20 ton Chiller compressors cycling too often?

[DeeKay]

How often should my chiller compressors be cycling? Right now I'm seeing them run for about 5 minutes, rest for 5-10 minutes and then start over again (Usually 10-15 minute cycles if that makes any sense).
I'm wondering if my water pump speed might need to be adjusted(they're on a VFDs)

Asking for advise here because the local Trane guy couldn't even figure out how to get the pump working in the first place (it wasn't working because the isolation valve was closed doesn't inspire much confidence)

 

[acmikee]

not nearly enough info so its 20 tons but number of compressors, load size, is the air handler on VFD or just the pumps, age or is it being commissioned now
if the isolation valve was closed and it was running where is the flow switch
what water are you trying to maintain
was it cycling on temp or a safety
need a lot more info is your trane guy residential or commercial

 

[DeeKay]

It's a process chiller so no air handler per se, but it does feed a shell and tube heat exchanger for cooling compressed air but I'll have to figure out how to calculate the load on it, leaving water temp set at 35-36°F, cycling on temp.

it's a 20 ton, x2 2 stage compressors, 2 chilled water pumps(redundant, only one runs at a time), 2 years old, given how bad the build quality of this unit was in the first place and the other problems it has had in it's short life I'm just worried pump speed hasn't been correct since it was installed and might lead to premature compressor failure if they're cycling too much.

It's a "custom" chiller Based on a Trane Odyssey split unit; Odyssey condenser unit on top of a "custom" chiller skid built by Trane Creative Solutions.
I'm Just wondering if there is any basic general rule of thumb as to how often is too often for compressors to cycle(assuming this was built/sized correctly in the first place).

The chiller itself wasn't running at the time, Trane tech had just replaced a pump that seized(that's a whole other unrelated story of ****** design) and was running the loop to check for leaks(you can run the pumps in local from the VFD without the flow switch being made evidently.) Anyway, they put the pump in, turned it on and left for lunch, I came by to check on things and noticed I had no glycol pressure and found the pump sitting there deadheading for who knows how long.

Trane guy is commercial

 

[American Locomotive]

What type of compressors are they? Scrolls, screws? We had 60 ton chillers with 4 15-ton Copeland scrolls. They were setup with variable hot-gas bypass, so they could modulate well.

However we ran into a weird edge case because of a strange programming glitch where at around 45-50 tons of load, one of the compressors would start short-cycling. I want to say it was around ~8 minute cycles. After a while the Copeland protection system would kick in and lock the compressor out.

Copeland says a maximum of 10 starts per hour for their small scroll compressors, but gives no defined limit on their bigger compressors. Instead they say the bigger compressors should just run long enough to make sure oil returns to the compressor. So I would say really you'd probably want at least 10 minutes on.

You could reduce cycling by increasing the load, or allowing a larger process temperature hysteresis.

 

[DeeKay]

What type of compressors are they? Scrolls, screws? We had 60 ton chillers with 4 15-ton Copeland scrolls. They were setup with variable hot-gas bypass, so they could modulate well.

However we ran into a weird edge case because of a strange programming glitch where at around 45-50 tons of load, one of the compressors would start short-cycling. I want to say it was around ~8 minute cycles. After a while the Copeland protection system would kick in and lock the compressor out.

Copeland says a maximum of 10 starts per hour for their small scroll compressors, but gives no defined limit on their bigger compressors. Instead they say the bigger compressors should just run long enough to make sure oil returns to the compressor. So I would say really you'd probably want at least 10 minutes on.

You could reduce cycling by increasing the load, or allowing a larger process temperature hysteresis.

These are Copeland Scrolls and I believe they have that variable hot gas bypass; there's a 2 wire plug on the side of the casing I'm assuming goes to some kind of solenoid inside.
The weird thing is, I found one of the compressors had that 2 wire plug plugged in, the other one was just dangling there and Trane can't give me an answer as to if they should both be plugged in. The other weird thing is that those 2 wire plugs don't show up anywhere on my drawings...of course neither do the condenser fans so no real surprise there I guess.

Unfortunately I can't change the hysteresis without Trane's Tracer software that they won't sell me.
I could increase my air compressor discharge temp by a little bit which should load the chiller more but that will bring my closer to my air compressor shutdown temp.
The chiller and heat exchanger are really just a giant refrigerated compressed air dryer so as long as I can maintain a similar Delta T I should be fine. I just have to avoid freezing water and avoid not getting cold enough to drop moisture out of my air.

