Reply to Thread

Id have to guess

The air flow to be around 350 to 450 per ton.

I don't have a air hood belometer, so I really don't know. One of these days I'll buy one of those hand held vane anemometers, I could figure it with one of those fairly easily.

I'll try and get some info from Trane to see what the cfm is at a give pressure drop across the coil.

It's good to hear your hard work is paying off and you didn't have to run the chiller any colder, at least for now.

Aside from the water-side cleaning and flow balancing I'd guess your results are due to lower airflow; could you hazard a guess what your CFM/Ton is?

Update on the chiller

I removed the glycol entirely. Added fresh water and 3 gallons of closed loop system treatment as recommended by the water chemist. Ran that 24 hours. Flushed out with fresh water again till clear, and tested a sample. It tested good. I then added 3 gallons of corrosion inhibitor and a pint or so of biocide. Ran that 24 hours to stabilize. Had the water chemist come and test, and all is okay. I will need to continue changing filters in the side stream set up till I can get the TDS in order. The chemist will make monthly checks, and adjustments as needed.

On all 12 air handler, I turned out the drive pulleys as far as possible. I imagine the air slowed by 25 to 30%. Bringing me much closer to where I want.

I balanced the flow to the air handlers as best I could ( not test points ) so I did it by distance from the house pump.

I set the flow through the evaporator bundle, and the flow for the system as a whole.

On start up the space was 78 DB and 39% RH

The chilled water set point is 47 degrees. The space temperature controls were set to 71. The next morning the the temp was 71.5 DB 38.9% rh

I reset the space to 72 and the RH went to 40% (within desired range)

Now these results were during mild weather, but it had rained that morning, so I am able to control humidity load as is, and not address humidity load below my desired RH in the space. I just need to see what will happen when it is hotter outside.

At this point I am confident I'll get the system to do exactly what I want. And, I'm also certain that I have this system dialed in better then anyone else ever has.

By the way, there is no outside air to these air handlers.

Thanks to all

P.s. There is no such thing as over thinking.

Originally Posted by heatingman

In your situation, were you bringing in outside air through the rtu?

I have no OA connected to the air handlers.

We were bringing in OA, or more precisely plant air, the clean room is in the middle of a bigger unconditioned production area. The amount of OA was a constant 10% to keep the room pressurized, but the temperature and humidity did vary somewhat. But we did see an average over time of 48F water versus 44F water and the colder water did better.

On the one hand we were having to dehumidify moist OA, but on the other hand the air entering the space had just been dehumidified, and pressurizing the space was preventing infiltration. Infiltration will wreck a humidity control scheme.

Originally Posted by dlove

Airflow across the cooling coil, coupled with sufficient run-time, has a significant effect on proper humidity removal. Too much air will result in poor dehumidification. Too little air can cause the ductwork to sweat in the unconditioned spaces .

but The size of the coil will play a part to help remove latent heat as well, that's what your asking how remove latent heat with out changing temp. But if your system was not designed this way it will be hard. Running your chilled water cold to ring out moisture and than a reheat to maintain a discharge air temp is common. to maintain constant temps and humidity does require some controls. from what you said you have chilled water what type of fan system do you have? does it have mixed air, outside,return dampers. reheat coils, does it have temperature sensors like humidity, SAT, RAT, MAT. ? if you just lower your water temp you will ring out moisture, but you will lower air temp. To achieve what you want its not just one thing.

And to top it all off I'm 90% positive the run-outs to the air handlers are too small (3/4'' good for 4 gpm at normal design, and I'll need at least 12gpm)

i hate repeating myself... but...now if your saying some of the lines are small and GPM isn't enough, Colder water temps, you surely need, but the GPM is your biggest issue. COIL size, GPM, airflow, return air, mixed air, all play major rolls. With your glycol being messed up how is the rest of the piping? If you plan to add more glycol after all this. I would do a major flush of the chilled water loop, check flows and balance. How has the chemical treatment program been going? sounds like you have your hands full.

Similar to what klrogers said. Figure out what your space dew point ought to be and control ahu valves to maintain that less 2f or so as your off coil temp. Then modulate fan between 15 and 60 Hz to maintain space temp. Decrease the set point further if you still have humidity issues. If you over cool, see if you can get a de superheater in the chiller and a HW loop off that or you could put in run around coils or wrap around heat pipes in the ahus for some reheat. Rawal is not applicable since this is not DX equipment.

Good luck with the hydronic issues. Sounds nasty!

Originally Posted by heatingman

If I don't need to address the sensible load, why run so cold?

From what I understand about psychometrics, the humidity load and the sensible load are completely separate.

