Linus Tech Tips makes entertaining videos about technology, including tech reviews, showcases and other content.
![[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 1](https://img.involve.asia/rpss/campaigns_banners/71956-eFC6Ck0kcofg71gNAWcu2SL8u8wogWnw.gif "[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 1")
![[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 2](https://img.involve.asia/rpss/campaigns_banners/68745-vDBjuNX2DjCpCalfkVHjRdeyUHGS3ehC.png "[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 2")
![[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 3](https://img.involve.asia/rpss/campaigns_banners/72005-XObolGg4VdCs0xHS6Ko5Udeq3PmITXiC.jpg "[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 3")
![[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 4](https://img.involve.asia/rpss/campaigns_banners/1667182864-2b73acS1d8qsyJgUPCe8OS7VN5v5CEmc.jpg "[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 4")
![[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 5](https://img.involve.asia/rpss/campaigns_banners/1671605925-qml0ZN4Rof9UcyAheIjCIRwcutuiWajE.jpg "[Video] I’ve been water cooling wrong for YEARS - $H!T Manufacturers Say 5")
[Video] I’ve been water cooling wrong for YEARS – $H!T Manufacturers Say
Linus Tech Tips makes entertaining videos about technology, including tech reviews, showcases and other content.
The Corona Lockdown Beard suits you. No more boyish looks 😜
Have radiators outside case, no heat from PC.
Put on back side with spacer feet static pressure fans that way you can have one two or three in the back
I don't think Corsair's problem wasn't necessarily having more than one radiator though. The fact that the hot air from the first radiator is the same air that's used to dissapate even more heat into is the main problem. Even though removing one increased temps by 5 degrees, therefor showing that it isn't completely useless, doesn't prove them wrong because the difference is way smaller than you'd expect from having double the radiators.
IC cooling is going to be obsolete when they come out with ultra large scale integrated graphene coplanar waveguide chips
Something not clarified in this video: direction of water flow. In this situation to run the water through top radiator then front radiator would be the best: First cool the water then cool it more with fresh cool air…
Wo wo wo new intro♥️
Well, for my case, it was fan placement. Many people said the best configuration for fans at the top of the case to be exhaust. I tried that (since my case can fit 3 120mm fans) it did create a negative pressure since at the front, I only have 2 fans (because I'm still using Blu-ray Disc drives) and the rear 120mm fan is the exhaust. Temperature is… okay, but the system is still warm because the air expelled from the graphic card actually get sucked into the CPU (using stock cooler, which is a radial heat sink with a down draft fan. Once I switch all my top fan to intakes, I get better temperature for my CPU, and it can sustain longer before thermal throttling.
So, lesson learned. Do lots of experiments. Many suggestions do have a valid point, but sometimes, it just won't work in your scenarios.
IMO radiaters should be mounted outside ones case as heat should be kept far away from sensitive components. I have a large but crappy Cooler Master CM Storm Trooper case. Have a Corsair AIO water cooler for the cpu where the rad sits at the very top of the case. Then about 5 cm below there is a Noctua fan blowing the hot exhaust air from the rad out of the case. (maybe the rad should exhause hot air straight out, don't know.) Would like to water cool my 1080TI but can't figure out how to externally mount the rad or how to place a fan/rad combo to exhaust the rad hot air, Possibly at the bottom with the hot exhausting outwards. But from having done our cottage plumbing using compression fittings I know, painfully, that my plumbing skills leave a lot to be desired.
Corsair: Trying to persuade people that their products are good enough to cooldown your pc with only one radiator.
Linus: "No! Just buy it, more is better."
Something not right……
So many people in this comment section are giving Linus shit, when it's super obvious they didn't actually watch the entire video.
But stacking the rads like in the minecraft server would not allow fresh air to mix.
I would think that the surface area of a 240 rad isn't nearly enough to saturate your setup. Adding a second 240 rad increases the surface area and still gives you a benefit even though hot air from the first rad blows through it. If you instead put on 2 480 rads the first 480 should be enough to take most of the heat out of your loop, the second 480 rad would then possibly not really take any heat out of the system at all and just add unnecessary air friction.
You probably could’ve handled this better
Those rads werent stacked. You suck
A watercoold radiator cooling the incoming air to the case. Whit all incoming air flow seald of except for true the rad. Combined Whit a big air cooler on the CPU. How would that cool in comparsion. The ability to put all Ester cooling components except the rad outside the case should decrease the ambient tempo in the case. No friktion heat from pump and pump motor in the case and No hot waterpipes in the case. Have Always thougt cooler masters cooling tower was a great idé in that regarde…
Laughs in 32°C
Can you watercool a home? Summer is coming!
