h240x vs h110 RESULTS!!!!

[hunkeelin]

So my h240-x came in 3 days ago and I am very excited to see the improve upon my h110.

I have to note, I am not using stock fans for this experiment. All Fans are replaced by noctua sp140 running at max 1500rpm in a pull configuration.

Results:

The h240-x is around 1~2c better. Let me explain in details. When I run real bench 264h test. With the h110, my cpu will hit 72C in 5 seconds, while the h240-x will be around 68~69. So However, after running 5 test. The max temp for h110 is 74 while the max temp for h240 is 73. A 1 C different.

Btw all test are tested in the same temp. I have a heater in my room that keeps my room at 75f constantly.

Conclusion:
It seems the difference between h240-x and h110 is not the copper vs aluminum radiator, nor the custom grade cpu water block vs AIO h110. It is the pump and the higher amount of water that came along with h240-x. It cycle water faster and make it easier to dissipate heat in the radiator. However, when the load test ran long enough the temp of all water in h240 is evenly spread; thus the swifttech yield similiar temps compare to h110.

TLDR; it takes a more time for h240-x to reach max temp, however at max temp h240-x is similar to h110.

 

[RampantAndroid]

So my h240-x came in 3 days ago and I am very excited to see the improve upon my h110.

I have to note, I am not using stock fans for this experiment. All Fans are replaced by noctua sp140 running at max 1500rpm in a pull configuration.

Results:

The h240-x is around 1~2c better. Let me explain in details. When I run real bench 264h test. With the h110, my cpu will hit 72C in 5 seconds, while the h240-x will be around 68~69. So However, after running 5 test. The max temp for h110 is 74 while the max temp for h240 is 73. A 1 C different.

Btw all test are tested in the same temp. I have a heater in my room that keeps my room at 75f constantly.

Conclusion:
It seems the difference between h240-x and h110 is not the copper vs aluminum radiator, nor the custom grade cpu water block vs AIO h110. It is the pump and the higher amount of water that came along with h240-x. It cycle water faster and make it easier to dissipate heat in the radiator. However, when the load test ran long enough the temp of all water in h240 is evenly spread; thus the swifttech yield similiar temps compare to h110.

TLDR; it takes a more time for h240-x to reach max temp, however at max temp h240-x is similar to h110.

If the temps end up the same, what is your ambient? Have you changed the fans on the rad? You've increased the surface area over which you dissipate the heat. I would expect a larger difference.

 

[thilanliyan]

If the radiator capacity is about the same then I would expect similar temps. If it is not then something is off, assuming fans are equal.

 

[hunkeelin]

the rad size is the same. The h240's rad is a big thicker but that's it. I swap out the stock fans for noctua for both h110 and h240.

 

[RampantAndroid]

the rad size is the same. The h240's rad is a big thicker but that's it. I swap out the stock fans for noctua for both h110 and h240.

I would expect a thicker rad to do better given it has more surface area.

 

[hunkeelin]

I think it's becuase i'm only on pull config. If i do push + pull i think thicker rad make sense. One thing I have to note. The H240 can't do push + pull. While the H110 can.

 

[RampantAndroid]

I think it's becuase i'm only on pull config. If i do push + pull i think thicker rad make sense. One thing I have to note. The H240 can't do push + pull. While the H110 can.

My first reaction was "why is push/pull not possible?" Then I looked at the design. Wow, swiftech...

You can still do 3 fans though, I guess? Just not the 4th.

 

[BonzaiDuck]

My first reaction was "why is push/pull not possible?" Then I looked at the design. Wow, swiftech...

You can still do 3 fans though, I guess? Just not the 4th.

Funny that never crossed my mind while following the news about the H240X.

 

[BonzaiDuck]

What model case are you using for the 5960X, Rampage etc.?

 

[SithSolo1]

My first reaction was "why is push/pull not possible?" Then I looked at the design. Wow, swiftech...

You can still do 3 fans though, I guess? Just not the 4th.

I was thinking the same thing.

Wouldn't half push/pull and half pull do slightly better than all pull? It has the screw holes for it.

 

[BonzaiDuck]

I was thinking the same thing.

