AIR COOLING A CPU WATER BLOCK

We have water cooled an air cooler but does it work the other way around. well i think you are going to like this one. Im taking my EK Supremacy evo water block and im going to hook it up to an air compressor and see what kinda CPU temps we can achieve.

[@taariqpetersen9883]

Major Hardware: Asking the questions you didn't know you have

[@gendragongfly]

You don't need the ice, let the air do the work. All you need is expansion. First the air is compressed in a compressor, that heats it up. Leave it in a tank for a while to cool it down. Then pump it into a waterblock (at high pressure, 70 psi or so should do it). Make sure that the waterblock has a tiny pinhole in the part you want to get cool. (If the hole is small enough you won't hear it hiss because its frequency is ultrasonic). The expansion of the air coming out of the hole will cool your CPU to sub ambient 😋

[@Klokopf52]

I have seen Air Power Tools freeze during hard use, so this could be enough indeed :)

[@snakeinabox7220]

The emoji makes me fell like you're a idiot

[@gendragongfly]

@@snakeinabox7220 "an" idiot 🤪

[@AntimatePcCustom]

thought the same. in worst case the 0-1 degree water could add heat to the air. rather than taking it out. haven't done the math. but indeed your observation is correct. but only IF the air is cooled before hitting the tank, like on a screw compressor on a workshop. don't think it's the case on an armature compressor. but hey. this isn't science. he is just playing around.

[@kai990]

Also add heatsinks to the compressor pressure vessel

[@DawidDoesTechStuff]

This is an awesome idea. That poor sacrificial PC has seen some shit.

[@MajorHardware]

when i bought it in 2011 it had no idea what it was in for

[@BTElectric]

Helium is more thermally conductive than air, maybe somehow try a closed loop of helium? Or maybe other gases?

[@arraybytes]

That is an interesting idea, perhaps just using a regular to pass helium from a tank across the cpu for a prof of consept.

[@Next-nb9pr]

It doesn't matter if you don't have enough compound to take the heat away,it's like stacking 5 another fans to gpu with shitty radiator it won't work

[@QuantumQuibbles]

Liquid helium is literally one of, if not THE coldest substance known/available to mankind, so I am positive this would work extremely well. However, it's not easy to obtain nor contain in it's liquid state.

[@magnusnilsson9792]

Isn't that what Linus always does, and thats the reason he talks lika a smurf? I bet he's got helium in his AC in his whole building.

[@bleepbloopblahp]

And then loop it back to the compressor... You are basically just building a phase change fridge then. Phase change cpu cooling is a thing.

[@Weichspueler]

Nobody: Absolutly nobody: Major Hardware: AIR COOLING A CPU WATER BLOCK

[@FrankTheTank107]

Sorry to be a party pooper, but there was sombody....a lot of sombodys in the comments from the last video.

[@B0BBYGAMER]

What would actually help is attaching a air cooler to a all metal waterblock not running air thru a waterblock

[@chincemagnet]

I’ve done this lol

[@arraybytes]

I would love to see how much pressure it could take, I would bet that on an open system like that 90-100 psi would not be out of the realm as the pressure would drop through the system by a great deal. I might have to attempt a test with my compressor as mine can do 90psi continues.

[@jakegarrett8109]

I’d bet 200 psi air at least (extreme low density fluid, the backpressure will be much lower than water, and backpressure is the main force the acrylic blocks will see), I’m not counting that wimpy pinhole nozzle, I’d say up to 400 psi for it due to its extremely low flow rate meaning the end of the hose before the block would equivalently be like 40 psi or rough napkin math Dnozzle/Dtube*Pressure) 150 PSI is what I use for cleaning a Titan Xp air cooler (90 psi for laptops) Water Blocks I’de use 150 psi for a quick cleanout.

[@pagaiba3]

The radiator is limiting the airflow. I would bet that if you completely omited that part and went straight to the water block with air, it would cool down a lot nicer.

