originally my Dads. When I opened it up to replenish the paste what was left was pretty much stained to the top of the CPU chip and was basically non-existent. Unfortunately I had shorted my CPU when I applied some new thermal paste. I even bought and installed a new thermal heatsink and fan for the CPU. I never got to finish fixing it. I had left it behind when I moved.
We've had Dell reach out to one of our customers and tell them if we didnt raise them temps in the data center they wouldn't honor RMS's because it was too cold. Definitely surprised when we heard that one.
Apparently a streamer actually did this in Alaska. And they had the same issue with the AIO so they switched to air cooling which worked and they clocked an Intel cpu at like 6.3 ghz or something
More recently, I had to do some maintenance on one of my older computers, a Gateway E-2000 from 2004. The computer was running Windows 2000 and trying to play Bejeweled 2, which caused a stop error (blue screen of death). It was rather late at night, but I still tried to do some "repair" to the system. As the stop error was caused by a page fault in a non-paged area (or most likely, the PC trying to play the game, which was having some difficulty), I was still able to restart the computer in normal mode and I backed it up, and tried to do an antivirus scan and disk defragment. Neither of the last two actions got completed, but why I bring this up is that I noticed the system was getting rather warm, and since I didn't know if I was going to just let it run while I was asleep, I didn't want it to overheat (as this was my first time doing this kinds of stuff), so I placed my powerful floor fan nearby the computer to try and cool it off. Wasn't the most effective tactic, but I thought I was at least being helpful. Also, if you decided to read this, I converted the units to Fahrenheit for everyone who uses those units. The coldest temperature is -40°F, and the hottest temperature is 356°F.
17:20 WTF was that cliffhanger. One of the best videos ever. Love your content guys also great camera work dealings with Linus and editing to increase the shock value.
heat isn't how you remove moisture (water). heat evaporates moisture which increases humidity which condensates back to moisture when the temp drops. you need to chill everything to single digits C or upper 30s F so it does condensate and can be wiped out or ideally drained at the evaporator. just like the water that drips from a car in the summer because the cold air conditioner condensates the humid air and it drains off.
I was in doubt about which 4090 model to buy and so I chose the Rog Strix after this video. Where I live not in summer we have temperatures like 40°c and this test was really insane.
ปีที่แล้ว +24
Alex and Linus janky cooling projects are my favorite type of LTT videos.
At least you’re getting a lot of mileage out of that temperature chamber. For a second I thought you actually sent a computer up to the North Pole with long-distance remote tracking.
A little advice from someone who uses environmental chambers daily. Shut it off before you open the door. The system is working overtime trying to correct for the sudden change in conditions. That will shorten the life of your chamber and also shorten the time to your first repair. Also, glycol does not work for crap at 100% concentration. The "antifreeze" you buy at the auto parts store is made to be mixed 50/50 with water. What you should do is stop cheaping out and just go buy a bottle of pre-mixed antifreeze. Also, don't get the "low tox" stuff if you are going to very cold temperatures. Finally, what was coming out of that bottle looked in no way like glycol based anti-freeze. Because you went cheap I'd bet you grabbed someone's half filled bottle - probably just water they had in their trunk for a roadside emergency. Research some of this stuff before you try to make a half-baked video and destroy a bunch of equipment.
I have 2 OCs saved in Ryzen Master. One for Winter which is basically the max it can do at 5Ghz and one for summer where I stress tested during 30C @ 4,825 Ghz and 1.3125 Volts. It can reach 35C here sometimes but those days are few and far between.
I'd like to mention heating a room up does not remove water from the air necessarily, relative humidity will lower but that's because warmer air can hold more water. As the temperature drops the air can hold less water making the relative humidity go up. So you can have the same amount of moisture in the air at two different humidities depending on the temperature.
Yeah that was absolutely pointless lol. Also an object sitting in a refrigerated room/chamber isn’t going to condensate when it’s not the evaporator itself lol.
While you were there, you should have tried removing the cooler altogether! Wouldn’t all that moving cold air in the chamber cool the cpu down enough? I would like to see a test with just a bare cpu in -40° Great video! I’ll be more cautious about eagerly installing upgrades that have been sitting out on my front porch in the winter! ** I understand that this experiment was about testing a gaming pc in extreme cold temperatures and that a cpu cooler is essential for any pc**
It wouldn't be enough on its own to keep the chip cool. One of the reasons why CPU coolers have to be tightened is because the additional pressure added by the mounting mechanism will better conduct heat away from the IHS (metals under stress will transfer excited electrons more freely). LTT had another video a few months ago where they used a vortex cooler aimed directly at the IHS, and the CPU still throttled immediately. th-cam.com/video/185VQ_d2bYU/w-d-xo.html
as an hvac tech thats not at all what an accumulator does at all it is basically extra storage for refrigerant because your evaporator and condensing coils are too small to store all the required refrigerant. because this thing can pull down so fast and get so cold you need alot of refrigerant running though the system because as refrigerant becomes saturated with a load either heating or cooling it loses its efficiency so one way to circumvent that problem is simply adding more refrigerant to the loop so that it keeps the evap coils cool/hot depending on the mode you are running it in but to do that you need a place to store the refrigerant which is what the accumulator is for. the thing that dries off the refrigerant is called a filter dryer but it normally just dries it once as fresh new refrigerent is being enterted into the system becuase there is a chance that air was mixed with the new refrigerant as you put it in via your hoses.
I remember you guys got a hold of the Dell Rugged Extreme G2. Would that laptop handle the temperatures? I think it’s not the temperature that would kill it but the rapid changes.
I like how there's a $2000 3080 ti in that build redux ad. Of course they have competitive prices with that. (I know it's probably just an outdated slide from the gpu shortage, but it's still funny to me)
I'm pretty confident that you just filled the cooler with water. I have never seen parts store automotive anti freeze that wasn't colored. I have seen anti freeze bottles filled with water because it's the closest container when you need to dillute fresh anti freeze or top off your radiator.
11:32 this is one of those times with a shity camerawork adds to the experience... Cuz you KNOW it's not the fault of the person holding the camera, it's just that literally no one thought this through and they're just rolling with it (no pun intended)
I skipped over this video when it first came out based on the thumbnail and title. I checked it out after watching the overvolted 4090 video from today. I think if the title and thumbnail for this video better represented the lab and the temperature controlled chamber, I might have not skipped it over the first time. This video was excellent. Entertaining, I learned something new, and was on the edge of my seat for a moment.