LINUS you know damn well what they meant… stacking radiators is stupid, heat from one gets recycled to the next which diminishes the difference between temps and minimizes the effectiveness. Radiators on the absolute top seem to be less effective as well… and you don't want exhaust one radiator inside the case and have another radiator sucking in that air. The point is to get the heat AWAY from the system, not recycle it to ambient around the components OR allow it to be partially reabsorbed into another radiator.
And then, after corsair some smart person emailed him and told "dude you look like 1:23, but you can look like 1:22". I'm glad you didn't ignore that too)
Corsair's recommendation may not hold true for a standard PC build, but you didn't test it with the server thing with multiple rads one after the other where it would hold true. That's a situation where you've worsened your mater cooling performance because the last rad will only get hot air from the rad in front. Worst case scenario.
Corsair is right on this one. I can't believe that Linus can't understand that?
I'm sure someone else will comment on this, but here it is anyway. All the best hot rod cars, drag cars and even race cars use "triple flow" radiators. That is, the water turns behind itself twice, to form three walls of water, with one common air flow. This does increase heat capacity, and you will have seen what that means to heat soak, BUT it also allows great "packaging" and that's what we're really talking about here. Optimised cooling VS packaging. If you want to fit cooling inside a case, multiple rads and lots o fans are your friend! You generally want the coldest air to touch the coldest water, i think that's called contra flow (it's been a while). The type of rad will affect this decision. Hot water will cool "better" on cold air, but, IF you have the capacity and thermal window, cold air will make the "cold" side, colder, so keep the system from heating as much. Ambient, surface area, air and water flow (as well as peak temps, temp difference and thermal input power) will change how you choose the config. For shits and giggles, try a control experiment. Do two external "bench" rads with the same fans (always the same number of fans). One test with serial rads, giving the cooling a "longer" pathway for the total surface area (this should work well for low temperature difference) and do a second test with the two radiators in parallel, using splitters to separate and join the water flow. This "slows" the water flow, allowing more time in contact with the rads. I think this works best in high temp difference scenarios (depending on your target cooling temp too? ). Then compare this with stacked rads, outside then inside the case. If you're super bored, try push-pull fan configs vs push-push. Short story, you work with the packaging you want to achieve. I'm sure we've all built that franken-cooler with three rads hanging from the outside of the case. Nothing beats fresh air. Also, one should always consider higher static pressure fans for rad cooling since they shouldn't suffer the flow losses through the radiator(s) AH! one more fun experiment: stack fans at the hottest part of the rads and see how that affects the numbers.
JayZ2cents has done tests on the video like a year ago. this video is puke-inducing.
Why not bench the Minecraft server (or similar) with 1 rad, 2 rad, 3 rad and compare the differences?
It seems though that this doesn't really test the radiator's efficiency at dissipating heat into the air. Having multiple radiators will mean that there is more radiator surface area for the water to dissipate heat into, so you have a greater overhead for cooling the water. But does the water continue to cool effectively when the radiators become saturated with heat? I think you would have to do a very long test that saturates the radiators with heat before you would notice if an air dissipation bottleneck is preventing your CPU or GPU from getting cool.
I'm also surprised that the fancy thermal camera didn't come out for this video. Why didn't they check the surface temperature on the radiators?
Still linus they are right. Its thermodynamics linus, of you stack you wont be gaining as much as you think.
There is a strong correlation between airspeed and temperature difference.
The more temperature difference there is between both air and water, and the more air velocity, the more efficiënt the radiator is.
Grain of salt and all i mean sure, but this is coming darn close to saying that science is wrong because we did something.
Like the laws of physics and thermodynamics dont change based on your feelings.
You did clarify a few points; not a perfect world, secondary air from other sources, and space restrictions leading to the inability to optimize cooling performance, so ill respect that
My issue is the format of this video is purposefully hostile when it could have been very educational
Hi do u still do those videos were u get sponsored by like asus or rog and u get a kid to build it and u secretly but jokingly make fun of them bc they didn’t spend their lives studying pc’s bc id like to be on one of those
…and everyone suddenly finds out why Air Conditioners work harder, are less efficient, and fail more during the summer heat. it's all about the deltaT. Science says you can't add cold, all you can do is add heat, (subtracting is a silly concept here.) so you need to find a cooler environment to add your heat too. Hot CPU to cooler liquid, now warmed liquid to cooler air (stacked now adds one pass cooled liquid to warmed air?). Instead of stacking, try isolating the fresh air into and out of the radiator from the other components.