[@CuttingEdgeRetro]

100% you need to add a blow off valve to that air loop hahaha. Also, missed opportunity to call your channel "Mad Science Hardware" of something ahah

[@diederickb1806]

indeed, the pressure in his block is probably waaay lower than the output pressure of the compressure because it can escape freely. Adding a blow off valve will increase pressure (read air density) and increase thermal capacity of the air in the block and thus probably increase cooling capacity if the compressor can keepup. Looking forward to part 2!

[@runninggames771]

@@diederickb1806 Doesn't compressing air heat it up? Maybe you could have a system where you decompress the air, then blow it over the water block.. Having an adjustable valve At the exit might work too, slight resistance could make it blow faster??? Idk. There's so many different things to try haha.

[@diederickb1806]

@@runninggames771 tru but the compressor will work as a big heatsing for most of that heat, the decompression that will happen in the 'cpuheatsing' could cause extra cooling i think. But keeping it compressed will probably give it better heat transfer capacity. So i would love to see which of these works best xD.

[@thekrautist]

This video brought to you by Home Depot's Homer's All-Purpose Bucket (5 Gallons).

[@Pandamonium004]

Here I was thinking it was going to be a water cooling setup with a fan blowing onto the waterblock... O how wrong I was lol :P great video, very interesting! :D

[@arrobamarcos]

You should definitively try to air cool your watercooled air cooler.

[@ianzhang8337]

You mean use watercooled air to cool his watercooled air cooler? Oh man my mind is looping.

[@Escalofrios29]

Instead of running water through try to run cold air

[@WahlVids]

Multistage cooling isn't a new thing.

[@beachuk2000]

what might work better is to attach the air line directly on input line and move the air valve to after the cpu block as in your setup you are not running the pressure you are saying as after the compressed air leaves the valve in your hand it's venting to atmosphere pressure and is mostly just flow. but if you had your valve after the cpu block you would have the full compressed air at that point untill it leaves the valve. more compressed air is more dense and might possibly cool better

[@Pheatrix]

True. And more compressed air is more likely to explode this setup. Which would make for an awesome video^^

[@MCMonsterbuilder]

Yes, and the way it's now more pressure means more flow thats why it got better with more pressure. You would need really high flow till the tubes would be restricting enough to actually build pressure in the system.

[@finraziel]

@@Pheatrix It would... as long as he would actually properly protect himself... I mean, yay for eye protection, but if that thing had blown up and shrapnel torn through his neck, those goggles wouldn't have done much good. Would need this setup to actually be behind some kind of shield (that you're sure will withstand shrapnel impact). I was kind of cringing when he was leaning to the side protecting his head because the thing might blow when he went to the highest pressure... if it actually had blown he could have gotten seriously injured.

[@johansvensson9429]

Most underrated techchannel!

[@HardwareHorizon]

I know this idea would never work in practicality, but I love the fact you test this stuff out for fun. Great to watch!

[@SuperSilvi1990]

With a open air loop you could run that air compressor at full. It wouldn't build pressure with an open end.

[@Macabri_2k10]

the whole pressure calculation he did is irrelevant since the headpressure of the pump is only to overcome height differences the loop in use is basically without pressure and the air can go through full blast it will do no harm whatsoever, fluid pressure and air pressure in an open system, who cares

[@teardowndan5364]

@@Macabri_2k10 Even in an open system you still have head pressure from drag and bends which will become significant at sufficiently high flow. Probably can't get there using a 3/8" hose of any useful length and a ~130psi consumer compressor though. Upgrade to a 4000psi scuba tank with a short 1/2" hose and the flow resistance should blow the CPU block apart.

[@peterpui7219]

You did a great research! I love your innovative idea.Hopfully in future we can have a new kind of CPU cooling system, and I believed if you using copper pipe to replace plastic or pvc tubing, as outlet or inlet that could be even better on heat dissipation.