I live in a country where 40+ degrees celcius is commonplace in the summer and was literally thinking today if you were going to someday go into what kind of cooling someone living in such temps would ideally need, and someone living in say, Alaska. Beat me to it!
If you are a normal human being, you use the same air or liquid cooling solutions as everybody does. It does not matter what temperatures you get outside. I presume you keep your PC in a room, not in the scorching sun or in the snow, when it’s freezing outside. In every normal household, the ambient temperature (where one would live and also keep a PC) is indpendent from the outside weather. No matter it’s +50C or negative 40C. If you don’t have that under control, owning a PC and being worried about it’s performance in regards to the ambient temperature should be the least of your worries.
@@ron200088 "the ambient temperature (where one would live and also keep a PC) is indpendent from the outside weather." The hell it is, if you don't have something like a heater/AC, that doesn't even make sense
@@leonardo25gabriel As I said, a normal, reasonable human beings would first create proper living conditions for themselves. I doubt that many keep a computer somewhere else than where they live. As I said, if you don’t have heating and/or AC you definitely have other things to worry about, than a PC.
Actually Linus, heating up the chamber is not going to do much for dehumidifying it unless you are trying to evaporate liquid water. You need a dryer of some sort or just let the humidity get attached into the walls. At -40 there is no much humidity that the air can carry anyway provided you don’t open the door too much.
😂 they clearly doesn't know what they are doing... As soon as the door opens, all the cold "dry" air just pouring out and warm air will get sucked in... This is one reason why those supermarket refrigerators with doors on top are very efficient, they can even have no doors on top if they have an air curtain.
The point of heating the chamber wouldn't have been to dehumidify it. It would have been to prevent the computer from shorting out due to the warm, humid outside air condensing on the cold computer parts.
My brother was on IT in military while he was doing his mandatory service. Apparently, they were so lax about things that they had their computers inside a very heavily cooled room that people that wanted to use them had to wear winter clothing, even during summer. So, I was always wondering what kind of effects this practice had to those computers and this video, somewhat, solved my questions.
I think someone re-used that bottle of antifreeze. That was just water Actual antifreeze is strongly dyed to differentiate it, and will also be notably more viscous when pouring. If you get any on your hands, it has a greasy feel to it. I'd love to see how that machine does with actual antifreeze in the loop
Also just lick it and if it's sweet then go have a shot of vodka. Also don't do taste it. For serious attempts though you definitely want to actually properly mix the ratios of antifreeze. I suggest propylene glycol as it's less toxic and can give you a little more water for the same freezing point of the mixture compared to ethylene glycol, though it is more viscous. You'll probably want the fairly standard 40% propylene glycol mixture that's good down to -45 C although you could sacrifice some cooling performance by even going all the way up to the eutectic composition at around 60% glycol and drop that down around -60 C.
@@Lonewolf_121 Pure ethylene glycol is more like a hot syrup than water. The bottle they had in particular was concentrate, not pre-mix, so I would have expected it to be both notably viscous, and present a refraction differential when poured into the water of the loop. I can't remember the term... But the swirlies when you mix 2 liquids together The freeze-behavior of the loop also brings doubt into my mind that it was antifreeze. Especially because it was an opened bottle
The accumulator on the compressor is because the system is way over size or the area you’re doing, and there’s a chance that some of the refrigerant will get back to the compressor as a liquid instead of as a gas. Accumulator is there is a buffer tank basically
Heating the chamber will not remove humidity at all. "Relative humidity" is based on the air's ability to carry water. As you heat the air, it can hold more water, so the % goes down, but that's just because you've increased the capacity, not because you've removed any water. The reason condensation occurs, is because you cool the air, and its capacity reduces far enough that it can't hold the water it already contains. The way active dehumidification works is to cool a small area below the dew point, so the water condenses and is drained away. Likewise, if you're using air driers on an air line to dehumidify, it only works because you're pushing wet air out and replacing it with drier air. It would be a lot more effective to draw air from the chamber, run it slowly through a small dehumidifier chamber with a coil held just above freezing, and a drain hole for the condensate, and then back into the chamber. The air will be cooled in the process, and the heating will take longer, but the air can be brought down almost completely dry,. Pumping air once through desiccant isn't likely to get it very dry, and you're going to be mixing it with the wet air, too. I've never messed with desiccant for drying air (except for small amounts for an airbrush), but I doubt very much it's going to accomplish much.
And I need to emphasize that you need to keep the airspeed low, so that the air gets cooled below its dew point where it touches the coils. if you push too much air through, you just lower its temperature without condensing the water. I often see the opposite problem taking care of datacentre air conditioning. HVAC people typically adjust things to a specific temperature drop at the outlet, as you might want in an office space, and this means the coils are a lot colder, and a lot more of the energy goes into condensation (HVAC people talk about latent vs sensible cooling) If I run a lot more air across the coils, it won't come out as cold, but there'll actually be more heat taken out of the room (think cooling 2000 litres of room air 5c, vs 1000 litres 10c) And less energy goes into condensing water, and into re-humidifying air that's been dried too much when it was cooled.
I used to be a tech at a computer shop in Anchorage in the mid-late 90s. Had a customer bring in a PC in the back of his truck, wrapped in plastic in mid winter (-40 F) from the mat-su valley (about a 2+ hr drive). Told him I would have to let it warm up to room temp gradually and would test it in about an hour or so, he wanted nothing to do with waiting and proceeded to hook it up and boot it in the waiting area. As you can guess, the frozen chip instantly heating up with condensation all over it blew with an audible POP! Customer was LIVID! had to have security remove him and legal ate his lunch weeks later LOL!
Yep, warm it up first and avoid breaking it however I like watching other people break their stuff. Getting cameras in there would be super cool and another challenge to conquer.
Now we need this, but have the liquid cooling tubes come out of the machine, and hook it up to the industrial chiller, that way, you have both ambient temperatures super low, and the liquid cool loop even colder.
the fun thing about -40° is that you don’t even have to specify F or C
-40 Kelvin? haha just kidding, nothing's that cold
i know right? kinda surprised they didnt mention that.
turns out they do at 5:35
And -40 Kelvin is not possible.
