Controlling room pressure in a closed loop (again)

[skuntz]

I have read past discussions on this topic and I understand the concept that there must be a flow differential between the supply and exhaust to maintain the desired pressure in space. I have also reviewed the background information on how to calculate this differential. What I have not seen is how to physically implement a system or technique that relates to a specific project that I am working on. Allow me to explain.

We are circulating air at approximately 1000 through a rather large sealed enclosure - let's just say it's the volume of several rooms. We are using a single blower in the loop. Air is supplied directly into the enclosure and on the return side it passes through a bank of filters (HEPA & gas adsorbers) before returning to the suction end of the blower. It is permissible to allow makeup and bleed air so long as no air escapes before filtration. The enclosure is to be held at between 0 and 0.1 inches w.g. Control will be based on the feedback of a pressure transducer at the enclosure.

May I please get some advice from HVAC or other "experts" who may have implemented a similar setup? Sorry to be a bit long winded but I think in these forums one needs to be very specific to get meaningful feedback.

 

[CountOlaf]

Sounds like you're on the right track. If you want to maintain a slightly positive pressure in the enclosure compared to outside the enclosure (0.0 to 0.1" w.g) , you are going to need to introduce make-up air (as you say), and of course, ideally, upstream of the fan intake and filters. If the enclosure is sealed really well (unlike some buildings), then you'll have to provide a means of relief (or bleed air as you call it). The relief damper might be just a simple device (i.e., no actuator) which opens up when the pressure exceeds 0.1" (they make these with a couterbalance that little weights are added/removed to set up for your specific pressure); and the make-up air duct may just be balanced with a manual volume damper.

If you want precise control of course, you should use actuated dampers on the make-up air duct and exhaust/relief duct which are controlled via a differential pressure sensor/controller measuring both inside and outside the enclosure (Magnahelic or others make these pretty commonly for theses type applications).

If you are conditioning the air besides filtration (heating or cooling), the make-up air will put an additional demand on those systems, although it won't be a whole lot by the sounds of it.

BTW, what units of measure was that "1000" of yours?

 

[skuntz]

CountOlaf-

Thank you. You response alerted me to an error I made in my original post. I should have written "0 to -0.1 in w.g." as I want a net vacuum in the enclosure. Sorry I forgot to specify my units as 1000 CFM.

What I was thinking was using two dampers, one at suction and one at blower discharge.

 

[CountOlaf]

OK. Yeah, positive or negative pressure, it should be the same basic set-up. If the make-up and relief volume is around 10% of the total flow rate or 100 CFM, you're not looking at very large ducts/dampers--maybe 6" square/round?

 

[ChasBean1]

Negative is a new ball game with this setup. You have a single blower - you'd have to relieve a portion of air on the supply side prior to being delivered back to the space. You'd take in no makeup air. If you want makeup air, you'd probably want a separate exhaust that removes slightly more than what you're making up.

 

[CountOlaf]

Chasbean1:

He's talking about putting a relief opening/damper downstream of the blower, but before it enters the space; and also a make-up opening/damper upstream of the filters and blower, again outside the controlled space. The DP switch would control to setpoint and would relieve/exhaust more ariflow volume than is made-up to maintain the vacuum. So, in terms of number of control dampers, fans, DP switches, and locations of devices, it's the same basic set-up, right? I'm only clarifying because I want to make sure I haven't missed something fundamental/critical.

 

[skuntz]

CountOlaf:

That is the setup I was thinking of exactly. I was looking for validation of this idea and any alternatives.

 

[mauricestoker]

I'd suggest using pressure independent air control valves with master-slave board, pretty much a standard Phoenix air valve configuration. If you look on the Phoenix web site, I think you will see an example of how it works.

As you are maintaining negative, I'd recommend setting the exhaust for whatever minimal ACH are required, for IAQ, process, heat load, whatever. Using a magnahelic or photohelic, adjust the master-slave air differential until you hit the differential pressure desired. Pretty much the same as for doing a damper balance except you have pressure independent operation.

