Compressed air is one of the MOST expensive forms of energy in an industrial plant coming in at almost 8 times that of electricity. Watt for watt cooling between a peltier cooler and vortex cooler, the vortex cooler would cost far more money to operate. That said, vortex coolers are still awesome.
The vortex cooler is optimal for applications where a Peltier would not be appropriate. You can channel the cooled air from the vortex cooler through insulated tubes into a machine while keeping the 'hot side' away. You can't do that with a Peltier, the 'hot side' will always be in close proximity to the machine. If the vortex cooler is constantly running, like it would in an industrial setting, the efficiency loss becomes less and less as the inner tube itself gets down to temperature.
Meh, peltier is around 20% efficiency and needs a ton of airflow to cool the hot part, which makes it more complex than using air throughout the entire system.
@@LtdJorge nah its way worse than that, expect under 10% if you want to cool cpu and see results. Tecs only look good on paper at zero T delta and zero watts of heat pumped. There is reason its niche used only if nothing else fits.
Yep, this thing is insanely inefficient. Napkin math: the vortex cooler is using 12,000 watts of energy to create the 850 watts of cooling. 14.1% efficient! That's not counting the efficiency of the air compressor, perhaps 85%, and pipe loses. McMaster where they bought it lists it next to refrigerant based A/C units, where it gets absolutely stomped. McMaster does not list CoP for these units, but 2 is a safe minimum bet. They would turn 1500 watt wall power into 3000 watts of cooling. The vortex tube would have very niche applications, perhaps where condensation, refrigerant leakage, or size constraints are a concern.
2:12 Hot air is not dense, cold air is denser. A vortex tube creates cold air and hot air by forcing compressed air through a generation chamber, which spins the air at a high rate of speed (1,000,000 rpm) into a vortex. The high speed air heats up as it spins along the inner walls of the tube toward the control valve. A percentage of the hot, high speed air is permitted to exit at the valve. The remainder of the (now slower) air stream is forced to counterflow up through the center of the high speed air stream in a second vortex. The slower moving air gives up energy in the form of heat and becomes cooled as it spins up the tube. The inside counterflow vortex exits the opposite end as extremely cold air. Vortex tubes generate temperatures as much as 100 deg F (56 deg C) below the inlet air temperature. The fraction of hot air exhausted can be varied to change the outlet cold air temperature, with more exhaust resulting in a colder cold air stream (with lower flow rate), and less exhaust resulting in a warmer cold air stream (and higher flow rate).
It's extremely hard to watch Alex when he is breaking down things technically. He fucks up so many things and gives out wrong information a lot. Glad LTT finally hired some real engineers, hope they get some screen time or at least can help the writers reign in Alex's bullshitting.
What confuses me is why the slower moving air gives up its energy, as in where does it go? As far as I can tell, the only thing that the slower air touches before it has cooled is the end of the tube where some hot air escapes as well as the outer hotter air vortex, so convection wouldn't be able to cool the inner air stream to lower temps than either of those, which should both be at or above ambient temperature. Is this a refrigeration-like effect because the inner stream becomes less dense somehow? That is the only way it would make sense to me, as it would be like a pressurized tank that gets depressurized and is much colder afterwards. But how is that pressure differential maintained? Just the centrifugal force of the air vortex spinning around? Or is there a physical divider between the two streams that isn't shown in the diagrams?
@@knexfan0011 You got it the wrong way round. Slow moving air is air that has given up its energy, that's why its slower. The idea is to split the faster, hotter part of the air from the slower, cooler part of the air. Faster air will create a vortex with a larger diameter than the slower air, same way it is harder to turn in a car when you are going faster.
I’ve used these at my job. I work at a nuclear facility and some times we have to wear protective clothing that is attached to an air compressor so we can receive fresh air. We use the vortex tubes as a form of heat stress mitigation. We can also flip the vortex tube around when we are dealing with cold weather. Quite a handy piece of technology.
yeah we use the same thing for abrasive sand blasting. Works well for a small unit, although the outlet can find its way in between your cheeks in tight spots and scold you a bit 😂
I also used it at my job as a spraypainter , in the summer the full face mask got to hot ,so we put something like this between the compressor and the mask...worked like a charm
you should attach it to a cpu water block, and leave the other fitting open. I think it will have a much better effect, as it will cool the copper block while allowing proper direction for the air to move.
8:30 "why is there frost only on the outside?" simple your compress air has an air dryer. so the air in the tube has very little humidity to create frost where as the outside is exposed to regular humidity air which has plenty of water to create frost. I'm sure the movement of air through the tube plays a role as well but I think that's the biggest factor to not creating frost on the inside.
An air compressor without an air drier still dries the air. Air molecules act like a sponge. It can only hold so much water, and when you reduce its volume by compressing or cooling, you reduce the amount of water that can fit between the fibers. When the sponge returns to its original size, there is less water in it.
@@seaneyo Yup, but remember compressing the air also heats it, and if getting "consumed" rapidly it's usually cooling down fast at the end of a smaller nozzle. I don't have a great filtering system on my shop air system but I still drain it daily from the filter and secondary petcock that are within a few feet of the tank, and see visible moisture every day. But also, Florida. *shrug*
"The frost is only on the outside... But how did they *do* that?" They didn't - you did. And also conveniently answered a question I had, while you were at it. You mentioned you used "filtered, oil-free air," but largely the problem with a compressor is less likely oil in the air and more likely water. Compressed air alone has *tons* of moisture, so much so that if you put your hand in front of a stream of air right off a compressor (at a safe distance, of course) it'd get damp basically instantly. Your filter is, like many, also a drier. No moisture in the outgoing air equals no water to create frost. Only the outside, in contact with air at typical ambient humidity, forms frost.
I see this all the time letting air out of tires at work. Most of the time the tire was inflated without any filter on the compressor, and as it deflates the valve will frost over (I've seen them clog with enough ice to stop letting air out).
u need a professional grade air dryer to feed your compressor for things like painting, bc u dont want water mixing in with your paint when spraying stuff like auto body panels for example. Its also not great for the internals of your air tools, which is why for air tools u use a lubricator which adds droplets of oil to the air stream.
@@freeflub What about simple at-home paint sprayers? The paint is water based, right? So they can just use the standard air and not have any problems, I'd think.
You can have dry air if you have a drier after the compressor motor. Negating a rusted out tank. And less purging of the tank. Although I think that having synthetic oil in the compressor tank will help negate rust. But that is not really likely but possible. Though due to the head contractions of the motor mounted on the tank can be an issue. unless it has a thick metal base plate for the motor that sits on top of the compressor. But you are still putting in hot air. Thus reducing your cfm.
A shroud needs to be made to direct the cold air more evenly over the fins of the heatsink. I know Alex wants to do this, don't let this be the last time we see this magic air cooler hunk of metal. And I know Linus wants to milk this as much as possible for how much that costs!
Agreed, turning adding a shroud will not only make the cooling more even, but also better direct the cold air. I expect that the test in this vid had a lot of cold going in unanticipated directions.
Random fact: JPL uses these to cool the Earth based twins of the Curiosity and Perseverance rovers on hot summer days. Since Earth is a lot hotter than Mars they have to be actively cooled here and vortex cooling is one of the few methods that's compact enough to fit inside the already jam packed rover. We just have to run a compressed air line along with all the tethers and other connections to the rover and voila, cooling!
any suggestions of the best video that explains the phenomenon? is it the same idea that dyson vacuums or mining sites use when centrifugal force in vortex sort the materials by weight? in this case hot (light) and cold molecules (heavy)?
These have been around since at least the 70’s. I’m a master auto tech. We used them to heat and cool carburetor choke springs to adjust properly. That price is insane. The last one I purchased was about 12-15 years ago for $25. Still use often to cool and heat temp sensors for diagnostics… And classic car chokes now and then
10:50 You are losing all the cooling power through the tube. In order for the vortex principles to work the cold site has to be a much smaller diameter than the warm site. As soon as you add a silencer or a tube the efficiency gets a lot worse.
doesn't really have anything to do with the size of the tube. The issue is the ability to transfer heat from the CPU efficiently. Just blowing cold air at the heat spreader is suuuuuper inefficient, there is a reason we use coolers with big fin stacks, airs thermal conductivity is pretty shit. The majority of the air is never in contact with the cpu heat spreader for long enough to actually transfer much heat. A much better way to do this is like use a small water loop, just a CPU block and some kind of reservoir, then put a heat exchanger in the reservoir, either just a coil of copper tubing or if you use a largish reservoir you could just submerge a radiator in it. Then have the cold air blow through the heat exchanger, thereby chilling the water. Since the device is capable of hitting subzero temps using a coolant other than water would probably be a good idea, something like an alcohol (isopropyl or ethanol, vodka cooling?) or just an antifreeze additive in the water like some kind of glycol.
You should be able to expand the cold site: the vortex cooler is a separator of high/low energy molecules in the air. That separation is done inside the tube with the radius that the two outlets are located. After that, it doesn't matter. The issue with the tube is just conduction with the cold air flowing through it
I've used a vortex cooler as a replacement when there was a leaking water circuit in a mold. You were just missing a water block or similar. Instead of directly blowing air at the cpu there needed to be a thermal mass to remove the heat from. If it works for 400F/204C plastic against a beryllium copper core pin, it'll work on a cpu.
Water block is probably also not optimal since it is optimised for the heat capacity and flow of water and not air. Probably would still do better than blowing against the IHS directly though...
Yeah that was my thought. The mass of air is not great, hence not a great conductor of heat. You'd either need a block of metal and then cool that block, or liquid cooler, and cool the liquid which is doing the same thing.
@@SianaGearz This is a bit of a guess but the chemical in heat-pipes also have an optimal temp range it works at. Could have been interfering in the phase change cycle of that chemical.
I've had to buy a couple of these back in the 80's. Used to buy 'em off the Snap-On truck. You use it to test a carburetor's Automatic-Choke on older vehicles. It allows you to quickly heat and cool the actuator-coil that opens and closes the carb's intake butterfly. They were only about $30 back then. Linus needs an inline Air-Drier Filter Combo, coming off that compressor.
I used to work in heavy industrial environments and we relied heavily on using vortex coolers to keep our equipment cool enough to operate reliably. Really the only reason we chose those over refrigeration or peltier coolers was the abundance of compressed air available and the reliability of the device. We generated all of our own compressed air for steelmaking so as far as cost goes, it was "essentially" free compared to the overall operating cost of the facility.