(nothing can go below 0 Kelvin)
@@manoz6194 negative Kelvin doesn't exist since 0k is the absolute zero
I know you were joking
@@shre6619 as far as we know/can prove today
-a scientist
Linus moving super quickly with the door open so that he can physically beat *check notes* MOLECULAR PHASE CHANGE is hilarious
I agree
That was like one of those ER drama shows with all the shouting.
Why didnt they just take the pc out and do the surgery there 😭😭😭
Now we just need to get you one to test the beans for the roadside boys.
@@sjorsangevare Because the PC would just get covered in water that way.
Alex is by far my favorite staff at LMG. This man can make anything amazing with a dremel and a dream.
His soldering is questionable but he is a good human
Just the right amount of knowledgeable jank
I also love when Linus asks if this is a good idea & he just does that look to camera like "its your money, whatevs lol"
Yes, but can he drop stuff?
He definitely gives the same vibes of a computer version of a Top Gear host.
Linus: *tries to keep the door opening small so the moist air won't condense on the testbench*
Also Linus: *BREATHES DIRECTLY ONTO THE TESTBENCH INSIDE THE CHAMBER*
Linus is a little dum dum /j
he is so wise in the ways of science!
My first thought for that problem was that they so easily could've just draped something (surely they have loads of moving blankets at LTT) over the door for Linus to duck under and mostly avoid all that drama.
Wouldn't be a proper Linus/Alex duo video without a moment of total chaos though, gotta admit it makes for a better vid to just wing it
@@mercklemore no it does not, it just wasted everyones' time
@@Piipolinoo oh my bad, guess I thought people are here for shenanigans. I must have missed LTT becoming a concise, no-nonsense consumer advice channel
As someone who lives in a tropical climate (the Yucatan peninsula in Mexico), I’m happy to see you guys test this, our electronics feels disposable and our phones are always hot. If I you want to play video games you need an AC for optimal performance. As an international viewer I’m excited to see what you will test next. I’m considering paying for floatplane.
Oh it gets hot in Northern Mexico too, although dry, yet 40° C ~ won’t stop me from playing airsoft 💪🏽
I used to live with summers like that. We'd move all our gear into basement where it was at least below 25
> If I you want to play video games you need an AC for optimal performance.
At such temperatures if you want to work you need AC for optimal performance.
I had for some time a heating problem in summer(last floor with black roof did make +40-45C(with 70-90% humidity) in the room and it was a nightmare) and with AC i have comfortable +20C in the room all the time.
@@deathmaster6117 well he did adress that in his comment
You lucky warm bastards 😉
That glycol looked pre mixed. Since their was water in the loop you probably got no better than a 25/75 mixture. Their are charts for antifreeze to water mixture to temperature tolerances easily available online so you can properly mix the antifreeze for ur car. Just use antifreeze which is designed to known standards and specifications and will produce repeatable results and mix as necessary to not freeze at -40. For -40 you require 52% antifreeze to 48% water mixture to prevent freezing. Premix the solution, completely drain the loop and fill with solution.
Just buy 100% Ethylene glycol and mix it yourself. But I agree, just going from the viscosity of the antifreeze when he was pouring it, it was definitely not just ethylene glycol.
Every time Alex is involved you know chaos is right around the corner
He always makes it fun 😂
@tapp-on-my.pic54byei
just the way we like it
His definition of "fine" is the same as my wife's
And with Jake laughing quietly in the background, you know it's going to stay that way lol
I lost it at the 'hot swapping' parts while trying to beat condensation. This was a wild ride. Love it!
So rushed 😅
It was stupid
Cold swapping 😎
I will see myself out...
yeah so cool using high end hardware to show us how much better and how rich they are
because lower end parts wouldnt have worked exactly the same
You're likely thermal shocking hardware if you're pulling from ambient to -40 and back again (~60c delta!). Ideally you should bring the chamber back up to ~50c to do a drying cycle before you come back down to ambient to negate the need to leave the system to dry out for a few days. There's a few fluids that you can use for "water" cooling that wont freeze at low temp (Engineered fluids EC110 is a good one - also dielectric).
Source: I design military servers that operate in -40 to +55c environments and make a lot of use of a very similar climate chamber!
- leave at 50°C for a while
- put in 10$ worth of dessicant, just a couple kgs of zeolithes from an aquarium petshop or sth
- put in a 5$ humidity meter
- cool down appropriately slowly while keeping an eye on the humidity meter
I love how there were no conclusions. Just went with it, ended it. Just because you know we only wanted for this to happen, not how or why. I love these cooling videos
Yup, i was going to comment "great!, i learnt absolutely nothing" but i thought i was too harsh, you way of saying it is more correct :)
I get that, but I would've loved for the video to go on for another 10 min and become gradually more conclusion- and explanation-focused.
I think they just wanted to display their ability to force computers to crash due to temps.
@@BlissToby when they fix it with a compressor they will 100% do that... They just can't give conclusions with non controlled tests which i can understand
More content like this is important to me. What good are the labs, if we can't see these things in action? These die hard extreme tests are great fun to watch!
@@elbeetlebeasto impatient? They've been setting up a new building for well over a year now and are most likely running behind schedule. They have every right to rush out videos for the sake of content. I'm sorry but would you rather wait another year for a video like this to come out?
Also keep in mind, what they do on camera is for entertainment purposes. Prototypes are funny to watch, any content creator with sense knows that a rushed prototype is funnier to watch then a fully fledged working project. In fact, they'll most likely have a follow up video where things go right.
Basically what I'm saying is they make money off of TH-cam videos and a tight ship that's running as intended just isn't funny to watch. They have to have both the more informative videos and more entertaining ones. Also I'd recommend you check out the livestreams, specifically the WAN show which talks a bit more seriously about tech news and such if this kind of content "enrages" you so much.
@@elbeetlebeasto Linus did explain exactly what they were doing wrong about temp management of the chamber, which means that this is just a for fun side project, rather than a serious take on the subject. It also means that they understand how they should do things but deliberately decide to ignore. As the other fellow said, they probably will have a follow-up where they make proper use of the equipment, and probably a third video once the compressor they're missing arrives to properly test humidity as well
@@elbeetlebeasto you're forgetting that labs will have its own website where the experts they hire will do proper testing. This is just a fun infotainment piece about the capabilities that they have now. So yes this is rushed and kinda bullshit but this was never supposed to be an in depth informative piece
no you are important content
Remember that the Lord Jesus Christ died on a cross for you because He loves you so much. He then rose up from the dead three days later
The Ten Commandments are called the moral law, (most of us are lying thieving blasphemous adulterer at heart and deserve hell) you and I broke the law, Jesus paid the fine. That’s what happened on that cross.