"I'm not saying that their cfd analysis or their testing is wrong or doesn't make sense …"
But you kind of are. It's in the title. "shit"
So BOTH parties Linus and Corsair are BRAIN DEAD SCIENTISTS! 1) What Corsair did wrong was forget about the water temperature heat gradient inside the block is the primary heat transfer surface (ie temperature gradient), and radiators are the second. 2) Linus also DID NOT COMPARE 2 radiators in PARALLEL both inputs which was the WHOLE POINT. THAT WOULD HAVE BEEN A FAIR COMPARISON. Instead he stubbornly compared 2 radiators with a second one being VERY inefficient (HOT EXHAUST TAKING HEAT FROM THE FIRST RADIATOR AND THE MB) and compared that with only one radiator. CONCLUSION we compared apples 2 RADIATORS in SERIES to frogs 1 RADIATOR UNFAIR UNSCIENTIFIC BOTH OF YOU! YOU FORGOT ABOUT THE DUAL INPUT DOUBLE RADIATOR TEST THAT IS F&%$ MISSING!!!!!
I like you’re beard looks nice
If the warmest water is ran through the warmest rad first then cascading through the stack to the intake rad where the coolest water sees the coolest air would be best practice.
Ok, thanks. When is that water cooled chair is coming …?
I have watercooled my Xbox for years, best descision in my life.
Sorry, Corsair is right and Linus is wrong. So many EE's and no mechanical engineers leads to stupid conclusions that stacking heat exchangers is "better". Only if better means thats it is possible to eject additional heat from the system and accepting the fact that it is always, always done with a massive loss in heat exchanger efficiency. So, do yourself a favor and buy a better case.
What if i use both radiator sets to pull ambient air in and exhaust at the back only? This setup should reduce the temps further, at least in theory.
Let's all cool down. No need for heated debates or to debit what brain freeze yields. Cooling (in omnibus formae) is a science in and of itself. Anyone lacking profound knowledge and near full comprehension of the matter shouldn't even consider tinkering with it. Marketing has got you all by the throats. I would like to call upon Alex The Greatish (and associates) to produce a one hour, in depth, comprehensive introduction to cooling—as the starter to a series of at least 5 more, shorter episodes. Heat your brains and increase your knowledge, not your liquidity problems.
you need to make a pie chart of water temp, ambient temp (the ambient temp of the air entering the radiator,) and position.
If you can add a second radiator whose air is cooler then your water temp you win. If the temp of the air you move over the radiator is hotter then the water, you add heat. If you can say, monitor the temp of the water coming off the water block for your processor, and the air moving out the top of the case, if the water is hotter coming off the block then the air flow out the top after impeding the air slightly, you still win, if not you loose.
In other words, every case is different, this is engineering a whole system. Stacks loose efficiency and gain resistance and heat soak. Also the delta between every stack gets worse with regards to the "ambient air" feeding each on the line.
Flow path could help that though. Feeding the rear most one first, so the hottest water gets cooled by the hottest air, feeds toward the front most getting cooled by the coldest before it heads out.
If you feed the front and come out the rear you are baking your cooling rack with itself.
id say for a dual rad set up it would be more efficient the corsair says it was done if the warmest air went into the exhaust rad first to pre cool the loop then into the primary rad to finish cooling the loop. this is done in chill water systems anywhere there is a demand for that type of heat exchanger.
what about the new nvidia tech they just unveiled ??
Isn’t it because of thermal mass?????????????? Tell me if I am stupid or not.
Nobody: me: the 13 year old in the comments who already knew the answer 😐
I'd assume that the air going through a radiator doesn't gain all that much heat, especially if you have good fans pushing it through quickly. As such, that "hot air" is most likely only "warm air" and still has plenty capacity to take heat from the second rad. Not that there isn't diminishing returns, but there are still returns.
Here's the the thing, if you have ambient temp cooling the water on a rad, the waste air is going to be cooler than the rad because it's function that involves the water and air temps. So the air going through the second rad will still be cooler than the water in it, always, sure the deltaT might not be as much as the first rad but it still is going to work. Also yeah if you have other fans bringing in cool air then you have a situation of WTF simulation were you running Corsair!?
By adding a second radiator you also drastically increase water volume. Which also helps dissipate more heat and likely counters the effect described by corsair.
Just something to consider.
you said smth like sagway t… then the sponsor i think taran messed up😂