Prefilled Water Cooling Just Got A Lot More Compact With Corsair H5 SF

[RedJaron]

Sucking in air from around the VRM is not the same as actively blowing air over them. I'll remain skeptical about how well this can handle VRM cooling until I see some reviews.

 

[George Phillips]

I haven't seen some really interesting design in PC for a while. This is one. They should tweak it and use it in larger motherboards.

 

[Sakkura]

Pointless compared to air cooling. More expensive and louder, without improving temperatures.

proof? I cant find any reviews yet. I suspect it will be on par with most huge air coolers though. That would be sad.

No proof, but it's rated for a pathetic 24 CFM which means it's going to be pretty poor.

 

[cub_fanatic]

I saw this the other day while trolling Newegg deals. At first, i thought this was one of their older, weird obsolete designs until I looked it up. This sitting on top of a tiny Z170 and unlocked Skylake i5 combined with a liquid cooled Fury X would make for a pretty sweet and quiet mITX console murdering system.

 

[jimmysmitty]

Pointless compared to air cooling. More expensive and louder, without improving temperatures.

proof? I cant find any reviews yet. I suspect it will be on par with most huge air coolers though. That would be sad.

No proof, but it's rated for a pathetic 24 CFM which means it's going to be pretty poor.

You are assuming that fans CFM is the end all be all for a CLC when it is not. The pump speed, radiator thickness and the static pressure will also be important, in fact static pressure is more important in a CLC than the CFM. You can have a high air flow fan with plenty of CFM, in fact more than a SP rated fan, yet it would have poorer performance than a SP rated fan.

 

[Lutfij]

Just because people don't like a point of view they go on a down voting spree

On that note jimmysmitty + 1 to those pointers.

 

[kiwicasper]

Yeah but at 84mm it will not fit into the coolermaster elile 110 case. ah heck..

 

[gilbadon]

Just because people don't like a point of view they go on a down voting spree

I would argue that is the point...
What is the use of the up and down vote if not to show how much the community agrees with what is being said?

I down vote what I do not like and up vote what I do like equally as to give an idea of how much consensus there is on a post. Maybe I am using it wrong.

 

[Vogner16]

@gilbadon Lutfij was referring to how sakkura downvoted every comment jimmy made after jimmy made a good argument to the performance of this water cooler not being entirely reliant to one spec, the fan air flow rate.

yes that's the point of the downvote, but sakkura used it in a backlash type manner to his being proved wrong, not the point of the downvote.

general consensus of this cooler is good and people are excited to see more unique cooling designs for mITX systems.

On a side note. yes this has poor vrm cooling, but don't most vrm's not need much cooling anyway? I have a GTX H100i exhausting out the top and a simple 140 exhausting out the back (no flow to vrm's) and have never seen temps from my probe above 45C. I don't think you really need to actively cool them or they would have fans on them...

 

[Sakkura]

@gilbadon Lutfij was referring to how sakkura downvoted every comment jimmy made after jimmy made a good argument to the performance of this water cooler not being entirely reliant to one spec, the fan air flow rate.

yes that's the point of the downvote, but sakkura used it in a backlash type manner to his being proved wrong, not the point of the downvote.

general consensus of this cooler is good and people are excited to see more unique cooling designs for mITX systems.

On a side note. yes this has poor vrm cooling, but don't most vrm's not need much cooling anyway? I have a GTX H100i exhausting out the top and a simple 140 exhausting out the back (no flow to vrm's) and have never seen temps from my probe above 45C. I don't think you really need to actively cool them or they would have fans on them...

I used it to downvote misinformation. He didn't prove anything except his own ignorance.

General consensus doesn't count for anything when people don't know what they're talking about.

In any case, I was the one who was downvoted first, for daring to break the circlejerk.

 

[RedJaron]

On a side note. yes this has poor vrm cooling, but don't most vrm's not need much cooling anyway? ... I don't think you really need to actively cool them or they would have fans on them...

