H60 Push/Pull - Is fan speed and CFM important?

[Luckbad]

I opted to go for the push/pull configuration for my H60. I'm pushing cold air from behind the case with the fan that came with the H60 and pulling warm air from the radiator into the case (then I exhaust it out the top).

 

However, the fans are different RPMs and CFM.

 

The H60 fan is up to 74.4 CFM @ 1700 RPM (the pusher) and my puller is up to 44.73 CFM @ 1200 RPM.

 

Will that be problematic? Do I have to shell out $15-$20 to get another of the H60 fans?

 

I'm wondering if it will even matter because I use the motherboard fan controller on my Asus P8Z68-V Pro Gen 3, so they probably won't match in speed anyway, right?

 

Anyway, just trying to see what others have experienced and maybe get a reply from a Corsair rep about it.

 

Thanks!

[babdi]

With differential flow rate as in your case, The air flow will create eddies as pull is less than push;the flow of air between the radiators has to be smooth and uniform for better cooling

[MSIINOZ]

You should try running the Corsair fan on the inside pulling air in which will help the slower external fan to push, At the moment your Corsair fan is pushing against a slower fan and basically restricting it

 

Try using just the Corsair fan on the inside and see if the temp changes? I am only using 1 Gelid thermally controlled fan running at 1000 RPM, When I fitted the Corsair fan it made too much noise and dropped the temp down 1 degree celcius, A waste of time really having all that noise, As long as there is air flowing through the radiator it works, No need for a noisy tornado going through it

 

One thing that gave me a drop in temp was a Coolingworks 120mm shroud

 

http://www.frozencpu.com/products/13805/ex-rad-29/Blue_UV_Reactive_Mini-CoolShroud_-_120mm.html?tl=g30c429s163

 

With the fan hard against the radiator there is a huge dead spot in the middle where air does not circulate, By moving the fan off the surface air can flow through the middle of the radiator, I fitted the fan on the outside of the shroud so my fan is now sitting about 30mm off the radiator

The base of this shroud is 130mm so it sucks a huge area of air into the fan, Much larger than the diameter of your fan which works great

 

http://www.aquastealth.com/ProductImages/watercooling/minishroud.jpg

[Luckbad]

Were you able to use the screws that came with the H60 with that sleeve? It seems like they could be too short.

[parsec]

You're right, the standard screws won't work when using a shroud.

 

The standard screws are 6-32 x 1 1/4", which is a common type of screw in the US. You can get longer screws of the same diameter and thread (6-32, 1/8") at hardware stores. The depth of the shroud determines how much longer the screws must be. It it was one inch, you would need 6-32 x 2 1/4" screws.

 

Finding screws of the exact length you need can be difficult, since stores don't carry every length possible. When I install shrouds, I buy screws that are longer than I need, and cut them to size with a hack saw. That way you can adjust the length of the screws for your particular application, if necessary.

 

A simple source for shrouds are old or dead fans. Just cut out the motor/fan with a large wire cutters and use the frame as a shroud. Works great. Or find the cheapest 120mm fan you can with a good frame to make a shroud.

[MSIINOZ]

I can measure the screw length for you, I will need to find out myself as I am adding another 120mm or swapping for a dual 120mm radiator to my system

 

The screws actually came with the shroud so you wont need to hunt them up, This place (BELOW) is dearer but shows that the shroud comes with screws the same as mine did from frozencpu, Forgot to mention it comes with a neoprene gasket too, Just the screws & gasket is nearly worth $5, Not really worth the time hacking up an old fan that will not give you the same amount of flow>

http://www.aquastealth.com/minicoolshroud.aspx

 

Edit> Found this pic googling but sidewinder does not have them according to their site? Frozencpu is the cheapest & they have them in stock

 

