The TECs did not draw the full amount of power expected because the supply voltage was too low. Looking at the data sheet for TEC1-12710, these modules will need up to 17V, but the supply used was only 12V.
Shall we talk about the failure to place the reservoirs above the tubing so that all the air naturally would migrate out or how about the use of metal bolts between the rails holding the peltier modules between pieces of metal shouldn't plastic nuts and bolts have been used to reduce the transfer of thermal energy through them oh well it was fun 😄
As an engineer I absolutely love all of this whacky engineering content, I would 100% love to watch longer format videos that keep dive into some of this stuff more.
"Engineering". Michael Reeves is more of an engineer than Alex is. So is Electroboom, Adam Savage, Simone Giertz, Integza, Joel Creates... heck, even Dan, on this very channel, is a better engineer.
His video ideas are the most interesting thing on the channel for sure. I'd be way more into the boring "we bought a mansion" videos if Alex was doing all the HVAC.
I spent a couple of years playing with TEC's, successfully building out a universal whole case cooler that dropped the CPU temp by 20c while maintaining a case temp of around 10c. So a few things, mounting pressure on the TEC plate is EXTREMELY important, you need even pressure across both sides or performance takes a serious hit. We ended up threading out own bolts with a very high thread density, counting the turns and confirming with calipers. You also need to have a neoprene gasket (cut a hole for the TEC and sandwich between the heatsinks) we used quarter inch high density neoprene, but your application may need something different. We also used 600w TEC's (24v @ 25a) which required a special switching PSU (30v@30a), and HUGE melcor nickle plated copper heatsinks that were cooled with 120mm 120cfm sillverstone fans. The HS's were 120x120x90mm, and each weighed about 5 pounds, with each tec using TWO. You can simply drop the cold side into a container of water, and use as many as you like. The other important thing, you're much better off using high power TEC's like the 600w we used, and undervolting it. At lower power, TEC plates are much more efficient. Basically you're better off using 8 600w TEC plates, running 75w each than you are using 1 plate running full load. I would have to check my 15 year old forum posts, but i'm pretty sure we found that 30% load was the sweet spot in terms of efficiency, though i'm really not sure if that's a flat metric or simply applied to our 600w plates. Either way, i'd suggest taking another run at this, you should be able to get much better results using much fewer parts and a less janky set up.
@@heygek2769 I'm sure it is, but they've taken a couple runs at TEC coolers for various applications, and in each instance they've done it rather poorly, ended up with bad, or at least insanely inefficient results, and perpetuate the idea that TEC's are useless. I freely admit, this is a personal pet peeve and this was clearly done for the entertainment value, but...if you're going to do something, you really should try to do it as well as you possibly can. One of the simplest things that is consistently overlooked, is using a TEC to cool the back of your CPU socket. There have actually been a few cases in recent years that have employed fans behind the motherboard tray for this purpose, using a TEC in this was is simply more beneficial, and less complicated than trying to directly cool the chip. Like i said before, i spent years playing around with TEC designs, and ended up with several very viable solutions that i simply didn't have the money to see through, and now i just don't have the time. These guys do, and i'd like to see them make a serious attempt.
Alex, Jake & Dan are perfect trio. Alex "should be fine" frankenstein inventions are lawful evil. Jake's persona is chaotic good. & Dan is the true neutral with pinpoint accuracy reactions. I can't imagine the LTT/LMG without them.
that WOULD be really cool to see if it could ever be really good at what it's supposed to do... but on the flip side, "optimized and efficient" isn't really the reason we click on these videos, is it? xD
There is a channel called "Этот Компьютер", in his playlists there is a playlist called "Пельтье", there he assembled his cooling system on peltier modules with a solution to the problems that Alex encountered(these videos are two years old)
Alex, the way you connected the pumps is not giving you double the head pressure, you are basically connecting them in tandem. This mean that one pump discharges just to feed the suction tank to the next. There is not pressure transmission doing that, so basically is just the last pump working, the first is just feeding the second tank, that’s it. If you want to connect the pumps in series to double the head pressure you need to connect directly the discharge side of pump 1 to the inlet side of pump 2. If you want to double the flow, then both pumps must be connected to a discharge pipe at the same time, but the total head is equivalent to one pump.
I think the whole bubble trapping thing could've been solved if they mounted each row in series and not all of them in parallel but i still love the idea!
Indeed. I remember his first videos... he was paired with Riley and I believe Anthony also had some videos here and there. They both sucked. Riley was always a natural. I bet he cheated. But now Alex has found his own style and it''s fantastic.
Still waiting for a video where they do sub-ambient cooling without electrical worries, by sealing the whole PC in a different atmosphere without water in it. Someone needs to do it!