By believing that Jesus died on the cross and rose up from the dead 3 days later and not just confessing your sin, but also repenting of all sin you have done and putting all your trust in Him in prayer, He will grant you everlasting life as a free Gift.
I’d love to see passive cooling solutions for those temperatures
Now that you mention it, i wonder if they still have access to, or have, the computer case that was effectively a giant passive cooler that they did a video on.
It probably wouldn't work either
Just like on the NH-D15, the heatpipes will freeze and that'd be it
That's my guess at least
@@agalerex I remember seeing 12" wide enormous copper headsinks for desktop processors like 10 years ago, they were for applications that needed to be silent like acoustics labs, I remember you could run an i7 with one, just not at super high power levels
Why not liquid cooling :p
@@anthonyjames7173 I wonder why 🤔
man really took pc cooling to a whole new level
To be fair, this is why so many data centers are in Norway
Have you seen how he plans to heat his pool 😂
@@bdbdkdfotbrveiw it’s ridiculous 😂
ong
more like new degree ... right ?
"Have we crossed the line from fun to stupid?" - I find that any time Alex is involved, the answer to this question is almost always yes.
The look on his face tells you all you need to know.
also Alex "It will be fine."
what line?
Linus shouldn't even be asking that. For a long time already, the answer has been "F yeah"
We think alike
It’d be great to have a website with all the tests info in one place, with the ability to search through those data sheets. I’d bookmark it instantly.
Great job on the editing - the emergency cooler swap had about the same feeling as a cardiac arrest scene in a medical drama ^^
that was the best part
Greys Anatomy style. All it needed was the beeps in the background lol.
Bro! I was thinking the same thing.
Or a Pit Stop
50 degrees Celsius stress test must be done by computer part manufacturers. Especially if they want to sell parts to Australia, India and other hot places.
Tropical PC bros need to use a room with AC just to not make their PC into an oven here
@@bimbim1885 and people say our middle european winter is depressing... :D
they usualy run at up to 60c and just get bigger cooled and run at slower speed and a lot of strong fans
@@bimbim1885 My poor laptop in summer needs cold air directly from the AC when I game
I agree. My Asus laptop died while playing at 43°c, but here in summer the temperature can go up to 50°c with really high humidity.
HP laptops here are famous for failing at temperatures over 35.
You can tell they have never used and enviro chamber before. You have to run a normalization cycle before you open the door! Bring the components back to the air temp of the lab before you open the door. It only adds a few minutes to the testing and you don't have to worry about condensation. Glad they figured it out at the end. Also, you can mix glycol 50/50 with water and it should still work fine for those testing temps.
Keep at it guys! It's possible. I've run my PC around -25°C. Left it in my semi during winter and wanted to see what it would do in Cinebench. 280mm corsair AIO, 5900x, strix x-570I. Can't remember the score but it's possible. Canadian winter overclocking is fun, free sub zero cooling lol.
People out here using liquid nitrogen or helium while they can just move to Canada
@@stylinsandwich Move to Canada no cpu cooler needed
Something you must remember your pc case (The Chamber) is actively cooling or heating the system inside to hold that 50c set point, we will not see a thermal run away because of that.
The GPU running at 90% fan and 80c is getting cooled by the chamber’s 50c air.
Yeah, you would have to keep the chamber running for a while and then keep the same power settings that it settles into with the pc inside.
Never a dull moment with Alex and Linus!
A lot of components will get extremely angry once they go below -10-0C. Primarily PSUs and SSD HDDs can act up. There's also sometimes a "non-op" and "operating" difference. For example if you bring a PC that is running down to -40C, it usually is easier than doing a "cold soak" at -40C with it turned off and then attempting to boot. Source: have done a crap load of testing MIL-STD 810 at an aerospace company
Makes sense, I remember there being a similar issue with liquid nitrogen cooling they did in a video a while back, they had to get a load on the CPU before starting to really cool it or it'd go too cold and just stop working, I'd imagine similar issues in a lot of components in the PC, they just wouldn't have worried about it, cause... well... They weren't submerging the PC in the stuff lol.
Angry is the best, non-technical way to describe it. ^ ^
Love from India ❤
During summers, the room temperature varies from 30°C in the night to about 35°C-36°C in the noon in India. Currently it's 4 in the morning and the ambient room temperature is 32°C. Testing at such temperatures is crucial for us who have tropical summers and kudos to the team for doing this. Thank you.
You woke at 4am to end up watching LTT lmao?
You must be living way up North. In Delhi NCR region it has already started to go above 36° C during noons
You located in Northern part of India?
@@iknowiminsane1949 must have woken up to prepare for an exam or something and ended up watching LTT
@@soumilswapna even in the south(tamilnadu) it is around 40c
I am living in the Philippines. The highest room temp I got was 35c during "summer" march to may. GoW and Witcher 3 was about 73c on the cpu and gpu around 76c. Both did not reach 80c. But during cold months around nov-feb. Temps would go below 70 during the day and almost 60 during nights
Linus and Alex testing tech on the way to the North Pole would be a worthy sequel to the Top Gear Polar Special.
(this video suggestion is on me today Linus)
that'd be hilarious
Up you go sir
Watching linus change a cooler before condensation kills the computer is like watching emergency open heart surgery lol
Condensation should be only a problem if your PC is cooler then the ambient air and not if it is warmer.
Reason being is the air has a humidity limit based on how warm it is. So the warmer it is the more water it can hold.
If you have that warm air that has humidity come in contact with something much cooler the air will cool in response and if it cools enough to drop below the humidity limit of that air at that temperature then it will release water (Form Condensation) in response.
If you take something warm and put it in a cold environment or warm something up in a cold environment that warming of the air makes it able to take on more water and instead will encourage the opposite of Condensation which is drying.
In otherwords you don't need to worry about your PC forming condensation in extreme colds.
However in this specific circumstance if you do freeze the air rapidly without releasing the moisture first you will get frost on everything in the testing chamber.
However the warmer components in the PC will be the slowest place that frost forms.
In summery though:
-Warm air + Cold Stuff = Condensation.