The issue is the heat transfer rate, using the bigger nozzle isn't going to make it cool a cpu better, you need a way to efficiently take the cold air coming from the device and transfer the heat from the cpu into it. Just blowing the cold air at the CPU isn't effective because most of the cold air hits the surface and bounces/flows away after absorbing very very little heat. A better way to do this would be to set up a water block and pump that is just circulating water from a smallish reservoir (bucket or something). In your reservoir place some kind of heat exchanger, like a coil of copper tubing or even just a water cooling radiator. Set up the vortex cooler so the cold air is blowing through some tubing and through the heat exchanger, that way it chills the water which is then used to cool the cpu. Since the device is able to hit sub zero temps using something with a much lower freezing point than regular water would be advisable, an alcohol like isopropyl or even ethanol (vodka or everclear!) would work, or you could use water with an additive like some kind of glycol (basically just automotive engine coolant, anything marketed as antifreeze). The alcohol option would probably be the easiest to clean up and have the lowest toxicity if that's something you care about.
I was thinking long time ago, if there is posible to use the same heat-pipes from a normal cooler and injet water or some of kind of cooling solution, like a watercooling a normal cooler, but in this case could be conecting every heat-pipe and leave the exahust free of the loop, conecting the vortex to the enter and then left the cool air pass inside every heat pipe, cooling all the structure. I think. maybe could work.
It’s not only a matter of transfer rate, the CPU without the cooler has very little thermal mass, so it gets hot too fast. In fact this is the real advantage of (normal) watercooling, your transfer rate is similar to (good) air coolers, but your thermal capacity goes through the roof, so your CPU gets to overclock a bunch more.
It would be better to encase radiator with some wisely placed vents. keeping original cooler defeats the point since we want to see if it can provide more beans.
@@ZesPak I thought that was the direction they would go after the failed air cooler trial. it would have been fairly simple to plumb in as well. I'd bet Alex sees this comment or one similar and tries it.
The issue is thermal transfer speeds from metal to air; it's not that efficient. Like if you're shot out an airlock in outerspace, the body takes hours to cool down even though the temperature is -455. If they're going for true exotic performance, it's going to eventually be some sort of ceramic. But even industry giants haven't figured that one out.
The recycling, steel making and mining industries use these to thermally protect cameras inside of explosion proof housings. They are durable enough for the knocks the housings take, with the side benefit of clean positive air pressure to keep out dust. I've used this kind of setup on a machine built to shred cars and appliances. Once setup they really don't need any maintenance as long as they get clean dryish air.
That's interesting. I would have thought just just compressed air would have done it. I work in automation in panels that don't need a lot of cooling but also can't get dusty we will have timed puffs of compressed air into the panel to do it
@@jim4556 normally, compressed air is plenty cold enough, but boiler cameras, or smelting cameras need either an absurd amount of normal air, or a vortex cooler. We used these a lot in the paper mill I worked at. In fact, some of these were cooling multiple devices simultaneously.
@@jim4556 Thermal cameras in particular were sensitive, and we used them for fire prevention/detection and machine problem detection on a huge scrap shredder. So they got as much protection as we could provide.
@@jim4556 Compressing air generates heat, the pipe between my compressor and the air tank will burn you instantly when it's running. If the air compressor is running for a while, the entire tank will warm up, then the outgoing air will be warm. For many industrial purposes, we can assume things are constantly running. The steel industry is generally working with very high temperatures with glowing hot metal, ambient airflow would not be sufficient.
@@javaman2883 and decompressing air cools... that's high school physics. You are not presurizeing the control cabinet you are just releasing air into so it purges the warm air with in the enclosure with cool and filtered air in areas a normal fan can't because the ambient air is still too warm or contaminated. Now with a device that consumes idk 4x the cfm to deliver 1/4 the cfm of very cold air the argument goes to.. is it better to have just replace the air in the cabinet 4x more often than to use a device like that. Then of course the other option is just a heat exchanger for the panel. That's usually the method you would use for larger enclosures. But the water tight panels still leak some air and in really contaminated environments.. I still see failures. But those panels generally make so much heat it's still more economical since compressed air is very very expensive.
The way I understand it: Heat is basically the rate/speed a particle/molecule move/shaking. so the outer spiraling vortex is getting hotter because of frictions with the metal that makes the air particle "shaking more". it moves really fast that made it only want stay outside because of the centrifugal. now the neat part is only the outer vortex can go outside of the hot end tube. but now we still have mass flow that cannot escape through the hot end and it will look through the other end. the inside vortex is slower because the high speed air forces to be on the outside diameter. the slower air creates a vortex in the opposite rotation because of the returning impulse effect of the hot end nozzle. the slower vortex/air molecules "crashes" with the outer vortex that made the inner vortex slower and slower. the slower particle that has crashed stays in the most inside part of the vortex while the faster particles move to the ouside because of the centrifugal again. this "speed loss" over time creates colder and colder air
Translation: The centrifugal force/innertia in the air is the thing creating a pressure difference and thus the outside layer of hot gas where there is high pressure, and the inside layer of cold where there is low pressure. The hot end opening seems to merely exist for the sake of ensuring new gas gets in there and keeps producing a vortex. Otherwise the pressure would equalise quickly and it'd just blow air from the compressor directly out of the cold end. Cooling the hot end tube with heatsinking should further accelerate the cooling effect, however it will reach a critical point where this effect breaks down given enough cooling i suppose. the reason it flows backwards in an inverse helical pattern is because if it were to flow in the same path as the hot end but in reverse, it would have a lot more resistence as it would directly counterract the hot air vortex. Instead it flows perpendicular to the hot vortex in an inverse helical pattern because it's the path of least resistence. An analogy if you will: Turning a car left or right going 60mph down a road is easier than pushing it backwards. am i getting this right so far?
Nah, it's the deal with the devil that makes it work. I used to work in one of those factories that produces these, and I can confirm, the most expensive part is deal with the devil, current rate is 100 of these for one soul, currently we hire some poor guy for minimum wage, treat him really badly, and when he wants to quit, we offer him year salary severance pay, if he signs non-disclosure agreement... and of course we sneak deal with a devil in there. It's not the cheapest way, but it attracts least attention, after all expenses, deal with the devil part costs about $200 per unit, another $100 per unit goes to treatment of alcoholism and depression of workers. But hey, at least I got good severance deal, when I quit... wait a minute!
It’s been a while since the last experimental cooling build, & you know it’s going to be some level of jank with Alex involved. But now that Alex has used the last of his original ideas before joining LMG, I wonder what new ones he’ll come up with?
It's only been a couple of days. Also, you're making the assumption that all of his ideas beforehand was from before he worked at LMG, which is very likely untrue.
@@AlmostSickBoy Exactly the opposite. He takes complicated engineering concepts and applies them to ordinary things. It's not always educational but always entertaining. As an engineer, I would love to do what he does for a job.
I'd be more interested to see this used to introduce colder air into an existing fan-based system. e.g. imagine a micro PC with a 3080 Ti and you're feeding the case fresh sub-zero air.
Works well enough, but if you want to use this cooler for your plan, you have to slow down the air. At that point, you just turned it into a hugely inefficient AC unit. And since the thing is using an air compressor anyways, it's easier to just put the compressed air through a long enough heat exchanger to cool it to room temperature, and then decompress the air into your case, to use the decompression cooling.
Try putting the cold air through an isolated tube going to a waterblock with nothing on the outlet. The immense amount of cold air must cool the microfins quite a bit and there would be a lot less cold air lost.
I was a painter in a factory, and we had hoods that feed us fresh air, we used the vortex tubes as air conditioning to our head. You could turn it up a bit and condensation would form on the inside of the plastic window. I then used a second one and placed it in a painters jacket with a rubber hose with holes to keep my upper body cool. I didn't want to come out because our plant was like 100+ through the summer.
When I worked in a factory in the paint department, we use vortex coolers to chill the breathable air supply when we were in the paint booths. made those long runs a lot more bearable!
I used one when working with hazardous waste. Wearing Tyvek suits in the summer sucks! But with a hood on and running one of those it was helped tremendously. Also, reverse the flow in the winter when it's 20 degrees out, stay nice and warm! They are great, but not cheap.
@@sithus1966 id say keeping people who have to work in moon suits because of the stuff they work with is no time to be cheap on safety or comfort. I of course do not envy the dudes who fill satellites with rocket fuel, its usually some form of Hydrazine. Great stuff for its job, reacts quickly and is extremely stable for long periods. Also terrifyingly toxic.
At my previous job we had 2 electrostatic turbines for painting automotive parts and each turbine had 4 of those vortex coolers on them. If the turbines weren’t spinning but the coolers had air to them, the turbine cups would ice up very quickly, very cool stuff.
You need an aluminum cooler that has a hollow pass through with the finns inside. Then attach the cool nozzle so the cold air is forced past the finns.
I would be interested to see how it performs if you blow the cold air through an insulated tube into a water block or something that can transfer the heat away from the cpu better. I think a lot of that cold air is just spilling into the case rather than making direct contact with the important bits
As a Mechanic of the 70s and beyond, we had a device very similar called a Choke Checker. With carburetors you need a choke mechanism to make the fuel ratio rich on cold starts and this device could safely spray cold or hot air on the bi-metal coil that moved the choke valve.
Since air acts similarly to a fluid, you should revisit this by directly attaching the vortex output to the inlet of a CPU waterblock (with insulated tubing)!
Sure a fluid but the dynamics of the vortex tube would vary quite a bit it would have to be redesigned to work efficiently with water which is defiantly a lot harder than it sounds
10:21 It could've been worked if you used it further away. Because of the pressure when you blow it from a distance the air near it follows the flowing air and you achieve way higher air volumes. It is like how you can't inflate a plastic bag in one go if you put your mouth directly to it. But it is very easy if you blow it from some distance.
so your telling me air mattresses should designed more like a windsock than having some tiny ass, loud, slow compressor... lol Go on internet, i wanna see this advertised now. :P
Hey I learned about this when wondering why modern diesel trucks have those weird looking exhaust tips. The openings before the tip allow fresh air to flow through following the exhaust significantly cooling the exhaust. It's a safety thing but supposedly also done for emissions testing.
We have big industrial air compressors in our plant. Let me tell you, you get water in those lines (Say from condensation on a hot day) you're going to have a very bad (and dangerous) day. That whole week was an absolute mechanical nightmare.
8:50 because the air groing through the inside of the nozzle is from the compressor which is filtered and oil-free as you've stated, so also moisture free
I would've tried to blow the cold air into an intel stock cooler, that way you might get enough mass to maybe cool the CPU. I feel that by cooling that big Noctua heatsink most of the cold air got lost in the ambient...
I studied this in my undergrad lab class in chemical engineering. It is also called a Hilsch Tube. When it comes down to it we don't really understand exactly how and why it works. It looks like it should violate thermodynamics because it separates a group of molecules into their faster and slower versions. It seems like it should be impossible. However it does work and it doesn't violate thermodynamics. If you do the calculations the total entropy of the system is increasing even with the separation into hot and cold streams. I LOVE these things from an engineering perspective and I love that a device that looks so simple is still beyond our current understanding.