[@ReveriKeenani]

Nobody: "so how much your pc boost up?" Major Hardware: "About 50 psi"

[@follow_rallo2009]

Hell yeah! Keep these wacky videos coming my man.. great content. 👍

[@AxR558]

I'd be interested to see how inefficient a tower cooler mated to a second cpu block (plate to plate) would be instead of a radiator in a cooling loop. So instead of Pump/res - CPU - Rad - Pump/res it'd be Pump/res - CPU - CPU+Aircooler bolted face to face - Pump/res Edit: Typo

[@rBennich]

Even if the air cooler can keep the cpu block (that's cooling the water) cooler than a fan can keep a radiator, theoretically it's a bit like making the intestines go in a straight line from the stomach to the asshole. Now if you can find a CPU block with a more advanced, radiator-like structure for the water flow for really great TDP's, that could be something else.

[@skatcat743]

mod a water block onto a stock cooler?

[@happycurty5502]

Dude. Your the mad scientist of the pc world. Keep it up. This is great!!!!

[@fabulousfabrications1985]

You need to have the nozzle of the compressor very close to the CPU so the compressed air can take the heat away as it expands in volume

[@MrNlce30]

If you want cold air look up a vortex tube/valve. Also as it is venting to atmosphere the pressure in the block won't be as high as you think. Great experiment. Completely impractical but still great. Keep up the good work.

[@GallopingWalrus]

IVE BEEN WONDERING THIS FOR YEARS. God I love this channel. Weird shit is the stuff we love inquire about.

[@Sidewinder1009oli]

If you were to force the air through a small hole, as close to the processor as possible, so that expansion happens at that point (absorbing heat energy from surroundings) it should be more effective. For the same reasons that aersols spray very cold

[@jameswolfe8065]

Increasing the pressure on your regulator just allowed air to move faster, pressure on that with an open end is negligible at best. A high volume, low pressure air pump like a mattress inflator would work better in a setup like that

[@paulstrealer5414]

A /lot/ of US Navy ship computer hardware is cooled like this! "electrical dry air", air at normal air compressor pressures (90-120psi) gets run through a drier to condense the water out of the air and reduce the pressure some, and gets used to chill hardware that would get hit enough to die without water cooling. It doesn't run through a radiator, but rather works by Boyles (iirc? The one about expanding gas temperatures) law. It's a neat, leak proof system.

[@FordGTmaniac]

What about filling a closed loop with Galinstan? It's a metal alloy that's liquid at room temperature, and it won't turn solid unless it's below zero. Only problem is that it's corrosive to aluminium, so copper or nickel plated copper will need to be used for the waterblock and radiator, along with a pretty strong pump.

[@erikburzinski8248]

There is only one major problem with doing that. The only pump that stands a chance is an electro magnetic one though that type of pump will unfortunately heat up the liquid metal. The reason a electro magnetic pump is required is do to liquid meatals low viscosity ( or speed a liquid moves when force is applied to it) and electro magnetic pumps are far better at moveing it then normal pumps. also pure murcury works as well for the same purpose.

[@FordGTmaniac]

@@erikburzinski8248 Good point, though the reason I picked Galinstan instead of Mercury is because it is non toxic and less dense.

[@holzwurm_hd7029]

Oh, i thought he was going to screw a heatsink to an Waterblock. Please try this too, i had this in my head for years

[@dpjazzy15]

The compressor generates heat with the compression. They usually use a copper tube/coil to dissipate some of that heat, before it goes into the tank. When the compressed air is released, it'll be chilled cooler than it was in the tank. If you had a more easily compressible gas, you'd see more of the affect and the compressor probably wouldn't have to work as hard. It'd be like a refrigerator compressor. But then you'd have to run your output back to the air intake on the compressor. Propane could be used. If you fed it into the compressor, you might see an improvement. Also, instead of a compressor, what if you used a set of pc fans in series with a duct, tapering the air flow down into the tubing. IDK what kind of pressure you could achieve, but it would be interesting to see how pressure is affected by using more fans, series or parallel.