-Cold air + Warm Stuff = Drying or in least no condensation.
if you have a contained volume of air and lower it past its dew point, it will condense. not directly on the parts, as you would get with cold equipment, but you'll get fog, and likely dew on everything. (evenly distributed condensation)
if you were feeding it a flow of already cold air, any condensation would have already happened in the cooling process, so no further condensation would occur.
Frost is simply condensation that freezes immediately.
@@nwimpney, yep is what I was saying that if it is trapped air cooling it will cause an even distribution of Condensation. But the warmer part of the computer will actually Condensat less technically.
I don't know if you have some laptops you were willing to destroy, but I feel like the cold test (even with moisture) could be a good one for laptops. It could kind of simulate taking your laptop from inside, transporting it in the cold, and then running it again. It's something I never really thought to worry about, but there were definitely times in college when I would walk to class in the cold and my laptop would be really cold and have some condensation on it. Now I'm wondering if that did any damage to it.
If you really want to know, you could see if your model of laptop has any MIL-STD-810 certification done on it. One of the parts of that spec has to do with thermal shock.
Thats actually really interesting i live in a tropical area but i would love to know if that will damage it long term
Ah my old laptop from college does show liquid damage signs and has for a while, despite having never spilled anything on it. I took that thing with me everywhere in my backpack though, through sun, wind, rain, dirt, and fog. Living in a relatively cool and humid forest area, I always assumed this was the case.
Why not just pop the laptop open and check for corrosion? That'd be a very easy thing to see and it doesn't take any real skill or anything.
Yeah I really want to see it with a ROG Strix laptop since they run very hot. I'd also like to see some more gradual cooling so they can find the sweet spot between perfect performance, dropping performance, and total failure.
Nitrogen cooled PC in -40C ambient temperature next? Maybe you can get some world records this way.
There was already that one Chinese streamer who put her pc in -50
How is your comment from 12min ago ? :o
@@champignon2916 yeah, didn't it just release...
@@champignon2916 yeah wtf
@@champignon2916 lmao
I remember trying to cool our server rack by pumping in outside winter air at around -10°C, but it seemed to warm up in the 5" pipe before getting there. Goal was to meet the old ATX recommendation of +40°C CPU temp max.
11:20 I know you mentioned the video was going to go off the rails but this is not what I expected. Crazy cooling projects with Alex never disappoint
Thank you for actually testing in the warm configuration. The thing about tech reviews is that here in India we do not get a clear picture of what performance we could expect at home in normal living conditions and whether a chip is efficient enough to not throttle down to half its actual performance. I trust your reviews to be more inclusive than anyone else's. Hoping you can do this for most of the relevant gadgets ❤
I wonder if this is why I see so much more love for ROG computers on forums than I see from reviewers?
We need noise cancelling headphones just to escape the fan noise watching a 4k video 😂
@@Saitama07 LMAO sounds like me whenever I play at max settings
Philippines here: A good case and a mix of Arctic F12 and P12 max fans should do it. I had overheating (above 80-85) or loud PC before too. Then I swapped all 5 case fans to Arctic F12 and P12 and since then I can sustain 70 degree on full Prime95 and 3D Furmark stress test on CPU and GPU. I was totally amazed how much a good airflow setup can make such a huge difference. Much more than in all the tests - since temps here are so much higher, than in the test setups of reviewer/youtuber!
Went from 2 cheap (=shit, since Arctic is cheap too but sooo much better!) in front and 1 cheap upper rear and good but too loud rear out, to 3 F12 front in, 1 P12 rear top in (directly onto CPU cooler) and one P12 rear out. around 12-15 degree difference and almost complete silent, unless I am gaming.
Of course water cooling would be ideal. But expensive here.
It was weird though. Idk about India, but here in Argentina we have a ton of humidity AND we do reach sometimes 40°C (or more) in certain places.
My CPU at idle has been 55 or 60 tops, never reaching 75 or more doing nothing. Perhaps it's because mine is a ryzen 2600 and this was a high end Intel. Still, my case sucks ass (it's a generic badly designed one) and I have the stock cooler.
the camera work on this seemed more cinematic with slow focus pulls and panning shots, I liked it
Cant believe you went to North Pole just for the intro, impressive Canadian resistance against cold, spectacular dedication
Shit comment, not funny like bait.
9:15 I love these moments when I get to see things I deal with on a daily basis in Linus's cooling videos.
Every video I say to myself this must be the craziest cooling idea possible. Then they outdo themselves. Inspiring honestly.
The suction line accumulator in your system linus is used as an expansion chamber in low temp systems where you run the chance of actually getting liquid refrigerant coming back from the evaporator. It uses the high volume of the accumulator to flash any liquid directly to a gas before it gets to the compressor
While you wait for an electric dehumidifier, you could use a kilo or 2 of heated Silica gel put inside the enclosure that is at room temperature. It will suck up the water before you freeze everything. It is a “desiccant,”
They’ve literally got a dehumidifier right there. The refrigeration system is exactly what a dehumidifier is. It naturally pulls moisture out of the air. They just don’t know what they’re talking about.
At these temps you might actually need to consider contraction and expansion of materials. Maybe thats why the temps were so high for the direct die cooled pc, because of the cooler not making good contact anymore?
If they cooled everything down and then applied the cooler, I bet they would get waaay better temps. (also probably not using frozen liquid would help)
My room got to 40 degrees (Celcius) this summer. So it's really nice to know if the type of thermal paste makes a difference.
It does not make any sort of difference in regards to high temperatures. Which is only natural. Normal thermal pastes are only affected by very low negative temperatures, when they freeze. I doubt that’s the case. Anyway, there are special thermal pastes, which can whitstand extremely low temperatures when LN2 overclocking.
@@ron200088 Oh that's is very helpful thank you. I was worried about playing games on max during the summer, I worried that my thermal paste would dry out quickly or something like that
get air conditioner
@@Skystrike70not everyone can just do that
@@antonhelsgaun fr, in the US its normal to have AC, but in other countries not so much.
That refrigeration setup is called cascade cooling, I service these chambers for an aerospace company in Florida. It's only 1 stage but it's two circuits, circuit 1 evaporator cools circuit 2 condenser, circuit 2 evaporator cools the chamber.
I have always wanted this video. Seeing a PC operating in a really cold room or a really hot one. It surprises me it took this long for you guys to do it. Great video and thanks for making it! ❤
I think you have the humidity and temp thing backwards. The total water the air can hold goes up with the temp, so by heating it first, you likely had more total water even though it was low relative humidity, since that's relative to max it can hold.