I think it works like that : the compressed air is forced into a vortex and doesnt escape at the cool end since the hole diameter is smaller than the diameter of the vortex. Then the valve at the hot end make it so that some air can bounce back in the middle of the vortex (that have less pressure due to centrifugal force forcing the vortex air to be compressed at the surface of the interior walls of the tube) where the bouncing air loose energy by friction with the vortex (think how water flow faster in the middle of a river and slower near the edge where there is friction) and can escape at the cool end because it is now aligned with the hole.
Maybe try using a water block to increase the surface area the cold air can interface with. You could also design a custom air block with all the equipment you have at your disposal.
Cooling sewing machine needles is also a common application, you could look into using vacuum generators (creates a vacuum by using only compressed air) as even noisier replacements for mechanical fans, but with no moving parts.
Or the point at which it was getting cooled was messing up the circulation of the gas. Probably would be best to find a liquid nitrogen setup and just blast the cold air into it or properly seal off the radiator portion of the cooler to blow the air over that. There's also the issue of surface area of the small pipes.
@@ashc3765 Yep. Heating one end and cooling the other makes them work. Very low pressure inside makes the heat carrier easy to boil. The hot end boils the liquid, the cool end condenses it, sintered metal on the wall wicks the liquid from the cool end back to the hot end. The problem arises when the cool end is cold enough to freeze the liquid. The ice can't wick back to the hot end, so all the carrier material piles up at the cold end and the heat pipe process breaks down. At that point, it's no better than a thin-walled copper tube. At low pressures, water consistently freezes at 0°C, so the vortex cooler (which is colder than that) runs the risk of freezing up and shutting down a typical heat pipe. An acetone-based heat pipe won't freeze up until -95°C.
@@ashc3765 Everything above the absolute zero contains heat. That's why bobski named the temperature. In terms of absolute zero waters freezing point is still plenty warm in terms of absolute physics. Therefore a lot of heat can still be extracted in theory and as of now many real world applications. Like the thing shown in this video, air conditions fridges and freezers.
We use them here at work. They have problems. Even with an oil air filter a fine oil mist will be deposited in everything. Most air compressor systems, unless they're oil-less will have oil in the air. Second they take ALOT of air.
Compressed air is one of the most expensive energy sources. No, you will not use this in an application that is viable for many other and better solutions.
I installed a vortex cooler on a LSV. It's a measuring device for a 8 axis robot arm. The company started as PRODEVCO. Now I think it's BURNCO. The robot has a 400 amp plasma head and 3d imagining camera to preform cuts to raw structure steel also lay out marks. Had to install the vortex cooler to keep the LSV cool to keep precise measuring. It's get super hotttttt here in okc. That device made a huge difference. Been installed for 5 years and kounting. Love this video.
This Old Tony was the first time I saw one of these. He made one in his machine shop. It worked, as it's just solid metal, but it seems much harder to perfect one to the point of being particularly useful lol.
at a buffet, i personally sneak corndogs into the buffet so others can enjoy them. I hide 6 corndogs in my jacket pockets. it then, is a joy for me to see other patrons of the establishment eat my corndogs thinking they were part of the buff
Reading this I realised that corndogs are actually nothing but impersonation of a metamorphic ratchet which is accelerating into the abyss of meatball franchise noctua thermal compound ingestion.
We’ve used these at work to cool equipment for years, because the technology is basic and durable, they can last pretty much forever as long as there’s compressed air. As long as we have electricity there’s always compressed air. Although in the recent past, larger equipment gets little air conditioning units, which are much more energy efficient. Those compressed air coolers do get condensation though, so that’s something to consider, would need insulation.
While certainly cool (no pun nor literal meaning intended), I think 99% of potenial applications are entirely impractical due to alternatives being more efficient (in cost and/or energy usage). It's really just for super niche cases where you can't use typical refrigeration loops, _nor_ passive cooling, _nor_ air convection cooling, nor liquid convection cooling, nor TEC if that's ever even a consideration.
2:12 As a mechanical engineer, when they brought out the whiteboard, I just needed to clarify something since I'm contracted right now to military aircraft propulsion at the moment. And and far as what he is saying, I have no idea because I am not an engineer or something. Air go *pffft* it get hot and fast and Linus go _tapetape_ on $500 piece of metal.
Vortex coolers don't work on liquids, but what you could do is create a closed-loop air cooling system where the cold output from the vortex cooler goes to the chip, while the hot output goes to the radiator.
When I was younger I knew about this and was coming up with ways to cool a pc with it. The problem was cost of the vortex tube on top of the constant need of compressed air. It's certainly doable but not the most practical when you figure up the volume of compressed air and how quickly you go through it. With a proper interface however it'd actually work unlike simply directing air towards a CPU lol.
@@MrDJAK777 basically the same as freon but a bit more risky to set up? some people have made freon coolers and they are really cool (figuratively and literally)
We used one for PHA plastic Biodegradable straws for Starbucks. When we make the straws its extruded out of a die and it goes into a 50ft water filled tank. PHA is very tricky to work with as it has two different materials that like different temperatures to crystallize. One crystal forms in hot water and the other forms in a cold water. So the first 2 feet was 35°C and the last 48ft was Cold water. Then we had an air cooler to help remove water and also cool the straw as quick as possible to make it firm enough to packaged. This process was running at about 700 straws a minute. It did a pretty good job cooling the plastic
I wonder if this would work better if they were to shoot it through the tubes of a liquid cooling system, or even cut some heat pipes open on an air cooler and fire the air through that.
These things are commonly used to cool electronic enclosures in the field in hot areas - and $150 is completely neglectible over 25-35 years it is expected to operate given compressed air cost or commonly 20 grand of electronics in a single box Also - it is safe to use in potentially explosive areas, as it is not creating any sparks
Ding ding . We have a winner this man knows what it's used for. Also used in my field welding in enclosed spaces for cold air conditioning so I don't blow a gasket in a tiny dangerous space.
4:21 That is an ASCO valve. There is a diaphragm with a spring and plunger inside that is activated by the solenoid which is the black square with the wires coming out of it. The vortex is very noise but extremely effective. I use them for cooling down components that I need to service so we can get then repaired and back into production faster. I also use the vortex to cool myself when I am working in hot areas of the plant. If you take a water bottle and drill a ton of small holes all over and then fill the water bottle with scotch bright or something similar then run the cold air into the water bottle you have what we refer to as a "nut cooler". I stick this device into my pocket while working around hot equipment or I just use the vortex's that has magnets attached to it and aim 2 or 3 vortex's at the work area to cool us off as we work.
It's fairly simple. Compressing air heats, expanding air cools. In a vortex, the outside is compressed and the middle is low pressure. The vortex tube taps into those specific areas inside the vortex.
I am pretty sure the way this could work is to have a tiny heatsink on the CPU that will allow for even distribution of the air. That way you can manage to keep the small heatsink cool while its working through standard means of thermal compound into the CPU.
Same concept as a standard freezer compressor. You compress the air, using a tiny amount of air to exhaust the heat in the compression process. On the other end, it is pure decompressed air, which loses heat as it expands. Your compressor does this without a vortex as air exits the nozzle. The air in the tank was already heated-up when you compressed it, then it cooled down to room temperature. When it decompresses, it is absorbing tons of heat, just not as focused as the vortex device. AC compressors depend on cooled, unused cooling, returning to the compressor, to help cool the pump down. That is why the "return line" is the cold line in an AC unit, and the "feed line", going into your house, is hot. It decompresses in the air-exchanger, to freezing temperatures, and exits, still cold and decompressed, back outside. The pump compresses it again, into a liquid, and pushes that searing hot liquid through the exterior exchange radiator, to cool it to outside temps. Even at 105F outside, you can cool the 200F+ liquid, and get a near-zero cooling inside the house. For a real test of value, try using just your nozzle, possibly with a focused "air-brush attachment", and blow it at the phone again. Use that as your base-line. (You just won't get "hot exhaust", doing that.) FYI, You can also make the vortex even cooler if you just cool down your compressor air a few degrees, right at the pump line which is entering the tank. If your pump stays running, the temp in the tank will get super hot and make the cooler less effective and the air in your room will get crazy hot, as well as the exhaust from the vortex. You are literally just using your AC unit, again, for third-party and fourth-party cooling at that point. It would be better to just hijack the "return line" from the AC unit, and get cooling directly from the source.
So heat is related to kinetic energy. The air has a random distribution of fast and slow (hot and cold) molecules, and fast moving ones have greater inertia. This greater inertia means that they are harder to turn so bounce around the sides of the tube when the lower inertia, slower (colder) molecules can swirl around in the middle. That's how the hot gets separated from the cold.
You could also think of it as the centrifugal force. Due to the high RPM, the outside edge of the tube is at very high pressure and so adiabatically heats up. Since energy is conserved, the gas closer to the center of the tube becomes cold. The hot-end of the tube is designed to leak out only the edge air, so the system always loses heat at one end, and what's left is very cold.
I love how this morning I was literally just looking at these on McMaster for a project at work and thought "huh, wonder how they'd do for PC cooling".
Interesting this idea has actually crossed my mind, except just blowing compressed air directly at a small cpu heatsink. I guess if this did not work that well my idea would not work that well either. Not that you'd want to use it over a fan, because of the noise. I had no idea these vortex coolers existed, that is a really interesting tech.
I used one of those style coolers for an automated drilling machine to keep the drill bits at the right temperature. We also used one of these for a larger scale electrical cabinet to keep it at the correct temperature instead of an air conditioners.
@12:20 🤷♀️usually you let the welds cool at room temperature so they have more time to relieve stress and not crack the welds… as for cooling the welding torch, they are liquid cooled with a radiator mounted under the power supply.
You guys should've put a heatsink on the CPU, just a basic one like you'd see in for example PoE switches. That would've probably worked a lot better due to the increased surface area allowing to pull more heat away from the CPU.
You need to try to connect this to a water cooler and run the air through the water block or create a mini chill box around the cpu where you can contain the cold air and not let ambient air heat it up.
I think it'd potentially work if you had a passive heatsink on the cpu, like the things that are just a block with fins. Problem with a the heat pipe / fan radiator could be that it's really good at equalizing with the surrounding air temp, effectively dissipating the cold elsewhere on the fins instead of delivering it to the cpu. You want to maximize the surface area that's being hit by the cold air, but also minimize the surface that could be getting warmed by ambient air and minimize the distance between the temperature extremes.
