I have an Antec Solo Mk1 case with two silent 120mm fans, one at the front bringing air in and one at the rear for exhaust, plus Arctic Cooler7 pro MK11. This has been running fine with a fanned PSU , but now replaced the PSU for a fanless and am concerned about the movement of air around the PSU. It seems to be isolated from the airflow in the case and is only vented at the top and some smaller vents to the back and side. The top is onlt about 20mm from the top of the case and I assume any hot air will rise into this space and become trapped.. I thought of chopping the case top and rigging up an exhaust fan, I can manage this OK but would rather not.
My other thought is to reverse the rear fan and have this too blow inwards, this would be fine for the cpu cooler I think plus it would along with the front fan give positive pressure in the case which ( this is where I am unsure ) push the warmer rising air to the top where it would vent out of the rear of the PSU grill. I could even internally mount a smaller silent fan to blow into the rear of the PSU to help the air flow that way.
The system is not Oclocked, has passive cooled GPU and runs cool anyway.
Your thoughts on this appreciated.
System consists of:
Antec Solo Mk1
Gigabyte GA 970A-DS3
Phenom11x4 955 BE (125w)
2x2Gb AMD performance ram
Samsung 128GB SSD
WD 500GB HDD sata2
LG Optical.
ATI 4380 passive cooled
Arctic Pro7 rev 2
PSU is a Silver Power 480 Gold certified. ( system requires 385 watts tops, with imaginary additional hard drive in the calculation)
Fans as described above
No gaming, no rendering.

Correct me if I am wrong, but you should have two front 92mm fans not one 120mm fan correct?

You have the germ of a good idea. The problem is that most fans don't respond well to fighting each other like you propose. By the way, the force of warmer rising air is very weak and will be irrelevant in your situation.

Before irretrievably chopping up your case, try this.

Get two of the higher speed Scythe Gentle Typhoon fans plus a fan speed controller. This one is good fan controller because it will always generate reproducible speeds, and it has a lot of speeds to select from:
http://www.performance-pcs.com/catalog/ ... s_id=25981

Then experiment.

Here is the experiment that I think will work. Block off the rear fan and the rear vents. Try just a single 120mm GT fan on the front. This will likely bring in more air than two fans fighting each other, and all the air will be forced out through your PSU. Keep in mind that the size of the case and the number of fans are not relevant. What is relevant is the amount of air flowing through the case and the amount of heat that air is able to carry away with it. Everything else is just a means to this end.

If there are two fans that can tolerate fighting each other with both pushing air in, it will be the Scythe Gentle Typhoon fans. The one area which they are the best in, is performing under high static pressure.

If you have the ability to mount two 92mm fans on the front, I would try this with two 92mm GTs. That might work even better than a single 120mm.... and will give you some redundancy in the event of a fan failure.

Can you flip it over so that the vent is facing downward?

Thanks for the replies, I will respond, although in reverse order, last first.
As for mounting upside down this is a strict no go. The bottom is a solid plate, 1mm Ithink so probably a bit of a heatsink as well, but being a solid face it will not allow any heat to escape upwards either. In theory I could flip it and use the now vented lower face for either an inlet or exhaust vent. But its too expensive an experiment for me given the high cost of the PSU. It is based on a Seasonic X series and exactly the same proviso is given by them re: the mounting.
The two fans not complimenting each other in a positive pressure set up.
The front fan is set at the bottom, the rear is above midway level with the CPU, so they are not facing each other and I think the air flows would merge, with the CPU fan lifting air upwards.( I am unsure as to whether or not I could remount the CPU cooler to blow in another direction, might be worth considering if it helps any particular configuration)
I have no way of knowing if any of this would work so thank you for you help, appreciated.
The rear fan does have a controller on a remote switch so I do have some 3 speed control, but would prefer the lowest setting for the quietest setting.
The front did have two 90mm fans originally but I found that the honeycomb grill did make more noise ( droning) than I was happy with so I took the snips and removed it, this left room for a 120mm, a tight fit with the bay trays, but doable. Much quieter and for some unknown reason far less dust attracting! ( the filter is still in place. )
Thinking now of setting the front 120 up at the rear of the PSU, perhaps with a duct, to blow air through and out of the rear of the PSU. I realise this will lose the positive pressure but I can see how I get along with that.
Thanks for responding Ces, this is something I have not really considered much before.

Do you have any clues on actual psu temps? Is it shutting down because of overheating or are you just concerned that it might overheat?

see also: viewtopic.php?t=60172

No issue with overheating anywhere, just building it up. Want to give it some serious thought thats all.

Thanks for the link did not come across that one. A few points about it:
The buzzing PSU sounds more like a oneoff problem, maybe an RMA .
The positive / negative airflows we have discussed a little and I would think a negative case would be hotter, especially with this PSU.
The top vent chop I can understand but it really would be a lst resort and I am sure there are easier ways to get a good result.
The last point: The PSU duct/ chamber. This I could fabricate in acrylic ( plexiglass to you I think )
I have had a dig in my boxes of bits and have two good ( and V quiet ) Xilence 90mm fans. I could use this pair for the front panel instead of the 120mm and use the 120mm to fabricate a PSU duct, in effect leaving the rest of the case as it is, which I know runs very cool from when I had the previous PSU in, and the fanlss PSU with its own airflow solution. I know it might seem a retrograde step, getting a fanless CPU and the rigging a fan solution for it, but its controllable, the PSU is way more efficient than a fanned unit anyway ( has to be ) and I can experiment with various fans to get the best solution quite easily if its remotely mounted...
I want to get it right and whats more love tinkering round with it like this anyway!

It isn't about flows, it is about pressure. Each fan is exerting pressure against the other, wherever they happen to be located, each forcing air to back flow against the other's blades. Basically each is generating unwanted static pressure for the other.

That is why one of your experiments, prior to resorting to the cutting of sheet metal, should involve sealing all the exits other than the PSU exit and trying bring air in from just one inlet, the front. All things equal, two 90mm fans will do the job better than one 120mm fan... If things are equal, small fans handle static pressure better than larger ones. Xilence fans are not GTs though.

In your case, I would at least get one 120mm Scythe GT and see what it can do, if you permit it alone to generate static pressure to push all the air out of your case through the PSU housing.