You miss the point that almost all stock coolers are down-draft, which do actively blow over/across the VRM and any VRM heatsink. Even a tower cooler will still incite a breeze across the area that goes with whatever crossflow the case already generates with its intake and exhaust fans.

The bigger point is that a CPU that needs cooling above what the stock solution can handle is drawing a lot more power through the VRMs, which heat them up. This cooler is also designed for ITX machines, and it's much easier for heat to build up in those smaller cases.

The bottom line is that in a small case where airflow is already limited, moving the majority of the airflow to a different location through liquid cooling and a radiator means the area with the highest power draw on the board is left to fend for itself.

 

[Vogner16]

In any case, I was the one who was downvoted first, for daring to break the circlejerk.

I appreciate that, people just disagreed with you so following your procedure, they downvoted what they disagreed with. opposing opinion creates discussion and creates solutions to problems others missed.

how is what jimmy said misinformation? he stated that rad surface area is just as important as air flow past the cooler and static pressure is more important than flow performance when pushing air through a confined surface such as a rad.

@redjaron your point is valid and I agree people with a water cooler will overclock, thus heating vrm's more. BUT you forget my example of my personal rig. I have no fans to vrms and have a 8350 overclocked to 4.8 ghz. those vrm's do not get hot. granted my motherboard has an excellent heatsink to the vrms, it still proves my point that they do not need a constant flow of air to keep them cool.

nobody makes a northbridge/ vrm combo that was ever hotter than the LGA 775 was and even those did not need air flow over them unless you overclocked the northbridge and cpu to "extreme" levels.

I think this solution from corsair will be able to vent quite well any case heat from other minor heat generating elements of the computer (southbridge, northbridge, vrm's, audio and ethrenet controllers etc..). as always however we will have to see. Tom's Hardware needs to do a good review on this when it comes out!

Cheers!

 

[turkey3_scratch]

I have to agree with Sakkura on this one, I just can't help it. When it comes to these mITX cases, we need to consider the volume of the case. The volume is only a fraction of an ATX mid tower, meaning all that dissipated heat is more compact and therefore the typical gas particles have a higher temperature (it could be double minus ambient). This particular unit points outward at the exhaust of the computer, so all that hot air that was just spurred out will get pulled right into that radiators.

Any single 120mm radiator cooler is, well, not going to do too much. a 2x120mm radiator gets far better cooling, and simply considering the positioning of the radiator as well as the fan (the fan seems to be underneath and intake more hot air that's inside the case) I don't see the liquid getting cooled very well at all.

What the average person underestimates is that the case fans are actually most vital to the cooling of anything.

 

[RedJaron]

@redjaron your point is valid and I agree people with a water cooler will overclock, thus heating vrm's more. BUT you forget my example of my personal rig. I have no fans to vrms and have a 8350 overclocked to 4.8 ghz. those vrm's do not get hot. granted my motherboard has an excellent heatsink to the vrms, it still proves my point that they do not need a constant flow of air to keep them cool.

When you first made that remark, you only listed your cooler, not your CPU, and you still haven't listed your mboard or case. Since you're fitting a 140mm fan and a 240mm radiator, I'll assume a full tower and full ATX board. The ATX board means you've got a lot more room for VRM heatsinks on the board itself. Bigger, thicker heatsinks can compensate for a lot for lack of airflow, which is why passive-cooled GPUs exist. That's not a luxury ITX boards have. You're lucky if you get even a tiny VRM heatsink there.

Next, you've got a large case. That means a lot of air inside that can suck the heat away, even if it isn't swirling around a lot. However, with at least three sizable case fans, you've got some decent ventilation happening here. Again, most ITX cases where this bulldog cooler would be used are limited to one medium fan or two smaller fans.

I'm not saying VRMs always need active cooling ( though it never hurts ). I'm saying that in the typical use case where this cooler would be used ( small ITX case, OC'd CPU ), the VRM cooling should not be an afterthought. And until I see some proper reviews of this, I'm going to remain skeptical of its VRM cooling ability.