I believe you ran the cooling to the peltiers in parallel, that would completely explain why the coolant was not getting there. Water always takes the easiest route so there was no need for it to go through the top ones, it could just go through the bottom ones. Also rethink the diameter or your water tubes, the tubes going to the waterblocks were really thick and the one going to the pumps was only a little thicker. No wonder there was almost no flow in some areas. You would probably be fine with those flimsy 5mm ish tubes going to the individual units on the hot side. Maybe for a second iteration of this split the cooling system up into groups of like 4 peltiers cooled in series with smaller separate radiators. Also have the water in the cold loop flow the other direction from that in the hot loop to make sure the water doesn't get reheated by the higher temperature peltiers (because they are in series the last one gets a lot warmer cooling water and thus will cool less far on the cold side). I think by hooking up the cold side in groups of 4 (or so) with their own pump and combining the output with a splitter to the CPU plus the aforementioned changes you could optimize this setup.
I feel like in a previous video, it was established that if you want to do this, you should first run the hot water through a radiator to get the temperature down for most of the way and then run it past the TECs to go sub-ambient.
1. Use a few distribution blocks (eg Phoenix PTFIX) instead of soldering 80 wires. Or good old Wagos. You'll hate yourself less. 2. I would do a 5s4p connection instead of 20p to save on tubing effort while having some redundancy. E: any manifold connection has potential for unequal flow so you need to watch out for that. Big benefit of series tubing is you have the same flow rate everywhere.
But that means potentially trying to remove 600W from a small 40mm x 40mm waterblock. I don't think you can get a fast enough flowrate through it to make that happen.
He could tune the flow rates with flow controls. reducing flow in the lower peltier devices would probably balance the flow up to the peltier devices that are much higher and thus need more head pressure. If you are feeding the same head pressure to each row the top will never get good flow.
You need a separate res at the top of the loop for this to work. pump at the bottom, res at the top, then you're not fighting gravity. come on guys this is water loops 101!
also the clamps idea was alright, but you can make mini clamps out of 2 sections of dowel and 2 wood screws. then you can make 1 adjustable clamp for each inlet and dial in the restriction perfectly!
For a manifolded assembly like that you want it to be horizontal rather than vertical so gravity affects each one the same. Then you can have radiators attached to the hot side drawing cool air from the cold tubes on the bottom and the cold side blowing chilled air over the hot tubes on top for a small performance boost in the setup.
these kind of videos are the most interesting you guys put out; just let Alex and the other engineers do whatever crap they come up with, its super entertaining
As not an engineer. But knowing something about something. I 3th this Also using different hardware for the distribution would've made it more efficient
i think you might get better results if you use a large high reservoir above each loop and let the pumps just pump the liquids up, then let gravity let the liquids flow through the coolers this would probably get equal flow through each cooling element. that way you could also store some cold liquid in the reservoir and run a benchmark from the stored cooling capacity of the reservoir.
You need a manifold that distributes to each of the lines with a constant volume flow rate. In other words, the fluid is taking the path of least resistance, like the lower TECs and shorter tubing runs.
You gotta connect those blocks on the peltier-s in parallel (at least in blocks). You could do four groups of TEC-s with each group having their own pump and radiator combo so you have higher crossection for the water to flow through. If you connected everything here in series, I would imagine, the pressure needed to push the coolant through the whole system is too much with just a tubes diameter of cross section for coolant flow.
As I understood it, they were connected in two parallel blocks of 10. The frame's left side supported 10 TEC-s, and the right supported the other 10. Front was cold, back was hot. Am I missing something? These aren't connected in series at all.
@@TorgieMadison They all had their own circuit, which makes for a very long pipe in sum. When you group 4, you almost cut that length by /4 -> less pressure is needed.
@@VealCalf1 If you make yourself familiar with these particular physics you will learn that the relation of pressure needed to length of the pipe is not linear. Spreading out the pumps to different positions in the circuit could help with that.
Hot side in series would only make most of them non-functional as they'd be "cooled" with hot water (heated by earlier ones). OTOH the "cold side" in series would freeze the antifreeze, as the cooling effects would keep stacking.
It’s great that they always happen at the onset of the new generation. Then legit coolers come out, then the next generation changes it up again and the cycle repeats
Always enjoy mad Alex engineering. You definitely need to revisit this with debauer's delidding tool and whatever tips were on the GN N2 overclocking stream! Also great seeing you guys host this content without Linus. It amazing how everyone has grown their presenting skills over the years!
I did the same but in smaller dimensions (about 400W). I wasn't happy with the results. So i tried the outdoor mod. Placing my 9x120mm radiator outdoors at about negativ 5 degrees celsius improved the result greatly. The cool side of the peltier element was extremely cooled and the cold side could get insanely cold (not gonna spoiler, try it out its crazy!!!. Its easy to realise If mother nature provides you with the right circumstances. Shouldnt be a problem in Canada i guess. I would love to see you stepping the project up with the "oudoor mod". Sorry for my terrible english and greetings from Austria
I used to have the "Ultra" brand peltier tower air cooler hybrid. It actually worked very well. With how it was set up it kept the CPU at one temperature when under load, didn't cool it too much or too little. Gimmicky, yeah. But it was cool as hell and still did a fairly good job. Not the best, but good. This idea takes that to the max. I like these ideas lol
You have to create strings and parallel streams with your cooling liquid. The way have it now you would need an open reservoir which presents NO pressure at all. Best way to to this would be to have the water coming from top into the reservoir like a small waterfall. With that much parallel water flow you actually hampered the whole performance. Also add some automatic air reliefs at the top to get the air out as fast as possible. We have the same concept in out town water system where I'm responsible of. You need strings and parallel flow correctly done or else you have no flow at all...
