install vsd on existing lead/lag chilled water system 4

[sye15929]

hi! need help to see if this strategy makes sense.
existing installation is 1 centrifugal chiller and 1 air cooled chiller. no secondary loop (only primary). supply to few ahu pneumatically controlled with 3 way valves (can isolate bypass side).

going to shut-off some of the bypass side for the 3 way valves at ahus making sure that enough flow is maintained for minimum flow even when 2 chillers operating. install dp sensor at end of loop. also install vsd for for each chiller pump.

control strategy.
1. run lead chiller till either temperature cannot be met or dp keep dropping.

any input is much appreciated.

 

[mtngreen]

I would also plan on installing motorized isolation valves on the chillers that is isolate the chillers when not needed.

 

[Drazen]

i am curious why would you want to link chiller on-off with dp? it is to be assumed that dp dropping would mean more ahu control valves open (if you have zone valves), but that does not necessarily mean that load is rising.

if you do not have zone control valves for chiller (meaning one chiller = one zone, you can call them isolating valves as mtngreen), that when do you expect dp to drop? if you balanced bypass with main loop, dp should remain the same.

moreover, you mention vsd in post title, but there is no mention of it in text. what is that? vfd, maybe?

 

[jkauwale]

What about a variable primary scheme?

http://www.trane.com/commercial/library/vol31_4/index.asp

Justin K, P.E.

http://www.engproguides.com

 

[ChasBean1]

See attached to determine if I understood the system correctly from your description.

If you have check valves where shown installed (in bold), you don’t need automatic isolation valves for each chiller.

Install a bypass line (also shown bold) with a modulating control valve set to maintain the required minimum flow rate for ONE chiller. Shut ALL your 3-way AHU bypass valves.

Install a flow meter (F), capacity should be just above full flow for two chiller operation.

Note: the pumps should be rated at the same flow and head if this installation is correct. Assuming it is, SOO (basic):

1. Modulate operating chiller pump VFD to maintain operating loop DP.

2. During 1 chiller operation, the new bypass line modulates to maintain minimum flow set point required by the chiller.

3. If T1 rises 3°F above set point for 15 min, start second pump and then second chiller. I would also suggest adding an algorithm that says outdoor air enthalpy must also be above X, e.g. 30 BTU/lb, to ensure a second chiller doesn’t falsely cycle on.

4. When two chillers operate in parallel, the two primary pumps modulate at the same speed to maintain loop DP set point. The minimum flow bypass is closed.

5. With two chillers in operation, when calculated tonnage (using T1, T2, and flow) fall below 80% of the capacity of a single chiller, shut off the lag chiller. 15 minutes later, shut off the lag CHW pump.

 

[sye15929]

hi all,

thanks everyone for the reply and inputs....

what i am trying to do here is to imitate a vpf scheme with minimum modification or cost to the existing installation. instead of by-pass line with control valve, i am leaving some of the existing 3 way valve at the ahus open, some close to have minimum flow instead.

the control for the vsd which i am going to put in i think will be based on the input reading from the dp sensor which i am going to install.

basically currently control scheme is if chilled water return temp is below setpoint 1 chiller operate operate with its pump. if return temp above setpoint, lag chiller operate in tandem.

my idea of the new strategy is:
1. chiller operation still based on chilled water return temp.
2. vsd speed based on dp. the vsd for the connected pump shall not run if its chiller is not turned on. Question: does the vsd speed need to be linked to chilled water return temp? Or only the dp?

not too worried about the check valves and such as existing installation would've taken care of that i believe.

sorry for the unorganised rambling. will try to put in drawing if i've enough time.

 

[jkauwale]

Hi sye15929,

To answer your question, the VSD should maintaing the proper flow to the AHU's based on the dp sensor and the chillers should be sequenced to maintain the chilled water temperature.

I am a little confused as to how you are controlling the AHU's? If you are closing off the 3-way valves, then how will you be modulating flow to the AHU's? Are you replacing them with modulating 2-way valves?


Justin K, P.E.

http://www.engproguides.com

 

[sye15929]

hi jkauwale!

thanks for the reply. i m closing some but not all the 3 way valves. will maintain enough 3 way valves just enough to maintain minimum flow. hope this makes sense.

 

[ozmosis]

Maybe this attachment might assist.

Disclaimer: I work for this company. I would normally try and find another company's data to provide some balanced information but there isn't any.

 

[DRWeig]

ozmosis, your link gives a "not found" answer. I'd like to see the pdf if you can fix it.

closing some but not all the 3 way valves

I took this the wrong way due to the word "closed" in that phrase. I think this means that you are valving off the bypass port. I have done that before too, but in a couple of cases the AHU became hard to control because the three-way linear or equal-percentage characteristic changes when the third port is blocked during modulation. In that case, we spent just a little bit of money to replace the 3-way valves with 2-way valves.

Attached is another excellent paper you should read. Gil is a Fellow of ASHRAE, and a great guy. Some other excellent engineers dispute some of the analysis, but not many anymore because the energy savings and successful chiller/control operation have been proven many times now. In other words, your idea is a good one.



Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[ozmosis]

http://mcliterature.danfoss.com/WebPublish/downloadPub.do?no=2

Sorry guys, travelling with my new iPad and still getting to grips with putting links in.
If this doesn't work, I'll wait until I get back to the trusty pc and do it properly.

 

[DRWeig]

You're just teasing us, arent' you?

"Error description: Parameter publication is missing"



Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[ozmosis]

grrr.
Back on the PC now.
Hopefully the wait is not an anti-climax!

 

[DRWeig]

That was worth the wait, oz.

I do a lot of training in controls and HVAC basics. I'll send Danfoss a kudo or two as well.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[ozmosis]

Thanks Dave.
You should ask your local Danfoss person for a copy of HVAC Planet. That's really informative and good for training. Too big to send but they should have the DVD/cd

 

[ChasBean1]

sye15929,

You can close off three-way bypasses instead of having a minimum flow control assembly (the latter is preferable, but not mandatory). You would run the loop at the balanced DP, which should be enough to just satisfy AHUs under demand conditions. With the system operating at that DP, line up flow to ONE chiller and command closed all AHU chilled water valves. Close AHU three-way bypass valves starting with the AHU closest to the chiller plant and working your way down the loop to the most remote units. The next AHU inline that has its bypass closed resulting in the chiller's evaporator flow switch trip should be re-opened.

Now the pump VFD(s) can operate to maintain the loop DP set point and the chiller should not experience low evaporator flow issues.

The CHW pump speed should only be linked to the DP, not the CHW return temperature.

And on cycling based on CHW return temperature, not recommended. You should stage to a second chiller based on the system not maintaining SUPPLY CHW temperature (X° above set point for X minutes). You should stage back down to a single chiller based on running load amps or chiller % load. If chillers in parallel fall below 80% load for X minutes, stop the standby chiller and X minutes later, stage off its pump.

Do you have the check valves there as per the sketch in my prior post? Are the flows for each chiller the same (same pumps/head ratings)?

Best,
CB

 

[sye15929]

hi chasbean1!

ya...same pumps (from what i remember) but slightly different head as one is serves as air cooled chiller and one a water cooled chiller (head difference only thru different chiller)...