Air "Pusher"?

[ERIMechEng]

Problem: I have a 20"x16"x7" box with some heat generating radio components inside. The aluminum box is sealed for EMI concerns, but the lack of heat exchange is causing the internal air temperature of the box to get too high, and the components are overheating and failing. As a modification, we installed EMI fan vents, which will allow some air to flow in and out, while keeping the EMI shielding intact. We installed some computer fans on the underside of the EMI vents to induce some air flow, but these are obviously not designed to "suck" air through something akin to an air filter.

Question: do "they" make any low profile fans that are designed to suck air through a filter? Mounting provisions are variable, though right now I'm working with an 80mm bolt hole square (size 6 screws). The reverse is do they make any fans that are designed to push air through a filter, or are conventional computer-type fans the best I'm going to find out there?

 

[IRstuff]

You post all this informaiton, yet say nothing about this "filter." What size openings are they, and what was the rationale for choosing their size? Is there some other "filter" you've added? The fact that the original premise was that you needed a solid metal box for EMI and now you allow holes suggests that the design is being winged. You should get a EMI consultant to properly size the openings.

Computer fans have no problems with the holes that come in computer cases, so you've left out a critical part of the problem.

TTFN

FAQ731-376

 

[ERIMechEng]

Fair enough. The openings are not fixed in size as they would be with a hole-punched computer case. The EMI vents are made up of a thin walled honeycomb configuration with (I'm guessing) about a 95% open area, which poses no problem. The "dust screen" that caps the EMI vent is the problem: the dust screen is made up of a series of weaved meshes that has a very low open area - lets say on the order of 10% open.

Here is the link to the EMI vents:

http://www.tech-etch.com/shield/8200series.html

Yes, the EMI vents have been added as an afterthought (winging it, as you put it), but the EMI vents are proving to be a pretty good solution. With regular computer fans installed as pushers and pullers, we're getting a 40 deg F drop in temp. What I'm looking for a is a more efficient way to move the air through these EMI vents, given that the % open area is quite small and is hindering airflow.

 

[IRstuff]

Ok, then you need to find a beefier fan. It's simply a matter of cost, power, and noise. Or, look at this dust filter and find something that better. People have used simple open-cell foam-type filters.

You only have a limited set of choices:
> Larger fan
> Lower resistance filter
> Larger overall opening

TTFN

FAQ731-376

 

[ERIMechEng]

Yes, those are the obvious choices.
Let me ask it a different way: Are there alternatives to conventional (axial) flow computer fans, and what are the trade-offs? I know of centrifugal fans, but I don't know the trade-offs. Any other way to circulate air?

 

[Compositepro]

You may simply be using inappropriate filters. However, if you must use filters that are so restrictive to flow then you must use higher pressure blowers rather than fans. Small squirrel cage style blowers are readily available.

 

[ERIMechEng]

CompositePro...you helped me before, and you've helped me again! Squirrel cage blowers seem to be a good solution. These can give a higher output pressures (compared to axial fans), but I wonder if I can get any data on how well they can overcome and inlet restriction, or are they the same thing?

 

[Compositepro]

All fans, blowers, and pumps have performance curves (generated pressure versus flow rate). Vendor should be able to supply that.

 

[IRstuff]

The first thing you need to do is to determine the characteristics of your FILTER. Only then can you even begin to size the fan.

Finding a fan is usually the easiest part:

http://www.ametekaerodefense.com/pr...tCAtegory_Id=4259&Product_Id=1086&SKU_Id=1279

The big question is whether your filter will pass any air with a 4" pressure loss.

TTFN

FAQ731-376

 

[BM4BRANES]

At risk of stating the obvious, we should assume that the convention heat sink methodoligy found in the electronis field is no practicle. If you had a constant source of clean dry air from a local compressor, you could diplace the heat while avoiding the maintenance/dust /fan failures issues. They(someone)make a vertex cooler that utilizes cda as well, they are a little noisy but get the job accompished.

 

[ERIMechEng]

BM: no source of CDA, so i'm stuck with sucking in ambient and pushing out hot ambient.
For those interested, we're going to end up with a combination of centrifugal fans on the inlet side and conventional axial fans on the exhaust, both pushing air through the EMI vents. I've got a call in to the manf of the EMI vents to make sure that we wont be loosing any sheilding effectiveness by axing the dust screen (it shouldn't matter, but better to double check). If the dust screen is non issue, we will cover all vents with wire mesh to guard against FOD.
Thank you to CompositePro, IRstuff, and BM4BRANES. Happy Holidays!

 

[CountOlaf]

bm4branes: yeah, they're called vortex cabinet coolers; we've used some from Exair but I suspect there are a lot of manufacturers...

 

[RossABQ]

You might consider some of the refrigerant cooling systems used for servers; APC and Liebert offer these. Your system is likely way down on the low end of their applications, but it's scalable. You put a coil inside the box, feed in cooled refrigerant, it vaporizes and removes heat. I'm sure you can find a dielectric joint for EMI converns.