I've never gotten cracked up by these sentence but when Dan came up to them and said "What the fuck is this abomination?", "What are you cooling?" and "What are you heating with this?" all at the same time I audibly laughed
I'm here for more of Alex's off-the-wall cooling ideas! As in, he comes up with the ideas by "throwing shit at the wall to see what sticks", then makes videos out of what does /not/ stick.
I can’t believe you guys are doing ANOTHER peltier cooler video right as I’m designing a stupidly crazy design for a giant observatory telescope chilling system. So thankful!!
Please don't use Alex as a starting point. He is making so many stupid mistakes. Main stupid thing was that he did not place the header tank on the top of the loop.
@@Running_Wakko don’t worry, we aren’t using this is a guide! Our system is going to be far more complex as it has to work in all orientations. The water blocks will be changing orientation as the telescope mirror moves around, so we have to account for everything. We’re gonna build a working model before we even try out the real thing.
Thank you Thank you Thank you LTT, 1 wk ago it was so close to buy the 1/10th of your test parts on Aliexpress, still try to convince myself, then you did it for me, saved he heaps effort, and made my mind to keep standard water cooling. thank you again...
Honestly, I love these types of videos, even if the result doesn't exactly pan out! Sometimes you get a wild and crazy result, and we all benefit by learning!
If i was Linus i would be so proud of Alex and his mad ideas. Also he got so much better in front of camera, i'm so happy to see him do this kind of stuff with this confidence. Go Alex!
I think the pumps needed to be split up among the different sets of coolers, so a pump and a "one in five out" manifold per level of that tower. Getting the water/coolant to cycle properly through all the blocks would probably make a huge difference, and if I understand how they work correctly, may actually get the TECs to draw more power. The reason for that being they have temperature limits on the hot side, once it reaches a certain point it just can't push more heat across, and will drop off.
I really was hoping they'd go all out and like machine a whole water block manifold for the in side and outside. I would love to see a rig they make that actually has some bananas cooling and could be used later on for other experiments. Kinda like their air conditioner cooler.
I think my Prometia cooler only drew like 800w and could keep a 500W load at -20C. I just got tired of dealing with condensation, so went with water chiller that keeps the loop at 1C above ambient.
This looks like a brain from a bad sci-fi movie from decades ago. It's lovely! So when do you design a case that the wonko cooler mounts to and call it "The Brain That Lives!" and make your own mini movie for Floatplane?
i think you should try improving this. you should mount the pump/reservoir up high to prevent air lock and pressure issues if you want to keep testing the in parallel like you are. i would recommend each row of coolers in series and then have the rows connected in parallel. it would really force the coolant to circulate through the coolers and likely require less force and make the water much cooler.
Exactly, never understood why most liquid cooling setups in typical computers have the res down lower either, should be the highest point - even higher than the rad - makes it much easier and would automatically deaerate
I had Two suggestions, to save you time and it is cheap Alex, Thread sealant paste is very inert works better then Teflon tape. I used it a lot to test 3-15lb and 4-28lb pneumatic actuators, pressure transmitters monometers for an old big boiler. My other suggestion is from when I worked for an electronic tool co is "haisstronica Heat Shrink Butt Connectors". These Connectors are Used with a heat gun and leave a marine grade electrical finish. They are fast and fairly cheap, heat guns are faster then soldering. you can also line them up and hit several at once. Good video, pumps go high in industrial loops unless it is hydraulic whole setup might cost $100 but for 50 threads its about 1/2 the time. I'm not sure what Alex gets paid but my guess is it is cost effective
I SEE IT. He has the engineer eyes now. You know, the sort of sunken, passion deprived gaze that is earned by answering emails and attending team meetings instead of doing ACTUAL work. Congratulations.
It is not just a pressure issue, it is a flow issue too! You need to assess the temperature drop of the coolers and then use Q = M.C.DT to work out the total flow required.... then assess the flow of the pumps against the head pressure of the assembly.....
from how I understand piping in general (I am not an expert) the reason there is so much air trapped is because they don't have any pressure release valve located on the highest point of the loop so the air just gets trapped on whichever place gets the water flow last... it's basically a one-way valve that traps liquid below it while release any air upwards without giving it a chance to get back into the loop (I think the show The Good Doctor made a decent demonstration of this)
You should have used the 'Tichelmann-System' for tubing. That's a plumbing/piping technique. This way you would have the same resistance in every loop so the Water would spread properly between all loops.
The way you do this is to buy larger water blocks. You can get blocks that are longer so that you can attach 4x or even 8x pet's per block. The correct way to do it is to add a pet inline wired up to an inverted furnace thermostate so that you can temp controle power on and off the pet's. Best way to go sub ambiant is with geo thermal heat sinking though. Your coolant should stay a constant inline temp between 10-15 degrees. Heres is another hint people condensation requires a temp differential based on humidity in the air. So if your pc was in an airtight humidity controlled enviroment like a mineral oil case or a freezer you can eliminate the opertunity for condensation to occur.