[@spacecy]

Nice video man. Your channel is really interesting. Seeing things here that i do not see anywhere else. Btw, i think you might be having issues with your audio. What windows version are you running?

[@jamesaron1967]

This gives a completely new meaning to the phrase 'reverse engineering'

[@TatharNuar]

I think it's imperative to find out the exact limits of what the block can handle by sacrificing it to the experiment gods.

[@nolo2k7]

You don't need the radiator actually. The air heats up when it's compressed, thats why the cars have intercoolers, to cool the compressed air. The compressor should have an aircooler as well. When you're decompressing the air it will cooldown due to pressure drop. The more pressure drops, the more itll cool. Adding the radiator and the extra hoses will cause pressure losses that will turn into less effect at the cpu block. You should try this without the radiator to see how much it affects it!

[@catriona_drummond]

Somebody who knows physics.

[@elipseplays3579]

Major hardware trying to simplify his explanation for doing this by using the even more complicated car intercooler explanation.

[@benjimc1]

it actually made more sense to me though... as a car guy

[@huerta25264]

this channel is gonna blow up if you keep doing stuff like this haha! very good!

[@TheGwarGaru]

Dude, I love your experiments. :D

[@AlphaCharlieFour]

The nozzle outlet is a restriction and the tubing shoved on the end is much bigger than the hole in the nozzle, causing a pressure drop. The pressure you are seeing is way less past the nozzle. this is because it has more space to flow out than it does to volume of air coming in. The pressure drop causes a temperature drop past the nozzle. so if you ran the higher pressure air through a cooler before the nozzle. It would take more of the heat from the compressed air out before the temperature drop at the nozzle, and if you put the nozzle in the line closer to the block you would have a temperature drop past the restriction where the pressure drops. Or...

[@TheBuLLe7Pr0oF]

Crazy idea: hook a radiator from an AIO to the heat pipes of an air cooler. Fill the radiator with water and stick a fan on it. So basically u get a Hyper 212 solder the heat pipes somewhere in/out the radiator and cool with a fan like an air cooler with the benefit of the water inside the radiator. That should work pretty well!

[@MyEconomics101]

AIR COOLING A CPU WATER BLOCK ... actually works. My BeQuiet Silent Loop 120mm AIO pump failed after 2 years use. 1 week ago. I removed the two fans from the radiator and placed them directly on the CPU block, making a roof. In an open case, I had 60-80 degrees depending on load (no gaming), i5 8400. Enough to still able to use the PC, and wait till tomorrow for Amazon next day delivery, Corsair AIO.

[@paulb8264]

This is sweet. I'm having flashbacks to my thermo and fluids classes. You should get a really old school block with the interchangeable jet plates and use on with the smallest holes possible. Also get rid of the door gun, it's a restriction the system, go with a ball valve instead.

[@veemyu]

Absolutely Amazing Experiment ✌🏻 Your PSU says MOO 9 :D 😂

[@rantos9]

Cooling an air cooler with an air compressor. I bet that would be pretty good. Our cooling a radiator with an air compressor. WE NEED THIS.

[@gendragongfly]

The weakest part of the waterblock is probably where the screws thread into the plastic, (acetal or plexi?) I have no information on the size of the screws otherwise I could give you an estimate of the pressure it will handle without it exploding 🙂 The construction looks rather solid though, good chance it can handle more than the 40 psi you used thus far. Although I don't recommend testing that with the block on a CPU. Wear safety glasses!