Only time I've had to use the relief damper (with pressure independent controls)is where design was altered from differential volume control to cascading control using fans in parallel. Had problems with difference in motor horsepower and ramp times, so only effective way to deal with it was to include a "dumper-damper" in the supply discharge. It's pretty much a standard now for ABSL-3+ animal laboratory design. If you have a single fan and pressure independent air valves, I don't think you should need a dumper damper. For a constant supply or exhaust, you could use a dP sensor to maintain constant supply against filter loading, and a photohelic to maintain desired differential by driving the exhaust VFD.

 

[Compositepro]

Your statement that "there must be a flow differential between the supply and exhaust to maintain the desired pressure in space" is not really true, or is, at least, fuzzy.
Particularly, when you say your enclosure is sealed.

Remember one of the most basic principles of Chemical Engineering: "Mass-in minus mass-out equals accumulation"

The mass rate of flow of air in and out are equal for your blower, your room, your loop. Your blower simply gives your air velocity and you can convert that to pressure by putting in a resistance to you air flow.

If you have a blower with a HEPA filter feeding your enclosure you will have a high pressure drop across the filter. You can vent high pressure air from your enclosure from the discharge of the blower and the pressure in your enclosure and at the suction of the blower will go down. If you open a vent on the suction side of the blower, the pressure on the discharge of the blower and in your enclosure will go up. This does assume that there is some resistance to flow, i.e., pressure drop between your enclosure and the suction of the blower.

For your application you would typically have a fixed vent on the suction of the blower and a controlled vent on the discharge side of the blower, to control the enclosure pressure. At steady state you do need any actual flow into or out of the enclosure, but temperature changes will prevent this. For an airtight enclosure both vents could go to an expansion bladder made of a large plastic bag, but not if there are people in there.

 

[Compositepro]

Sorry, that last paragraph should be "at stead state you do NOT need any actual flow".

 

[ChasBean1]

How sealed is the space? Much can depend if you’re trying to achieve -0.001” or -0.095” (both of which are within your stated tolerances). With "the size of several rooms" and 1,000 cfm for pressurization you might find yourself returning 900 cfm, bringing in 100 fresh for your air exchange, delivering 200 cfm back into the space and relieving 800 on the supply side if it’s anything close to typical construction and if you want something midway, say -0.05”.

If you care about the air quality (sounds like you do with HEPAs and gas absorbers), consider also that the difference in volume for pressurization is infiltration air, which may not be likewise treated. This might be a good candidate for hiring an HVAC designer to look/think it over and devise a sketch and some specs.

 

[Zesti]

In my experience "sealed spaces" do not exist.

Construction companies will tell you they make a perfectly sealed room and then afterwards you come to the conclusion that more is leaking from/into the room than expected.

I would reserve a flow of at least 1 times the volume of the room per hour for leaks.

Skuntz: how large is the room and how is it constructed ?

Having HEPA & gas adsorbers on the return-side sounds to me like some dirty stuff is going on inside the room. Therefore, the quality of the air going/leaking into the room might not be that important. It might depend on this room being inside a larger building-envelope (in which the air leaking in might already have been treated elsewhere) or if the room has walls on the exterior of the building.

 

[skuntz]

Zesti:

Because of the nature of the project I cannot get into some specific information, but let us say that the volume is greater than 50,000 but less than 100,000 cubic feet. Construction is of a particular type of airtight fabric (yes it is a very large tent).

I am sure that there will be some leakage.

 

[Zesti]

Well, a "tent", that's a new one to me.

So the volume will be around 75,000 cubic feet...

Circulating 1000 cubic feet per minute = 60,000 per hour.
This is close to 1 ACH. Might be enough to allow for air leaking in at all the seams and "doors"...

Another thing: don't underestimate the forces acting on the tent because of the pressure-difference.
Might not be a bad idea to have some fail-safe system in place for when pressure control fails: higher pressures might totally collapse the tent/structure.

Keep us posted on your experiences. I'm quite curious to know what the actual leakage will be.