You can use an old 120mm fan but an old fan shell does not have the same outer diameter as the Coolingworks shroud base (130mm), It is very little difference but a huge difference in terms of flow+cooling & for the price I think it is worth it, The shroud gives you a velocity/cone shape effect pushing a larger area into your fan blades, Using an old fan shell will work to remove the dead spot in the middle but still leaves a dead spot all the way around the edge where the fan shell is mounted due to the thickness of the fan shell which is fairly thick especially in the corners, The proper shroud is thin plastic and does not sit on the actual radiator core allowing air to flow through ALL of the core, Buy a proper shroud & you will be pleased, There is other brands out there but this is the only one that has the larger base and gives the velocity/cone effect

 

Happy Cooling

[parsec]

I didn't say a fan frame is a perfect shroud, it is simply an option for making one yourself. I've found that fan frames are not all identical, some have less material in the frame to accommodate larger impellers, so more of the radiator is exposed to air flow. Of course, the small amount of material in the corners of the frame does block some of the radiator.

 

I was curious about the 130mm size of the Coolingworks shroud, and why it seems rather thin on the sides. I noticed this description of it from FrozenCPU's page:

 

The Blue UV Reactive mini-CoolShroud - 120mm is a perfect match for enclosing a single 120mm fan onto a radiator or chassis wall. This shroud hides the fan itself and is UV Blue reactive.

 

So the original intent of this product seems to be as a cover for a fan mounted on a radiator, that encloses and hides it. I thought that the thin sides of this product might not work well when compressed between a fan and a radiator, particularly with screw holes only on one side. That design matches the purpose of a cover, and makes sense.

 

But you have used this product as a typical shroud successfully, that's great. Exposing all of the radiator to air flow as you described is certainly an improvement.

[MSIINOZ]

Yes as you mentioned it is supposedly intended to cover the fan which is ridiculous and it is sold as a shroud???, Maybe that is a mistake made by some marketing people that have no idea about the product, Maybe they are asian or speak another language other than english and got there wires crossed but either way this works really well as a shroud (Not as a fan cover), I think it was designed & manufactured as a shroud?

Title says one thing description says another???

 

If this was used as a fan cover it would totally defeat the purpose and the air would be literally going backwards & forwards inside the shroud & I think it would actually increase the temperature using it as a fan cover, I wonder how many people have tried that? I could not even be bothered trying it that way because it is just plain silly covering up a fan which has no gain at all

 

It works fine as a shroud as mentioned but I would not let a monkey tighten the fan to the radiator with this shroud in between, Just enough pressure so it all does not move is fine, It does not feel brittle enough to just simply snap but some people out there well.......you know

 

The only downfall with the base being 130mm is that I cant run 2 of these side by side on a dual 120mm radiator due to the 15mm mounting hole gap in between which really bugs me, I am thinking about just simply adding another radiator into the loop next to the stock H60 radiator, I have found very very similar almost identical 120mm radiators on ebay for under $25 including delivery, I may have to go with that option?

 

I am going to try an external reservoir (Swiftech micro) first & see how that goes with the temp?

[parsec]

The use of this shroud as a cosmetic feature, to cover or hide a fan, may fit in the paradigm we see in some mother boards, with the (IMO) bizarre chipset heatsink covers. The gun magazine and literally faux cartridges used as adornment are ludicrous, and a fan cover is actually mild in comparison. Perhaps it "looks cool" when enclosing an LED fan, and is UV reactive too. Someone may be trying to pioneer a new accessory, fan covers... always something new to buy (sell.)

[MSIINOZ]

I dont think they are pioneering, These seem to have been out for a long time unless someone knows better?

Getting off track here but this is strange, The box has on it, Note the date>

 

(copyright logo) 2004 Armelin Widget Corporation?

 

http://www.coolingworks.com = http://www.armelin.us/

 

What the heck???

[Luckbad]

Avoiding starting a new thread for this question since you guys seem to know what's up:

 

I decided to get another of the Corsair H60 fans so I have identical speeds and CFM.

 

Now the question is: How do I make sure they are operating at the same speed at all times?

 

I have an Asus P8Z68-V Pro/Gen 3.

 

It has both CPU_FAN and CPU_FAN_OPT.