That's what I came here to say too. Heating the chamber increases the capacity of the air to hold moisture, so the RH% goes down, but the absolute amount of moisture in the air remains the same (unless you introduced more when you opened it).
This is why humidity is an issue and your AC removes a ton of it in the summer when it's running (because lowering the temperature raises the humidity), and also why indoor air gets so dry in cold climates in the winter (heating the indoor air doesn't remove moisture, but warmer air holds more moisture, so RH% goes down).
Yeah.. wondered that too
I literally screamed this at the screen lol. Glad I'm not the only one
This is why you should probably use dew point instead of RH for this stuff IMO
One of them few times I've tried to visualize a psychrometric chart in my head...
When heated, relative humidity would lower, but the absolutely humidity/humidity ratio stays the same assuming the test chamber is a closed volume with only thermal energy crossing the boundary.
If anything, precooling as low as possible and literally toweling up the condensate, then raising to a desired cold temperature could work.
What a dream job to test hardware and software, good job bringing more of it to the world Linus!
You were really close to the job of the accumulator. Instead of moisture in the system that would turn to acid, the accumulator stores and traps liquid refrigerant that hasn't turned to a low pressure gas before entering the compressor turning it into a high pressure gas. Getting liquid in your compressor is the number 1 reason a compressor fails, so this is a way to prevent that especially in bigger systems. But it is cool lately you've been kinda doing some research on how air conditioning works on some of your chillers. Keep up the great work.
I'm a hvac guy and I was looking for this comment! I was yelling accumulator before they said it hahaha
Fun Fact Alex actually wanted to go to the North Pole to do the test, but Linus said no because they have to makes use of the money they spent.
destroying a PC is still cheaper than flying to the north pole
Where did you hear this? Was it said in a floatplane exclusive, WAN or somewhere else?
They don't even need to go there. -40 is pretty normal in a lot of places in Canada during the winter.
yeah, I mean they're in Canada north pole is just right to their north
Probably Linus: "Alex, we dont need the North Pole, we have the North Pole at home!"
you all need to run that test with a non heat pipe air cooler, something like the slightly more beefy intel stock coolers the ones that are triple height and have a big slug of copper in the middle, that way there are no heat pipes to freeze over, and no water or pump to seize up with the cold either, that would be the best bet for running at -40 with it running without issues.
pretty sure i saw one of those intel coolers I'm talking about on a video you previously did, might have been comparing intel coolers over the years or something.
I know the coolers I'm talking about were meant for workstations, I got one years ago bundled with my i7 970 and it actually performed well enough that the CPU was usable in Aussie summers, though was noisy and I ended up upgrading it pretty quick
When I would go to the desert it would often get into the 115 to 120s so I think it's awesome to see stuff like this. I've had more than one phone stop entirely cause of the heat
120? How did you not boil alive?
Imagine walking around trying to find an Eevee in the desert, it's 115 degrees, and your phone just tells you "nope"
This is not interesting. My phone dies due to heat in southeast qld aus all the time.
I wonder if it would of worked if the CPU cooler was swapped with an aluminum or copper heatsink with no pipes, and the paste swapped with a graphite pad.
but it wouldn't be practical enough for common people
Sometimes the gas for the heater would empty before it could be refilled again. Playing while there was no heat on my PC was actually nice. The ambient low temperature prevented my PC from overheating. It was an old PC but it was my first PC. I would play games outside it's specs on low graphics with low framerates. The heat generated by my PC could warm up my hands when they got clammy. I swore it also performed better. It regularly reached 40-57 on hot days even in Canada. Like you guys said poor ventilation in both the room and case would cause the internals to heat up to a surprisingly high temperature. One day it was shutting down whenever it would reach a certain temperature. This corrupted the BIOS. The poor thermal paste hadn't been changed in over 10 years. The PC was originally my Dads. When I opened it up to replenish the paste what was left was pretty much stained to the top of the CPU chip and was basically non-existent. Unfortunately I had shorted my CPU when I applied some new thermal paste. I even bought and installed a new thermal heatsink and fan for the CPU. I never got to finish fixing it. I had left it behind when I moved.
We've had Dell reach out to one of our customers and tell them if we didnt raise them temps in the data center they wouldn't honor RMS's because it was too cold. Definitely surprised when we heard that one.
Apparently a streamer actually did this in Alaska. And they had the same issue with the AIO so they switched to air cooling which worked and they clocked an Intel cpu at like 6.3 ghz or something
Lol
More recently, I had to do some maintenance on one of my older computers, a Gateway E-2000 from 2004. The computer was running Windows 2000 and trying to play Bejeweled 2, which caused a stop error (blue screen of death). It was rather late at night, but I still tried to do some "repair" to the system. As the stop error was caused by a page fault in a non-paged area (or most likely, the PC trying to play the game, which was having some difficulty), I was still able to restart the computer in normal mode and I backed it up, and tried to do an antivirus scan and disk defragment.
Neither of the last two actions got completed, but why I bring this up is that I noticed the system was getting rather warm, and since I didn't know if I was going to just let it run while I was asleep, I didn't want it to overheat (as this was my first time doing this kinds of stuff), so I placed my powerful floor fan nearby the computer to try and cool it off. Wasn't the most effective tactic, but I thought I was at least being helpful.
Also, if you decided to read this, I converted the units to Fahrenheit for everyone who uses those units. The coldest temperature is -40°F, and the hottest temperature is 356°F.
17:20 WTF was that cliffhanger. One of the best videos ever. Love your content guys also great camera work dealings with Linus and editing to increase the shock value.
heat isn't how you remove moisture (water). heat evaporates moisture which increases humidity which condensates back to moisture when the temp drops. you need to chill everything to single digits C or upper 30s F so it does condensate and can be wiped out or ideally drained at the evaporator. just like the water that drips from a car in the summer because the cold air conditioner condensates the humid air and it drains off.
I was in doubt about which 4090 model to buy and so I chose the Rog Strix after this video. Where I live not in summer we have temperatures like 40°c and this test was really insane.
Alex and Linus janky cooling projects are my favorite type of LTT videos.
I just watched Alex do the tea cooled build yesterday 😅
At least you’re getting a lot of mileage out of that temperature chamber. For a second I thought you actually sent a computer up to the North Pole with long-distance remote tracking.
