Closed Loop Chilled Water flow issues 1

[RustyPipez]

Hi all.

I have a glycol system that includes 3 air-cooled chillers with 3 pumps on a loop. The pumps are controlled by VFD's controlling pump speed using differential pressure. Off the loop are various fan coil units. I run minimum one chiller and one pump around the clock, adding chillers and pumps for seasonal changes in load.

In the past, I would have 575 GPM at 75% speed on the pump, with a differential pressure of 24 psi. The flow and pump speed were very stable since my load is pretty stable. Suddenly, with no changes to the system, my flow was greatly reduced, varying between 250-400 GPM with the same DP. The load has not changed, and doesn't change other than slight changes due to seasonal temperature differences. If I change my DP setpoint to 33 psi,the pump will speed up, and I can get flow in the range of 330-470, but the pump speed is now varying between 83%-90%. The reduced flow is also causing my return glycol temp to increase.

In it's present conditions, I need to run the pump at 100% to get the 575 GPM flow I had, but then the pressures are much higher than they should be.

This situation is the same regardless which chiller/pump combination I use.

My question is, Does this sound like a possible restriction in the loop? Or am I barking up the wrong tree?

Thanks for any assistance,

 

[EnergyProfessional]

more information needed. 2-way or 3-way valves on the fan-coil units?
and the 3 pumps in parallel, or one for each chiller? A schematic?
how are you measuring flow? How are the FCU and valves controlled? is that under your control, or some campus situation where you runt he plant, but not the building?

 

[LittleInch]

Aaah, the joys of faultfinding from afar [polishes crystal ball and peers in]

Nope, still can't see what you're looking at.

Please add a diagram and notes as there are multiple possibilities here and it's not clear what sort of system you've got.

To understand why something isn't working the way you think, first you need toi understand how it is supposed to work. To do that you need some sort of schmetaic / PFD/ P &ID and a control narrative saying what is controlling where and how.

I don't fully understand what you mean by " Off the loop are various fan coil units" - Secondary pump system or a bridge between flow and return header?

If it's happened suddenly and from your current description, best guess is that a valve which is supposed to be open or controlling flow is shut and hence all flow is going through the AHU?? This bypass would keep the return temp lower.

Need a schematic and answers to questions to understand your system.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[RustyPipez]

Thanks for the replies. Hopefully, I have the answers to your questions. I also uploaded a quick drawing of the system. I have more FCUs than shown, but you’ll get the idea.

The FCUs all have 2-way valves controlled by local controls/sensors, with setpoints under my control.

The flow is measured by flow meters on the CHW supply; one on each chiller.

The 3 pumps are in parallel, so I can use any pump/chiller combination.

Pump speed is controlled by a DPT (shown on the drawing) on the loop to maintain a DP that I set.

The balancing valve shown on the drawing is not a modulating valve. It’s a butterfly valve that I presume was set during commissioning of the plant. It’s also in an inaccessible spot where no-one would be able to mess with it.

The only modulating valves on the system are the isolation valves on each chiller and the valves in the FCUs.

Thanks again guys.

 

[LittleInch]

Ok, that's much better.

For me your options are:
1) despite being "inaccessible", clearly the balancing valve was accessible at some point and for whatever reason someone turned it off or it has become blocked
2) Despite not being shown are there any other sectioning valves on the loop, either supply or return, which could result in flow having to go a long way around if someone has suddenly closed them.
3) Has someone changed the location of the FCUs being used or has blocked off some manually - do the ones not in use still allow water to flow through, even by accident, but not contribute to load because there is no air going through the FCU??
4) Maybe the previous set-up was the incorrect flow and pressure and something has now acted to increase pressure losses??
5) There is a blockage or restriction somewhere.

1 to 3 are realitvely easy to check, 4 & 5 a bit harder.