The Vortex cooler is amazing I took one apart and its a tube with a plastic flat round thick washer that has grooves in it that are angled so the air spins the washer and last a metal washer with a small hole in the middle. This washer was probably used for pressure increase to the V groove metal part. So basically the setup took the cold out sending it one way and expelled the hot out the end. the air is separated by a bent metal sheet (in a V shape) in the tube. hot air comes out the bottom and cold comes out the top. This unit was for a fresh air breather setup for painting industrial coatings on large Jets. What happened to mine was that bent metal part slipped into the tube and the cold stopped. There is no electricity only compressed air going to the unit. It was not unusual to hang my hood up and come back and the face shield was frozen or on hotter days covered in condensation and very cold. The unit expelled hot air that would burn you. I got it working again once I figured out there the metal part went. The only moving part was the 3/8th inch washer with the grooves into it. Taking it apart was not what I expected on the inside.
I wonder if sending the air from a cold side to a pipe, then trough a waterblock, with the outlet not connected (not a loop) wouldn't do better. Air is way less thermally conductive, but considering that the flow is way higher and the temperature way lower, it may work out!
Wow, I've never heard of such a thing. I can't help but think of ways to make it work in the application you were trying. I wonder who the genius was who originally thought of cooling this way. That's thinking WAY outside of the box.
These are known as Ranque-Hilsch coolers. When I was at RPI these were known as Hilsch tubes, then there was an engineering improvement that makes it much more difficult to understand. (That's the Ranque part with the air reflected and the cold air more concentrated.) The Hilsch tube itself is pretty easy to understand. Create a circular chamber with a hole on one side near the edge, and a hole at the center on the other side. Now add a third port that is at the edge of the circular part and blowing at right angles to create the vortex. Like in a centrifuge, the heavier stuff goes to the outside. In this case that is the colder air. I saw one Hilch tube unit for putting in the anchor cable (mooring rope for you not naval types*). For a sailboat, it would provide cooling to a refrigerator and optionally heating to the cabin. (NOT for use in winter climate, but it could extend the sailing season a few weeks on either end. I wouldn't try to put it into any of the sheets--the ropes that pull the lower corner of the sail away from the mast--when the boat was under sail. * Special vocabulary makes things much more concise. For sailing it helps when you just have a skipper and one crew, and makes it possible to manage a ship with hundreds of lines. For example, the skipper is the person holding the wheel or rudder, unless the ship has a wheel with multiple sets of spokes--and people holding them. Then the skipper is the OOD. Note that who the skipper is is very well defined, even on modern ships--everyone within range of the skipper's voice must know the sound of the voices of anyone who might be in that role. Oh, and the call "Captain on the Deck!" should have everyone saluting. But the salutes are not the point, they are the acknowledgment that the Captain is in a position where he may override other orders. Of course, on the bigger ships, this culture is limited to the bridge and the (CIC combat information center).
That is a very good hint. Basically, from the point of view of an optimal heat transfer, pretty much everything that can be done wrong has been done wrong here. ;)
I have actually known about Ranque-Hilsch vortex tube's for a long time. I do think this video could have done a little better job of testing this concept with more testing methods, the end was a bit of a letdown. Still I love this channel and content and look forward to more great videos, thanks.
I think it would be better to make a "air block" specifically for the CPU. Basically a heatsink with ducting to give the air enough contact time with the heatsink. The problem with using the cooler with the heat pipes if you are actually getting the pipes down to freezing the water in the pipes which is the working fluid may be freezing.
When converting electrical power to compressed air, the amount of energy recieved is 94% heat and 6% compressed air. A compressor is basically a room heater that also happens to make a little bit of compressed air. It's the least effective way of converting electrical power to energy. So using compressed air for cooling a seperate different electrical component that also consumes power is beyond stupid.
Any way to somehow utilize and capture this heat in a sterling engine or water boiler + turbine to somehow create more electricity so it's a bit more efficient?
@@saosaqii5807 Your idea is basically WHR (waste heat regeneration) or TERS (thermal energy recovery system) This has been used in Formula 1 and LeMans racing alongside KERS (Kinetic energy recovery system) This is as of this day the most efficient way of converting heat into electrical energy. Very expensive but very effective. A sterling engine is not efficient at all. It has very low torque and power output. A modernized steam engine like on a nuclear powerplant is in theory possible, but comes with other problems like size i.e.
I do not know if that was already mentioned, but to mitigate the condensation problems I think this could help: - Cool down the compressed air through a radiator which can drop the temperature below dew point for that pressure and relative humidity. Extract the water with a water tramp. Some compressors have even a built-in one. Being external means you can apply other chilling sources to cool it down even more beyond ambient temperature. - Put a chemist-based filter like the ones sold for refrigerators/wardrobes. Easyly passing the air through a piece of a big enough pipe or even a recycled gas cylinder filled with this pearls. They will dry completely the remaining moisture. The chemist can be recycled by reheating it up. Just for fun because it's a very very expensive way to refrigerate but with obvious advantages. Anyways dry cooled air is usfeul for many applications. I'd try how long does a beer to get frozen so xD. As always loving your videos, thanks again!!
This definitely needs a specialized rig to take advantage of that kind of cooling. just a random thought... maybe this on an old school stock cooler? Smaller surface area means the cold air might be more effective. Oh, and keep the fan on. lol
Seal the case with duct tape and attach that to the front and open a small vent in the back and let it run … effectively “ air conditioning the entire case “ Or maybe even build a enclosure around the heat sink fins and do the same thing so you can control the air flow
Thought the same. Maybe even running the water from the block through a heat exchanger specifically designed for their available airflow and temperature.
I have seen these used about 25 years ago at a Staley's plant it was used to keep the electronics cool in a control cabinet. These things use a lot of air.
We use these all over the place at my work. They work great at keeping cameras and other electronics cool even though they're exposed to high heat from being next to a furnace melting and refining steel. Amazing technology
The first time I saw one was in 1980. It was used to cool air for a sandblaster guy's suit and helmet. We have 2 in the shop now to keep cutter tools cool on the lathe and mill.
I suggest building a custom heat sink with hollow copper pipes, and blow the frosty air through the copper pipes, similar to how old fridges work, but using cold air instead of refrigerant gas. You can build an adapter that allows you to connect the cooler cold air exit to the heat sink copper pipes.
You should do a video showing how to disable every single nonessential background process in windows. I literally can't find anywhere on the entire internet how to do this and I think a lot of other people would like to know as well.
![[LIVE] : ONE ลุมพินี 88 | คู่เอก "ป้อมเพชร vs อัสลามจอน"](http://i.ytimg.com/vi/1ZqTVVbMgbo/mqdefault.jpg)
Compressed air is one of the MOST expensive forms of energy in an industrial plant coming in at almost 8 times that of electricity. Watt for watt cooling between a peltier cooler and vortex cooler, the vortex cooler would cost far more money to operate. That said, vortex coolers are still awesome.
The vortex cooler is optimal for applications where a Peltier would not be appropriate. You can channel the cooled air from the vortex cooler through insulated tubes into a machine while keeping the 'hot side' away. You can't do that with a Peltier, the 'hot side' will always be in close proximity to the machine. If the vortex cooler is constantly running, like it would in an industrial setting, the efficiency loss becomes less and less as the inner tube itself gets down to temperature.
Meh, peltier is around 20% efficiency and needs a ton of airflow to cool the hot part, which makes it more complex than using air throughout the entire system.
@@shadowtheimpure I totally agree with you. I was just making an observation. Most people don't know just how expensive compressed air is.
@@LtdJorge nah its way worse than that, expect under 10% if you want to cool cpu and see results. Tecs only look good on paper at zero T delta and zero watts of heat pumped. There is reason its niche used only if nothing else fits.
Yep, this thing is insanely inefficient. Napkin math: the vortex cooler is using 12,000 watts of energy to create the 850 watts of cooling. 14.1% efficient! That's not counting the efficiency of the air compressor, perhaps 85%, and pipe loses.
McMaster where they bought it lists it next to refrigerant based A/C units, where it gets absolutely stomped. McMaster does not list CoP for these units, but 2 is a safe minimum bet. They would turn 1500 watt wall power into 3000 watts of cooling. The vortex tube would have very niche applications, perhaps where condensation, refrigerant leakage, or size constraints are a concern.
2:12 Hot air is not dense, cold air is denser.
A vortex tube creates cold air and hot air by forcing compressed air through a generation chamber, which spins the air at a high rate of speed (1,000,000 rpm) into a vortex. The high speed air heats up as it spins along the inner walls of the tube toward the control valve. A percentage of the hot, high speed air is permitted to exit at the valve. The remainder of the (now slower) air stream is forced to counterflow up through the center of the high speed air stream in a second vortex. The slower moving air gives up energy in the form of heat and becomes cooled as it spins up the tube. The inside counterflow vortex exits the opposite end as extremely cold air. Vortex tubes generate temperatures as much as 100 deg F (56 deg C) below the inlet air temperature. The fraction of hot air exhausted can be varied to change the outlet cold air temperature, with more exhaust resulting in a colder cold air stream (with lower flow rate), and less exhaust resulting in a warmer cold air stream (and higher flow rate).
This is so awesome
This is also why there was no frost on the one side. The hot air forms a thin film around the part. The cold air is only touching air.
It's extremely hard to watch Alex when he is breaking down things technically. He fucks up so many things and gives out wrong information a lot. Glad LTT finally hired some real engineers, hope they get some screen time or at least can help the writers reign in Alex's bullshitting.
What confuses me is why the slower moving air gives up its energy, as in where does it go?
As far as I can tell, the only thing that the slower air touches before it has cooled is the end of the tube where some hot air escapes as well as the outer hotter air vortex, so convection wouldn't be able to cool the inner air stream to lower temps than either of those, which should both be at or above ambient temperature.
Is this a refrigeration-like effect because the inner stream becomes less dense somehow? That is the only way it would make sense to me, as it would be like a pressurized tank that gets depressurized and is much colder afterwards. But how is that pressure differential maintained? Just the centrifugal force of the air vortex spinning around? Or is there a physical divider between the two streams that isn't shown in the diagrams?
@@knexfan0011 You got it the wrong way round. Slow moving air is air that has given up its energy, that's why its slower. The idea is to split the faster, hotter part of the air from the slower, cooler part of the air. Faster air will create a vortex with a larger diameter than the slower air, same way it is harder to turn in a car when you are going faster.
I’ve used these at my job. I work at a nuclear facility and some times we have to wear protective clothing that is attached to an air compressor so we can receive fresh air. We use the vortex tubes as a form of heat stress mitigation. We can also flip the vortex tube around when we are dealing with cold weather. Quite a handy piece of technology.
and now used to set freeze seals on pipe also. In fact this beats the heck out of having to use liquid N2 and constantly replenish it
yeah we use the same thing for abrasive sand blasting. Works well for a small unit, although the outlet can find its way in between your cheeks in tight spots and scold you a bit 😂
Your handy "technology" was invented in the 1790s by an Italian physicist.