 

From the rather short reply to a similar question on Asus' site, the webmaster there said "They are both PWM controlled, and linked together."

http://vip.asus.com/forum/view.aspx?board_id=1&model=P8Z68-V+PRO&id=20110514042117226&page=1&SLanguage=en-us

 

If this is true, does that mean I can connect one of the H60 fans to CPU_FAN and the other to CPU_FAN_OPT and they will always operate at the same speed? That would be ideal.

 

If not, what are my options?

 

Is there some way to slave one fan to another? Would it be safe to use a Y-splitter and plug them both into CPU_FAN?

 

Thanks all!

[MSIINOZ]

Yep just use a splitter & they should run at the same speed, I dont think you will gain any temp drop at all though using 2 fans on the H60 but let us know if it somehow does?

 

http://www.ebay.com/itm/GELID-PWM-Y-Cable-Splitter-Adapter-Share-12V-Case-Fan-/180746442099?pt=LH_DefaultDomain_0&hash=item2a1553d573

 

As mentioned above using a shroud will give you an actual gain & drop the temp and also without adding more noise, Those stock H60 fans are damn noisy, I still have a brand new one sitting here that I will never use

[parsec]

Yep just use a splitter & they should run at the same speed, I dont think you will gain any temp drop at all though using 2 fans on the H60 but let us know if it somehow does?...

 

What's this, a push-pull heretic? Blasphemy! How could you question a technique used by liquid cooling/radiator users since the dark ages, as well as on some straight air-cooled CPU coolers? ;):

 

Adding a pull fan will usually buy you a few degrees lower CPU temps. That is what has been seen in reviews many times, it's almost always not a huge difference. I could see it helping more if two low RPM, ultra-quiet fans were used, but that is rarely done.

 

My thoughts on matching the fans or fan speed in a push-pull fan setup is also in the heretic category. When only a push fan is used, the air flow created by the fan after passing through the radiator is significantly changed, it is reduced due to the physical resistance created by the radiator. The radiator creates a small amount of back-pressure on the push fans airflow, further reducing the resulting airflow, which is why high static pressure fans are recommended for use with radiators. The airflow out of the radiator has negligible if any resistance due to the PC case, unless the case is configured as a well sealed, very high positive pressure cooled case.

 

Adding a pull fan whose speed or air flow level is known to be less than the push fan (not restricted by a radiator) assumes that the pull fans airflow is less than the resulting airflow through the radiator created by the push fan. It also assumes that the pull fans airflow, or suction in this case, will still be a greater resistance to the air flowing out of the radiator in the push fan only configuration.

 

This makes sense IMO, although I have not tested it, and I'm not saying this is true. The main question is at what point does a pull fan create a greater resistance to airflow out of a radiator than nothing at all? When the pull fans airflow rate is less than that exiting the radiator. Given the airflow rate of the push fan is reduced due to the radiator, it seems to me that a lower speed/airflow pull fan will not always be restricting the airflow through a radiator.

[MSIINOZ]

To cut it short (Edit, Tried), I agree that adding a pull fan will/SHOULD help reduce the temp, Firstly why anyone would setup with just a fan pushing through the radiator is beyond me?

 

If you only have one fan put it inside the case or either way use it as a pull fan & you will certainly gain a temp drop, All cars & trucks etc etc have used this method on their radiators light years before PC water cooling was even thought of, Also all cars now use fan shrouds too!

 

Edit just found this looking for PC cooling info, Automotive but Same deal>

 

"Fans work the best when used in the "pull" configuration. In other words the fan should be mounted on the backside of the radiator facing the engine."

http://www.a1electric.com/spal/faninfo.htm

 

Also mentioned above, Using a fan shroud & pulling air through the entire surface area of the radiator core will certainly work much more efficiently than a fan hard against the core & ONLY pushing air through the radiator exactly where the fan blades are which is a much smaller area than the core itself & a much smaller area than 130mm that the coolingworks shroud pulls air through

 