 

[karoc]

Is it reaching setpoint, does your display temps match water temps? Maybe low on charge if that has been problem in past. But like my old instructor use to say"Leave the drum on truck"

 

[American Locomotive]

These are Copeland Scrolls and I believe they have that variable hot gas bypass; there's a 2 wire plug on the side of the casing I'm assuming goes to some kind of solenoid inside.
The weird thing is, I found one of the compressors had that 2 wire plug plugged in, the other one was just dangling there and Trane can't give me an answer as to if they should both be plugged in.

It could be the compressor crankcase heater, but it could be the stage unloader valve. Copeland makes two different types of compressors with unloaders. One type is variable, allowing the compressor to be modulated 10-100%. The other is fixed 2 stage - high/low capacity. There shouldn't ever just be dangling wires, though.

So just to clarify - 20 ton chiller with 2, 10-ton, 2-stage modulating compressors. That should theoretically give you 4 different capacity values. It sounds like your chiller might be a little oversized for the application, however?

 

[DeeKay]

Is it reaching setpoint, does your display temps match water temps? Maybe low on charge if that has been problem in past. But like my old instructor use to say"Leave the drum on truck"

Yeah, I have temp gauges on the both glycol lines and they show within 1-2° of what the display shows.

It could be the compressor crankcase heater, but it could be the stage unloader valve. Copeland makes two different types of compressors with unloaders. One type is variable, allowing the compressor to be modulated 10-100%. The other is fixed 2 stage - high/low capacity. There shouldn't ever just be dangling wires, though.

So just to clarify - 20 ton chiller with 2, 10-ton, 2-stage modulating compressors. That should theoretically give you 4 different capacity values. It sounds like your chiller might be a little oversized for the application, however?

Correct, I'm pretty sure these are the 2-stage ones but I'll have to take a couple of pictures when I'm there tomorrow be sure. I want to say these compressors have the little band style crankcase heater that wraps around the bottom of the compressor case.
I've been trying to get a good set of drawings from Trane for about a year now with no luck, but I think I'm going to end up having to tug and trace wires and make my own set of accurate-ish drawings. The drawings I do have show the compressors being staged by the UC600 trane PLC thing but nothing controlling unloaders which makes zero sense.

I don't think it's oversized because this unit replaced an old 25ton from the 90s that never seemed to have any problems.

 

[DeeKay]

Ok so the compressors are 2 stage units (Copeland ZPS122KCE-TFD-305), band heater on the bottom which means the thing that was unplugged has to be the unloader. I couldn't find much help on emerson's site does anyone know if these unload to the 67% when energized or are they 100% loaded when energized?

 

[Full Throttle]

I take it it's a CGAM chiller. Need to change delta T design. Usually they are set for 10* but reality might be different. Delta T is the max differential between in and out water. This will slow staging. You can also adjust compressor staging dead band to allow more rise in temp before another compressor is brought on. Also raising differential to start will help with short cycling. These adjustments depend on the process and how critical exact temps are required.

Other than that I need flow readings and design specs.

I work for Trane and just so happen to be a chiller tech. Process machines get beat to snot due to what they have to do.

 

[Full Throttle]

Get me the serial number for the machine please. I can get you any info you would like.

 

[DeeKay]

Get me the serial number for the machine please. I can get you any info you would like.

I wish it was a CGAM, then I'd have accurate drawings, documentation and probably a lot less issues. We had a CGAM at the brewery I used to work at and aside from one dead flow switch had zero problems in 5 years of it running.

This unfortunately is a custom chiller from Trane Creative Solutions and the serial number only comes up with the Odyssey unit up top . Its run by a UC600 with a TD7 display which is basically useless because the only setpoint/parameter I can change is the water temp. Anything else has to be changed using the Tracer software that I can't buy.

I just found out from one of our engineers today that we have a few of these units around the country that are also having issues, so hopefully we can get someone from the factory to sort some of this out.

 

[Full Throttle]

I wish it was a CGAM, then I'd have accurate drawings, documentation and probably a lot less issues. We had a CGAM at the brewery I used to work at and aside from one dead flow switch had zero problems in 5 years of it running.

This unfortunately is a custom chiller from Trane Creative Solutions and the serial number18261617TA only comes up with the Odyssey unit up top . Its run by a UC600 with a TD7 display which is basically useless because the only setpoint/parameter I can change is the water temp. Anything else has to be changed using the Tracer software that I can't buy.

I just found out from one of our engineers today that we have a few of these units around the country that are also having issues, so hopefully we can get someone from the factory to sort some of this out.

Holly ****! I feel your pain now. ''Trane Creative Solutions'' You are now working with an engineer, who designed an abortion on a piece of equipment that was never designed to do what it was designed to. You have been given a solution in a language you don't understand to a problem you didn't know you had.


