Cube of HVAC filters w/ PC inside cube?

Enclosures and acoustic damping to help quiet them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Devonavar

Post Reply
Michael-Inet
Posts: 25
Joined: Sat Oct 31, 2015 3:03 pm
Location: Austin or Nashville
Contact:

Cube of HVAC filters w/ PC inside cube?

Post by Michael-Inet » Wed Aug 08, 2018 2:42 pm

Hi All,

I’ve had this thought for several years, maybe close to two decades, but I never really explored it.

Facts we agree upon?:

- The more unrestricted airflow inside a case the lower the temps.
- Leaving the case open allows very good airflow.
- An open case accumulates dust pretty rapidly.

So the thought was to build a box to put your PC in that’s made out of standard HVAC filters that can be bought cheaply at the Lowes, Walmarts, Costco’s of the world. Then add one, or more, low noise exhaust fans [1] to the top and disconnect the no longer needed case fans. Leaving the CPU/GPU fans active.

A search here pulls up something similar by slipperyskip [2]:

viewtopic.php?f=14&t=56127

Which is way overkill, as I’m not looking for ‘pretty,’ just very effective.

- Would it provide enough ‘clean air’ to work?
- Would some other air filter material provide better filtering, more sound deadening, cheaper costs?
- Other thoughts or drawbacks?

Best,
Michael

[1]
http://www.xoxide.com/360mm-case-fan.html

~$25
Dimensions: 360 x 360 x 30mm
Rated Voltage: DC12V
Rated Current: 500±10% mA
Locked Current: 600±10% mA
Rated Input Power: 6±10%W
Operation Voltage: DC9~13V
Starting Voltage: ≤4.5V
Rated Speed: 600±5%rpm
Maximum Air Flow: 126.7 CFM
Maximum Static Pressure: 12.3mmH2O
Noise: <21.7 dB(A)
Lead Wire: 4PIN
Work Temperature: -10~85°C
Motor Temperature: ≤ 85°C

http://www.thermalright.de/en/fans/46/ty-147-a

~$18
Manufacturer Number TY 147 A
Dimensions 140 x 152 mm
Installation Depth 26.5 mm
Voltage Range ~ 5 - 13 V
Fan Speed 300 - 1.300 U/min
Air Flow 28,7 - 125 m³/h (16.9 - 73.6 CFM)
Noise 15 - 21 dB(A)
Life expectancy 50,000 hours
Bearing Enhanced Hyper-Flow Bearing
Connector 4-Pin PWM

[2]

You can see many more of slipperyskip’s designs at, http://slipperyskip.com/page23.html

whispercat
Posts: 376
Joined: Tue Jul 22, 2008 6:05 pm
Location: Calgary, AB, Canada

Re: Cube of HVAC filters w/ PC inside cube?

Post by whispercat » Thu Aug 23, 2018 12:23 pm

Interesting. But you wouldn't necessarily need to make the whole PC case out of HVAC filters. Why not just place HVAC filters where the case openings are?

Michael-Inet
Posts: 25
Joined: Sat Oct 31, 2015 3:03 pm
Location: Austin or Nashville
Contact:

Re: Cube of HVAC filters w/ PC inside cube?

Post by Michael-Inet » Fri Aug 24, 2018 7:11 am

whispercat wrote:Why not just place HVAC filters where the case openings are?
Short answer: Airflow.

If you only place filters on the case openings, you'd restrict the airflow so much your case would become a furnace. To overcome the filter resistance, you most likely would have to turn your fans into aircraft engines to achieve the prior case temperatures.

Removing the case entirely and adding massive amounts of filter surface area should have reasonably equal, and possibly much better, heat extraction from your components as the original case/fan design.

The probable upsides of a 'filter case' are:

- Your components are now pretty much 'dust' free.
- Depending on which fans are used/removed/replaced your PC
- - is significantly quieter.
- - is significantly cooler.

Best,
Michael

Here's a link on filter airflow, a bit technical:
https://www.aivc.org/sites/default/file ... _12995.pdf

whispercat
Posts: 376
Joined: Tue Jul 22, 2008 6:05 pm
Location: Calgary, AB, Canada

Re: Cube of HVAC filters w/ PC inside cube?

Post by whispercat » Sun Sep 16, 2018 1:22 pm

Not sure I follow: if the filter is so air resistive, then making the whole case out of the same resistive material would make the whole case just as resistive, no?

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: Cube of HVAC filters w/ PC inside cube?

Post by CA_Steve » Sun Sep 16, 2018 2:16 pm

whispercat wrote:Not sure I follow: if the filter is so air resistive, then making the whole case out of the same resistive material would make the whole case just as resistive, no?
filter is less restrictive than a solid piece of metal. :)

Michael-Inet
Posts: 25
Joined: Sat Oct 31, 2015 3:03 pm
Location: Austin or Nashville
Contact:

Re: Cube of HVAC filters w/ PC inside cube?

Post by Michael-Inet » Sun Sep 16, 2018 3:09 pm

whispercat wrote:Not sure I follow: if the filter is so air resistive, then making the whole case out of the same resistive material would make the whole case just as resistive, no?
No.

A fundamental of fluid dynamics is that the more volume a fluid has to pass through the greater the total flow rate will be for the same amount of ‘pumping’ force. We’re only caring about keeping the same fluid volume (air flow in CFM in this instance) going into and out of the ‘case’ to maintain the same case temperature. [1]

Vastly and horribly oversimplifying and these numbers are guaranteed to be wrong...

The TY 147 A fan has 73.6 CFM airflow when completely unrestricted.