To sum it up pulling air through the core with (Or without) a shroud will work much better & Im certain better than 2 fans pulling & pushing

 

If you dont mind heaps of noise (& are possibly overclocking for a world record:roll:) then go ahead & use 2 fans but I am working on cooling with the least amount of noise & still be efficient as some out there that have a zillion fans sounding like a tornado, My H60 (Modded) cooled HTPC is pretty much silent which is partly why I wanted to go with water cooling, Introducing more fans & noisy fans seems to defeat the purpose of water cooling

 

I do agree pretty much with everything you are saying persec but I think there is a limit to the airflow that is needed for a radiator to work

As shown in my thread>

http://forum.corsair.com/forums/showthread.php?t=106013

I gained a 1 degree temp drop using the stock H60 fan spinning at full (very noisy) speed compared to my Gelid TC fan spinning at half the speed, I was actually surprised to find that airflow does not exactly equal more cooling

 

It would be good to hear back from luckbad on the subject, Please trying running your stock fan on its own pulling air in through the radiator & report back if there is any difference in the temp? I think you will find you dont even need the other fan, Especially if you were to fit a shroud also

[MSIINOZ]

I just found this thread which backs exactly what I was trying to explain, More air does not mean more cooling & it actually made it worse

 

http://www.tomshardware.com/forum/272176-29-push-pull-configuration-producing-heat-pull-configuration

 

Luckbad, As your name states it may be badluck/A waste of time adding another fan but please let us know your results?

 

I know the best setup from experimenting & thats obviously why I am trying to recommend getting a fan shroud

[parsec]

Very interesting MSIINOZ, and of course goes against the standard "knowledge" of using push fans with PC cooling radiators. Well worth trying a pull fan only setup. I will add my commentary on your good post, simply discussion and analysis.

 

Pull fans in vehicles makes sense, since push fans would obstruct the airflow caused by a moving vehicle. In the old days before sideways mounted engines, fans were attached to the engines. Electric fans did not exist in most vehicles prior to front drive cars, and then became a necessity.

 

Consider the worst case cooling situation for a vehicle, not moving in traffic on a hot summer day. The shrouded pull fan method works. One might argue that since push fans simply aren't applicable in vehicles, we really don't know which is better, and the comparison is apples to oranges. That is a valid argument, but I can't dismiss pull fans on that notion alone.

 

The fact remains that PC liquid cooling product manufactures tell us to use push fans. I would hope there is a good reason for that, but you never know. I am always open to new and different ideas, and pull fans only is one worth testing, IMO.

[sickfiddydee]

Very interesting MSIINOZ, and of course goes against the standard "knowledge" of using push fans with PC cooling radiators. Well worth trying a pull fan only setup. I will add my commentary on your good post, simply discussion and analysis.

 

Pull fans in vehicles makes sense, since push fans would obstruct the airflow caused by a moving vehicle. In the old days before sideways mounted engines, fans were attached to the engines. Electric fans did not exist in most vehicles prior to front drive cars, and then became a necessity.

 

Consider the worst case cooling situation for a vehicle, not moving in traffic on a hot summer day. The shrouded pull fan method works. One might argue that since push fans simply aren't applicable in vehicles, we really don't know which is better, and the comparison is apples to oranges. That is a valid argument, but I can't dismiss pull fans on that notion alone.

 

The fact remains that PC liquid cooling product manufactures tell us to use push fans. I would hope there is a good reason for that, but you never know. I am always open to new and different ideas, and pull fans only is one worth testing, IMO.

 

Actually, if I am reading the correct portion of the linked page...

Q. Can I reverse the power wires and make my "pull" fan work as a "push" fan?

 

A. Literally, yes it can work. But it will not work very well. Spal fan blades are designed to work in one direction. So our recomendation is "Don't do it".

... it would appear not to be a general condemnation of "push" fans, but rather a notation of the idiosyncrasies of this particular manufacturer's fan, due to its blade profile.

 

Also note that it is a discussion about reversing the fan's direction of rotation, which is not done when using a fan as a "push" fan on an H60 radiator.