[@alexandreferreira7362]

I've just watched 2 min, and come here to say, this is the most crazy video about air-cooling a water cooler or whatever, and I've just come from the water on the 212 video

[@MajorHardware]

Hope you like the random experiments

[@thesupercheerio]

You should do this again, but change things up. Try using an air mattress pump, large fish tank air pump, and 3d print an adapter for a computer fan down to the hose size. Than mix up the cooling, like add salt to the water/ice mix, do a 80% crushed ice to 20% water and same ratio with salt water vs. cube ice, or go to as low as using cheap vodka and dry ice. THAN haha try it all using the air cooler that you converted into a water cooler, but use it in this manner, run super chilled air through it and see how little of air you actually need to use. Imagine if a super cooled salt water mixture, and a fish tank pump, or fan adapter, could keep temps down extremely efficiency. Could also insulate the exterior of the air cooler, you converted into a water cooler, that you're running chilled air through (it needs a name lol) and see if that helps. Also you can pick up a cheap infrared heat gun that might make things neat to see. Oh and try cooling mineral oil so it doesn't degrade the cooler! Dual water coolers, closed loop oil, in ice bath.

[@nyftn]

the coldest point is the nozzle where high pressure becomes lower pressure . so get that as close as possible to the waterblock

[@lexzbuddy]

Well, if you want to try something different, try this... Get a big metal water tank, a boiler for instance, and bury it in the ground. Hook up you watercooling loop to it and be blown away. All the heat is passed into the ground which is effectively an infinitely large heat sink. I first heard about this over a decade ago at least. You want extreme, this is it. Dig up your basement folks! It works, it's passive and as such completely silent with the most extreme overclocking potential outside of nitrogen. Hell, add a peltier if you like and go the little bit further. No compromises, I won't believe you until I see the jackhammer and shovel !

[@CaptainHoldenian]

Love these experiments, keep 'em coming :^)

[@coachnutt61]

Basically what you have there is a liquid to air exchange intercooler! Same setup I run on my super snake and eagle talon.

[@NeoVoodooTech]

I wonder what rpm that impeller is spinning at lol, still a neat and silent way to cool a work station using a tube from the next room over!

[@jakegarrett8109]

My water loop hisses and screeches (no air bubbles) from extreme high flow rate (I don’t know if you’ve used an air compressor for blowing next to sharp thin objects like air vents, but blasting through a block full of fins is really bad for noise!). You could just have a low pressure water cooler about like a trickle and coil it much better (without using 300w compressor, use a 1 watt water pump, also no ice). Or you could use a passive air cooler with zero power wasted...

[@termin8ter4562]

I had no hope for this. I figured you would have been sitting close to thermal limit. What a neat and wierd experiment. The shit you come up with lol

[@TheStigma]

There is nothing wrong at all with aircooling via hoses and blocks. I'm not sure why anyone would think that except for it just "not being normal". Well, that and because open coolers are easier to construct (and thus cheaper), much less hassle to install due to no hoses, fittings ect. The main reason why this isn't on par with a good air-cooler is simply that the block is designed for water rather than air. An air-block would need to be larger (closer to a traditional air-cooler) - simply because air has much lower thermal mass and thus you need a lot more volume moving through without running into any pressure issues (although with an open loop that would take a lot to happen). A size or two larger hoses would also help optimize it, although I expect 90% of the restriction in this setup is the tiny block and not the comparatively large hoses. And ultimately a setup like this has some benefits over traditional open-air: - You can vent heat perfectly to wherever you want without need for case-fans to do the job. - You don't need a radiator since a pump sucking in ambient air would effectively equal a 100% effective radiator. - Leaks are obviously a non-issue. Oh, and you definitely don't need an air-compressor. A pressure-optimized air-pump would do the job just fine - at least if you weren't going all out extreme.. Basically something like a beefed up aquarium-ventilator. So that part wouldn't necessarily need to be super loud or expensive. The only real reason why you need a compressor in this example is as I said the system (and especially blocks) are designed for water and are waaay too restrictive. An air-optimized block wouldn't need anywhere these pressures to get the same volume of air through (and be a lot safer as a result too). The main expensive part here would be the blocks - just as with air-cooling, except they would probably be a tad more expensive due to higher material costs from simply the extra size needed. It would be great fun to see just how effective you could make a system like this if you actually had proper blocks designed for the task - but that's not a very easy thing to make happen without access to and familiarity with high-end milling equipment. Allthough .... now that I think about it, it should be fairly doable to just mod a common tower-cooler with a casing very similar to the watercooled-aircooler experiment. The only thing is that I think it would be CRITICAL for performance to force the air through the actual fins rather than just "being around the cooler". That flaw in the original experiment was a minor issue on water, but I think it would be vastly more severe on air. But that's not really that hard to DIY together - especially when you don't have to worry about leaking from imperfect seals. @Major Hardware Hop to it I say! :D