 

[gilbadon]

@gilbadon Lutfij was referring to how sakkura downvoted every comment jimmy made after jimmy made a good argument to the performance of this water cooler not being entirely reliant to one spec, the fan air flow rate.

yes that's the point of the downvote, but sakkura used it in a backlash type manner to his being proved wrong, not the point of the downvote.

general consensus of this cooler is good and people are excited to see more unique cooling designs for mITX systems.

On a side note. yes this has poor vrm cooling, but don't most vrm's not need much cooling anyway? I have a GTX H100i exhausting out the top and a simple 140 exhausting out the back (no flow to vrm's) and have never seen temps from my probe above 45C. I don't think you really need to actively cool them or they would have fans on them...

Oh lol. I should probably read contextual evidence... Thanks

 

[jimmysmitty]

@gilbadon Lutfij was referring to how sakkura downvoted every comment jimmy made after jimmy made a good argument to the performance of this water cooler not being entirely reliant to one spec, the fan air flow rate.

yes that's the point of the downvote, but sakkura used it in a backlash type manner to his being proved wrong, not the point of the downvote.

general consensus of this cooler is good and people are excited to see more unique cooling designs for mITX systems.

On a side note. yes this has poor vrm cooling, but don't most vrm's not need much cooling anyway? I have a GTX H100i exhausting out the top and a simple 140 exhausting out the back (no flow to vrm's) and have never seen temps from my probe above 45C. I don't think you really need to actively cool them or they would have fans on them...

I used it to downvote misinformation. He didn't prove anything except his own ignorance.

General consensus doesn't count for anything when people don't know what they're talking about.

In any case, I was the one who was downvoted first, for daring to break the circlejerk.

You literally said, in your own post, "no proof but it will suck".

I stated plenty of well known facts about CLC. There is more than just one factor for CLCs and saying that CFM is the only factor is misinformation as CFM is not the only nor the most important part. You could throw 80CFM into it but if the water pump is too slow it won't make a difference. You could throw 80 CFM at it but if it has low static pressure it might only perform as well as a high rated static pressure fan 1/4 the CFM rating.

There is a reason why all CLCs, and even custom loops, use high rated SP fans over high rated CFM fans.

My only point is that you are judging the product with only one piece of information and no facts to back it up at all.

 

[gilbadon]

This conversation right now...

 

[Sakkura]

I appreciate that, people just disagreed with you so following your procedure, they downvoted what they disagreed with. opposing opinion creates discussion and creates solutions to problems others missed.

how is what jimmy said misinformation? he stated that rad surface area is just as important as air flow past the cooler and static pressure is more important than flow performance when pushing air through a confined surface such as a rad.

The other things matter too, but if the airflow is terrible the cooling performance cannot be good. It's also worth noting that static pressure in itself doesn't matter at all; it's just a factor influencing the airflow, but we already have the airflow rating.

And I didn't complain about people downvoting my comment. I will complain if I'm not allowed to downvote what I disagree with while others are allowed to do so when they disagree with me.

 

[Sakkura]

You literally said, in your own post, "no proof but it will suck".

I stated plenty of well known facts about CLC. There is more than just one factor for CLCs and saying that CFM is the only factor is misinformation as CFM is not the only nor the most important part. You could throw 80CFM into it but if the water pump is too slow it won't make a difference. You could throw 80 CFM at it but if it has low static pressure it might only perform as well as a high rated static pressure fan 1/4 the CFM rating.

There is a reason why all CLCs, and even custom loops, use high rated SP fans over high rated CFM fans.

My only point is that you are judging the product with only one piece of information and no facts to back it up at all.

The airflow is a necessary component in the performance. One weak link in the chain is enough to hurt performance - it doesn't really matter if it's low airflow, low flow rate in the water loop, a poor cold block or a bad interface between the cold block and the CPU heatspreader. Any one of those is enough to seriously harm cooling performance.

And again, the only thing that matters with regard to the fan is airflow (well maybe noise too, but that's another issue). Static pressure is just a factor influencing what the airflow will be in a given application. High static pressure means nothing if the airflow is low. And an airflow-optimized fan is only bad on a radiator because the actual airflow in that configuration will be low.

