Water Cooling Fans - Suck or Blow?

[wardp 19]

I am in the process of building a new Haswell 4770K system on a Gigabyte GX87X-OC board using a Corsair H110 Water cooling kit and it's all going in a Corsair Vengeance case which has the mounting points for the H110 radiator in the top grill.

In the past I have always gone with the idea that case fans at the front and side of the case are intake fans and those at the top and back are exhaust. However in the H110 water cooler instructions it says it works better if the radiator fans are intake. If I did as they say to balance the intake/exhaust fans I would have to change the side panel fans to exhaust which seems all wrong for the airflow.

This is my first venture in water cooling and any advice would be welcome, what should I do about the radiator fans - suck or blow ??

Thanks
Phil

[westman 257]

Push or ( blow ) is better.

Read this tests on differnt fan configs.

 

http://martinsliquidlab.org/2012/01/15/radiator-shroud-testing-v2/3/

[Illegitimi non carborundum 4]

Phil

It will depend on the configuration of all the fans as if you say some opposing each other you could create a negative pressure effect with little effective cooling. 

Say with a liquid cooler if you are pulling air into the case then it has to go through the warm/hot radiator and you are then flooding the mobo, ram, gpu etc with air that is already warm/hot and then relying on any exhaust fans to get rid of it.

I always set mine to blow outward and then I'm drawing in ambient cooler air and blowing/exhausting hot air.  That is provided that you have good ingress air access to the case.  Make sure you clean any filters regularly.  I also use a HEPA filter device to keep dust down.

Its a balance and the bigger the case the more efficient the cooling.

pH

 

[pmanhart 176]

For the same reason that a traditional Heatsink and Fan has the fan push the air through the fins of the heatsink, you want the fan to push through the radiator.

 

Pulling through the radiator is only efficient if you have enough force already pushing through from the other side (such as using fans on both sides).

[Illegitimi non carborundum 4]

Phil

Good point - I'd forgotten on my H80i there are fans either side of the radiator and you have to make sure the air flow is in the same direction!!

pH

[martin-w 3,234]

I am in the process of building a new Haswell 4770K system on a Gigabyte GX87X-OC board using a Corsair H110 Water cooling kit and it's all going in a Corsair Vengeance case which has the mounting points for the H110 radiator in the top grill.

In the past I have always gone with the idea that case fans at the front and side of the case are intake fans and those at the top and back are exhaust. However in the H110 water cooler instructions it says it works better if the radiator fans are intake. If I did as they say to balance the intake/exhaust fans I would have to change the side panel fans to exhaust which seems all wrong for the airflow.

This is my first venture in water cooling and any advice would be welcome, what should I do about the radiator fans - suck or blow ??

Thanks

Phil

 

Corsair advise this because by configuring the radiator fans as intake, you are passing cool outside air over the radiator, thus lower CPU temp.

 

However, by doing this you are also blowing hot radiator air into your case, thus higher enclosure temp.

 

Many ignore Corsair's advice, and stick to the conventional orientation. So fans blowing enclosure air through the rad and outside. CPU temp will be slightly higher, but not much to be honest.

 

So it depends on your desires. Coolest possible CPU, with somewhat warmer enclosure temp, thus slightly warmer GPU and motherboard. Or slightly warmer CPU, but maintain a cool motherboard and GPU.

 

Either way there's not a huge difference, but for me, if I had a Corsair, I'd probably endeavour to keep the enclosure cool.

[wardp 19]

What I hadn't realised until I got the Corsair H110 that the fans have to be fitted between the case and the radiator as the bolts pass through the case and fan then terminating in the radiator and not as I envisaged passing through the radiator then secured with a nut. So to extract the air out of the case the air would be pulled through the radiator whereas air from outside would be pushed in through the radiator which is as I have it at the moment and it seems to be OK but I am still working on it. 

 

Thanks everyone

 

Phil

[RWFeldman 159]

Whats cooler. The air inside the computer or outside the computer? I always have used ambient outside air through the cooler, air conditioned room. And multiple case fans to expel from the case to the outside. CPU always cool.

[ant_1984 20]

Are you using a Carbide 500R case?

 

That's my case and I have my fans pulling air through the radiator exhausting out the top of the case to avoid heat build up.

 

Performs perfectly well for me, although I do get slightly lower temps if I have air pulled in from outside the case. However with the number of intake fans I have now it really is neither here nor there as far as temps are concerned.

[Andydigital 93]

The most important thing is to make sure that your fans are designed to work in close proximity to a radiator, if they aren't designed for it the blades will stall and become inefficient when blowing into a radiator, its the same when sucking the air through but its not quite as inefficient. I have my Apaches sucking air through the rad at the top of my 800D case.

[Bob Scott 7,710]

I run my radiators external to the computer case, and run tubing in through the back of the case (with a drip loop to prevent a cooling plant leak from running down the tube into the case). This keeps all but the four connections to the CPU and GPU cooling blocks outside the case, something that saved my butt when a radiator started leaking from a bad solder joint on one of the fins a year or two ago.

 

But back to topic, whether to install the fans to suck or blow across the radiator, a few years back I did some experiments here using cigar smoke to watch airflow and a thermometer monitoring radiator input/output temps, and I concluded that if you have a good, well-sealed shroud that keeps the fans a few inches away from the radiator, sucking produced a more even flow and slightly better temps. Without the shroud, in the "suck" config there were a lot of dead spots, especially in proximity to the fan hubs, and pressurized flow from blowing across the radiator worked better. If you have air leaks around the edges of the fan(s), or run without a shroud, then I concluded blowing through worked better, but only marginally so.

 

I'd surmise that in a more traditional configuration with the radiator internal to the case, airflow considerations through the case as a whole would trump the fan-radiator configuration issue.

 

Regards