16:06: I like how they black out the password, like anyone wants to break into their warehouse, ignore all the expensive equipment, and take a motherboard covered in vasoline 🙄
The TECs did not draw the full amount of power expected because the supply voltage was too low. Looking at the data sheet for TEC1-12710, these modules will need up to 17V, but the supply used was only 12V.
Makes sense indeed. I was sure something was off. This was it. Thanks for confirming
Holy shit, theres more
Shall we talk about the failure to place the reservoirs above the tubing so that all the air naturally would migrate out or how about the use of metal bolts between the rails holding the peltier modules between pieces of metal shouldn't plastic nuts and bolts have been used to reduce the transfer of thermal energy through them oh well it was fun 😄
@@allenfunstuff HAHAHA I was sure something was tickling my brain when the water could not reach the top and the answer *MORE PUMP* looked dog
What, you think these guys are experts or something?
As an engineer I absolutely love all of this whacky engineering content, I would 100% love to watch longer format videos that keep dive into some of this stuff more.
Really hoping they release more on Floatplane for things like this! Would love more details.
Floatplane probably
"Engineering". Michael Reeves is more of an engineer than Alex is. So is Electroboom, Adam Savage, Simone Giertz, Integza, Joel Creates... heck, even Dan, on this very channel, is a better engineer.
@@rherydrevins well ofc. This is just a whacky project and nothing to serious
@@rherydrevins maybe, but also Alex does have an engineering degree.
Alex's cooling ideas are honestly my favorite videos
When he actually tries and don't just half ass it, they are nice
His video ideas are the most interesting thing on the channel for sure. I'd be way more into the boring "we bought a mansion" videos if Alex was doing all the HVAC.
Still not as ambitious or cool as whole room water cooling.
next time they just dump the whole pc into ln2
agreed
I spent a couple of years playing with TEC's, successfully building out a universal whole case cooler that dropped the CPU temp by 20c while maintaining a case temp of around 10c. So a few things, mounting pressure on the TEC plate is EXTREMELY important, you need even pressure across both sides or performance takes a serious hit. We ended up threading out own bolts with a very high thread density, counting the turns and confirming with calipers. You also need to have a neoprene gasket (cut a hole for the TEC and sandwich between the heatsinks) we used quarter inch high density neoprene, but your application may need something different.
We also used 600w TEC's (24v @ 25a) which required a special switching PSU (30v@30a), and HUGE melcor nickle plated copper heatsinks that were cooled with 120mm 120cfm sillverstone fans. The HS's were 120x120x90mm, and each weighed about 5 pounds, with each tec using TWO. You can simply drop the cold side into a container of water, and use as many as you like.
The other important thing, you're much better off using high power TEC's like the 600w we used, and undervolting it. At lower power, TEC plates are much more efficient. Basically you're better off using 8 600w TEC plates, running 75w each than you are using 1 plate running full load. I would have to check my 15 year old forum posts, but i'm pretty sure we found that 30% load was the sweet spot in terms of efficiency, though i'm really not sure if that's a flat metric or simply applied to our 600w plates.
Either way, i'd suggest taking another run at this, you should be able to get much better results using much fewer parts and a less janky set up.
I think a big part of why this video even happened is because he used stuff that they already had or bought cheap off AliExpress.
@@heygek2769 I'm sure it is, but they've taken a couple runs at TEC coolers for various applications, and in each instance they've done it rather poorly, ended up with bad, or at least insanely inefficient results, and perpetuate the idea that TEC's are useless.
I freely admit, this is a personal pet peeve and this was clearly done for the entertainment value, but...if you're going to do something, you really should try to do it as well as you possibly can.
One of the simplest things that is consistently overlooked, is using a TEC to cool the back of your CPU socket. There have actually been a few cases in recent years that have employed fans behind the motherboard tray for this purpose, using a TEC in this was is simply more beneficial, and less complicated than trying to directly cool the chip.
Like i said before, i spent years playing around with TEC designs, and ended up with several very viable solutions that i simply didn't have the money to see through, and now i just don't have the time. These guys do, and i'd like to see them make a serious attempt.
The Jank is part of the appeal though.
@@heygek2769 sqws
some people, yanno, have a life lmao
I absolutely love these things that us regular nerds just dream of like "this idea is dumb, I want to test this" but you guys actually get to do it :D
The Alex and Jake combo is pure gold definitely would love to see more of them together
The kids just running wild in their dad's garage
They definitely were great together
It was chaotic and I love it
Alex, Jake & Dan are perfect trio.
Alex "should be fine" frankenstein inventions are lawful evil.
Jake's persona is chaotic good.
& Dan is the true neutral with pinpoint accuracy reactions.
I can't imagine the LTT/LMG without them.
It really seemed unplanned too haha love it!
We need more Alex and Jake, such a great duo, both of them are smart and janky at the same time
Lol, smart and janky is the best description!