[Luckbad]

I'll do some experiments when I get in the extra Corsair fan. I currently have a slightly lower CFM fan inside the case pulling and haven't noticed appreciably different temperatures from when I just had the one Corsair fan pushing by itself.

 

Frankly, I had good temperatures just running the Corsair fan as exhaust as well, but that was just when testing the CPU and not the entire case under load (which raises the ambient temperature of the entire case).

[MSIINOZ]

Well if you have a lower CFM fan pulling inside the case and the temp has not increased that is good and goes with the point I am trying to make

 

When you say "different temperature" has there been an increase or decrease?

Im guessing you have possibly dropped 1-2c?

 

If you dont already have a 120mm fan at the top of your case? you could try the corsair fan there for an experiment, It will help the pull fan draw more air through the radiator & may even work better than using the extra corsair fan as a push fan

 

You will drop a few degrees just using a shroud on the radiator, It will be good to see how it works out for you either way

[babdi]

Msiinoz

Your concept of pull may not work always as I found out. When I installed my 1st H50 I pulled outside air in. Then I changed to pull hot air out. I got 2 degrees drop in temp. You should understand that people allover use these. In tropical countries for example there is not very much difference in outside the case and inside the case temps in idle.Summers in our country reach 46 deg C in many places,so having to blow outside air inside would make case more hot.

[MSIINOZ]

Yes Babdi I totally understand what you are saying, I used to (thank god) live in an opal mining town in outback Australia near bourke (see link below) & the temp would get near 50 deg C

 

http://www.gungahlinweather.com/articles/factsandfigures.htm

 

My point I am trying to get across is that using the fan to pull air through the radiator is more efficient than using it as a push fan, Whether you are pulling air into the case or pulling air out of the case, That is another whole debate again which is exactly as you mentioned a case of ambient temp being better or worse than inside your case? That is the deciding factor

[babdi]

Pulling is better than pushing,has a better mechanical advantage,don't know if it applies pneumatics :)

[parsec]

While the discussion in that link was not the best explanation regarding the use of pull fans for cooling radiators, IMO that concept deserves testing. The link serves more to describe the concept rather than show whether or not push is superior to pull for cooling PC radiators.

 

This reminds me of a debate about fans being designed for use as push or pull. There seems to be impeller optimizations for both applications, but we use standard PC fans for push and pull use all the time in PC cases, for example. A PC fan being designated for a specific use (push/pull, intake/exhaust) is very rare.

[parsec]

Removed my push fan, leaving the identical pull fan, PWM type, max RPM ~2000 RPM (not stock) on my H60 radiator. Nothing else changed, left all shrouds in place. Used on a stock clocked, SpeedStep enabled i7-930 in intake mode, exhausting into a well ventilated case.

 

Idle temps after a few hours use are at most 1 degree C higher, from 27.5C to 28.5C. IBT results were identical, or no more than 1 degree C higher with the pull fan only. Quieter of course with a single fan, but I'm tempted to say the tone of the noise is different.

 

The push-pull fans were connected to the board with a PWM splitter cable, and were identical. Have I reached the max heat dissipation capability of this radiator with two fans, which is a bit below that with one fan, and a pull fan too? The stock fan could also function at or just below the max dissipation capability of the radiator.

 

I would need to go back to a push only test to make a true comparison. I had bought into the "push fan is superior" notion, which is the general liquid cooling standard as far as I can see. You live and learn...

 

I would not hesitate to recommend at least trying using just a pull fan if that was most convenient with a PCs configuration. It's clear to me that it is not greatly inferior to push mode fan mounting, they seem to me to be very close in cooling performance, although I cannot call it superior.

[babdi]

Parsec

IMO the radiator thickness plays a great part in the fan performance in push or pull combination. I feel in H60/H100 the difference may not be significant while in H80 it will.

In H80,pull setting the air hits the radiator 1st and gets diffused,while in push setting the air from the fan is focused at the radiator entry and defused or dissipated at the exit.