L pfp
A little advice from someone who uses environmental chambers daily. Shut it off before you open the door. The system is working overtime trying to correct for the sudden change in conditions. That will shorten the life of your chamber and also shorten the time to your first repair. Also, glycol does not work for crap at 100% concentration. The "antifreeze" you buy at the auto parts store is made to be mixed 50/50 with water. What you should do is stop cheaping out and just go buy a bottle of pre-mixed antifreeze. Also, don't get the "low tox" stuff if you are going to very cold temperatures. Finally, what was coming out of that bottle looked in no way like glycol based anti-freeze. Because you went cheap I'd bet you grabbed someone's half filled bottle - probably just water they had in their trunk for a roadside emergency. Research some of this stuff before you try to make a half-baked video and destroy a bunch of equipment.
Them having fun heating up the PC is just my annual summer struggle
I have 2 OCs saved in Ryzen Master. One for Winter which is basically the max it can do at 5Ghz and one for summer where I stress tested during 30C @ 4,825 Ghz and 1.3125 Volts. It can reach 35C here sometimes but those days are few and far between.
You know, I used to use environmental chambers, and we had secondary doors with holes to reach in while it's running
no i didn’t know
I walked into that one…
I just realised: Could those giant wall-mounted gloves work, or would the temperatures kill them?
...then again, that leaves fine control issues.
I'd like to mention heating a room up does not remove water from the air necessarily, relative humidity will lower but that's because warmer air can hold more water. As the temperature drops the air can hold less water making the relative humidity go up. So you can have the same amount of moisture in the air at two different humidities depending on the temperature.
Yeah that was absolutely pointless lol. Also an object sitting in a refrigerated room/chamber isn’t going to condensate when it’s not the evaporator itself lol.
If a video has Alex in it, you know it’s gonna be a good video 😂
Except this wasn't a good video
Nah usually sketch
While you were there, you should have tried removing the cooler altogether! Wouldn’t all that moving cold air in the chamber cool the cpu down enough? I would like to see a test with just a bare cpu in -40°
Great video! I’ll be more cautious about eagerly installing upgrades that have been sitting out on my front porch in the winter!
** I understand that this experiment was about testing a gaming pc in extreme cold temperatures and that a cpu cooler is essential for any pc**
It wouldn't be enough on its own to keep the chip cool. One of the reasons why CPU coolers have to be tightened is because the additional pressure added by the mounting mechanism will better conduct heat away from the IHS (metals under stress will transfer excited electrons more freely). LTT had another video a few months ago where they used a vortex cooler aimed directly at the IHS, and the CPU still throttled immediately. th-cam.com/video/185VQ_d2bYU/w-d-xo.html
Not an open cpu, but a large metal fin heat sync, with a good fan. That would be interesting to see.
The lab is already so sick!! I can't wait for it to be fully functional... 🤯
as an hvac tech thats not at all what an accumulator does at all it is basically extra storage for refrigerant because your evaporator and condensing coils are too small to store all the required refrigerant. because this thing can pull down so fast and get so cold you need alot of refrigerant running though the system because as refrigerant becomes saturated with a load either heating or cooling it loses its efficiency so one way to circumvent that problem is simply adding more refrigerant to the loop so that it keeps the evap coils cool/hot depending on the mode you are running it in but to do that you need a place to store the refrigerant which is what the accumulator is for. the thing that dries off the refrigerant is called a filter dryer but it normally just dries it once as fresh new refrigerent is being enterted into the system becuase there is a chance that air was mixed with the new refrigerant as you put it in via your hoses.
I remember you guys got a hold of the Dell Rugged Extreme G2. Would that laptop handle the temperatures? I think it’s not the temperature that would kill it but the rapid changes.
Haven't even watched it yet and I already know it's going to be a favourite because it has Linus, Alex and extreme cooling
I've been wondering about how computers handle 50C (122 F) because the temperatures where I live have started to approach that.
Linus never disappoints and will always be supporting you guys work. I can't wait to work in the ICT sector after university. 🙂
I like how there's a $2000 3080 ti in that build redux ad. Of course they have competitive prices with that.
(I know it's probably just an outdated slide from the gpu shortage, but it's still funny to me)
I'm pretty confident that you just filled the cooler with water. I have never seen parts store automotive anti freeze that wasn't colored. I have seen anti freeze bottles filled with water because it's the closest container when you need to dillute fresh anti freeze or top off your radiator.
Yeah, I thought the same.
hahahaha, this is fantastic! cant wait to see more shenanigans with Alex. it always seems like when Alex is in a video, its going to get crazy haha
Linus ,thank you very much for interesting videos you make with your crew . Especially big thanks converting temperatures to Celsius 👍
yes, since fahrenheit to me is pretty useless when it comes to stuff like this and international affairs
The Canadians™ seem to have a rather weird mix of units. Maybe they use C anyway?
@@5Andysalive the mix of C and F is actually crazy in Canada.
Can you do a video where you weld, using different compounds, a CPU to a cooler? I thin that would be interesting.
Glad to see fun videos like these. I may already know too much about building PCs but having these "what's gonna happen if.." is very cool!
11:32 this is one of those times with a shity camerawork adds to the experience... Cuz you KNOW it's not the fault of the person holding the camera, it's just that literally no one thought this through and they're just rolling with it (no pun intended)
I skipped over this video when it first came out based on the thumbnail and title. I checked it out after watching the overvolted 4090 video from today. I think if the title and thumbnail for this video better represented the lab and the temperature controlled chamber, I might have not skipped it over the first time. This video was excellent. Entertaining, I learned something new, and was on the edge of my seat for a moment.
I’m really excited for this extreme temperature testing chamber to see which lithium batteries will work best in both temperatures.
I live in a country where 40+ degrees celcius is commonplace in the summer and was literally thinking today if you were going to someday go into what kind of cooling someone living in such temps would ideally need, and someone living in say, Alaska. Beat me to it!
If you are a normal human being, you use the same air or liquid cooling solutions as everybody does. It does not matter what temperatures you get outside. I presume you keep your PC in a room, not in the scorching sun or in the snow, when it’s freezing outside. In every normal household, the ambient temperature (where one would live and also keep a PC) is indpendent from the outside weather. No matter it’s +50C or negative 40C. If you don’t have that under control, owning a PC and being worried about it’s performance in regards to the ambient temperature should be the least of your worries.