I really hope that there's a followup video where Alex fixes the worst issues with this cooling apparatus... there's soooo much room for optimization.
that WOULD be really cool to see if it could ever be really good at what it's supposed to do... but on the flip side, "optimized and efficient" isn't really the reason we click on these videos, is it? xD
Imagine spending that much money on tools and equipment, and this is the results...
There is a channel called "Этот Компьютер", in his playlists there is a playlist called "Пельтье", there he assembled his cooling system on peltier modules with a solution to the problems that Alex encountered(these videos are two years old)
I would also love to see them get the full power and tie in a GPU block to the loop.
Der8auer has a video where he built and tested something like this and it worked pretty well iirc.
Alex, the way you connected the pumps is not giving you double the head pressure, you are basically connecting them in tandem. This mean that one pump discharges just to feed the suction tank to the next. There is not pressure transmission doing that, so basically is just the last pump working, the first is just feeding the second tank, that’s it. If you want to connect the pumps in series to double the head pressure you need to connect directly the discharge side of pump 1 to the inlet side of pump 2. If you want to double the flow, then both pumps must be connected to a discharge pipe at the same time, but the total head is equivalent to one pump.
I think the whole bubble trapping thing could've been solved if they mounted each row in series and not all of them in parallel but i still love the idea!
i was thinking the same thing
or if they mounted the reservoir at the top :)
Then they'd need even higher head pressure and even more D5 pumps.
@@hapybratt8640 is that a problem tho?
@@Nils__pltr yes they already using 3 :D
I love all of these "technically feasible but practically inadvisable" projects. Give Alex's mad science lab a bigger budget!
literally the best shit on yt
they should give him his own channel to do this shit
"technically feasible but practically inadvisable" sounds like a mantra for engineering school xD
"Technically feasible but practically inadvisable" basically describes overclocking generally. ;-)
And make him wear a white lab coat
Alex went from nervous engineering nerd to full on video presenter in a few short years. It's a beautiful thing to see
Just like Linus at the start, a worthy successor.
@@shanekhiu9884 Alex for LMG 2100
now he's a cooling mad genius and we love it
Indeed. I remember his first videos... he was paired with Riley and I believe Anthony also had some videos here and there. They both sucked. Riley was always a natural. I bet he cheated.
But now Alex has found his own style and it''s fantastic.
I remember his introduction with the laser cutter! But that's wayy early on!
Alex doing silly cooling projects are genuinely my favorite videos on the channel, I hope we see more in the future.
I assume we already see as much as possible 😀
These projects take up so much time and are so damn much work to realize
Still waiting for a video where they do sub-ambient cooling without electrical worries, by sealing the whole PC in a different atmosphere without water in it. Someone needs to do it!
That’s how I found out LTT
I believe you ran the cooling to the peltiers in parallel, that would completely explain why the coolant was not getting there. Water always takes the easiest route so there was no need for it to go through the top ones, it could just go through the bottom ones. Also rethink the diameter or your water tubes, the tubes going to the waterblocks were really thick and the one going to the pumps was only a little thicker. No wonder there was almost no flow in some areas. You would probably be fine with those flimsy 5mm ish tubes going to the individual units on the hot side. Maybe for a second iteration of this split the cooling system up into groups of like 4 peltiers cooled in series with smaller separate radiators. Also have the water in the cold loop flow the other direction from that in the hot loop to make sure the water doesn't get reheated by the higher temperature peltiers (because they are in series the last one gets a lot warmer cooling water and thus will cool less far on the cold side). I think by hooking up the cold side in groups of 4 (or so) with their own pump and combining the output with a splitter to the CPU plus the aforementioned changes you could optimize this setup.
I feel like in a previous video, it was established that if you want to do this, you should first run the hot water through a radiator to get the temperature down for most of the way and then run it past the TECs to go sub-ambient.
I would love to see a sequel to this that uses the same idea but in a more optimized way.
using a server 12V only server PSU, and maybe using cheap air coolers on the peltier hot side to half the effort on the cooling runs might help!
Then you should watch der8auers Video on cooling with multiple peltier elements ;)
1. Use a few distribution blocks (eg Phoenix PTFIX) instead of soldering 80 wires. Or good old Wagos. You'll hate yourself less.
2. I would do a 5s4p connection instead of 20p to save on tubing effort while having some redundancy.
E: any manifold connection has potential for unequal flow so you need to watch out for that. Big benefit of series tubing is you have the same flow rate everywhere.
5s4p with each string on its own pump would make a whole lot more sense.
But that means potentially trying to remove 600W from a small 40mm x 40mm waterblock. I don't think you can get a fast enough flowrate through it to make that happen.
Add some RGB. You forgot about that.
He could tune the flow rates with flow controls. reducing flow in the lower peltier devices would probably balance the flow up to the peltier devices that are much higher and thus need more head pressure. If you are feeding the same head pressure to each row the top will never get good flow.
@@ShieTar_ Don't underestimate the heat conductivity of water. 600W on 40x40 sounds like a lot but is extremely manageable
Alex was so preoccupied with whether or not he could, he didn't stop to think if he should
That thought has rarely if ever entered his mind.