 

[Lutfij]

Again as stated by jimmysmitty, the points about conventional watercooling will always apply to either a custom loop, a kitted solution, a semi AIO solution or an AIO because they are all dealing with a radiator, that dissipates heat, the fan which will need to overcome the fins density, the fluid used and the flow of the coolant within the loop as well as the pump. So here we have people poking the product when it hasn't even been reviewed nor disassembled to inspect its workings to lay criticism to its build, purpose and effectiveness in a SFF environment.

The makers of that cooler are Asetek and they have been known to provide these exact similar solutions for data centers and other large corporations so essentially speaking they have their niche market - for some though it may be highly questionable.

Disagreements are good but its best served if you have information to back up what you are trying to refute.

 

[Sakkura]

Disagreements are good but its best served if you have information to back up what you are trying to refute.

We have the full specifications on the fan. That's information, and it backs up what I said.

 

[JackNaylorPE]

Silly.

 

[Vogner16]

You literally said, in your own post, "no proof but it will suck".

I stated plenty of well known facts about CLC. There is more than just one factor for CLCs and saying that CFM is the only factor is misinformation as CFM is not the only nor the most important part. You could throw 80CFM into it but if the water pump is too slow it won't make a difference. You could throw 80 CFM at it but if it has low static pressure it might only perform as well as a high rated static pressure fan 1/4 the CFM rating.

There is a reason why all CLCs, and even custom loops, use high rated SP fans over high rated CFM fans.

My only point is that you are judging the product with only one piece of information and no facts to back it up at all.

The airflow is a necessary component in the performance. One weak link in the chain is enough to hurt performance - it doesn't really matter if it's low airflow, low flow rate in the water loop, a poor cold block or a bad interface between the cold block and the CPU heatspreader. Any one of those is enough to seriously harm cooling performance.

And again, the only thing that matters with regard to the fan is airflow (well maybe noise too, but that's another issue). Static pressure is just a factor influencing what the airflow will be in a given application. High static pressure means nothing if the airflow is low. And an airflow-optimized fan is only bad on a radiator because the actual airflow in that configuration will be low.

@redjaron once again valid points and your assumptions are correct as 8350 only really fits in midtower or full tower atx case.
I have a mid tower case and crosshair V motherboard (excellent vrm cooling, but passive and connected to my overclocked northbridge via heatpipe.

@sakkura a SFF case is optimized for a console experience or simply a small location. usually near the end user. thus a quiet case is best solution for these situations to not interfere with the audio of the games. my work SFF pc sits right next to my moniter and im so glad dell put a super quiet cooling solution for my cpu. I never hear it spin up. this cooler would be better than the stock dell solution and be quieter for hotter cpu's.

as an engineer I know and have done calcs showing flow does not mean better cooling/heating. things such as internal cooling substance, interface thickness, material, surface area, and exterior cooling substance make major changes to how heat transfers from one point of a system to another. there is an exponential curve for when air flow increases do not improve heat transfer. when talking about the instance of a water to air heat transfer there is always a certain amount of heat dissipation due to convection of air that you will find is massively improved during the first small improvements to external cooling substance flow rate.

the work that goes into these coolers is waaayyy above your ability to say it will suck in a SFF system without having done the calcs of every heat transferring surface on the cooler. you look at one spec and imply that it will suck. I may not be the smartest man but I do know heat transfer

FYI: we don't have enough information to know if this will be a good solution without having one and examining it with extreme scrutiny, or simply testing it ourselves.

 

[Sakkura]

Ceteris paribus, less coolant flow means less cooling. All the evidence we have points to this probably being a low performance cooler, not the high-performance overclocking champion people seem to expect (based on absolutely no evidence).

 

[Vogner16]

you don't see that this might not be designed for a super overclocking SFF pc? 24 CFM may be able to cool a simple 95W i7 to safe temps while remaining whisper quiet.

less coolant flow does not mean less cooling. how do passive coolers work? you cant turn a 10 element equation into a simple 1 element equation.