Water vapor could not care less how far below the dew point the coil is. If the coil is 48 db and the incoming wet bulb is over 48, water will condense. Now slowing down the air I definitely agree with. The increased dwell time of the air will allow for more contact time with the cold surface.

I'll certainly test it both ways colder water vs warmer. But, at this point in the job it's moot. I'm still in the process of cleaning up the closed loop. The glycol did a real number on this thing, and who ever did the piping was a real dip@#$%

They ran the risers up made an inverted trap, then dropped down to the headers, then to make it worse, they ran the system in parallel instead of reverse return, so balancing will be fun. And to top it all off I'm 90% positive the run-outs to the air handlers are too small (3/4'' good for 4 gpm at normal design, and I'll need at least 12gpm) All the air handlers are 20 feet up. Thank God the place has a scissor lift. I spent 10 hours there today, and am still fighting air.

Your over thinking this.

I live in a very humid climate and we have to run our water no higher than 44 in the summer for humidity control.

Just like a DX system, 48 deg (or 81 PSIG on R-22) doesn't dehumidify as well.

FXB80

In your situation, were you bringing in outside air through the rtu?

I have no OA connected to the air handlers.

You would think that colder water wouldn't make much difference, but it does.

We ran into this last year with a couple of 80 Ton AHUs whose primary task is dehumidification of a clean room. Target conditions were Room 74F @ 50% RH, 44F chilled water, 371 CFM/Ton. We had some issues with the chiller and water temp was 48F at times before we got it dialed in. We were datalogging performance from start-up and there was a marked improvement in dehumidification with the cooler water. For our situation, if they wanted the room cooler than 74 or drier than 50%, we would have to have 42F water. Fan speed control was not an option for us due to required air changes through the filter banks.

We commissioned during hot weather and there was a big sensible load. I usually let the reheat use the heat setpoint, which I put 2F lower than the cooling setpoint. There was enough load (100F outside all four walls and the roof) that it wouldn't cool the room to the 72F reheat setpoint; we had room temp of 72.5F at 55%. Pysch chart showed that if the room were 74 the RH would be 50%, so I made the reheat setpoint the same as cooling setpoint during dehumidification and it did what the chart predicted; Room 74F @ 50% RH. You're probably going to need reheat from somewhere at some point for RH that low, either internal load or a reheat coil.

Too high for what?

Originally Posted by R123

Lower your chilled water temp to 40 to 42 as stated above. 48 is too high!

If I don't need to address the sensible load, why run so cold?

From what I understand about psychometrics, the humidity load and the sensible load are completely separate.

Water vapor could not care less how far below the dew point the coil is. If the coil is 48 db and the incoming wet bulb is over 48, water will condense. Now slowing down the air I definitely agree with. The increased dwell time of the air will allow for more contact time with the cold surface.

I'll certainly test it both ways colder water vs warmer. But, at this point in the job it's moot. I'm still in the process of cleaning up the closed loop. The glycol did a real number on this thing, and who ever did the piping was a real dip@#$%

They ran the risers up made an inverted trap, then dropped down to the headers, then to make it worse, they ran the system in parallel instead of reverse return, so balancing will be fun. And to top it all off I'm 90% positive the run-outs to the air handlers are too small (3/4'' good for 4 gpm at normal design, and I'll need at least 12gpm) All the air handlers are 20 feet up. Thank God the place has a scissor lift. I spent 10 hours there today, and am still fighting air.

Originally Posted by klrogers

IMO the de-humidification improvement with the RAWAL/APR devices is much more to do with the increased run times than limiting evaporation temperature.


Kevin

EDIT: Make sure the condensate drains are clear, because you will see condensate flow like you never seen before.

ANOTHER EDIT: Make sure the coils are sparkling clean, heat transfer is the key here.

From speaking with Al B. the regional sales engineer for rawal, they accomplish increased run times by limiting the evap temperature.

Lower your chilled water temp to 40 to 42 as stated above. 48 is too high!

Originally Posted by heatingman

I only spent about an hour there on friday. I'll be going back on Monday.

It's a bastadrized process chiller applied for comfort cooling/ process dehumidification. It's a printing operation, so humidity is way more critical then cooling.

60 ton screw machine. With 36,000 cfm on the air side from 12 air handlers (3000 cfm per)

I'm going to try to slow the air way down to try and get closer to 350 or 400 cfm per ton.

The chiller had an open tank for water feed, but was filled with a glycol mix. The mix went septic during the off season due to air infiltration. So I've got a lot of clean up to due before I even start the chiller.