So they took an Odyssey Condensing unit (made for split system cooling) and have piped it into a heat exchanger to make chilled water. Guess the wheel was not round enough. And your right I cant get info unless I call the factory, then they are gonna question why I am working on a piece of equipment outside my area.

Your problem is heat exchange or water flow. Stay on top of it to get the short cycling taken care of or your gonna be putting compressors in on your dime.

 

[TRWham]

Ok so the compressors are 2 stage units (Copeland ZPS122KCE-TFD-305), band heater on the bottom which means the thing that was unplugged has to be the unloader. I couldn't find much help on emerson's site does anyone know if these unload to the 67% when energized or are they 100% loaded when energized?

I was still at Copeland when this technology was developed (called 2 step back then) but I do not recall which way the solenoid works. It came from the AC side of the house and I was in a different division. You could tell by reading motor amperage and cycling the solenoid to see which setting drops the current.

 

[DeeKay]

Holly ****! I feel your pain now. ''Trane Creative Solutions'' You are now working with an engineer, who designed an abortion on a piece of equipment that was never designed to do what it was designed to. You have been given a solution in a language you don't understand to a problem you didn't know you had.


So they took an Odyssey Condensing unit (made for split system cooling) and have piped it into a heat exchanger to make chilled water. Guess the wheel was not round enough. And your right I cant get info unless I call the factory, then they are gonna question why I am working on a piece of equipment outside my area.

Your problem is heat exchange or water flow. Stay on top of it to get the short cycling taken care of or your gonna be putting compressors in on your dime.

Yeah unfortunately for me a 20 ton CGAM is too big to fit in the area we needed it to so they came up with this garbage.
My initial thought was that the pump speed is too high to get decent heat transfer in my heat exchanger. But now I'm wondering if this thing is actually staging the compressors like it is supposed to, if it's even staging them at all. Or it could be a combination of the two.

I was still at Copeland when this technology was developed (called 2 step back then) but I do not recall which way the solenoid works. It came from the AC side of the house and I was in a different division. You could tell by reading motor amperage and cycling the solenoid to see which setting drops the current.

Good idea, I have a trending meter I can throw on for a few hours and see when, and if it's staging.

 

[acmikee]

so the compressors can unload w/hot gas bypass are they unloading if they are then they are way over sized how much load do you have is the pump cycling or just ramping down depending on the load can you access the control panel you should be able to read amps-pressures-stages and adjust load settings.
what is cycling the pumps on the VFD is it temp or pressure

 

[DeeKay]

so the compressors can unload w/hot gas bypass are they unloading if they are then they are way over sized how much load do you have is the pump cycling or just ramping down depending on the load can you access the control panel you should be able to read amps-pressures-stages and adjust load settings.
what is cycling the pumps on the VFD is it temp or pressure

Read the responses above.

If the old 25 ton unit ran for years with no problems I doubt this 20 ton unit that can stage and unload is oversized. Suspect that this is a combination of problems; pump speed is too high, compressors are not unloading which would either be a design or controls issue. I'll dig deeper when I'm onsite today to see if it's staging or unloading at all.

The only reason the pumps are on VFDs is so we can adjust the speed to get the design flow rate through the evaporator and to eliminate the need for a balancing valve...they said they did that for power savings. There is no active speed adjustment on the pumps, they run at whatever speed they're set at 24/7.

 

[American Locomotive]

So here's what the documentation from Copeland says:

"When the solenoid valve is energized the compressor is in full-load and when de-energized the compressor is in part-load".

So if you have one compressor unplugged, that means at most you've been getting 16 tons of cooling, as the compressor can only run at 65% with the solenoid de-energized. If you're down capacity by that much, and your compressors are only spending 5 minutes out of every 15 minutes, it sounds more like you need a ~10 ton chiller, not a 20 ton.

Either way, regardless if the chiller is oversized, it probably shouldn't be short-cycling like that. It should just be able to run on one compressor and modulate capacity as necessary.

So if I understand your set up, it's: Chiller --> Refrigerant/Water HX --> Water/Process Air HX? Is all of that packaged inside the chiller skid, or is the process air HX located remotely?

 

[DeeKay]

So here's what the documentation from Copeland says:

"When the solenoid valve is energized the compressor is in full-load and when de-energized the compressor is in part-load".

So if you have one compressor unplugged, that means at most you've been getting 16 tons of cooling, as the compressor can only run at 65% with the solenoid de-energized. If you're down capacity by that much, and your compressors are only spending 5 minutes out of every 15 minutes, it sounds more like you need a ~10 ton chiller, not a 20 ton.