[@zedorda1337]

Water and air are both fluids just at different pressures and temperatures. I wonder how well a large fish tank bubbler pump would do?

[@aizu-san8945]

*Phone Rings* Major Hardware : Yeah , Hello? EKWB: we want you to head our new product line of EK-Air Blocks , Come in as soon as you can if you choose to accept it. *Call Ended*

[@benjimc1]

i don't think the pressure actually matters because it's an open ended loop. all you're doing by increasing pressure is increasing the flow rate of the air through the loop. you would have to have some high pressure, verging on thrust, to actually blow anything i guess.

[@JiArumaiti]

One thing I was thinking about is taking a 12V fridge and there's the intake and exhaust of the radiator that cools the fridge. If you weld them to a copper CPU block or just adapt fittings. It could have impressive results. Or the simple way installing the PC inside the fridge

[@jakegarrett8109]

Phase cooling, it’s very common for extreme overclockers, particularly our “regular daily machines” (-60c isle and -35c load testing with FX at 1.7v).

[@portedbikes]

You should not be worried about PSI, as your hose on the other end is open, so you can crank up that PSi a lot higher, while there is no obstruction on the exit you could go way higher than 40, actually your compressor should not be able to blow you setup at full PSI, while the exit on the end stays open, for a really cool test you should add a water spray on the mix, like those fans at the parks with the fine mist of water adding cooling to the moving air, that is not going to be an easy task but you temps should drop drastically, or maybe make the compressor suck air from a bucket filled with Dry Ice. Just some crazy ideas.

[@vidrax3481]

its simple. One forget to cut to the shortest lenght possible the heat scape tube, straigth up 10 cm tube for the exit would be twice effective than this bench. Basically the hot air is being mixed back with the cold for its constantlly tring to push back up the tube on that curve.

[@BrandenMcKinney]

duct tape was actually purpose built by the Army Corp of Engineers to seal ammo boxes, and yes, it was originally called duck tape because it was waterproof

[@ZaganStoanes]

Since it's an open air system there's no real blow apart limit since you aren't building pressure up in the block like you would a closed loop.

[@DerMaje]

You could 3d print a mount for a fan, and make it a full circle. Without the need of a compressor.

[@plasmar1]

look into refrigeration, but I believe it's based on the concept of pumping air through a cooler:P(course they replace air with an alternative gas that works better but similar concept)........ there's something I"m not completely in the know of regarding diameter of pipes(capillary tube I think is tiny and I think the return being a bigger diameter pipe) and some junk relating to the end of the compressor hose becoming really cold when the air comes out

[@youkofoxy]

If you make the air compress and expand you can create a heat pump, that means sub ambient temperatures.

[@cncgeneral]

This may actually be useful for big loops. Rather than having water go through cpu and gpus. You could just turn up the pressure and split it so everything gets cold air

[@emildavidsen1404]

uhm... going from air to water is a hands down thermal convective improvement - the other way around is well... the other way around :-) On one hand it is just good fun seeing you tinker with things just for the heck of it, on the other hand, you are kinda' reinventing the wheel here and finding out that a circle shape is pretty close to optimum. TLDR: You can cool with whatever fluid you feel like as long as your temperature difference is, continuously, sufficient.