Wouldn't half push/pull and half pull do slightly better than all pull? It has the screw holes for it.

Here's where the idiotically simple discovery or conclusion that I and at least one other forum member made can count for something.

"Case pressurization contributes to heatpipe or radiator performance."

Also, "Smaller cases are easier to pressurize with smaller fans, even if larger cases (even a "fat" midtower) can do it with a smaller number of larger fans."

Essentially, you'd want your AiO or custom radiator deployed for exhaust. The pressurized case would essentially replace at least part of the effectiveness lost by fans that are either push or pull, but not both. And adding a third fan to an oddball design like the H240X can't hurt; more likely it would help.

 

[XavierMace]

Essentially, you'd want your AiO or custom radiator deployed for exhaust. The pressurized case would essentially replace at least part of the effectiveness lost by fans that are either push or pull, but not both. And adding a third fan to an oddball design like the H240X can't hurt; more likely it would help.

I'm not sure how you've reached the conclusion that you want the radiator deployed for exhaust. To get optimal performance from a radiator, you want as cool air as possible going through it.

 

[xthetenth]

The idea is that the flow rate gain from a pressurized case forcing air through the rad offsets the slightly hotter case air being used, hopefully more than enough. Considering it seems the H-240x is hurting for flow it might be able to get good use out of that sort of layout.

 

[BonzaiDuck]

I'm not sure how you've reached the conclusion that you want the radiator deployed for exhaust. To get optimal performance from a radiator, you want as cool air as possible going through it.

We've been at this before. It depends on the overall airflow through the case. But except for graphics card(s), the essential remaining items at issue for cooling are motherboard VRM and NB components.

It's already been proven, though, with WGusler's C70/4790K/H80 project in an older thread. After setting up his H80 with ducted exhaust, he had deployed the sensors for an Aquaero 5 monitor/controller throughout his system, and then removed the case side-panel. Temperatures increased on the CPU by a few degrees.

Logically, you'd want the coolest air moving through the radiator, but the difference between room- and case-interior ambient would have so little effect that the factor of radiator-cooling with good CFMs would leave the exhaust CFM as the only significant factor.

 

[kstud]

Interesting thread.

Could you take both coolers out of the case and run it as in an "open bench"? This should remove case/airflow as a limiting factor.

Since you have it in a pull configuration, I'm assuming both coolers were exhausting warm air from the case.

The Swiftech can be consider a custom loop with a weaker pump than stand alone water cooling pump...and the h240 cpu waterblock is among the best. I can understand your point on water equilibrium but to eliminate other factors, the testing should be in an "open bench" environment.

 

[BonzaiDuck]

Interesting thread.

Could you take both coolers out of the case and run it as in an "open bench"? This should remove case/airflow as a limiting factor.

Since you have it in a pull configuration, I'm assuming both coolers were exhausting warm air from the case.

The Swiftech can be consider a custom loop with a weaker pump than stand alone water cooling pump...and the h240 cpu waterblock is among the best. I can understand your point on water equilibrium but to eliminate other factors, the testing should be in an "open bench" environment.

Sure. That's the purpose for external standalone radiators or heat-exchanges like the MO-RA3 or the old Zalman Reserator. On the down side, you've created a configuration with umbilical connections, a need for quick-release connections and other cumbersome things. If you want it, you get it. If those things are big factors to a person, then they either find a way around it or internalize cooling to the case.

Even so. You can hang a radiator on the case at the rear -- preferably ducted to the ubiquitous exhaust fan in its common locations, perhaps to eliminate a dead spot for a fan too close to the fins.

But I disagree with the "open-bench" idea. Sure you can rate radiators and other things by using a consistent open bench test-bed; you can also test devices relative to one another with any other common environment, like the same case. But there's no denying that pressurization with radiators in the counterintuitive location of an exhaust will provide more effective radiator cooling because the pressure and perhaps air density provides an advantage greater than the loss of a few degrees above ambient.

Look at it this way. an H80 AiO is made for case interior installation. You can test it without any ducting or consideration for the dead spots; you can test it without much case airflow -- in which case -- case interior will simply be warmer. But the greater airflow would mitigate the difference between internal ambient and room ambient, forcing the former closer to an asymptotic limit of the latter. If you also pressurize the case under that scenario, -- well -- WGusler's project proves it, and my own cases and cooling solutions prove it.