Either way, regardless if the chiller is oversized, it probably shouldn't be short-cycling like that. It should just be able to run on one compressor and modulate capacity as necessary.

So I checked today and with the Chiller running, both compressors run at the same time and neither solenoid is energized meaning it's operating as a 13 ton unit. Now that I think about it I bet they sized this as 20 ton unit because of the extreme temperature swings we get here; I'll see days as hot as 105-110° in the summer and then get below 0 in the winter, not to mention we're at around 6000 ft here.
I agree though I would think it should be able to run on one compressor only and function as a 6.5 or 10 ton chiller, this is something I'll have to take up with Trane. I tried to load the chiller as much as I could today by raising my process air temp up to 90° (which is dangerously close to my shutdown) but I could only get the chiller's cooling calc up to 50% so I'm not sure if this thing ever actually calls for 100% cooling or not. I'll have to wait for a warmer day since it only got up to like 20° today.

So if I understand your set up, it's: Chiller --> Refrigerant/Water HX --> Water/Process Air HX? Is all of that packaged inside the chiller skid, or is the process air HX located remotely?

Correct, HX is remote; see the PID attached.

 

[DeeKay]

This should be easier to see

 

[American Locomotive]

Neither solenoid being energized is really whacky.

Unfortunately, at this point I don't think you are going to get any further without a serious controls retrofit/upgrade by Trane.

If the main process HX is remote, is there any particular reason why you didn't go with a packaged chiller unit from a company like KLH, Dimplex or ThermalCare? They basically specialize in units that do exactly what you're looking for minus the weird cobbled-ness.

https://www.dimplexthermal.com/products/chiller-solutions/w-series/

 

[acmikee]

sounds like they didnt properly commission the equipment. with the cold weather try shutting of one compressor at the breaker.
the unit should switch lead/lag on the compressors and stage down but its not
talk to the service manager about your complaints and have him send a different tech out to address the problems

 

[DeeKay]

Neither solenoid being energized is really whacky.

Unfortunately, at this point I don't think you are going to get any further without a serious controls retrofit/upgrade by Trane.

If the main process HX is remote, is there any particular reason why you didn't go with a packaged chiller unit from a company like KLH, Dimplex or ThermalCare? They basically specialize in units that do exactly what you're looking for minus the weird cobbled-ness.

https://www.dimplexthermal.com/products/chiller-solutions/w-series/

Technically this is a packaged unit, just not a very good one haha.
The company I work for has been around in one way or another since the 1800s and have been working with Trane since forever, my guess is we get pretty good pricing on their equipment with whatever crazy contract we have with them.

I actually liked the old system better, instead of pumps on VFDs run
by the chillers fancy useless controller we had the 25ton Trane CGA that would actually stage, and one single speed skid pump controlled by the plant.
To me the VFDs on the new setup are kind of a waste of money since there's no active SP adjustment and flow never changes, just size the pump trim correctly in the first place and we wouldn't have to worry about changing the speed... but what do I know, I'm no engineer

 

[DeeKay]

Has anyone had any experience with the Trane UC600 controller and TD7 display?
Does it seem normal that I can't change any setpoints at all? They programmed an "override" that allows me to change the water temp setpoint but that's it, and if I or someone else accidently press "release overides" it disappears and Trane has to come back out to program in the override again.
Or is this just the way the local Trane office is ensuring future call outs? Just seems odd to me.

 

[acmikee]

do you have the password to access the controller
if you do go thru the program and check set points and operations

 

[Full Throttle]

Has anyone had any experience with the Trane UC600 controller and TD7 display?
Does it seem normal that I can't change any setpoints at all? They programmed an "override" that allows me to change the water temp setpoint but that's it, and if I or someone else accidently press "release overides" it disappears and Trane has to come back out to program in the override again.
Or is this just the way the local Trane office is ensuring future call outs? Just seems odd to me.

It depends on how its programmed. You should be able to change more than setpoint. However it really is up to who wrote the program in the UC600.

I have some Trane Horizon MAU built by KCC solutions with UC600s. 1 I have full access the next beside it I have to hook my laptop and use my TU software.

I sent the picture you posted to my shops TCoE (training guy) his reply was WTF!

You need to escalate this to get corrected. I can't help as it is obviously an engineering issue be it control or mechanical and you have no access to setpoints

 

[ant.foste]

Read the responses above.

