Anyone doing psychrometric control for dehumidification? 1

[ROCHESTER]

I'm a mechanical engineer for a small Bio/Pharm manufacturing company. Over the last 10 years I've developed some designs to improve cost efficiency of the air handling systems. The next, much bigger, plant is being designed. I've reviewed tha A&E's proposed design and would like to respond with suggestions for energy savings. I'm not a HVAC specialist and I'm posting here hoping for comments from the professionals on my ideas in the following areas:

* Psychrometric control for modulating AHU's cooling/dehumidification temperature as a function of OA humidity load.

* Feed forward loop control to improve temperature tracking with variable AHU DAT(50-70F)

* Dehumidification of OA(make-up air)prior to mixing with RA. Eliminates sensible cooling of RA to dew-point for dehumidification.

* Locate Fan(s) outside the air stream to reduce cooling load of the system.

I'd appreciate any discussion and/or comments on these topics. I feel reluctant to present these ideas to the professional HVAC engineers given my limited experience and qualifications.

 

[phisatho]

You mean chemical dehumidification?

I don't think it is worth.
The sensible cooling of OA at FAHU will reduce cooling load on AHU

If the AHU is located in conditioned area, relocating the motor will not make any difference.
If it is located outdoors, just do a bit of calculation, you may find it economical to improve the insulation and control leak rather than relocating the motor.

I haven't seen a single system where cooling valve is controlled based on OA H/T which is working flawlessly.

I didnt understand the feed forward you proposed. Can you elaborate?

 

[ROCHESTER]

Hi phisatho,

No, latent cooling not chemical dehumidification.

Not sensible only cooling but latent(sensible required). Take the water out before it's mixed with the RA. This is for clean room system with about 8 SCFM/S.F. floor area. The RA comes back with good W and only 2-3F > room set-point. Standard office AHU's gain no benefit from pre-dehumidification as the moisture removal is free with the low DA temps required for low SCFM's delivered to the spaces.

Our systems are built for < 1% leakage already and any further improvement isn't cheap. For the above clean room system, with the AHU outside and a 80 HP fan motor moving 30,000 SCFM, the motor does introduce a fair amount of heat into the system. I don't understand why this cooling load isn't placed outside the system?

I control system on Humidity Ratio(W) and T. There are RH/T sensors outside, on DA, and RA. I do this:

[ol][li]Calculate W for OA, DA, and RA[/li]
[li]Determine if OA and/or RA require dehumidification[/li]
[li]If dehumidification required, drive OA coil to dew-point for required moisture removal(set-point W)[li]
[li]If dehumidification not required, use MA coil to cool only to T required by fan-motor and space loads.[/ol]

The &quot;feed forward&quot; reheat control improves the PID loops' control of space temperature over the range of the AHU's DA temps. I've had very good results doing the following:

[ol][li]Set the AHU DA T for low range of operation(i.e. 52F after fan)[li]After steady state, record reheat valves % open position[/li]
[li]Repeat step 2. for 68F[/li]
[li]Generate for each reheat coil a &quot;m&quot; slope for DAT vs. valve % open position[/li]
[li]Feed the calculated valve position forward to the PID loop &quot;bias&quot;
[/ol]

This relults in faster tracking of DA temperature changes as well as tighter control fo space set-point.

 

[quark]

Bricklayer!

The question of dehumidification is subjective to your required conditions, ambient and person (likes trouble or not). One of the top experts of this forum always defered with me in this regard. But I always prefer desiccant dehumidifiers because the RH requirement in pharma companies (in which I work) is lower and bio load is considerably reduced because of regeneration heat. (with cooling coils, we found that, bio load is very high). Using cooling safe level of RH to maintain is 45% and not below that.

We have standardized keeping motor out of the air path to reduce contamination as well as heat load.

Treating fresh air is always better,for I believe in reducing the problem at source. This definitely will reduce your ADP.

Regards,

Repetition is the foundation of technology

 

[ROCHESTER]

Thanks quark,

I appreciate your comments. Are the air handlers you use custom designs, or can you point me toward manufacturers that make units with the features you mentioned? The designs I'm reviewing use roof top units @~ 40,000 CFM to maintain 70F, 40-50RH(winter/summer), location: South Carolina, US.

Thanks,

Craig

 

[quark]

All the AHUs are made up of custom made double skinned GI precoated puf insulated sheets with extruded aluminium section and with thermal break. But the construction is tailor made to suit your geometrical requirements viz., duct orientation, floor space and cooling coil position etc.

These are generally available everywhere. (In our country manufacturers import the panels from Italy and are very good).

40 to 50% RH is a good range and you should not have any attaining it with cooling heating application. My suggestion is to use hotwater (smoother heating)with RH control near the return grills in the room or (if used for multiple rooms) in the return ducting.

If it is a tablet manufacturing facility return ducting is generally avoided for the possible cause of powder settlement on sensor and subsequent malfunctioning.

Hope this helps.

Regards,

Repetition is the foundation of technology

 

[ROCHESTER]

quark,

Yes...locating the Temperature sensors in the RA ducts was one of my first lessons. Temp sensors located on walls in a cleanroom result in unacceptable response time in the reheat control loops. They're also subject to corrosion and mechanical damage from mopping the walls.

Do you use psychrometric cotrol for sensible vs. latent cooling? If so, have you seen resources on the web with details?

I wish I had this forum as a resource 10 years ago.

Thank you very much,

Craig

 

[quark]

Bricklayer!

I couldn't totally understand &quot;psychrometric control for sensible vs. latent cooling&quot;. In broad sense AHU is itself a psychrometric controller.

We use the following simpler control system. Cooling coil flow is controlled by a threeway modulating valve (or a twoway valve with variable speed pumps). This valve gets signal from a temperature sensor which is kept in the supply air duct after reheating.

RH (and reheating) is controlled by either steam or hotwater coil which again has a modulating control valve which is driven by humidistat. (Electrical heating is not a good option as far as my experience is concerned for RH control)

Note that you have to maintain constant chilled fluid medium temperature in this case. (vaariation of 2 deg.C is ok)

Hope this helps.

Regards,

Repetition is the foundation of technology

 

[ROCHESTER]

Hi quark,

I'd like to continue to discuss the &quot;psychrometric&quot; control of dehumidification further. We are having an FDA inspection and I'm having to focus on the review of our cleanroom designs. This should be complete by the end of the week and I'll have more time then.

Short response:

We inject steam for humidification. Given that air flows and DA temperatures are OK, the steam valve control is straight forward and reliable.

Dehumidification has been more difficult to do efficiently. The original design in the clean rooms had fixed discharge air(total HEPA CFMs) temperature(DAT) set-point at ~51F(after fan in flow stream). This worked, but required large sensible cooling even when OA introduced little or no latent cooling load. I implemented a controls scheme that modulates the AHU's DAT to control on humidity ratio(W):

1. Switch to &quot;dehumidification mode&quot; when OA moisture content(W) or space return air(RA) > W set-point(W @ 70F DB, 45% RH).

2. Switch to &quot;sensible cooling and/or humidification mode&quot; when OA and RA W < W set-point. The AHU's DAT typically settles at ~ 68F in this mode.

On newer systems I use an additional chilled water coil on the OA make-up. This requires more complex control which I'd like to discuss in more depth when I have time.

Thank you,

Craig