It’s dimensions are 140 x 152 mm (21,280 mm2) and we’ll suppose the case opening(s) feeding that fan are 1 square foot (92,903 mm2). Think of the fan having a 1 square foot hose feeding it.

Adding a filter to the case opening will decrease the ‘volume’ of the 1 square foot hose by something like 90% (very fuzzy, and completely dependent upon the specific filter used), so now your fan is being fed by an opening of 9,290 mm2. Cutting your feed hose to less than half of the fan’s dimensions will reduce the CFM the fan produces to ???. (I’ve got no idea, but you can see the issue. It’s the same as putting a piece of cardboard over half your fan.)

Enclosing the entire space your PC is in with four, 2ft x 2ft filters will give you a total filter surface area of 16 square feet (1,486,448 mm2). Reducing that by the same 90% gives you 148,644 mm2 of fan feeder hose. Under this your fan should still be producing 73.6 CFM.

# # #

If this doesn’t help you, then I’ll have to punt you to your nearest University’s physics department...

Best,
Michael

[1]
CFM, cubic feet per minute
ft2, square feet
mm2, square millimeters

Michael-Inet
Posts: 25
Joined: Sat Oct 31, 2015 3:03 pm
Location: Austin or Nashville
Contact:

Re: Cube of HVAC filters w/ PC inside cube?

Post by Michael-Inet » Sun Sep 16, 2018 4:27 pm

Steve, Abula,

Do either of you know what your actual case airflows are and what temps that keeps your case? I think I'll write this up and post it on stack physics and see if I can get real world numbers...

Best,
Michael

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: Cube of HVAC filters w/ PC inside cube?

Post by CA_Steve » Mon Sep 17, 2018 7:08 am

Sorry, I don't own an anemometer.

One thing you haven't considered is how the case factors in meeting FCC Part 15 rules. Moving to a non-metal case may raise the PC's EM radiation enough to impact other devices (BT, Wifi, your cellphone, etc...).

Here's a nice webpage on shielding. The bit on apertures is beneficial.

Michael-Inet
Posts: 25
Joined: Sat Oct 31, 2015 3:03 pm
Location: Austin or Nashville
Contact:

Re: Cube of HVAC filters w/ PC inside cube?

Post by Michael-Inet » Wed Sep 19, 2018 6:41 am

Hi Steve,

FCC shouldn't be an issue. Several reasons:

- All the components are bought retail, so they individually have FCC compliance.

- We’re still using a standard case. Granted it’d be the cheapest we can find (~$20) and it would be left ‘open’ with all the panels off. I’ve used open cases in the past, and known plenty of people who have had open cases, and I’ve never really heard of anyone having a problem with it. (Other than dust accumulation.)

- Absolute worst circumstances, adding an internal layer of conductive material would address this [1]. Grounding it to the case as needed/desired (the case being grounded by it’s power connection). You're not likely to make a true Faraday cage, but you should get enough attenuation to meet your needs.

# # #

On the needed airflow, I think I can reverse engineer that from max power draw (Watts) being converted to heat and then figure out the airflow volume needed to maintain say a 5 degree rise over ambient temperatures ‘inside the case.’

Ah, fun with physics ;)

Best,
Michael

[1]
Chicken wire, fine metal mesh, etc. But not something like aluminum foil as, duh, that’d block our airflow ;)

Reference (copy/paste, not validated):
The wavelength of a cell phone signal of a frequency 1800 or 1900 MHz is around 15 cm.

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: Cube of HVAC filters w/ PC inside cube?

Post by CA_Steve » Wed Sep 19, 2018 10:27 am

<shrugs> sounds like a fun experiment. If you do go with a wire mesh Faraday cage, I think a mesh diameter of less than 1/10th the wavelength of the offending source should provide decent filtering. A 2GHz signal is 15cm. So, if that were the culprit, 1.5cm diameter mesh. Realistically, it's the 2.4GHz band to be concerned with (wifi, BT, every unlicensed dongle out there..). So, ~1.25cm mesh. Uh oh, now we're in the range of what all those fan enclosure honeycombs are...:)

Michael-Inet
Posts: 25
Joined: Sat Oct 31, 2015 3:03 pm
Location: Austin or Nashville
Contact:

Re: Cube of HVAC filters w/ PC inside cube?

Post by Michael-Inet » Sun Sep 23, 2018 7:58 am

CA_Steve wrote:So, ~1.25cm mesh. Uh oh, now we're in the range of what all those fan enclosure honeycombs are...:)
lol, not a chance, those things are expensive :o

For those playing the home game, you want something like aluminum bug screen. You'll need to caliper the opening, but they usually have about a 1mm aperture as well as being cheap and easy to find at Home Depot, Lowe’s, Walmart?, etc. Just make sure it's metal and not fibreglass.

Olle P
Posts: 711
Joined: Tue Nov 04, 2008 6:03 am
Location: Sweden

Re: Cube of HVAC filters w/ PC inside cube?

Post by Olle P » Tue Sep 25, 2018 5:05 am

I wouldn't bother about any EM concerns.
Made my own box of board seven years ago and had no problems whatsoever.
Used an older cheap case as the base. Removed all covers, any fan mesh and made a big hole in the bottom.
Then placed the remains inside the wooden box to reduce any noise.

Some pictures:
The box during build. There is a noise dampening channel for intake air below the bottom fans and an also dampened exhaust down behind the rear wall.
Image

Exterior of the finished box.
Image

Front doors opened.
Image

Inside, with computer parts.
Image

Post Reply