@@SomeOldGamers venturi
I also used it at my job as a spraypainter , in the summer the full face mask got to hot ,so we put something like this between the compressor and the mask...worked like a charm
you should attach it to a cpu water block, and leave the other fitting open. I think it will have a much better effect, as it will cool the copper block while allowing proper direction for the air to move.
This is what I came here to say.
3D print an shroud for the heatsink so all the air is forced past the fins.
This
That actually sounds really smart
exactly what i wanted to say
8:30 "why is there frost only on the outside?" simple your compress air has an air dryer. so the air in the tube has very little humidity to create frost where as the outside is exposed to regular humidity air which has plenty of water to create frost. I'm sure the movement of air through the tube plays a role as well but I think that's the biggest factor to not creating frost on the inside.
Cry about it
An air compressor without an air drier still dries the air. Air molecules act like a sponge. It can only hold so much water, and when you reduce its volume by compressing or cooling, you reduce the amount of water that can fit between the fibers. When the sponge returns to its original size, there is less water in it.
Lotta air-flow going on on the inside too, right? Hard to accumulate moisture when you have that kinda velocity, I'd imagine...but I'm no expert.
+ there's the o-ring seal?
@@seaneyo Yup, but remember compressing the air also heats it, and if getting "consumed" rapidly it's usually cooling down fast at the end of a smaller nozzle.
I don't have a great filtering system on my shop air system but I still drain it daily from the filter and secondary petcock that are within a few feet of the tank, and see visible moisture every day. But also, Florida. *shrug*
"The frost is only on the outside... But how did they *do* that?"
They didn't - you did. And also conveniently answered a question I had, while you were at it. You mentioned you used "filtered, oil-free air," but largely the problem with a compressor is less likely oil in the air and more likely water. Compressed air alone has *tons* of moisture, so much so that if you put your hand in front of a stream of air right off a compressor (at a safe distance, of course) it'd get damp basically instantly. Your filter is, like many, also a drier. No moisture in the outgoing air equals no water to create frost. Only the outside, in contact with air at typical ambient humidity, forms frost.
I see this all the time letting air out of tires at work. Most of the time the tire was inflated without any filter on the compressor, and as it deflates the valve will frost over (I've seen them clog with enough ice to stop letting air out).
u need a professional grade air dryer to feed your compressor for things like painting, bc u dont want water mixing in with your paint when spraying stuff like auto body panels for example. Its also not great for the internals of your air tools, which is why for air tools u use a lubricator which adds droplets of oil to the air stream.
@@freeflub
What about simple at-home paint sprayers? The paint is water based, right? So they can just use the standard air and not have any problems, I'd think.
You can have dry air if you have a drier after the compressor motor. Negating a rusted out tank. And less purging of the tank. Although I think that having synthetic oil in the compressor tank will help negate rust. But that is not really likely but possible. Though due to the head contractions of the motor mounted on the tank can be an issue. unless it has a thick metal base plate for the motor that sits on top of the compressor. But you are still putting in hot air. Thus reducing your cfm.
@@tOSdude I just have a propane torch nearby whenever changing tires because of that.
A shroud needs to be made to direct the cold air more evenly over the fins of the heatsink. I know Alex wants to do this, don't let this be the last time we see this magic air cooler hunk of metal. And I know Linus wants to milk this as much as possible for how much that costs!
Yep cnc a shroud to afix around the air cooler
they should try a open bench and use a CO² 7m³ tank outside.
now that, that will be crazy.
In and out🤔Maybe you can use a water cooler on in cold and out take airflow goes out
@@lvl3592 3d printed would be cheaper quicker and easier
Agreed, turning adding a shroud will not only make the cooling more even, but also better direct the cold air. I expect that the test in this vid had a lot of cold going in unanticipated directions.
Random fact: JPL uses these to cool the Earth based twins of the Curiosity and Perseverance rovers on hot summer days. Since Earth is a lot hotter than Mars they have to be actively cooled here and vortex cooling is one of the few methods that's compact enough to fit inside the already jam packed rover. We just have to run a compressed air line along with all the tethers and other connections to the rover and voila, cooling!
any suggestions of the best video that explains the phenomenon? is it the same idea that dyson vacuums or mining sites use when centrifugal force in vortex sort the materials by weight? in this case hot (light) and cold molecules (heavy)?
@@cubertmiso I think thats moreso granular convection
that's interesting AF
@@thecoon3193 thanks dan. didn't even know that that was a named thing.
@@cubertmiso this old tony has a really good video about how they work.
These have been around since at least the 70’s.
I’m a master auto tech. We used them to heat and cool carburetor choke springs to adjust properly.
That price is insane. The last one I purchased was about 12-15 years ago for $25. Still use often to cool and heat temp sensors for diagnostics… And classic car chokes now and then
Isn't that just typical for McMaster Carr?
you should make some money... i believe you just saying...
It is a tube with a venturi in it $25 even sounds pricey.
love you answer, thanks
This guy obviously has not looked at gas prices............
10:50 You are losing all the cooling power through the tube. In order for the vortex principles to work the cold site has to be a much smaller diameter than the warm site. As soon as you add a silencer or a tube the efficiency gets a lot worse.
This is the flir camera through tempered glass all over again
kinda make sense, I noticed that too
@@4rd17 FLIR
doesn't really have anything to do with the size of the tube. The issue is the ability to transfer heat from the CPU efficiently. Just blowing cold air at the heat spreader is suuuuuper inefficient, there is a reason we use coolers with big fin stacks, airs thermal conductivity is pretty shit. The majority of the air is never in contact with the cpu heat spreader for long enough to actually transfer much heat. A much better way to do this is like use a small water loop, just a CPU block and some kind of reservoir, then put a heat exchanger in the reservoir, either just a coil of copper tubing or if you use a largish reservoir you could just submerge a radiator in it. Then have the cold air blow through the heat exchanger, thereby chilling the water. Since the device is capable of hitting subzero temps using a coolant other than water would probably be a good idea, something like an alcohol (isopropyl or ethanol, vodka cooling?) or just an antifreeze additive in the water like some kind of glycol.
You should be able to expand the cold site: the vortex cooler is a separator of high/low energy molecules in the air. That separation is done inside the tube with the radius that the two outlets are located. After that, it doesn't matter.
The issue with the tube is just conduction with the cold air flowing through it
I've used a vortex cooler as a replacement when there was a leaking water circuit in a mold. You were just missing a water block or similar. Instead of directly blowing air at the cpu there needed to be a thermal mass to remove the heat from. If it works for 400F/204C plastic against a beryllium copper core pin, it'll work on a cpu.
That was my first thought, that a water block would better utilize the high psi cold air than a traditional cooler
Water block is probably also not optimal since it is optimised for the heat capacity and flow of water and not air. Probably would still do better than blowing against the IHS directly though...
Yeah that was my thought. The mass of air is not great, hence not a great conductor of heat. You'd either need a block of metal and then cool that block, or liquid cooler, and cool the liquid which is doing the same thing.
Why didn't the heatpipe heatsink work?
@@SianaGearz This is a bit of a guess but the chemical in heat-pipes also have an optimal temp range it works at. Could have been interfering in the phase change cycle of that chemical.
I've had to buy a couple of these back in the 80's. Used to buy 'em off the Snap-On truck. You use it to test a carburetor's Automatic-Choke on older vehicles. It allows you to quickly heat and cool the actuator-coil that opens and closes the carb's intake butterfly. They were only about $30 back then. Linus needs an inline Air-Drier Filter Combo, coming off that compressor.
I used to work in heavy industrial environments and we relied heavily on using vortex coolers to keep our equipment cool enough to operate reliably. Really the only reason we chose those over refrigeration or peltier coolers was the abundance of compressed air available and the reliability of the device. We generated all of our own compressed air for steelmaking so as far as cost goes, it was "essentially" free compared to the overall operating cost of the facility.
The issue is the heat transfer rate, using the bigger nozzle isn't going to make it cool a cpu better, you need a way to efficiently take the cold air coming from the device and transfer the heat from the cpu into it. Just blowing the cold air at the CPU isn't effective because most of the cold air hits the surface and bounces/flows away after absorbing very very little heat.
A better way to do this would be to set up a water block and pump that is just circulating water from a smallish reservoir (bucket or something). In your reservoir place some kind of heat exchanger, like a coil of copper tubing or even just a water cooling radiator. Set up the vortex cooler so the cold air is blowing through some tubing and through the heat exchanger, that way it chills the water which is then used to cool the cpu.
Since the device is able to hit sub zero temps using something with a much lower freezing point than regular water would be advisable, an alcohol like isopropyl or even ethanol (vodka or everclear!) would work, or you could use water with an additive like some kind of glycol (basically just automotive engine coolant, anything marketed as antifreeze). The alcohol option would probably be the easiest to clean up and have the lowest toxicity if that's something you care about.
Im still astonished they havent used glycol more. Thats what we use here to heat driveways.
id say blow the cold air through a waterblock and see what happens. there is a lot more surface area in them than a CPU heat spreader.
I was thinking long time ago, if there is posible to use the same heat-pipes from a normal cooler and injet water or some of kind of cooling solution, like a watercooling a normal cooler, but in this case could be conecting every heat-pipe and leave the exahust free of the loop, conecting the vortex to the enter and then left the cool air pass inside every heat pipe, cooling all the structure. I think. maybe could work.
This Old Tony has a video explaining Vortex Tubes th-cam.com/video/Hn8hDY4bvpI/w-d-xo.html
It’s not only a matter of transfer rate, the CPU without the cooler has very little thermal mass, so it gets hot too fast.
In fact this is the real advantage of (normal) watercooling, your transfer rate is similar to (good) air coolers, but your thermal capacity goes through the roof, so your CPU gets to overclock a bunch more.
You should make a mostly sealed case. Stick the cold end inside the case, hot end out side, and run regular CPU/GPU cooler and see if it helps.
It would be better to encase radiator with some wisely placed vents. keeping original cooler defeats the point since we want to see if it can provide more beans.
I feel like this should work with the right contact area on the CPU, maybe with some special heat pipe rigging
Just use a waterblock to put the cold air in, should do the trick.
@@ZesPak I thought that was the direction they would go after the failed air cooler trial. it would have been fairly simple to plumb in as well. I'd bet Alex sees this comment or one similar and tries it.
The issue is thermal transfer speeds from metal to air; it's not that efficient. Like if you're shot out an airlock in outerspace, the body takes hours to cool down even though the temperature is -455. If they're going for true exotic performance, it's going to eventually be some sort of ceramic. But even industry giants haven't figured that one out.