If the old 25 ton unit ran for years with no problems I doubt this 20 ton unit that can stage and unload is oversized. Suspect that this is a combination of problems; pump speed is too high, compressors are not unloading which would either be a design or controls issue. I'll dig deeper when I'm onsite today to see if it's staging or unloading at all.

The only reason the pumps are on VFDs is so we can adjust the speed to get the design flow rate through the evaporator and to eliminate the need for a balancing valve...they said they did that for power savings. There is no active speed adjustment on the pumps, they run at whatever speed they're set at 24/7.

So here's what the documentation from Copeland says:

"When the solenoid valve is energized the compressor is in full-load and when de-energized the compressor is in part-load".

So if you have one compressor unplugged, that means at most you've been getting 16 tons of cooling, as the compressor can only run at 65% with the solenoid de-energized. If you're down capacity by that much, and your compressors are only spending 5 minutes out of every 15 minutes, it sounds more like you need a ~10 ton chiller, not a 20 ton.

Either way, regardless if the chiller is oversized, it probably shouldn't be short-cycling like that. It should just be able to run on one compressor and modulate capacity as necessary.

So if I understand your set up, it's: Chiller --> Refrigerant/Water HX --> Water/Process Air HX? Is all of that packaged inside the chiller skid, or is the process air HX located remotely?

Chillers are a component of my background, but my new unit learning really ended around the time the York / JCI YCIV/YCAV/YVAA models began to hit the market. As I'm not too familiar with the controller you're mentioning, I'll ask you this: Can you download Trane TechView and hook into the controller? TechView may give you more access than you currently have.

Moving beyond that, but from the comfort of my dining room 900 miles away, your pump speed certainly is suspect as well as sizing. You mention the old 25 ton unit did fine, but how much scaling and efficiency loss was that unit working under, as well as plenty of other variables like HX sizing and water capacity. And sometimes chillers will have a problem staying synced up with a quickly changing heatload such has an industrial air compressor; adding more water to the loop with a large tank can help smooth the up/down swings of the return water temperature.

Something that is a bit of a band aid but a worthwhile solution is to increase the size of your loop. Add a 500-1000 gallon tank as a buffer which will extend the run time of the compressor at a lower tonnage while also extending the off-time of the compressor between cycles. I agree, you need to get those cycles per hour reduced.

Another band aid option is to add heat load through a basic water-cooled electric boiler. PM me if interested.

Meanwhile, continue working to figure out the staging. If I'm reading your posts correctly it sounds like both compressors are cycling on/off together? That is definitely not ideal; hopefully there's a setting in the controller to move these to a primary/secondary configuration based on manual (human) option or balancing hours/starts. Either way, I would not do anything to push your air compressor harder. The chiller is there to serve the needs of the air compressor, not the other way around.

A lot of what AL has posted above is good info, very accurate. And escalate with Trane.

Does this unit perform janky during the hot summer months as well? Or is this only a cold weather problem?

 

[Balvar24]

You need a chilled water storage tank.

 

[u3b3rg33k]

Technically this is a packaged unit, just not a very good one haha.
The company I work for has been around in one way or another since the 1800s and have been working with Trane since forever, my guess is we get pretty good pricing on their equipment with whatever crazy contract we have with them.

I actually liked the old system better, instead of pumps on VFDs run
by the chillers fancy useless controller we had the 25ton Trane CGA that would actually stage, and one single speed skid pump controlled by the plant.
To me the VFDs on the new setup are kind of a waste of money since there's no active SP adjustment and flow never changes, just size the pump trim correctly in the first place and we wouldn't have to worry about changing the speed... but what do I know, I'm no engineer

your issues aside -
I tell my boss, never buy something custom when off the shelf will do. "they" seem to do a good job of building and debugging systems that they can mass produce, but one-offs are another ballgame. -

VFDs can have a lot of value even if you don't need to operate a machine at varying speeds. they function as soft starters/motor starters, can limit inrush to RLA instead of LRA. they can also be good for error checking - the ones we use store the last handful of non-commanded shutdown codes, so you can find out WHY something had a problem when the guy running it has already cleared the error and has it running again.

 

[DeeKay]

It depends on how its programmed. You should be able to change more than setpoint. However it really is up to who wrote the program in the UC600.

I have some Trane Horizon MAU built by KCC solutions with UC600s. 1 I have full access the next beside it I have to hook my laptop and use my TU software.

I sent the picture you posted to my shops TCoE (training guy) his reply was WTF!