@@ron200088 "the ambient temperature (where one would live and also keep a PC) is indpendent from the outside weather."
The hell it is, if you don't have something like a heater/AC, that doesn't even make sense
@@leonardo25gabriel As I said, a normal, reasonable human beings would first create proper living conditions for themselves. I doubt that many keep a computer somewhere else than where they live. As I said, if you don’t have heating and/or AC you definitely have other things to worry about, than a PC.
11:00 This is legit the most intense moment in an LTT video. It's like an action suspense thriller.
Actually Linus, heating up the chamber is not going to do much for dehumidifying it unless you are trying to evaporate liquid water. You need a dryer of some sort or just let the humidity get attached into the walls. At -40 there is no much humidity that the air can carry anyway provided you don’t open the door too much.
😂 they clearly doesn't know what they are doing... As soon as the door opens, all the cold "dry" air just pouring out and warm air will get sucked in...
This is one reason why those supermarket refrigerators with doors on top are very efficient, they can even have no doors on top if they have an air curtain.
The point of heating the chamber wouldn't have been to dehumidify it. It would have been to prevent the computer from shorting out due to the warm, humid outside air condensing on the cold computer parts.
do you actually need active cooling in -40? I'd be interested to see no fans in the case performance
You probably do, because air doesn't conduct heat very well.
My brother was on IT in military while he was doing his mandatory service. Apparently, they were so lax about things that they had their computers inside a very heavily cooled room that people that wanted to use them had to wear winter clothing, even during summer. So, I was always wondering what kind of effects this practice had to those computers and this video, somewhat, solved my questions.
I think someone re-used that bottle of antifreeze.
That was just water
Actual antifreeze is strongly dyed to differentiate it, and will also be notably more viscous when pouring. If you get any on your hands, it has a greasy feel to it.
I'd love to see how that machine does with actual antifreeze in the loop
They specifically said glycol in the video, ie ethylene glycol, it's clear, and watery.
Also just lick it and if it's sweet then go have a shot of vodka. Also don't do taste it. For serious attempts though you definitely want to actually properly mix the ratios of antifreeze. I suggest propylene glycol as it's less toxic and can give you a little more water for the same freezing point of the mixture compared to ethylene glycol, though it is more viscous. You'll probably want the fairly standard 40% propylene glycol mixture that's good down to -45 C although you could sacrifice some cooling performance by even going all the way up to the eutectic composition at around 60% glycol and drop that down around -60 C.
@@Lonewolf_121 Pure ethylene glycol is more like a hot syrup than water. The bottle they had in particular was concentrate, not pre-mix, so I would have expected it to be both notably viscous, and present a refraction differential when poured into the water of the loop. I can't remember the term... But the swirlies when you mix 2 liquids together
The freeze-behavior of the loop also brings doubt into my mind that it was antifreeze. Especially because it was an opened bottle
The accumulator on the compressor is because the system is way over size or the area you’re doing, and there’s a chance that some of the refrigerant will get back to the compressor as a liquid instead of as a gas. Accumulator is there is a buffer tank basically
Heating the chamber will not remove humidity at all. "Relative humidity" is based on the air's ability to carry water. As you heat the air, it can hold more water, so the % goes down, but that's just because you've increased the capacity, not because you've removed any water.
The reason condensation occurs, is because you cool the air, and its capacity reduces far enough that it can't hold the water it already contains.
The way active dehumidification works is to cool a small area below the dew point, so the water condenses and is drained away.
Likewise, if you're using air driers on an air line to dehumidify, it only works because you're pushing wet air out and replacing it with drier air.
It would be a lot more effective to draw air from the chamber, run it slowly through a small dehumidifier chamber with a coil held just above freezing, and a drain hole for the condensate, and then back into the chamber.
The air will be cooled in the process, and the heating will take longer, but the air can be brought down almost completely dry,.
Pumping air once through desiccant isn't likely to get it very dry, and you're going to be mixing it with the wet air, too. I've never messed with desiccant for drying air (except for small amounts for an airbrush), but I doubt very much it's going to accomplish much.
And I need to emphasize that you need to keep the airspeed low, so that the air gets cooled below its dew point where it touches the coils. if you push too much air through, you just lower its temperature without condensing the water.
I often see the opposite problem taking care of datacentre air conditioning. HVAC people typically adjust things to a specific temperature drop at the outlet, as you might want in an office space, and this means the coils are a lot colder, and a lot more of the energy goes into condensation (HVAC people talk about latent vs sensible cooling)
If I run a lot more air across the coils, it won't come out as cold, but there'll actually be more heat taken out of the room (think cooling 2000 litres of room air 5c, vs 1000 litres 10c)
And less energy goes into condensing water, and into re-humidifying air that's been dried too much when it was cooled.
I am loving these highly technical, informative videos with controlled enviroments. Great job!
Props to Andy getting us the best shot possible whatever is happening.
I used to be a tech at a computer shop in Anchorage in the mid-late 90s. Had a customer bring in a PC in the back of his truck, wrapped in plastic in mid winter (-40 F) from the mat-su valley (about a 2+ hr drive). Told him I would have to let it warm up to room temp gradually and would test it in about an hour or so, he wanted nothing to do with waiting and proceeded to hook it up and boot it in the waiting area.
As you can guess, the frozen chip instantly heating up with condensation all over it blew with an audible POP! Customer was LIVID! had to have security remove him and legal ate his lunch weeks later LOL!
You know its a good day when linus puts a pc in -40C and overclocks it
@Carl Gunderson oh yeah, eh
Jake sitting behind the camera laughing was the best part
Even in Portugal, we have quite a few days over 40ºC during the summer, so that's good to know.
Could you passively cool a PC with no fans at Commonwealth Bay Antarctica? Where the wind is regularly recorded at 150mph
Just when I think you guys know what you're doing... you reset my expectations.
Yep, warm it up first and avoid breaking it however I like watching other people break their stuff. Getting cameras in there would be super cool and another challenge to conquer.
2:56 oh linus, you poor guy, watch odisha's Titlagarh instead, you will find regular summer is 50°C 🥴😵💫😵💫😵
I love that the most concerning thing at LTT is Linus yelling "everything is fine!"
Now we need this, but have the liquid cooling tubes come out of the machine, and hook it up to the industrial chiller, that way, you have both ambient temperatures super low, and the liquid cool loop even colder.
Can't believe they're already testing stuff in the lab... crazy, man. 👍🏽