@@ZesPak yes, also the condensation he referred to would also aid in removing heat.
In space there's no air to conduct heat away, it can only radiate away.
The recycling, steel making and mining industries use these to thermally protect cameras inside of explosion proof housings. They are durable enough for the knocks the housings take, with the side benefit of clean positive air pressure to keep out dust. I've used this kind of setup on a machine built to shred cars and appliances. Once setup they really don't need any maintenance as long as they get clean dryish air.
That's interesting. I would have thought just just compressed air would have done it. I work in automation in panels that don't need a lot of cooling but also can't get dusty we will have timed puffs of compressed air into the panel to do it
@@jim4556 normally, compressed air is plenty cold enough, but boiler cameras, or smelting cameras need either an absurd amount of normal air, or a vortex cooler.
We used these a lot in the paper mill I worked at. In fact, some of these were cooling multiple devices simultaneously.
@@jim4556 Thermal cameras in particular were sensitive, and we used them for fire prevention/detection and machine problem detection on a huge scrap shredder. So they got as much protection as we could provide.
@@jim4556 Compressing air generates heat, the pipe between my compressor and the air tank will burn you instantly when it's running. If the air compressor is running for a while, the entire tank will warm up, then the outgoing air will be warm. For many industrial purposes, we can assume things are constantly running. The steel industry is generally working with very high temperatures with glowing hot metal, ambient airflow would not be sufficient.
@@javaman2883 and decompressing air cools... that's high school physics.
You are not presurizeing the control cabinet you are just releasing air into so it purges the warm air with in the enclosure with cool and filtered air in areas a normal fan can't because the ambient air is still too warm or contaminated. Now with a device that consumes idk 4x the cfm to deliver 1/4 the cfm of very cold air the argument goes to.. is it better to have just replace the air in the cabinet 4x more often than to use a device like that.
Then of course the other option is just a heat exchanger for the panel. That's usually the method you would use for larger enclosures. But the water tight panels still leak some air and in really contaminated environments.. I still see failures. But those panels generally make so much heat it's still more economical since compressed air is very very expensive.
The way I understand it: Heat is basically the rate/speed a particle/molecule move/shaking.
so the outer spiraling vortex is getting hotter because of frictions with the metal that makes the air particle "shaking more". it moves really fast that made it only want stay outside because of the centrifugal. now the neat part is only the outer vortex can go outside of the hot end tube.
but now we still have mass flow that cannot escape through the hot end and it will look through the other end.
the inside vortex is slower because the high speed air forces to be on the outside diameter.
the slower air creates a vortex in the opposite rotation because of the returning impulse effect of the hot end nozzle. the slower vortex/air molecules "crashes" with the outer vortex that made the inner vortex slower and slower.
the slower particle that has crashed stays in the most inside part of the vortex while the faster particles move to the ouside because of the centrifugal again.
this "speed loss" over time creates colder and colder air
Translation:
The centrifugal force/innertia in the air is the thing creating a pressure difference and thus the outside layer of hot gas where there is high pressure, and the inside layer of cold where there is low pressure. The hot end opening seems to merely exist for the sake of ensuring new gas gets in there and keeps producing a vortex. Otherwise the pressure would equalise quickly and it'd just blow air from the compressor directly out of the cold end.
Cooling the hot end tube with heatsinking should further accelerate the cooling effect, however it will reach a critical point where this effect breaks down given enough cooling i suppose.
the reason it flows backwards in an inverse helical pattern is because if it were to flow in the same path as the hot end but in reverse, it would have a lot more resistence as it would directly counterract the hot air vortex. Instead it flows perpendicular to the hot vortex in an inverse helical pattern because it's the path of least resistence.
An analogy if you will: Turning a car left or right going 60mph down a road is easier than pushing it backwards.
am i getting this right so far?
Nah, it's the deal with the devil that makes it work. I used to work in one of those factories that produces these, and I can confirm, the most expensive part is deal with the devil, current rate is 100 of these for one soul, currently we hire some poor guy for minimum wage, treat him really badly, and when he wants to quit, we offer him year salary severance pay, if he signs non-disclosure agreement... and of course we sneak deal with a devil in there.
It's not the cheapest way, but it attracts least attention, after all expenses, deal with the devil part costs about $200 per unit, another $100 per unit goes to treatment of alcoholism and depression of workers. But hey, at least I got good severance deal, when I quit... wait a minute!
It’s been a while since the last experimental cooling build, & you know it’s going to be some level of jank with Alex involved. But now that Alex has used the last of his original ideas before joining LMG, I wonder what new ones he’ll come up with?
Seems like Alex is the worst addition to the team. He is just ordinary, average go-by-at-a-right-time Joe.
Thats why i usually like videos with alex in it 😂 for the pure jank and stank
It's only been a couple of days. Also, you're making the assumption that all of his ideas beforehand was from before he worked at LMG, which is very likely untrue.
@Si he wasn’t talking to you directly?
@@AlmostSickBoy Exactly the opposite. He takes complicated engineering concepts and applies them to ordinary things. It's not always educational but always entertaining. As an engineer, I would love to do what he does for a job.
I'd be more interested to see this used to introduce colder air into an existing fan-based system. e.g. imagine a micro PC with a 3080 Ti and you're feeding the case fresh sub-zero air.
I'd love to see that, too, but I'm pretty sure it'd kill something in the computer from condensation
Rust
Jay2cents does it with a airco system and boxes, maybe it's not sub zero but almost .
steamdeck
Works well enough, but if you want to use this cooler for your plan, you have to slow down the air. At that point, you just turned it into a hugely inefficient AC unit.
And since the thing is using an air compressor anyways, it's easier to just put the compressed air through a long enough heat exchanger to cool it to room temperature, and then decompress the air into your case, to use the decompression cooling.
We use them in industrial equipment to keep the PLC enclosure cool in place of a phase exchange cooler.
Try putting the cold air through an isolated tube going to a waterblock with nothing on the outlet. The immense amount of cold air must cool the microfins quite a bit and there would be a lot less cold air lost.
this
Exactly what I was thinking.
Glad I'm not the only one with this idea. Didn't J2C "cool" a CPU this way in the past? (minus the vortex cooler of course, just raw air)
so long as a water cooling block can take that much pressure
You need to isolate the cold air. Its like a fan, it drags A LOT of room temp air with it when you use it in the open.
That's what I was thinking, needs to be sealed then maybe a laminar flow screen to smooth and disperse the air.
I was a painter in a factory, and we had hoods that feed us fresh air, we used the vortex tubes as air conditioning to our head. You could turn it up a bit and condensation would form on the inside of the plastic window. I then used a second one and placed it in a painters jacket with a rubber hose with holes to keep my upper body cool. I didn't want to come out because our plant was like 100+ through the summer.
When I worked in a factory in the paint department, we use vortex coolers to chill the breathable air supply when we were in the paint booths. made those long runs a lot more bearable!
I used one when working with hazardous waste. Wearing Tyvek suits in the summer sucks! But with a hood on and running one of those it was helped tremendously. Also, reverse the flow in the winter when it's 20 degrees out, stay nice and warm! They are great, but not cheap.
@@sithus1966 id say keeping people who have to work in moon suits because of the stuff they work with is no time to be cheap on safety or comfort. I of course do not envy the dudes who fill satellites with rocket fuel, its usually some form of Hydrazine. Great stuff for its job, reacts quickly and is extremely stable for long periods. Also terrifyingly toxic.
At my previous job we had 2 electrostatic turbines for painting automotive parts and each turbine had 4 of those vortex coolers on them. If the turbines weren’t spinning but the coolers had air to them, the turbine cups would ice up very quickly, very cool stuff.
You need an
aluminum cooler that has a hollow pass through with the finns inside. Then attach the cool nozzle so the cold air is forced past the finns.
theirs a half-dozen ways they could have used this as an effective cooling solution. really unfortunate to see them wiff this bad
Heatsync - It synchronizes the heat... with something.
I was expecting a custom mount made to direct its air for effectively harnessing its dispersion 🧐
I would be interested to see how it performs if you blow the cold air through an insulated tube into a water block or something that can transfer the heat away from the cpu better. I think a lot of that cold air is just spilling into the case rather than making direct contact with the important bits
As a Mechanic of the 70s and beyond, we had a device very similar called a Choke Checker. With carburetors you need a choke mechanism to make the fuel ratio rich on cold starts and this device could safely spray cold or hot air on the bi-metal coil that moved the choke valve.
Since air acts similarly to a fluid, you should revisit this by directly attaching the vortex output to the inlet of a CPU waterblock (with insulated tubing)!
Sure a fluid but the dynamics of the vortex tube would vary quite a bit it would have to be redesigned to work efficiently with water which is defiantly a lot harder than it sounds
Yes but No...
I would let a Heat Exchanger do the job to make cool liquid out of cold air.
Otherwise - put this thing in a watercooled loop fs!
That's a really smart idea. It will keep the air concentrated on the die and the CPU block has micro fins so they'll get more surface area too.
Air IS a fluid. You probably mean liquid but no, a gas absolutely doesn't act like a liquid in any scenario that includes compression.
Similarly to a liquid
10:21
It could've been worked if you used it further away.
Because of the pressure when you blow it from a distance
the air near it follows the flowing air and you achieve way
higher air volumes.
It is like how you can't inflate a plastic bag in one go
if you put your mouth directly to it. But it is very easy
if you blow it from some distance.
so your telling me air mattresses should designed more like a windsock than having some tiny ass, loud, slow compressor... lol
Go on internet, i wanna see this advertised now. :P
The warmer surrounding air will reduce the cooling slightly but not much.
Ey, the bernoulli effect
Hey I learned about this when wondering why modern diesel trucks have those weird looking exhaust tips. The openings before the tip allow fresh air to flow through following the exhaust significantly cooling the exhaust. It's a safety thing but supposedly also done for emissions testing.
But wouldnt that pull in ambient temp air? Which would heat up the airstream
We have big industrial air compressors in our plant. Let me tell you, you get water in those lines (Say from condensation on a hot day) you're going to have a very bad (and dangerous) day. That whole week was an absolute mechanical nightmare.
8:50 because the air groing through the inside of the nozzle is from the compressor which is filtered and oil-free as you've stated, so also moisture free
Not moisture free, unless they have a pretty overkill filtration system. Moisture reduced maybe..
I would've tried to blow the cold air into an intel stock cooler, that way you might get enough mass to maybe cool the CPU. I feel that by cooling that big Noctua heatsink most of the cold air got lost in the ambient...
"This is a solenoid, it will go click when you click the button."