You need to escalate this to get corrected. I can't help as it is obviously an engineering issue be it control or mechanical and you have no access to setpoints

WTF is a lot of people's reaction hahaha. I'll keep bugging our Trane Contact, if push comes to shove I'll have to get our procurement people involved to lean on Trane "Creative" solutions a little harder.
But at this point I've narrowed it down to the controller never actually staging anything, I can't do much from my end though since Trane won't sell me the Tracer TU software.
Thanks for all the help! I'll update when/if we get some movement from Trane on any of this.

 

[DeeKay]

Chillers are a component of my background, but my new unit learning really ended around the time the York / JCI YCIV/YCAV/YVAA models began to hit the market. As I'm not too familiar with the controller you're mentioning, I'll ask you this: Can you download Trane TechView and hook into the controller? TechView may give you more access than you currently have.

Tech view won't work with the UC600, evidently Tracer TU is the only software that works, aside from BACnet but that only allows you to change addressing and stuff like that

Moving beyond that, but from the comfort of my dining room 900 miles away, your pump speed certainly is suspect as well as sizing. You mention the old 25 ton unit did fine, but how much scaling and efficiency loss was that unit working under, as well as plenty of other variables like HX sizing and water capacity. And sometimes chillers will have a problem staying synced up with a quickly changing heatload such has an industrial air compressor; adding more water to the loop with a large tank can help smooth the up/down swings of the return water temperature.

I'm sure the old unit was pretty inefficient and tired towards the end of it's life but I do know it actually did stage the compressors like it was supposed to. I wish my trends went back to when the old unit was new so I could see how it behaved then but I can only go back like 10 years.
Oddly enough the old system had no storage tank, and the new system has a 100 gallon tank. The heat load actually stays pretty consistent; the compressor is a 3 stage centrifugal with the aftercooler fan on a VFD, Even on the hottest or coldest days air temp usually stays within 10° of my SP. Pressure stays at 110-113psi 24/7 and flow sits around 60KCFH 24/7.

Something that is a bit of a band aid but a worthwhile solution is to increase the size of your loop. Add a 500-1000 gallon tank as a buffer which will extend the run time of the compressor at a lower tonnage while also extending the off-time of the compressor between cycles. I agree, you need to get those cycles per hour reduced.

Another band aid option is to add heat load through a basic water-cooled electric boiler. PM me if interested.

I appreciate the suggestions but being this is a process plant I can't just add stuff without going through an intense MOC(management of change) process...which is a massive pain in my ***, hours and hours of paperwork and meetings plus this setup is on a skid so the space I have to work with is limited( that's why we ended up with this stupid custom unit and not an off the shelf CGAM) If I start having a problem before Trane can figure this out I'll probably just pull the fuses for my second compressor, effectively making it a 10 ton unit temporarily.

Meanwhile, continue working to figure out the staging. If I'm reading your posts correctly it sounds like both compressors are cycling on/off together? That is definitely not ideal; hopefully there's a setting in the controller to move these to a primary/secondary configuration based on manual (human) option or balancing hours/starts. Either way, I would not do anything to push your air compressor harder. The chiller is there to serve the needs of the air compressor, not the other way around.

That's correct, the whole unit is operating in an On/off state without staging either compressor. My air compressor shutdowns are pretty far away from the actual danger zone of the unit so I can run it up there, I'm just on call for this site 24/7 so I'm trying to avoid those 1am Saturday morning call outs though I do like OT lol
A lot of what AL has posted above is good info, very accurate. And escalate with Trane.

Does this unit perform janky during the hot summer months as well? Or is this only a cold weather problem?

Yeah, looking back at my trends from last summer it had the same issues, both compressors cycling at the same time. I don't know however if the solenoids for the unloaders were energizing or not. So it could have been running as a 20ton at full capacity instead of unloading to a 13T unit. I'll have to wait for a warm day to double check that.

Thanks for the help

 

[DeeKay]

your issues aside -
I tell my boss, never buy something custom when off the shelf will do. "they" seem to do a good job of building and debugging systems that they can mass produce, but one-offs are another ballgame. -

VFDs can have a lot of value even if you don't need to operate a machine at varying speeds. they function as soft starters/motor starters, can limit inrush to RLA instead of LRA. they can also be good for error checking - the ones we use store the last handful of non-commanded shutdown codes, so you can find out WHY something had a problem when the guy running it has already cleared the error and has it running again.