Nobody likes dust, but dust in electronics is a fact of life with the exception of "clean rooms." In "general" electronics, airflow cools the devices, no airflow or stale air reduces the life of components, and dust is a fact of life no matter what you do.

 

[XavierMace]

It's already been proven, though, with WGusler's C70/4790K/H80 project in an older thread. After setting up his H80 with ducted exhaust, he had deployed the sensors for an Aquaero 5 monitor/controller throughout his system, and then removed the case side-panel. Temperatures increased on the CPU by a few degrees.

What does removing the side panel have to do with putting the radiator on an intake port? Did I miss where he installed it on a front intake port as suggested? I just read the thread again and I don't see it. I'm not disagreeing on the effectiveness of the ducting or positive pressure.

Until he installs it on one of the front intake ports, he hasn't proven anything in regards to radiators and intake vs exhaust. Not to mention there's literally countless threads on the internet about using radiators on intake ports vs exhaust.

Just to be absolutely clear, I 100% agree on your suggestions of ducting and positive pressure if you want to achieve the absolute max cooling effectiveness (air or water). But using a radiator as an intake instead of an exhaust doesn't prevent you from doing that if you've done your shopping right.

His setup does what he needs it to do which in the end is all that matters. However, at no point was what is nearly universally accepted as an optimal front mount setup (ducted or not) tested.

404 - Proof not found.

If he keeps his fan setup 100% as is and just moves the radiator to the front of the case, and gets worse performance, I will happily apologize to the both of you. But until then, I haven't seen one ounce of proof that the exhaust setup is superior in an otherwise equal intake setup.

 

[BonzaiDuck]

What does removing the side panel have to do with putting the radiator on an intake port? Did I miss where he installed it on a front intake port as suggested? I just read the thread again and I don't see it. I'm not disagreeing on the effectiveness of the ducting or positive pressure.

Until he installs it on one of the front intake ports, he hasn't proven anything in regards to radiators and intake vs exhaust. Not to mention there's literally countless threads on the internet about using radiators on intake ports vs exhaust.

Just to be absolutely clear, I 100% agree on your suggestions of ducting and positive pressure if you want to achieve the absolute max cooling effectiveness (air or water). But using a radiator as an intake instead of an exhaust doesn't prevent you from doing that if you've done your shopping right.

His setup does what he needs it to do which in the end is all that matters. However, at no point was what is nearly universally accepted as an optimal front mount setup (ducted or not) tested.

404 - Proof not found.

If he keeps his fan setup 100% as is and just moves the radiator to the front of the case, and gets worse performance, I will happily apologize to the both of you. But until then, I haven't seen one ounce of proof that the exhaust setup is superior in an otherwise equal intake setup.

That was a decently long thread, and he may have considered other deployments, but he put -- first -- the H60 -- and then the H80 - at the rear exhaust.

After he'd finished with the modest ducting mods, he did indeed test it with side-panel installed and then with it off, and marked the CPU temperature increase under the same load with an open case.

My point is fairly simple. There are many ways to skin a cat. There are often a variety of "good" solutions that work. The only problem I see with using the radiator at intake to achieve good results is that it would benefit more from a double push-pull configuration -- to move more air through the radiator. It is also not pressurized air.

If there was an advantage of using the radiator at exhaust, it would be that the hopefully higher case intake CFM and case pressure might reduce or negate the need for double push-pull, but he does indeed use a pusher and puller fan in that setup, so he has it better both ways.

That strategy is equivalent to using the case as a sort of bellows. But as I said, with good airflow, the delta between interior and exterior case ambient declines. In that case (pun), it's not a major factor.

But it would easily be a bigger factor if the builder-user takes a different concept for his work. Only observing over several years the projects people post on web-sites sporting their computer projects and water-cooling, folks have a penchant for nearly isolating the case from "what's outside." We have a dust phobia, we have a noise phobia.

So there are case windows instead of vents, or fish-tank, museum-showcase projects. Yet, ultimately, air is the final means of heat-removal, no less for water-cooling.