I studied this in my undergrad lab class in chemical engineering. It is also called a Hilsch Tube. When it comes down to it we don't really understand exactly how and why it works. It looks like it should violate thermodynamics because it separates a group of molecules into their faster and slower versions. It seems like it should be impossible. However it does work and it doesn't violate thermodynamics. If you do the calculations the total entropy of the system is increasing even with the separation into hot and cold streams. I LOVE these things from an engineering perspective and I love that a device that looks so simple is still beyond our current understanding.
It does what?
This is like this new internet meme where Christian Bale says
- Why isn't it possible?
- It's just not.
- Why not, you stupid bastard?
Sounds like a subject they'd drop on you in your third or fourth year to have an existential crisis and recheck your math a dozen times.
I think it works like that : the compressed air is forced into a vortex and doesnt escape at the cool end since the hole diameter is smaller than the diameter of the vortex. Then the valve at the hot end make it so that some air can bounce back in the middle of the vortex (that have less pressure due to centrifugal force forcing the vortex air to be compressed at the surface of the interior walls of the tube) where the bouncing air loose energy by friction with the vortex (think how water flow faster in the middle of a river and slower near the edge where there is friction) and can escape at the cool end because it is now aligned with the hole.
if the calculations hold, then our current understanding is enough to englobe this
Will it work with water?
Maybe try using a water block to increase the surface area the cold air can interface with. You could also design a custom air block with all the equipment you have at your disposal.
YES THEY SHOULD USE THIS AGAIN
I ALSO AGREE AND IM GOING TO SHOUT MY AGREEMENT VERY LOUDLY TO THE WORLD.
@@Kholaslittlespot1 YES I AGREE WITH YOUR AGREEMENT. AND FOR SHOUTING IT OUT TO THE WORLD.
@@ab15h THANKS, I AGREE WITH THE AGREEMENT THAT YOU'RE AGREEING TO. I'M GLAD WE AGREE. AHHHHHHHHHHHHH
Cooling sewing machine needles is also a common application, you could look into using vacuum generators (creates a vacuum by using only compressed air) as even noisier replacements for mechanical fans, but with no moving parts.
I’d imagine freezing the heat pipes would have just made them non effective.
Depends what they're using as the heat carrier. I've read some pipes use acetone instead of water, freezing point -95°C (-139°F).
Or the point at which it was getting cooled was messing up the circulation of the gas. Probably would be best to find a liquid nitrogen setup and just blast the cold air into it or properly seal off the radiator portion of the cooler to blow the air over that.
There's also the issue of surface area of the small pipes.
@@bobski3333 they require heat for proper circulation cycle.
@@ashc3765 Yep. Heating one end and cooling the other makes them work.
Very low pressure inside makes the heat carrier easy to boil. The hot end boils the liquid, the cool end condenses it, sintered metal on the wall wicks the liquid from the cool end back to the hot end. The problem arises when the cool end is cold enough to freeze the liquid. The ice can't wick back to the hot end, so all the carrier material piles up at the cold end and the heat pipe process breaks down. At that point, it's no better than a thin-walled copper tube.
At low pressures, water consistently freezes at 0°C, so the vortex cooler (which is colder than that) runs the risk of freezing up and shutting down a typical heat pipe. An acetone-based heat pipe won't freeze up until -95°C.
@@ashc3765 Everything above the absolute zero contains heat. That's why bobski named the temperature.
In terms of absolute zero waters freezing point is still plenty warm in terms of absolute physics. Therefore a lot of heat can still be extracted in theory and as of now many real world applications. Like the thing shown in this video, air conditions fridges and freezers.
This is the kind of cooling that I would use for my entire server rack in my remote tech room. I intend to have a whole setup like Linus' home setup
@Dart Runner for fun
@Dart Runner he has superpower and it is being rich that's why.
We use them here at work. They have problems. Even with an oil air filter a fine oil mist will be deposited in everything. Most air compressor systems, unless they're oil-less will have oil in the air. Second they take ALOT of air.
Compressed air is one of the most expensive energy sources. No, you will not use this in an application that is viable for many other and better solutions.
I installed a vortex cooler on a LSV. It's a measuring device for a 8 axis robot arm. The company started as PRODEVCO. Now I think it's BURNCO. The robot has a 400 amp plasma head and 3d imagining camera to preform cuts to raw structure steel also lay out marks. Had to install the vortex cooler to keep the LSV cool to keep precise measuring. It's get super hotttttt here in okc. That device made a huge difference. Been installed for 5 years and kounting. Love this video.
This Old Tony was the first time I saw one of these. He made one in his machine shop. It worked, as it's just solid metal, but it seems much harder to perfect one to the point of being particularly useful lol.
I knew id see it somewhere before!
at a buffet, i personally sneak corndogs into the buffet so others can enjoy them. I hide 6 corndogs in my jacket pockets. it then, is a joy for me to see other patrons of the establishment eat my corndogs thinking they were part of the buff
Wtf did I just read
Wtf🤣🤣🤣
Whaaaaaaaaaaa
That is actually evil
Reading this I realised that corndogs are actually nothing but impersonation of a metamorphic ratchet which is accelerating into the abyss of meatball franchise noctua thermal compound ingestion.
Glad you're Flir camera is getting so much use, the team over here loves watching the creative ways your team uses them.
We’ve used these at work to cool equipment for years, because the technology is basic and durable, they can last pretty much forever as long as there’s compressed air. As long as we have electricity there’s always compressed air. Although in the recent past, larger equipment gets little air conditioning units, which are much more energy efficient.
Those compressed air coolers do get condensation though, so that’s something to consider, would need insulation.
The best thing about this thing is that there's no moving parts in it. Making it somehow even harder to understand the magic.
Thanks for the job training! These devices are used on one of the machines at my job. I knew it made cool air but had no idea what it was called.
It would be dope if you could implement it as an emergency cooling system, you turn it on only when it is needed in an emergency or something
You mean in case of overheating?
What would you consider an emergency? It would be annoying if it turned on when CPU is above 65C, as it would just turn on and the off repeatedly.
Immediately before thermal throttling would be the ideal spot in my opinion.
@@JulianSildenLanglo yea that too
@@Neoxon619 overheating could be an emergency to some people yea
I've been fascinated by vortex tubes for years; a lot of really interesting applications they could be used for.
While certainly cool (no pun nor literal meaning intended), I think 99% of potenial applications are entirely impractical due to alternatives being more efficient (in cost and/or energy usage). It's really just for super niche cases where you can't use typical refrigeration loops, _nor_ passive cooling, _nor_ air convection cooling, nor liquid convection cooling, nor TEC if that's ever even a consideration.
2:12 As a mechanical engineer, when they brought out the whiteboard, I just needed to clarify something since I'm contracted right now to military aircraft propulsion at the moment. And and far as what he is saying, I have no idea because I am not an engineer or something.
Air go *pffft* it get hot and fast and Linus go _tapetape_ on $500 piece of metal.
I would love to see an experiment with this somehow cooling a water cooler radiator.
Or something like sub zero liquid cooling with anti freeze
Vortex coolers don't work on liquids, but what you could do is create a closed-loop air cooling system where the cold output from the vortex cooler goes to the chip, while the hot output goes to the radiator.
When I was younger I knew about this and was coming up with ways to cool a pc with it. The problem was cost of the vortex tube on top of the constant need of compressed air. It's certainly doable but not the most practical when you figure up the volume of compressed air and how quickly you go through it. With a proper interface however it'd actually work unlike simply directing air towards a CPU lol.
A closed loop propane cpu cooler would be interesting
@@MrDJAK777 basically the same as freon but a bit more risky to set up? some people have made freon coolers and they are really cool (figuratively and literally)
I love "Alex Videos." 😆Seeing the random stuff they find is one of my favorite types of videos.
Maybe a video idea for SmarterEveryDay, explaining the vortex tube
5:00 Harnessing the power of FARTS
We used one for PHA plastic Biodegradable straws for Starbucks. When we make the straws its extruded out of a die and it goes into a 50ft water filled tank. PHA is very tricky to work with as it has two different materials that like different temperatures to crystallize. One crystal forms in hot water and the other forms in a cold water. So the first 2 feet was 35°C and the last 48ft was Cold water. Then we had an air cooler to help remove water and also cool the straw as quick as possible to make it firm enough to packaged. This process was running at about 700 straws a minute.
It did a pretty good job cooling the plastic
I wonder if this would work better if they were to shoot it through the tubes of a liquid cooling system, or even cut some heat pipes open on an air cooler and fire the air through that.
These things are commonly used to cool electronic enclosures in the field in hot areas - and $150 is completely neglectible over 25-35 years it is expected to operate given compressed air cost or commonly 20 grand of electronics in a single box
Also - it is safe to use in potentially explosive areas, as it is not creating any sparks
Ding ding . We have a winner this man knows what it's used for. Also used in my field welding in enclosed spaces for cold air conditioning so I don't blow a gasket in a tiny dangerous space.
We use the vortex cooler to cool the electric enclosures near a big melting furnace
4:21 That is an ASCO valve. There is a diaphragm with a spring and plunger inside that is activated by the solenoid which is the black square with the wires coming out of it. The vortex is very noise but extremely effective. I use them for cooling down components that I need to service so we can get then repaired and back into production faster.
I also use the vortex to cool myself when I am working in hot areas of the plant. If you take a water bottle and drill a ton of small holes all over and then fill the water bottle with scotch bright or something similar then run the cold air into the water bottle you have what we refer to as a "nut cooler". I stick this device into my pocket while working around hot equipment or I just use the vortex's that has magnets attached to it and aim 2 or 3 vortex's at the work area to cool us off as we work.
It's fairly simple. Compressing air heats, expanding air cools. In a vortex, the outside is compressed and the middle is low pressure. The vortex tube taps into those specific areas inside the vortex.
🤜👍
I'm not vortex tube expert but I bet this is one of those explanations that sounds okay but doesn't actually hold up to scrutiny
I am pretty sure the way this could work is to have a tiny heatsink on the CPU that will allow for even distribution of the air. That way you can manage to keep the small heatsink cool while its working through standard means of thermal compound into the CPU.
Like some other people have said, just blow it through a waterblock.
Same concept as a standard freezer compressor. You compress the air, using a tiny amount of air to exhaust the heat in the compression process. On the other end, it is pure decompressed air, which loses heat as it expands. Your compressor does this without a vortex as air exits the nozzle. The air in the tank was already heated-up when you compressed it, then it cooled down to room temperature. When it decompresses, it is absorbing tons of heat, just not as focused as the vortex device.
AC compressors depend on cooled, unused cooling, returning to the compressor, to help cool the pump down. That is why the "return line" is the cold line in an AC unit, and the "feed line", going into your house, is hot. It decompresses in the air-exchanger, to freezing temperatures, and exits, still cold and decompressed, back outside. The pump compresses it again, into a liquid, and pushes that searing hot liquid through the exterior exchange radiator, to cool it to outside temps. Even at 105F outside, you can cool the 200F+ liquid, and get a near-zero cooling inside the house.