I hear ya, but a 3hp pump on 480 won't exactly dim the lights if you know what I mean haha . Most of our larger motor loads are soft start or on VFDs though. Fortunately for me I'm the only guy running and working on this plant so I usually don't have to worry about people clearing faults without looking at what happened in the first place. Now previous jobs though, that was one of the biggest headaches; operators having a problem and then cycling power a bunch of times to try to "fix" it. Then they'd call me out to troubleshoot after all the evidence was gone

 

[RPH]

We used Dimplex, Koolant Coolers, and Dry Coolers for induction machines at plants without central cooling water. The refrigerant compressors on the units that had them would phase in as the load increased. Induction heating is a very unique heat load. When on it is sending extremely hot water back but when off cold water. Took quite awhile for the engineering staffs at the cooling company to understand how we loaded their systems. For the most part it worked out, still had flyers though.
On dry cooler keypads there is a lock switch on the backside to key inquiring minds from changing things.

 

[ant.foste]

Thanks for the help

I appreciate the detailed responses. You're on to something with the lack of staging, unless staging is actually programmed to be done through unloaders (which I doubt). You're working with a Frankenstein and that comes with its own set of challenges in understanding the designers and programmers thought process and subsequent unit logic.

I'd consider sticking a Fluke 1750 on the unit then on each compressor and watching the amperage trend over a 24 hour period.

 

[DeeKay]

I appreciate the detailed responses. You're on to something with the lack of staging, unless staging is actually programmed to be done through unloaders (which I doubt). You're working with a Frankenstein and that comes with its own set of challenges in understanding the designers and programmers thought process and subsequent unit logic.

I'd consider sticking a Fluke 1750 on the unit then on each compressor and watching the amperage trend over a 24 hour period.

If only I had an extra $15K to buy a 1750 I think I can stretch my Fluke 289 out to log 24hrs if I drop my sample rate a little bit.
Yesterday and today I played around with the pump speed and it definitely needs to be increased, I'll have to find the sweet spot next week when it gets cold outside again.

I did actually get a little bit of movement from Trane and was able to get some info from them. According to the sales engineer the pump speed is static and never changes once set and both of the compressors should cycle on and off at the same time. If the controller determines the need for more cooling then it energizes the unloaders for full capacity. So this thing supposedly can only operate as a 13T or a 20T....pretty stupid design if you ask me. If I have extra outputs on the UC600 I'm going to see if we can get this thing re-programmed to actually stage the two compressors.

However, there's still confusion on their end. One guy thinks the pump speed is supposed to modulate to help maintain setpoint and the other says the speed should never change once it's set. So I guess I'll just have to keep pushing for more info.

 

[metlmunchr]

So, they sell you a condensing unit capable of 6.5, 10, 13, 16.5, and 20 tons, and controls capable of only 13 and 20 tons. Kinda like selling you a 4 door sedan with 2 doors and the trunk welded shut, isn't it?

Then you have the question of what sort of idiot designs a system with 2 compressors starting at the same time. I saw Trane start downhill when they were acquired by American Standard. Obviously, Ingersoll Rand has continued their tradition of making any company they acquire worse in their quest to squeeze out another nickel.

My dad was a commercial and industrial HVAC contractor for 40 years, and I bid and ran jobs for him for 20 years. Got burned by Trane in the early 90's on a church addition with a 40 ton condensing unit and split chiller. The CU was one of their fairly early ones using four 10 ton Copeland scrolls. CU mounted on the roof. Church on the edge of a residential area.

The damn thing was so loud that the neighbors complained immediately when we started it up. Couldn't blame them, as you could walk 2 blocks down the street and still hear the damn thing sounding like a jack hammer in the distance.

Trane's solution... Cut all the sheet metal out of the bottom of the unit and cover the interior of all the louvered side panels with open cell sound absorbing foam rubber. The unit was elevated a couple feet on a steel frame, so we also fabricated an expanded metal "tray" to sit about 6" off the roof and covered it with sound foam to try to kill the noise coming out of the open bottom of the unit. A genuine Rube Goldberg contraption from hell. It got the noise level down to where the neighbors didn't complain, but it still made about twice the noise of an equivalent size unit with a couple cast iron recips.

Fortunately for us, the equipment was exactly what the engineer had spec'd for the job. But Carrier was an acceptable alternate in the specs and I wished a hundred times I'd spent the couple thousand $$ difference and bought Carrier equipment. Dad had used a number of Trane centrifugal chillers on larger jobs over the years, and never had any problems with them. But, after than nightmare, we never spent another cent with them regardless of price. And, if we looked at a job that was flat spec'd (no alternate brands) on Trane, then we just rolled up the plans and no quoted it.

 

[DeeKay]

So, they sell you a condensing unit capable of 6.5, 10, 13, 16.5, and 20 tons, and controls capable of only 13 and 20 tons. Kinda like selling you a 4 door sedan with 2 doors and the trunk welded shut, isn't it?

