News Pumpless 240mm AiO cooler is only 8C behind traditional pump-powered rival in tests

Toggle sidebar Toggle sidebar
Status
Not open for further replies.
ezst036 said:
I would like to see this tested fanless, and pair it up with 35w to 65w processors. Or even less than 35w if such processors are available.
Click to expand...
Is it feasible to use a radiator without fans? I didn't find anything about the matter. I'd suppose you need at least a bit of airflow unless you manage to put it outside the case and use convection.

For a 35W processor I suppose a regular tower cooler would be okay
 
I can't hear any pump noise from my Corsair H150i even set on extreme with all fans off unless my ear is inches away from the block, same with my previous two Corsair liquid coolers, so noise isn't (or shouldn't be) an issue for AIOs, but eliminating the pump does remove one of the only two limiting factors of AIOs, the other being the (very) slow loss of fluid over time.

It will be interesting to see how it scales with smaller 120mm and larger 360mm radiators, and how tolerant of fluid loss over time it is. If it matches or exceeds pump driven AIOs in those respects, it'd be worth paying a slight premium for since there have been multiple issues in the past, and a couple of lawsuits surrounding AIO pumps for patent infringement.
 
  • Like
Reactions: The Historical Fidelity
Alvar Miles Udell said:
I can't hear any pump noise from my Corsair H150i even set on extreme with all fans off unless my ear is inches away from the block, same with my previous two Corsair liquid coolers, so noise isn't (or shouldn't be) an issue for AIOs, but eliminating the pump does remove one of the only two limiting factors of AIOs, the other being the (very) slow loss of fluid over time.

It will be interesting to see how it scales with smaller 120mm and larger 360mm radiators, and how tolerant of fluid loss over time it is. If it matches or exceeds pump driven AIOs in those respects, it'd be worth paying a slight premium for since there have been multiple issues in the past, and a couple of lawsuits surrounding AIO pumps for patent infringement.
Click to expand...
I used a cooler master AIO for a gaming build I made for a friend and that pump was so loud I had to turn down the pump speed to make it bearable. I have always used Corsair AIOs in my own builds and I agree with you that I have never noticed noise coming from their pumps.
 
As far as I understand it, Thermosyphon technology is essentially a heat pipe that uses a separate return channel. I assume that makes it somewhat orientation-dependent. I wonder if it works better, when mounted vertically.

The Historical Fidelity said:
Taping off one of the fan sections does not make a 360 rad behave like a 240 rad
Click to expand...
You can see the radiator is clearly 240 mm. I'm pretty sure text was referring to taping off the extra fan port of the case that would be needed for a 360 mm radiator. I don't know why they even mentioned that part, unless using the mounting position for 360 mm radiators is somehow less advantageous (i.e. more centered over the CPU) than the normal 240 mm position.
 
bit_user said:
As far as I understand it, Thermosyphon technology is essentially a heat pipe that uses a separate return channel. I assume that makes it somewhat orientation-dependent. I wonder if it works better, when mounted vertically.


You can see the radiator is clearly 240 mm. I'm pretty sure text was referring to taping off the extra fan port of the case that would be needed for a 360 mm radiator. I don't know why they even mentioned that part, unless using the mounting position for 360 mm radiators is somehow less advantageous (i.e. more centered over the CPU) than the normal 240 mm position.
Click to expand...
According to the article, he used an H150i 360mm AIO to compare temperatures. But I share your question about orientation restrictions, etc. if it is too restrictive, that will destroy its potential market share.
 
  • Like
Reactions: bit_user
I'd like to see this tested against the IceGiant. It's also a thermosiphon, but built into a beefy heatsink that bolts directly over the CPU and is designed to work in two specific orientations; vertical (length-wise) like it'd be in a standard case, and horizontal, like on a testbench. It's also roughly analogous to a 240mm radiator.
 
  • Like
Reactions: bit_user
terabite said:
Is it feasible to use a radiator without fans?
Click to expand...

I have no idea, I doubt many people have given it a try. Since this one has some sort of built in motion, it might work. Hopefully a real review would delve into the matter.

Yes, 35w processors work just fine fanless with (some, not all) tower coolers, mostly those coolers on the larger side. My curiosity ran a little more toward the 65w side.
 