[@chincemagnet]

I’ve actually done something like this before, 8 or 9 years ago I had an Asus Striker II Extreme I retired from my daily system, that board has a water block on it and I didn’t want the hassle of liquid cooling just the board, so I modded a fan to fit inside a funnel and ran tubing from the end of the funnel through the water block, it worked, I wouldn’t try to cool a CPU like that though

[@Jure1234567]

You probably want to try this setup without the radiator and ice, which will increase airflow. Oh and detach that outbound pipe off the cpu so the air is easier to make its way out.

[@Aabel1998]

Just as a fyi, 5 psi water pressure it more than 5 psi air pressure. For everyone one foot ( of hose ) water pressure roughly changed .45 pounds air pressure does not change.

[@Zayos_]

Can you try to cool your cpu with Thermoelectric Cooler Peltier please ?

[@vipervidsgamingplus5723]

lo sab already been done in fact there was a company who did that and made a socket 775 cooler

[@darrenmessick4971]

You should try building a pressurized helium loop, it can carry a lot more heat than air can. braze a box around a air cooler, input line and orifice on top suction line on the side, maybe use an old refrigerator compressor to move the gas and an automotive ac evaperator to radiate the heat after the compressor, add enough gas that the hot side can run at 60 - 100 psi

[@WouterVerbruggen]

Awesome crazy project!

[@official_commanderhale965]

This man is Overclocking an air compressor. 🥲

[@MotoCat91]

If you're confident enough to try sub-ambient cooling I want to see how far you can get with thermoelectric chillers (peltier). Basically it's just a square pad with a hot side, cold side, and some wires Sandwich it between the CPU and waterblock, cold side down (obviously :P ) Plug it into 12V power while cooling the hot side with a standard (but more radiator than normal) water cooling system and see how well it can cool under heavy load. I've seen plenty of liquid nitrogen, dry ice and even phase change (like airconditioners) but rarely does anyone try with TECs

[@mr.pineapple7620]

You should put the "intercooler" in between the compressor and the storage tank not after the storage tank. The expanding air already is sub ambiant. So you might even be warming it up

[@boelle1234]

Also try to put the regular fans on, and make a plastic casing with air holes, and use dry ice.. No water

[@Labeled90]

Dude, get a blower style fan, the look like snails and are used to inflate like bounce houses and stuff they're like 55-100 on Ebay, it would work better than an air compressor. You'd have to voice over however they're really loud

[@bobhumplick4213]

@ Major Hardware you should get a 55 gallon metal drum and fill it with water and use that as a resevoir and see if it can water cool without a radiator

[@hi_tech_reptiles]

So transfering the heat from a concentrated area to air as opposed to the spread out copper and aluminum fins of a heat sink to air. Interesting stuff. I think you found your thing - weird tests doing things people would never do in a build or never tried even on a test bed lol

[@plebetopro5786]

As I'm sure dozens of others have said, you are NOT pushing 30PSI through that block. You are pushing CFM. You can't hit "pressure" with an open hole. Also, lose the hose on the back side. "Hot side". You have an open exhaust, and your "feed" is the other room. Also, get a regulator for your compressor, run the tank at ~110, use the regulator on the exhaust of the compressor. Then the compressor might down cycle occasionally. I sell industrial pneumatics all day every day. And it is not "just" air. The air if it is not properly filtered is actually slam full of water, and oil from the compressor. And the air going into your bucket is actually quite warm from the compression. So the ice is actually a critical component.

[@MajorHardware]

I'm a mechanical engineer not a pneumatics engineer

[@plebetopro5786]

@@MajorHardware didn't think you were. Ergo, trying to share some experience/knowledge. Didn't mean to come off snippy.

[@tfkoincognito]

Idea: 2 waterblocks in the same loop no rads, one block on cpu air cooler on other block. Add more blocks/air coolers for more cooling potential.