Then you have the question of what sort of idiot designs a system with 2 compressors starting at the same time. I saw Trane start downhill when they were acquired by American Standard. Obviously, Ingersoll Rand has continued their tradition of making any company they acquire worse in their quest to squeeze out another nickel.

My dad was a commercial and industrial HVAC contractor for 40 years, and I bid and ran jobs for him for 20 years. Got burned by Trane in the early 90's on a church addition with a 40 ton condensing unit and split chiller. The CU was one of their fairly early ones using four 10 ton Copeland scrolls. CU mounted on the roof. Church on the edge of a residential area.

The damn thing was so loud that the neighbors complained immediately when we started it up. Couldn't blame them, as you could walk 2 blocks down the street and still hear the damn thing sounding like a jack hammer in the distance.

Trane's solution... Cut all the sheet metal out of the bottom of the unit and cover the interior of all the louvered side panels with open cell sound absorbing foam rubber. The unit was elevated a couple feet on a steel frame, so we also fabricated an expanded metal "tray" to sit about 6" off the roof and covered it with sound foam to try to kill the noise coming out of the open bottom of the unit. A genuine Rube Goldberg contraption from hell. It got the noise level down to where the neighbors didn't complain, but it still made about twice the noise of an equivalent size unit with a couple cast iron recips.

Fortunately for us, the equipment was exactly what the engineer had spec'd for the job. But Carrier was an acceptable alternate in the specs and I wished a hundred times I'd spent the couple thousand $$ difference and bought Carrier equipment. Dad had used a number of Trane centrifugal chillers on larger jobs over the years, and never had any problems with them. But, after than nightmare, we never spent another cent with them regardless of price. And, if we looked at a job that was flat spec'd (no alternate brands) on Trane, then we just rolled up the plans and no quoted it.

Yep my thoughts exactly. I'm going to keep bugging them about a controls upgrade but I doubt it will get anywhere. When I mentioned to the sales engineer that this thing could in theory have 5 different capacities he replied: "This unit unfortunately only has 2 compressors so you can only have 2 capacities." I don't even know how to reply to that.

This turd was built when Trane was part of IR back in 2018. Hopefully Trane can improve on some of their problems now that they're independent again.

Unrelated to the chiller but related to IR screwing up any company they acquire, the air compressor that the chiller serves is a 400hp Cameron(Joy & Cooper Turbo) Centrifugal. Cameron used to be great, IR now owns Cameron, Cameron is now ****. The Compressor showed up with a 1/16" misalignment between the motor and the airend, it's supposed to be more like ±.009" At least we checked before starting the thing. Lots of loose hardware, cooling fan is out of balance and shakes like crazy.
IR's reply to the misalignment was that it must have shifted in transit from the factory.

 

[ant.foste]

If only I had an extra $15K to buy a 1750 I think I can stretch my Fluke 289 out to log 24hrs if I drop my sample rate a little bit.
Yesterday and today I played around with the pump speed and it definitely needs to be increased, I'll have to find the sweet spot next week when it gets cold outside again.

I did actually get a little bit of movement from Trane and was able to get some info from them. According to the sales engineer the pump speed is static and never changes once set and both of the compressors should cycle on and off at the same time. If the controller determines the need for more cooling then it energizes the unloaders for full capacity. So this thing supposedly can only operate as a 13T or a 20T....pretty stupid design if you ask me. If I have extra outputs on the UC600 I'm going to see if we can get this thing re-programmed to actually stage the two compressors.

However, there's still confusion on their end. One guy thinks the pump speed is supposed to modulate to help maintain setpoint and the other says the speed should never change once it's set. So I guess I'll just have to keep pushing for more info.

I apologize but am short of time right now. You can rent the 1750 from ProTec for $1300/month with leads and CT's. I can't speak to their weekly pricing from memory.

https://www.protecequip.com/products/fluke-1750-three-phase-power-recorder

 

[TRWham]

...When I mentioned to the sales engineer that this thing could in theory have 5 different capacities he replied: "This unit unfortunately only has 2 compressors so you can only have 2 capacities." I don't even know how to reply to that...

Apparently more sales than engineer in that one.

 

[Balvar24]

How often should my chiller compressors be cycling? Right now I'm seeing them run for about 5 minutes, rest for 5-10 minutes and then start over again (Usually 10-15 minute cycles if that makes any sense).
I'm wondering if my water pump speed might need to be adjusted(they're on a VFDs)

Asking for advise here because the local Trane guy couldn't even figure out how to get the pump working in the first place (it wasn't working because the isolation valve was closed doesn't inspire much confidence)

How much water in the system? Check the volume against the IOM recommendation.