Last edited:
Watched the video this morning and couldn't help but wonder if this will become a product. They definitely need to do more work on the coldplate and likely the radiator as I'd imagine having bulges on both sides will lead to potential mounting issues. If they can match the performance of a high end air cooler without the cost being a lot higher I could see the benefit. I'd consider getting one for my 24/7 system where I'd never really consider getting an AIO or using custom liquid cooling.
 
thestryker said:
They definitely need to do more work on the coldplate and likely the radiator as I'd imagine having bulges on both sides will lead to potential mounting issues. If they can match the performance of a high end air cooler without the cost being a lot higher I could see the benefit. I'd consider getting one for my 24/7 system where I'd never really consider getting an AIO or using custom liquid cooling.
Click to expand...

I'm pretty sure the final radiator design will look more like the crossflow radiators from Alphacool, with integrated tanks on both ends also acting as air dams to prevent air from leaking out the sides as in the prototype. Whether or not they'll also have the extra ports depends on if those were for vacuum/filling purposes or just part of prototyping.

I also do have to wonder if performance can also scale with radiator thickness as an alternative to adding another ~120mm or so of rad length. The IceGiant prototype was extra wide before being slimmed down to where it is now, but the core finstack is still pretty thick compared to a regular AIO radiator, being closer to the 35-45mm thick radiators available.

As well, I'm curious if the fluid dynamics involved allow the radiator to also work in a vertical (length-wise) orientation (like the front mount and side mount) in addition to horizontal on the top mounts of the case, or if the "engine" that drives the thermosiphon is located in the CPU block itself.
 
  • Like
Reactions: bit_user and thestryker
While 8 degree C don't sound like a lot, its actually threading into the cooling performance of air coolers. A good twin tower air cooler may even perform as well as some 240mm fan. So why do I want to buy a AIO liquid cooler that may potentially cool no better than an air cooler, that costs more, and run the risk of leaks of evaporated liquid that you cannot replace easily?
 
  • Like
Reactions: yngndrw
terabite said:
Is it feasible to use a radiator without fans? I didn't find anything about the matter. I'd suppose you need at least a bit of airflow unless you manage to put it outside the case and use convection.

For a 35W processor I suppose a regular tower cooler would be okay
Click to expand...
Yes, as long as you have enough surface area. I used to use a fanless PA120.3 with a P4 back in the day. It ran fine even while playing Crysis. I think that loop also did my GPU, but I can't actually remember which GPU I had back then.
 
newtechldtech said:
This idea is stolen , Raijintek made the first prototype of it back in 2016. This is what happens when you dont patent your invention !

link

https://www.anandtech.com/show/10398/raijintek-shows-off-pumpless-liquid-cooling-system

View: https://www.youtube.com/watch?v=HqFLykiWbu4
Click to expand...
Thermosiphons existed before Raijintek existed as a company. They were not the first to show off a thermosiphon at a trade show, and far from the last.

The big problem everyone has to deal with is working leakage vs. working fluid thermal performance. Regular copper heatpipes (monolithic thermosiphons) work with water at low pressure. They can do this because the copper is nice and impermeable over regular human timescales, so you can use water (excellent thermal performance) and not worry about it leaking out or external atmosphere leaking in.
When you switch to nice long flexible tubes, now you either have the problem of atmosphere leaking in (at which point your water-based thermosiphon will rapidly see its boiling point rise and eventually hit 100°, which is also the temperature the heatblock will be at) or you use a low boiling point fluid like the various HFCs used in immersion cooling (e.g. Fluorinert), which do not require low pressures. However, these also have the problem of fluid and vapor leaking out - both through more-permeable-than-you'd-prefer tubing and through degrading o-ring seals - but on top of that the thermal performance of these fluids are far worse than water (both specific heat capacity and enthalpy of vaporization).

For specific applications where the position between the heat source and heatsink are both known and fixed, these problems can be sidestepped by using a hard sealed system (all copper or other metal, no non-soldered/welded joints). And in these applications, thermosyphons work well, e.g. spacecraft, military hardware, etc.
But think about how much of a hassle it would be to try and sell what is effectively a welded together hardline watercooling setup, let alone try and fit it!
 
  • Like
Reactions: bit_user
edzieba said:
Thermosiphons existed before Raijintek existed as a company. They were not the first to show off a thermosiphon at a trade show, and far from the last.