Alex was so preoccupied with whether or not he cold, he didn't stop to think if he should*
You need a separate res at the top of the loop for this to work. pump at the bottom, res at the top, then you're not fighting gravity. come on guys this is water loops 101!
also the clamps idea was alright, but you can make mini clamps out of 2 sections of dowel and 2 wood screws. then you can make 1 adjustable clamp for each inlet and dial in the restriction perfectly!
Lots of room for improvement on this one. Would love to see a rev 2
There's nothing that makes me happier than a video where Alex makes some crazy contraption. It is very relatable.
The trio we never knew we wanted, but deserve a series of videos with: Alex, Jake, and Dan!
Jake and Alex in the same room is my new favorite thing. Pure nonchalant chaos.
For a manifolded assembly like that you want it to be horizontal rather than vertical so gravity affects each one the same. Then you can have radiators attached to the hot side drawing cool air from the cold tubes on the bottom and the cold side blowing chilled air over the hot tubes on top for a small performance boost in the setup.
these kind of videos are the most interesting you guys put out; just let Alex and the other engineers do whatever crap they come up with, its super entertaining
As a future engineer, I love how overly complicated this is.
As a future engineer, I hate how grossly inefficient this is.
I second this...
As not an engineer.
But knowing something about something.
I 3th this
Also using different hardware for the distribution would've made it more efficient
as a engineer i could do better xD
Why do u think technology is going backwards? Higher power usage, less efficiency, high temperature 🌡️??
As an engineer, this a aprove of this. this video kills me
Dan's reaction was incredible 😂😂
AbOmInAtIoN
Dan has been the best addition to LMG since Anthony.
@@Michael_mki233 he's slowly started to become my favorite LMG member for a while now lol his reaction was priceless
I love Alex's cooling shenanigans.
I never watercooled or overclocked anything in my life and Alex' videos about this stuff are just fascinating, love it!
i think you might get better results if you use a large high reservoir above each loop and let the pumps just pump the liquids up, then let gravity let the liquids flow through the coolers this would probably get equal flow through each cooling element.
that way you could also store some cold liquid in the reservoir and run a benchmark from the stored cooling capacity of the reservoir.
A mad thumbnail for a madlad engineer. Love it
definitely
Alex needs to long beard and long hair it up with a labcoat. Would be awesome.
You need a manifold that distributes to each of the lines with a constant volume flow rate. In other words, the fluid is taking the path of least resistance, like the lower TECs and shorter tubing runs.
This is the right answer.
Restrict the lowers, use separate pumps, or series up several of the TECs
just have loops for few tec's each going to a tank and a different loop from the tank to the cpu.
@@lasskinn474 makes more sense
I failed thermofuids 3 times and even I was frustrated he didn't lift the red above the techs
I played too much factorio/satisfactory to know this 🤣
You gotta connect those blocks on the peltier-s in parallel (at least in blocks). You could do four groups of TEC-s with each group having their own pump and radiator combo so you have higher crossection for the water to flow through. If you connected everything here in series, I would imagine, the pressure needed to push the coolant through the whole system is too much with just a tubes diameter of cross section for coolant flow.
As I understood it, they were connected in two parallel blocks of 10. The frame's left side supported 10 TEC-s, and the right supported the other 10. Front was cold, back was hot. Am I missing something? These aren't connected in series at all.
@@TorgieMadison They all had their own circuit, which makes for a very long pipe in sum. When you group 4, you almost cut that length by /4 -> less pressure is needed.
I mean, yeah. But if you do that and lay the frame down so that you don't need the pumps to lift the coolant 3 feet it would work a whole lot better.
@@VealCalf1 If you make yourself familiar with these particular physics you will learn that the relation of pressure needed to length of the pipe is not linear. Spreading out the pumps to different positions in the circuit could help with that.
You do know you can get 40x120 aluminum water blocks so you can reduce you hose usage by a factor of nearly 4. J/s Hilarious fun build though! 🤘🏼🤘🏼
You should connect the blocks in series. So you solve the pressure problem and equalize blocks temps.
Agreed. Would've made it really easy to bleed the whole system
Hot side in series would only make most of them non-functional as they'd be "cooled" with hot water (heated by earlier ones). OTOH the "cold side" in series would freeze the antifreeze, as the cooling effects would keep stacking.
@@muffinconsumer4106 Would require a lot of smaller radiators, couldn't use these big liquid cooling radiators they did.
It's that time again, janky water cooling with Alex, I love it, these are my favorite videos LTT makes, and I hope the series never ends :)
It’s great that they always happen at the onset of the new generation. Then legit coolers come out, then the next generation changes it up again and the cycle repeats
Time for my daily dose of absolute certified craziness presented by Alex.
Always enjoy mad Alex engineering.
You definitely need to revisit this with debauer's delidding tool and whatever tips were on the GN N2 overclocking stream!
Also great seeing you guys host this content without Linus. It amazing how everyone has grown their presenting skills over the years!