[@MrMartinSchou]

If you ever repeat this, place a small toy fan in front of the exit air-hose to give us a visual indication of air movement.

[@MrMartinSchou]

Oh, and try cryo-cooling by submerging the radiator in liquid nitrogen and running the compressor off of pure nitrogen.

[@TravisFabel]

It's been over a year but I just want to point out that there is almost zero pressure in the water block. You have to cap off the end in order to have pressure in there. Right now there's almost no pressure as the air can leave completely.

[@azimalif266]

Just what I was thinking of the other day.😁

[@firefighterJerry]

you should try a fish tank air pump with smaller air tube and a valve to keep a avg psi in the tube. Hydrofarm Active Aqua Commercial Air Pump, 12 Outlets, 112W, 110 L/min

[@EnvAdam]

I'd love to see you coil the tubes in water and see if that makes a differnce vs the radiaor? what about AC air even?

[@joewilliams3352]

You could also connect the water hose to it.

[@laharl2k]

You should try adding a mister, like those used in cars. The water mist should help evaporatively to cool even more

[@MajorHardware]

i had an 03 Mitsubishi evo 8 and there was a switch near the shifter that sprayed the inter cooler with water, i never noticed any difference in performance but who knows

[@laharl2k]

Major Hardware By your description it sounds like its for when you are on track, it should help with knock under load, otherwise the intercooler would not be able to keep so much compressed air cool enough for the engine not to knock, hence the computer wont need to limit the fuel to avoid it. the idea is to use something like that to inyect some fine mist into the air so that it contacts with the waterblock and instantly evaporates coolimg a lot more than just air. And its self regulating as water's vapor presure gets low enough at about 10-15*C that it will barely evaporate anymore so it will never condensate and in theory should cool more than just liquid water.

[@brokenone22]

I am not an expert but just my thought its not really getting the full pressure because its an open system so even thought its set 40psi because the block is venting right away. Where as in a closed loop it would see a lot more pressure. I think if you get rid of the vent tupe it would have even less back pressure. I have some experience with tinkering with turbo cars and if you have even a small leak you build much less boost

[@thetruejay20]

Love your videos, you crazy SOB!

[@theepicslayer7sss101]

the thing about the pressure is that you have almost no build up in "resistance" on the other side. your compressor is trying to push 40 PSI but the resistance is probably like 3 at most... well at the compressor side and the first fitting... not that i am an expert but i can easily guess that it is not building up those 40 PSI of pressure!

[@Puffalupagus360]

The air we breathe is a fluid, a fluid that has abysmal thermal transfer characteristics, but a fluid nonetheless.

[@placeholder7054]

Air is a fluid so it makes sense that this would work, however, the specific heat of water is very high so it makes sense that it doesn't work very well.

[@craigdiaz4143]

huh... I thought it would be a water loop but with a normal cpu heatsink+fan mounted on the water block, assuming it was a sideways fed waterblock instead of front fed like this one is

[@StingGTO]

Should of just took the return line off the CPU. It would of helped cool it more.

[@caspersonit454]

Hey, what about adding a 360mm Rad and fans to your water cooled air cooler loop instead of iced water and see if it can dissipate the heat via that?

[@neildutoit5177]

Hey man. I want to see this without the ice box. Just straight ambient air. Because then we can have an apples to apples comparison of air and water

[@watchthe1369]

Kind of late, but how about running a tap off the refigerator water supply and running plain tap water through the cooling system like a swamp cooler? Can you heat your shower water with it or just see how much goes down the drain?

[@MerolaC]

Dat poor CPU. LOL Nice experiment. I am surprised it worked at all. Great video as always, Major!

[@MajorHardware]

I was surprised as well

[@jakobrusbjerg9696]

Use oile in your wartercooling rig, insted of water or air, The good thing is, on risk of short in case of a leak.