Air flow exceed rated CFM for HEPA filter

[immigrant]

Hi,
We have installed HEPA filter 24x24x12" in the duct. This filter is rated for 2000 CFM, but we have actual 3800 CFM air flow at this point.
The filter is collapsed for some reason. My question is, if it is happen due the exceeding CFM or no? Does somebody have an experience.

Yury

 

[ChasBean1]

It's happening due to the pressure across the filter from the additional flow. The pressure across the filter is about 3.6 times higher at 3800 cfm vs. 2000 cfm. You should think about doubling the size of the filter bank.

 

[CountOlaf]

Yeah, I would think that's the problem. You've basically increased the face velocity (fpm) from 550 fpm to 1000 fpm. Most filters have a face velocity limitation bewteen 200 and 800 fpm with 300 fpm being a typical upper limit for inexpensive throw-aways. Even if a filter doesn't collapse at high face velocities, you'll probably ending up blowing through dust/dirt you're trying to capture or bypassing a lot of the air/dirt at the edges.

If you plan on keeping the same airflow, expand the ductwork for the filter section (i.e., taper out to an appropriate size (maybe 42" x 24")) and back in after the filter section. Check what size of HEPA filters you can get of course and design accordingly.

 

[mauricestoker]

It should be a linear relation, so you need 1 SF area per 1,000 CFM flow. You need to roughly double your area. Along with that, the leakage/bypass should be tested for the HEPA regardless of whether it is properly sized or not, as CountOlaf has indicated, bypassing crap defeats the HEPA.

I haven't seen an off the shelf HEPA that handles 2,000 CFM/SF, but I've never really looked for a custom job. Static pressure would be high. If you haven't done so already, I'd recomend a prefilter and dP across each filter.

 

[Zesti]

Chasbean1,

You are assuming a "squared" relation between flow and pressure drop, which is of course true for most pressure losses.

However, this is not the case for a HEPA filter. These are near linear and so a doubling of the flow will result in about double pressure loss.

(Don't ask me about the physics on this one...probably got something to do with the way the filter actually captures the fine grains etc.)

 

[ChasBean1]

Maurice, might you mean 500 cfm per ft2? 1000 fpm is a high face velocity.

Zesti, really? That sort of defies logic for me...

 

[Zesti]

Chasbean1,

Yes, I know, it does not seem logical but it is true.

I just picked a "random" document from the TROX range:

http://www.troxtechnik.com/xpool/do...ts/filters/leaflets/f7_33_partic_fi_cells.pdf

Have a look at the graphs on page 3 and 5: linear relation between flow and pressure differential.

 

[Cniper]

Is it possible to put another filter in parallel, hence accommodating your flow, keeping similar pressure drop and also keeping within your DF range

 

[sridhar1312]

In most of the Air handling applications the first thing to be done is tuning your speed of the FAN to bring it closer to the designed cfm.

The design static will be higher considering the worst case of dirty filter conditions for all filters. In case of HEPA filter when it is dirty the static will be more, whereas initially the static will be less than the design contributing to higher cfm and consequently the velocity across filter is more than design and will make the filter performance topsy turvy.

Hence it is essential to fine tune either with VFD to reduce the speed or change the motor pulley close to the design cfm and have pressure differential measured to clean or replace the HEPA filter when it becomes dirty or not cleanable situation.

 

[mauricestoker]

HEPA's are available in both 500 and 1000. When possible, I like to use 500 on supply and 1000 on exhaust.