For a real test of value, try using just your nozzle, possibly with a focused "air-brush attachment", and blow it at the phone again. Use that as your base-line. (You just won't get "hot exhaust", doing that.) FYI, You can also make the vortex even cooler if you just cool down your compressor air a few degrees, right at the pump line which is entering the tank. If your pump stays running, the temp in the tank will get super hot and make the cooler less effective and the air in your room will get crazy hot, as well as the exhaust from the vortex. You are literally just using your AC unit, again, for third-party and fourth-party cooling at that point. It would be better to just hijack the "return line" from the AC unit, and get cooling directly from the source.
So heat is related to kinetic energy. The air has a random distribution of fast and slow (hot and cold) molecules, and fast moving ones have greater inertia. This greater inertia means that they are harder to turn so bounce around the sides of the tube when the lower inertia, slower (colder) molecules can swirl around in the middle. That's how the hot gets separated from the cold.
You could also think of it as the centrifugal force. Due to the high RPM, the outside edge of the tube is at very high pressure and so adiabatically heats up. Since energy is conserved, the gas closer to the center of the tube becomes cold. The hot-end of the tube is designed to leak out only the edge air, so the system always loses heat at one end, and what's left is very cold.
I definitely know that cold air is definitely more dense. Other than that I'm interested in seeing how exactly it works in a real world application!
I love how this morning I was literally just looking at these on McMaster for a project at work and thought "huh, wonder how they'd do for PC cooling".
Alex at 4:38 trying to figure out something to say to defend himself, but failing miserably 😁
Interesting this idea has actually crossed my mind, except just blowing compressed air directly at a small cpu heatsink. I guess if this did not work that well my idea would not work that well either. Not that you'd want to use it over a fan, because of the noise. I had no idea these vortex coolers existed, that is a really interesting tech.
I used one of those style coolers for an automated drilling machine to keep the drill bits at the right temperature. We also used one of these for a larger scale electrical cabinet to keep it at the correct temperature instead of an air conditioners.
1:17 for all you gun nerds. Better get the kit before the AFT hears about it.
Well, I'm glad I didn't have my earbuds in for this one, lol. ouchie.
@12:20 🤷♀️usually you let the welds cool at room temperature so they have more time to relieve stress and not crack the welds… as for cooling the welding torch, they are liquid cooled with a radiator mounted under the power supply.
You guys should've put a heatsink on the CPU, just a basic one like you'd see in for example PoE switches. That would've probably worked a lot better due to the increased surface area allowing to pull more heat away from the CPU.
We need to see this at full blast! Make it happen please :)
Linus pointing that compressed air nozzle at his face is one of the most anxiety inducing things I've watched in a long time. SO UNSAFE
You need to try to connect this to a water cooler and run the air through the water block or create a mini chill box around the cpu where you can contain the cold air and not let ambient air heat it up.
I think it'd potentially work if you had a passive heatsink on the cpu, like the things that are just a block with fins. Problem with a the heat pipe / fan radiator could be that it's really good at equalizing with the surrounding air temp, effectively dissipating the cold elsewhere on the fins instead of delivering it to the cpu. You want to maximize the surface area that's being hit by the cold air, but also minimize the surface that could be getting warmed by ambient air and minimize the distance between the temperature extremes.
The Vortex cooler is amazing I took one apart and its a tube with a plastic flat round thick washer that has grooves in it that are angled so the air spins the washer and last a metal washer with a small hole in the middle. This washer was probably used for pressure increase to the V groove metal part. So basically the setup took the cold out sending it one way and expelled the hot out the end. the air is separated by a bent metal sheet (in a V shape) in the tube. hot air comes out the bottom and cold comes out the top. This unit was for a fresh air breather setup for painting industrial coatings on large Jets. What happened to mine was that bent metal part slipped into the tube and the cold stopped. There is no electricity only compressed air going to the unit. It was not unusual to hang my hood up and come back and the face shield was frozen or on hotter days covered in condensation and very cold. The unit expelled hot air that would burn you. I got it working again once I figured out there the metal part went. The only moving part was the 3/8th inch washer with the grooves into it. Taking it apart was not what I expected on the inside.
I wonder if sending the air from a cold side to a pipe, then trough a waterblock, with the outlet not connected (not a loop) wouldn't do better.
Air is way less thermally conductive, but considering that the flow is way higher and the temperature way lower, it may work out!
Wow, I've never heard of such a thing. I can't help but think of ways to make it work in the application you were trying. I wonder who the genius was who originally thought of cooling this way. That's thinking WAY outside of the box.
These are known as Ranque-Hilsch coolers. When I was at RPI these were known as Hilsch tubes, then there was an engineering improvement that makes it much more difficult to understand. (That's the Ranque part with the air reflected and the cold air more concentrated.) The Hilsch tube itself is pretty easy to understand. Create a circular chamber with a hole on one side near the edge, and a hole at the center on the other side. Now add a third port that is at the edge of the circular part and blowing at right angles to create the vortex. Like in a centrifuge, the heavier stuff goes to the outside. In this case that is the colder air.
I saw one Hilch tube unit for putting in the anchor cable (mooring rope for you not naval types*). For a sailboat, it would provide cooling to a refrigerator and optionally heating to the cabin. (NOT for use in winter climate, but it could extend the sailing season a few weeks on either end. I wouldn't try to put it into any of the sheets--the ropes that pull the lower corner of the sail away from the mast--when the boat was under sail.
* Special vocabulary makes things much more concise. For sailing it helps when you just have a skipper and one crew, and makes it possible to manage a ship with hundreds of lines. For example, the skipper is the person holding the wheel or rudder, unless the ship has a wheel with multiple sets of spokes--and people holding them. Then the skipper is the OOD. Note that who the skipper is is very well defined, even on modern ships--everyone within range of the skipper's voice must know the sound of the voices of anyone who might be in that role. Oh, and the call "Captain on the Deck!" should have everyone saluting. But the salutes are not the point, they are the acknowledgment that the Captain is in a position where he may override other orders.
Of course, on the bigger ships, this culture is limited to the bridge and the (CIC combat information center).
They could've 3d-printed a funnel to interface with the tower so that the fins get properly fed with the cold air.
That is a very good hint. Basically, from the point of view of an optimal heat transfer, pretty much everything that can be done wrong has been done wrong here. ;)
I love these weird videos where we learn about the extremes of computers.
I have actually known about Ranque-Hilsch vortex tube's for a long time. I do think this video could have done a little better job of testing this concept with more testing methods, the end was a bit of a letdown. Still I love this channel and content and look forward to more great videos, thanks.
I think it would be better to make a "air block" specifically for the CPU. Basically a heatsink with ducting to give the air enough contact time with the heatsink. The problem with using the cooler with the heat pipes if you are actually getting the pipes down to freezing the water in the pipes which is the working fluid may be freezing.
When converting electrical power to compressed air, the amount of energy recieved is 94% heat and 6% compressed air.
A compressor is basically a room heater that also happens to make a little bit of compressed air.
It's the least effective way of converting electrical power to energy.
So using compressed air for cooling a seperate different electrical component that also consumes power is beyond stupid.
You found a secret ingredient of successful videos. Watching people fail since video was a thing.
Any way to somehow utilize and capture this heat in a sterling engine or water boiler + turbine to somehow create more electricity so it's a bit more efficient?
@@saosaqii5807 Your idea is basically WHR (waste heat regeneration) or TERS (thermal energy recovery system)
This has been used in Formula 1 and LeMans racing alongside KERS (Kinetic energy recovery system)
This is as of this day the most efficient way of converting heat into electrical energy. Very expensive but very effective.
A sterling engine is not efficient at all. It has very low torque and power output.
A modernized steam engine like on a nuclear powerplant is in theory possible, but comes with other problems like size i.e.
A similar setup is used to generate either cold or warm air to the protective hoods used in industrial sandblasting.
I do not know if that was already mentioned, but to mitigate the condensation problems I think this could help:
- Cool down the compressed air through a radiator which can drop the temperature below dew point for that pressure and relative humidity. Extract the water with a water tramp. Some compressors have even a built-in one. Being external means you can apply other chilling sources to cool it down even more beyond ambient temperature.
- Put a chemist-based filter like the ones sold for refrigerators/wardrobes. Easyly passing the air through a piece of a big enough pipe or even a recycled gas cylinder filled with this pearls. They will dry completely the remaining moisture. The chemist can be recycled by reheating it up.
Just for fun because it's a very very expensive way to refrigerate but with obvious advantages. Anyways dry cooled air is usfeul for many applications. I'd try how long does a beer to get frozen so xD.
As always loving your videos, thanks again!!
This definitely needs a specialized rig to take advantage of that kind of cooling. just a random thought... maybe this on an old school stock cooler? Smaller surface area means the cold air might be more effective. Oh, and keep the fan on. lol
you can build heat exchanger for CPU's
are they lazy?
Seal the case with duct tape and attach that to the front and open a small vent in the back and let it run … effectively “ air conditioning the entire case “
Or maybe even build a enclosure around the heat sink fins and do the same thing so you can control the air flow
Hearing Alex yell:
"OK IT'S IMMEDIATELY THERMAL THROTTLING. IT'S GOING REAL SLOW." Made my whole day.
Could have tried to run the cold air through a CPU water block
Thought the same. Maybe even running the water from the block through a heat exchanger specifically designed for their available airflow and temperature.
Yes but then how do they cool it before they run the vortex cooler?
I have seen these used about 25 years ago at a Staley's plant it was used to keep the electronics cool in a control cabinet. These things use a lot of air.
"the hot air because it's denser"
Now I'm certainly no physics expert but I thought hot air rose and cold air sank?
We use these all over the place at my work. They work great at keeping cameras and other electronics cool even though they're exposed to high heat from being next to a furnace melting and refining steel. Amazing technology
The first time I saw one was in 1980. It was used to cool air for a sandblaster guy's suit and helmet. We have 2 in the shop now to keep cutter tools cool on the lathe and mill.
0:32 I thought it's gonna say "Including your mum" 😂😂😂
Wonder how much they had to limit the sound in editing as I imagine that is pretty loud.
Microphones only pick up so much volume.
I suggest building a custom heat sink with hollow copper pipes, and blow the frosty air through the copper pipes, similar to how old fridges work, but using cold air instead of refrigerant gas. You can build an adapter that allows you to connect the cooler cold air exit to the heat sink copper pipes.
You should do a video showing how to disable every single nonessential background process in windows. I literally can't find anywhere on the entire internet how to do this and I think a lot of other people would like to know as well.
There's a script out there that does it....