Alex is by far becoming my fave LTT member
again the scariest things an engineer can say is "it should be fine" or "it will be fine"
I did the same but in smaller dimensions (about 400W). I wasn't happy with the results. So i tried the outdoor mod. Placing my 9x120mm radiator outdoors at about negativ 5 degrees celsius improved the result greatly. The cool side of the peltier element was extremely cooled and the cold side could get insanely cold (not gonna spoiler, try it out its crazy!!!. Its easy to realise If mother nature provides you with the right circumstances. Shouldnt be a problem in Canada i guess. I would love to see you stepping the project up with the "oudoor mod".
Sorry for my terrible english and greetings from Austria
Austrians and their cooling solutions smh
I used to have the "Ultra" brand peltier tower air cooler hybrid. It actually worked very well. With how it was set up it kept the CPU at one temperature when under load, didn't cool it too much or too little.
Gimmicky, yeah. But it was cool as hell and still did a fairly good job. Not the best, but good.
This idea takes that to the max. I like these ideas lol
I love how positive Alex’s energy is
He loves cooling of course he's a chill dude
You have to create strings and parallel streams with your cooling liquid. The way have it now you would need an open reservoir which presents NO pressure at all. Best way to to this would be to have the water coming from top into the reservoir like a small waterfall. With that much parallel water flow you actually hampered the whole performance. Also add some automatic air reliefs at the top to get the air out as fast as possible.
We have the same concept in out town water system where I'm responsible of. You need strings and parallel flow correctly done or else you have no flow at all...
POV: You have the Linus budget and a crazy idea. But it somehow works at the end.
I've never gotten cracked up by these sentence but when Dan came up to them and said "What the fuck is this abomination?", "What are you cooling?" and "What are you heating with this?" all at the same time I audibly laughed
Dan's great
Alex must be having the time of his life because we all know weird ideas pop into our heads but everyone get the chance to implement them.
This was so insane and felt fast-paced, I'm amazed that this video exists
I love this idea. Also Dan coming in was gold.
Alex is the best learning source of how to do water cooling, by pointing out ALL of the "don't do this" examples. A true master of his own craft.
Stuff like this is why I'm subscribed. By far my favourite clip from LTT in ages.
Just a pity Linus wasn't there to drop it.
I'm here for more of Alex's off-the-wall cooling ideas! As in, he comes up with the ideas by "throwing shit at the wall to see what sticks", then makes videos out of what does /not/ stick.
I can’t believe you guys are doing ANOTHER peltier cooler video right as I’m designing a stupidly crazy design for a giant observatory telescope chilling system. So thankful!!
Please don't use Alex as a starting point. He is making so many stupid mistakes. Main stupid thing was that he did not place the header tank on the top of the loop.
@@Running_Wakko don’t worry, we aren’t using this is a guide! Our system is going to be far more complex as it has to work in all orientations. The water blocks will be changing orientation as the telescope mirror moves around, so we have to account for everything. We’re gonna build a working model before we even try out the real thing.
This was great, we need more Alex and Jake making amazing jank
Thank you Thank you Thank you LTT, 1 wk ago it was so close to buy the 1/10th of your test parts on Aliexpress, still try to convince myself, then you did it for me, saved he heaps effort, and made my mind to keep standard water cooling. thank you again...
Honestly, I love these types of videos, even if the result doesn't exactly pan out! Sometimes you get a wild and crazy result, and we all benefit by learning!
Alex: Something is not okay with this computer
*Also Alex making Alien stuff just to cool the CPU*
alex is easily my favorite person to see leading a project
If i was Linus i would be so proud of Alex and his mad ideas.
Also he got so much better in front of camera, i'm so happy to see him do this kind of stuff with this confidence.
Go Alex!
Okay this is way cooler than I was expecting. You need a better thumbnail.
Alex's introduction to tecs and cables has probably taught me more than the 8 years of physics class I had so far
what?
I think the pumps needed to be split up among the different sets of coolers, so a pump and a "one in five out" manifold per level of that tower. Getting the water/coolant to cycle properly through all the blocks would probably make a huge difference, and if I understand how they work correctly, may actually get the TECs to draw more power. The reason for that being they have temperature limits on the hot side, once it reaches a certain point it just can't push more heat across, and will drop off.
Those tecs need 17v, they were giving them 12v. So the devices were underpowered, thus under delivering.
I love these engineering videos. Please keep them going in any way you can.
I really was hoping they'd go all out and like machine a whole water block manifold for the in side and outside. I would love to see a rig they make that actually has some bananas cooling and could be used later on for other experiments. Kinda like their air conditioner cooler.
The only thing missing is the "What the f* are you doing and HOW MUCH DID I PAY FOR IT?" Linus cameo.
Yes! 😂
Should have had those tec modules in series and parallel behind each other cooling the hot side of the other tecs for a multi-stage
I think my Prometia cooler only drew like 800w and could keep a 500W load at -20C. I just got tired of dealing with condensation, so went with water chiller that keeps the loop at 1C above ambient.
0:43 oh-oh, watch me **Double-tap**
This looks like a brain from a bad sci-fi movie from decades ago. It's lovely! So when do you design a case that the wonko cooler mounts to and call it "The Brain That Lives!" and make your own mini movie for Floatplane?