If I built a system with Corsair Carbide Air 540, I'd have a front panel radiator at intake, and the top-panel radiator for exhaust. The reason we examine other possibilities derives from a need to fit the cooling solution to the case. So in a C70 case, I'd think the single-radiator at exhaust works pretty well for that.

 

[XavierMace]

That was a decently long thread, and he may have considered other deployments, but he put -- first -- the H60 -- and then the H80 - at the rear exhaust.

After he'd finished with the modest ducting mods, he did indeed test it with side-panel installed and then with it off, and marked the CPU temperature increase under the same load with an open case.

You've completely ignored my point. Re-read what you just said. Every single one of his configurations was with the radiator at the rear. At no point was the radiator mounted in the front as an intake as basically everyone, including the manufacturer recommends. Therefore there is ZERO proof that an exhaust setup is superior as you claim. All you've proven is that static pressure and forced airflow is superior which is a pretty widely accepted fact. I'm not sure where the confusion is here.

I am not arguing ducting improved his efficiency.
I am not arguing the removing the side of the case increases temps.
I am not arguing that it can benefit from static pressure.

I'm arguing there's ZERO evidence to support your claim that setting the radiator up as an exhaust is superior to intake. Removing the side panel has absolutely NOTHING to do with this. Again, if he tests that and proves it wrong, I will happily apologize to the both of you. But he did zero testing that I can find with the radiator mounted on the front. I'm not aware of you doing any actual testing on the subject either.

I'm sorry if I'm coming across as rude here, but I get really annoyed when the term "proof" is used when there actually is none. You have a theory on why there's a possibility that it could be preferable. There's a big difference between theory and proof.

 

[BonzaiDuck]

You've completely ignored my point. Re-read what you just said. Every single one of his configurations was with the radiator at the rear. At no point was the radiator mounted in the front as an intake as basically everyone, including the manufacturer recommends. Therefore there is ZERO proof that an exhaust setup is superior as you claim. All you've proven is that static pressure and forced airflow is superior which is a pretty widely accepted fact. I'm not sure where the confusion is here.

I am not arguing ducting improved his efficiency.
I am not arguing the removing the side of the case increases temps.
I am not arguing that it can benefit from static pressure.

I'm arguing there's ZERO evidence to support your claim that setting the radiator up as an exhaust is superior to intake. Removing the side panel has absolutely NOTHING to do with this. Again, if he tests that and proves it wrong, I will happily apologize to the both of you. But he did zero testing that I can find with the radiator mounted on the front. I'm not aware of you doing any actual testing on the subject either.

I'm sorry if I'm coming across as rude here, but I get really annoyed when the term "proof" is used when there actually is none. You have a theory on why there's a possibility that it could be preferable. There's a big difference between theory and proof.

But in my last post, I thought I had dispelled this misunderstanding.

I never said -- as a matter of "exhaust versus intake" deployment -- that exhaust is superior. I said that the difference touted as defeating the exhaust option dwindles to nothing with adequate case airflow.

This had been a discussion in other threads where we'd crossed swords.

I would think it could be absolutely marvelous if we could simply deploy all our radiators of any size and in any case-chassis at intake, but certain conventions in case design such as the forward placement of lower drive cages makes it more difficult.

So now we see some departures from the usual ATX case-design that assumes a wider market for air-cooling -- like the Carbide 540, certain more expensive boutique cases including "Mountain Mods." In fact, alternative cases came earlier from Silverstone and Lian-Li, if I remember.

I also had the chance to see some older Dell machines recently, with the ATX motherboard in a reverse position and airflow ducted directly from intake to an air-cooled heatsink without pipes -- taking advantage of intake air.

And I am simply saying that exhaust-deployment of EITHER a heatpipe-cooler or a radiator does not lead to some fearsome loss of potential if there is some attendance to decent airflow.

But I would suggest to folks not yet part of the argument who are building Haswell-K systems like WGusler's that the temperature problem seems more than it is. If I can get IBT "Maximum" temperatures of about 71C with heatpipes, or WGusler can get stress-load temperatures of 70 with a 4790K @ 4.6Ghz with an H80 AiO, then any "worries" about exhaust deployment are likely groundless.