Someone must know whether anyone did anything to the system to make it change so fast, maybe they just don't want to come forward and say so...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EnergyProfessional]

what you show as a balancing valve looks more like it serves as a bypass (for minimum flow?)? Is it manual, or an actuated bypass?
Your system flowrate will depend on what FCUs are working and how wide their valves are open (assuming that bypass/balancing valve is fixed). did you verify valve positions?

your air eliminator is piped for temporary operation, usually it should be in line (all flow goes through all the time). not sure what "XT" is

Do you have cooling deficiency? or why do you think you NEED to have 575 gpm? if the FCU provide enough cooling, they have enough flow. More flow just lowers the dT (and wastes energy). Obviously the chillers have minimum flows.
what makes you think the load didn't change seasonally or over time? Is that an office, server rooms? hospital?

 

[BronYrAur]

Definitely looks like the balancing valve's purpose is to guarantee a minimum flow through the chillers. A butterfly valve is a poor choice for this if you ask me. I would double-check that it is functioning correctly.

Do you have the automation to drive all the fan coil valves open to see flow vs DP under a design condition?

You mention that your glycol is coming back warmer than normal. This is generally a good thing. Most places have low delta-T. Could this reduced flow be a better operating condition for you?

Did you introduce any chemicals that could have caused scale to flake off and clog strainers?

I'm trying to think of what would cause a sudden change. That bypass valve would be my first guess.

 

[lukaiENG]

Did any controls guys/ chiller techs update any firmware lately? They may have updated something in the software to optimize the operation of their part of the system.
I agree with BronYrAur that the new operating point may well be more efficient.

 

[RustyPipez]

Sorry I haven't replied to any of your follow-up questions. I've been offsite for a few days. I have some additional info, and if you think it's irrelevant, I'll go back and answer all the follow-ups.

I checked into it a little more today, and I find a Bermad Model 730 valve in the bypass loop. The butterfly valves are only for isolating the bypass loop. Unfortunately, I was initially going with info from the as-built and word from other guys around here as far as what equipment was in the ceiling.

The Bermad documentation states,
When installed in-line, it sustains minimum pre-set,
upstream (back) pressure regardless of fluctuating flow
or varying downstream pressure.
When installed as a circulation valve, it relieves excessive
line pressure when above maximum pre-set.

I'm assuming since this is installed in-line in the bypass loop, it is serving the 1st purpose...like a circuit setter. Also, I don't get much of a difference in flow if I close the butterfly valves to isolate the bypass loop. That leads me to believe the Bermad valve is closed and is not performing its function of regulating pressure (and therefore flow). It seems to me this is the most likely cause of all my problems. What do you think?

If so, does anybody have any tips for testing the function/non-function of this valve without tearing it apart?

Thanks again, everybody for your patience and your help.

 

[lilliput1]

Check all strainers. Hopefully each are provided with blowdown valve so you will not need to disconnect piping. Check also pressure drop across chiller to check for plugged tubes.

 

[LittleInch]

rusty pipez,

If there is little difference when you isolate the regulating valve then it sounds like it is closed. ~The question we can't answer is whether it is doing its job now and wasn't before (i.e. trying to maintain a min back pressure - I can't see how this would work when a lot of the flow is going in a different route) or it relieves excess pressure build up and hence maintains flow at some set point. The latter seems more like it to me as the FCUs when in parallel will operate using an identical pressure drop. As flow through the FCUs increase, the pressure falls and the bypass should then close to let more water through the FCUs.

The operating manual gives you a few clues as to what you could do (attached)

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Drazen]

Local controllers causing two-way valves closing would always reduce flow on constant dp, but in that case return temperature should not rise significantly, which points at clogged strainers as the most likely cause.

You can tentatively play with pump speed control, if you change mode from constant dp to proportional, flow should rise to some new working point (as dp likely rose at previous flow), but again that points toward clogging.

 

[John2025]

Is it just a sketch error or are the VFD Pumps really fed in series from the FCU loop? Has someone added/adjusted a thermostatic valve on the Pumps? That could reduce flow and raise return temp.