I keep coming back to alex's videos. They are always so full of controlled chaos that they are the most fun videos of this channel.
Exposed wiring, lots of water everywhere. Loving it.
i think you should try improving this. you should mount the pump/reservoir up high to prevent air lock and pressure issues if you want to keep testing the in parallel like you are. i would recommend each row of coolers in series and then have the rows connected in parallel. it would really force the coolant to circulate through the coolers and likely require less force and make the water much cooler.
actually u just write my toughts 😅😅😅
Exactly, never understood why most liquid cooling setups in typical computers have the res down lower either, should be the highest point - even higher than the rad - makes it much easier and would automatically deaerate
We are slowly getting to a point, where our CPU coolers will need their own coolers.
@Ahmed Jones but If we get a cooler for the CPU cooler's cooler, then we'd also need another cooler for the cooler for the CPU cooler's cooler.
I had Two suggestions, to save you time and it is cheap Alex, Thread sealant paste is very inert works better then Teflon tape. I used it a lot to test 3-15lb and 4-28lb pneumatic actuators, pressure transmitters monometers for an old big boiler. My other suggestion is from when I worked for an electronic tool co is "haisstronica Heat Shrink Butt Connectors". These Connectors are Used with a heat gun and leave a marine grade electrical finish. They are fast and fairly cheap, heat guns are faster then soldering. you can also line them up and hit several at once. Good video, pumps go high in industrial loops unless it is hydraulic whole setup might cost $100 but for 50 threads its about 1/2 the time. I'm not sure what Alex gets paid but my guess is it is cost effective
This is absolutely insane and im so glad you guys made it. 9:21 best part
These guys could never be plumpers! Or anything else than UA-camrs! 😏
Jake getting sprayed after asking if he can just crack it real quick had me cackling
I can't look away from your cooling ideas, just like you said.
I love them.
Please, keep doing what you do, Alex. It's great entertainment!
love alex's strange, wierd and amazing cooling ideas! best one yet!
I SEE IT.
He has the engineer eyes now.
You know, the sort of sunken, passion deprived gaze that is earned by answering emails and attending team meetings instead of doing ACTUAL work.
Congratulations.
It is not just a pressure issue, it is a flow issue too! You need to assess the temperature drop of the coolers and then use Q = M.C.DT to work out the total flow required.... then assess the flow of the pumps against the head pressure of the assembly.....
Only time I've used a peltier was for a DIY dehumidifier, they're terribly ineficient, just don't use them.
Alex, Jake and AliExpress in one Video.
And its exactly as insane as expected!
Alex and Jake are a funny team doing Frankensteiny stuff. More please!
the amount of chaos in this video is incredible. Im down for more alex and jake videos like this haha
12:14 "The D5 pump loosener has escaped prison and the police are on hunt to find him!"
Alex is the mad scientist of LMG 😂
I think with the new AMD cpu and Nvidia gpu all the time the user will be thinking about the Electricity bill
'Borrowing' from a friend.
then again, you wouldn't need heating at all during winter, so the expenses might cancel out
Peak Alex. Best addition to the team. Been watching since the days of just Linus and luke and having Alex onboard feels like a breath of fresh air.
from how I understand piping in general (I am not an expert) the reason there is so much air trapped is because they don't have any pressure release valve located on the highest point of the loop so the air just gets trapped on whichever place gets the water flow last... it's basically a one-way valve that traps liquid below it while release any air upwards without giving it a chance to get back into the loop (I think the show The Good Doctor made a decent demonstration of this)
NGL this thing looks like one of those steel bearing musical machines, but for computers.
Bloody fantastic! Alex has the best and craziest idea's, the things we have all thought of doing but would never do, But he pulls them off. Love it
Alex, Jake and Dan doing mad science cooling. Love it!
You should have used the 'Tichelmann-System' for tubing. That's a plumbing/piping technique.
This way you would have the same resistance in every loop so the Water would spread properly between all loops.
build one and post the video…
@@macking104 hm?
Dan reactions is so funny
what the fuck is this abomination?
it's a work of art
The way you do this is to buy larger water blocks. You can get blocks that are longer so that you can attach 4x or even 8x pet's per block.
The correct way to do it is to add a pet inline wired up to an inverted furnace thermostate so that you can temp controle power on and off the pet's.
Best way to go sub ambiant is with geo thermal heat sinking though. Your coolant should stay a constant inline temp between 10-15 degrees.
Heres is another hint people condensation requires a temp differential based on humidity in the air. So if your pc was in an airtight humidity controlled enviroment like a mineral oil case or a freezer you can eliminate the opertunity for condensation to occur.
9:26 I bet Lulu wants a taste then
16:06: I like how they black out the password, like anyone wants to break into their warehouse, ignore all the expensive equipment, and take a motherboard covered in vasoline 🙄
Jeez, & I thought the RTX 4090 was a power guzzler.
Best combo, Alex‘s wacky builds an Jakes jokes and ideas go a long way for great entertainment!
14:30
Thank you for such a selfless sacrifice to bring us such precise scientific data