But, I know it never performed, which is why glycol was added. They figured colder water would provide better humidity control, but I think it likely caused the sensible load to satisfy way before the humidity load was met leaving them with too much humidity.

Here's how I would address the situation, your customer might not like the cost but if the site requires it so be it.

1. Remove the glycol. Glycol reduces the heat transfer between the coil and the water.
2. Run the chiller at around 40°F set point
3. Run the chilled water valves wide open.
4. Fit VFD's to all 12 AHU's

Run the fans at minimum speed (around 20Hz) and with 40° water and valves fully open you will be getting the best de-humidification you can get with the installed equipment. Increase the fan speed if you really need to increase the sensible cooling but you will be reducing the de-humidification. If running the fans at this speed you are still overcooling the space without reaching the RH set point then re-heat is the only option.
The fan speed will make a bigger difference to the latent vs sensible heat removal than anything else you can do with the present equipment, if you could replace any of the coils then go with a much deeper coil 4,6 or even 8 row coils will de-humidify much better than a 1 or 2 row coil.

IMO the de-humidification improvement with the RAWAL/APR devices is much more to do with the increased run times than limiting evaporation temperature.


Kevin

EDIT: Make sure the condensate drains are clear, because you will see condensate flow like you never seen before.

ANOTHER EDIT: Make sure the coils are sparkling clean, heat transfer is the key here.

Originally Posted by beenthere

Slowing the blower decreases sensible, increases latent. So it will dehumidify more.

As the coil temp and dewpoint temp come closer to each other, moisture removal decreases. Same as if the coil temp was close to the drybulb air temp, less heat transfer takes place.

Right on. The colder the coil temp, the more moisture is removed. The slower the air flow, the more moisture is removed. If ducts sweat, insulate the wet spots with closed cell foam. Reheat is needed when the sensible cooling load is low or none.
With small systems, it is much easier to add a supplemental dehumidifier. The setup on the a/c is same to max the latent removal. During low/no cooling loads the dehumidifier removes moisture while providing free reheat.
Regards TB

Originally Posted by heatingman

If I lower my chilled water loop, that will dehumidify, but it will also cool the air too much.

The way I see it, as long as the chilled water is below the dew point I will dehumidify.

The closer the spread between the chilled water and the incoming air, the less sensible transfer there will be.

With the entering air at 72db and 45% rh, the dew point is about 49 degrees. So as long as the coil is below that, I will remove humidity.

If the coil was 40 or 45 that would dehumidify, but the spread between the coil temp and the incoming air is that much more, which will cause the the sensible load to be addressed even more.

Slowing the blower decreases sensible, increases latent. So it will dehumidify more.

As the coil temp and dewpoint temp come closer to each other, moisture removal decreases. Same as if the coil temp was close to the drybulb air temp, less heat transfer takes place.

Make shure coil is piped correctly coldest water/ coldest air, and how much building infiltration is their.

i agree with all posts!

i am usually or always working on vav stuff with vfd,s so lowering fan speed and maintaining proper duct static is not usually much of an option for me?

i have labs that run the heat all night and start cranking on the air an hour before they start their process!!!!!!!!!!

this is not efficient but helps the labs!!!!!!!!!

Evap temperature.

If I lower my chilled water loop, that will dehumidify, but it will also cool the air too much.

The way I see it, as long as the chilled water is below the dew point I will dehumidify.

The closer the spread between the chilled water and the incoming air, the less sensible transfer there will be.

With the entering air at 72db and 45% rh, the dew point is about 49 degrees. So as long as the coil is below that, I will remove humidity.

If the coil was 40 or 45 that would dehumidify, but the spread between the coil temp and the incoming air is that much more, which will cause the the sensible load to be addressed even more.

Bring your water temp down to 40, and slow the blowers.

Originally Posted by heatingman

But, I know that the relative humidity is high leaving the discharge.

The humidity reset once it is re-acclimated to the surrounding air. Thus lowering overall humidity in the space.

Part of what I understand the APR to do is set a low limit evaporating temperature to prevent over cooling the air and allow for longer run cylces.

I want to apply the same thinking to a chilled water system to remove more humidity without over cooling the space.

The customer in this case wants to maintain "40 to 45% RH in the low 70's (db)"

So to get that, I'm thinking my chilled water set point needs to be around 48 degrees. giving me a theoretical discharge air of 58 db 70% rh

Let me know if my thinking is off on this.

I do not believe you will be able to do that without reheat.
We are in a very humid area and have several clean rooms with tougher conditions. Basically all the stars have to be aligned, cfm ducting etc.
We drive the chw vlv wide open and temper it with reheat. The trick is to have very slow control loops (if DDC).