The big problem everyone has to deal with is working leakage vs. working fluid thermal performance. Regular copper heatpipes (monolithic thermosiphons) work with water at low pressure. They can do this because the copper is nice and impermeable over regular human timescales, so you can use water (excellent thermal performance) and not worry about it leaking out or external atmosphere leaking in.
When you switch to nice long flexible tubes, now you either have the problem of atmosphere leaking in (at which point your water-based thermosiphon will rapidly see its boiling point rise and eventually hit 100°, which is also the temperature the heatblock will be at) or you use a low boiling point fluid like the various HFCs used in immersion cooling (e.g. Fluorinert), which do not require low pressures. However, these also have the problem of fluid and vapor leaking out - both through more-permeable-than-you'd-prefer tubing and through degrading o-ring seals - but on top of that the thermal performance of these fluids are far worse than water (both specific heat capacity and enthalpy of vaporization).

For specific applications where the position between the heat source and heatsink are both known and fixed, these problems can be sidestepped by using a hard sealed system (all copper or other metal, no non-soldered/welded joints). And in these applications, thermosyphons work well, e.g. spacecraft, military hardware, etc.
But think about how much of a hassle it would be to try and sell what is effectively a welded together hardline watercooling setup, let alone try and fit it!
Click to expand...
all physics existed before all ideas . using it in PC in a special way is called engineering . and it can be patented.
we had small water pumps before Asetek as well and this did not stop them from patenting their pump inside AIO. and so on !
 
newtechldtech said:
all physics existed before all ideas . using it in PC in a special way is called engineering . and it can be patented.
we had small water pumps before Asetek as well and this did not stop them from patenting their pump inside AIO. and so on !
Click to expand...
That's a nice Gish Gallop, but completely unrelated to thermosiphons.

Other prior art includes the Akasa Revo from 2007, Innovation Cooling showed of a prototype in 2005 but never brought it to market, etc. And that's just in the consumer space, industrial and datacentre use stretched back much further.
 
  • Like
Reactions: bit_user
newtechldtech said:
Both Akasa and Raientik were wrong for not patenting it ...
Click to expand...
You can't patent obvious applications of an exsiting device, and "using a cooler to cool something" is an obvious application of a thermosiphon. As per 35 U.S.C. 103:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 , if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains.
Click to expand...
 
  • Like
Reactions: bit_user
edzieba said:
You can't patent obvious applications of an exsiting device, and "using a cooler to cool something" is an obvious application of a thermosiphon. As per 35 U.S.C. 103:
Click to expand...
yes you can . for example Asetek patented its pump above the water block in their AIO , and thats why ALL others either pay Asetek for every unit sold using that method or put the pump near the radiator or in between the radiator and the water block (on the tubes) to avoid lawsuits from Asetek. .

The AIO cannot be patented as a water loop + pump + rad , but the design can be patented .

and by the way AKASA cooler and Raijentek coolers are using same concept but different designs BOTH can be patented at the same time , for the design is totally different between the two while using the same physics.
 
newtechldtech said:
yes you can . for example Asetek patented its pump above the water block in their AIO , and thats why ALL others either pay Asetek for every unit sold using that method or put the pump near the radiator or in between the radiator and the water block (on the tubes) to avoid lawsuits from Asetek.
Click to expand...
A patent that is widely disputed, and has already been rendered invalid outside of the US.
newtechldtech said:
but the design can be patented
Click to expand...
A Design Patent is a completely different beast to a regular patent.
 
  • Like
Reactions: bit_user
edzieba said:
A patent that is widely disputed, and has already been rendered invalid outside of the US.

A Design Patent is a completely different beast to a regular patent.
Click to expand...
and most want to sell inside USA ... which is a huge market. no company ignores the US market.

Even PC case fans that are slot in and without wires is patented do you know that ? That stopped most case makers from making hot swap fans .... few years ago here on Tomshardware some one asked coolermaster why they dont make hotswap fans without the need for all these wires and for easy maintenance , they replied : it is patented !

https://www.tomshardware.com/reviews/cooler-master-ama-toms-hardware,3550-7.html

Q. Why don't you make all the fans on the case wireless?

A. This is a great idea; unfortunately it is patented by one of our competitors.
Click to expand...

and this is a simple thing and yet patented ! I guess all servers cases with hotswap fans are paying for the one who patented it !
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom