Chiller Trip at Low Condenser Water Temp

[jfschwarz3]

Have a project I'm working on where the chiller will trip and simply turn off at low CW temp (~65°F). Trying to optimize the plant so lower temps from the tower was where I was looking to start... Any ideas as to why this would happen? ~2500 ton centrifugal Trane unit

[chillrdude]

Oh boy this is going to be fun to watch

[jayguy]

Oh boy this is going to be fun to watch

Yes it will!

you did not provide why the Chiller decided to turn off...it will tell you. It is a ton of fun watching engineers and BAS techs try to control something they don’t understand properly. You probably heard that lower condenser water temps will lower energy costs (the super-cool word of the day is “Optimize”)...so lower must be better, right?

the architect who engineered the Brooklyn bridge, went to a wire manufacturer to learn how to make wire and specify it properly. A Chiller is significantly more complicated than wire and yet most engineers treat it as a part number and say that it is the service techs problem to figure out why it won’t work.

i am not saying any of the above applies to you specifically, but more as a general rant to those who search out why their designs won’t work.

[Healey Nut]

Time to put my uniform on again

[Healey Nut]

Have a project I'm working on where the chiller will trip and simply turn off at low CW temp (~65°F). Trying to optimize the plant so lower temps from the tower was where I was looking to start... Any ideas as to why this would happen? ~2500 ton centrifugal Trane unit

Do you know what lift is ?

[pecmsg]

Finish all 500+ pages and get back to us...…………………..

[jfschwarz3]

Guessing the reaction to the question is because it’s almost impossible to answer without more specifics?

Apologies if I’m in the wrong spot

[chillrdude]

Guessing the reaction to the question is because it’s almost impossible to answer without more specifics?

Apologies if I’m in the wrong spot

You are in the right spot, the problem is we have no idea who or what you are, your profile says "engineer" that's pretty vague for us, are you a mechanical engineer or are you a control guy working for one of the big box control companies trying to perform some kind of "tower optimization" or "chilled plant optimization"?

What specifically are you wanting to achieve? Your original question was so vague it leads a person to believe you are in way over your head already, not trying to bust your balls but most of us have been around the block with all of this way to many times before. Some engineer or a control guy with no idea or clue on how, gets tasked with saving money in a chiller plant.

Don't get me wrong there are always opportunities to save energy in chiller plants but a person needs to know and look at the entire system, tower optimization is just one piece of the puzzle, there are other pieces that have to be considered. Variable flow on the primary and/or secondary chilled water, variable flow condenser water, variable speed chillers etc.

Your 2500 ton Trane duplex chiller can operate just fine with 65 degree tower water if it has the rest of the pieces of the system to make that happen.

[JCN]

What chillrdude said.....

We have all spent way too many days shaking our heads as morons try to "optimize" a plant, then complain because now the plant doesn't run right. Must be the equipment's fault....right?

Give us the overview and then let us help you help yourself.

[jfschwarz3]

Yep I can see how that’s obnoxious. So... I’m a mechanical engineer consultant and we focus on central plant design primarily. Large university campuses, hospital districts, etc. Mostly chilled and HW/steam production with occasional CHP. This question is in regards to a project serving a university up in the northeast. Plant is CHW only, has 5 ~2500 ton chillers, 5 1-1 counterflow towers, and primary secondary pumping with all pumps and tower fans on VFDs. Chillers are all CS. Project goal is overall plant operation so one thing I’ll look into is transitioning to variable primary. As far as my original question, operators have mentioned they want to operate chillers with lower cond water temps but can’t wothout chillers tripping. Was trying to get some ideas as to why this would happen.

I’m sure there’s still details I’m leaving out so just let me know and I’ll provide. Just looking to the forum to get some ideas as to why this would happen. All I could come up with was freeze protection of some sort. Thoughts?

[JCN]

The secret here is lift across the compressor. A rule of thumb is that the entering condenser water has to be a minimum of 15 degs warmer than the leaving chilled water temp. Keep in mind that it is only a rule of thumb to use in the absence of actual design data, but it is a good place to start. As a design engineer, I would contact the factory for the info on that serial number. It will give you minimum and maximum chilled and condenser flows, plus the minimum and maximum condenser water temps. Once you have those, you will know how low they can run the towers.

[chillrdude]

JCN and Healey are both right on the money, it's ALL about lift. Lets make a few assumptions on your Trane 2500 ton chillers, your chillers are "probably" designed around an ashrae standard, yours "may" be different and you'll need the design specs to verify but for this lesson lets use the most common design specs.

Most of the chillers are designed around these numbers at full load unless otherwise noted.

CHW in 54*
CHW out 44*
Saturated evap temp 42* = -8.61 psi
CHW flow rate of 2.4 gpm per ton= 6000gpm

Cond in 85*
Cond out 95*
Saturated cond temp 97* = 4.97 psi
Cond flow rate 3.0 gpm per ton= 7500 gpm

Centrifugal compressors are designed by "lift" the volume of refrigerant the impellers can pump or amount of work the impellers can do, in the above case the lift is the difference between the evap pressure -8.61 plus the cond pressure of 4.97 so the lift across the impellers is 13.52 psid.

If the lift or volume of refrigerant the impellers are pumping becomes greater than the impellers can lift, the compressor goes into a surge condition, the refrigerant reverses flow backwards across the impellers and takes thrust off the bearings, this is not good.

If the lift goes to low, then you have the possibility of losing oil, you need a minimum amount of lift to keep the oil where it needs to stay and on some of the older chillers you need the minimum lift to maintain motor cooling.

So the key is keeping your lift under control, you go too high the chiller surges and tears up bearings, you go too low and you lose oil or your motor gets hot.

Trane has several bulletins out where they say the minimum lift is 3psid. I like to use 5psid. If we use your 65* entering cond water with the example above the sat evap pressure is -8.61 but our sat cond pressure now is -4.41 so our lift is down to 4.2 psid, barely above our minimum.

The key to running cold tower water is to not pump so much of it, if your design is 7500 gpm at 85* and you lower the temp to 65* you dont need 7500 gpm. Most big chiller plants that I take care of utilize variable everything, variable cond water flow, variable tower fans, variable primary and variable secondary chilled water, they monitor the compressor lift and only pump the amount of water they need to maintain temp and lift across the compressor.

Thats a basic lesson on what your looking at, there's a lot more to it to make it work, the numbers are not exact but will get you in the ballpark, you'll need the design specs of the chillers, pumps, towers, min/max flow rates etc to really nail it down.

It's been my experience that most of the engineering firms and big box control companies really don't have a very good idea of how it all works together, they really don't understand the design limitations of the chiller. The above example applies to Trane low pressure centrifugals, High pressure Yorks and Carriers have their own limitations.

[jfschwarz3]

Thanks guys. I really do appreciate yall taking the time. So to be more specific, client has a 2010 duplex Trane CDHF using R-123. It looks like the lift you estimated assumes R-123 too. Live data on the chiller:

CW 77/74 6660 GPM (min 3440 GPM)
CHW 46/42 3780 GPM (min 1890 GPM)

Bundle refrig sat temps are 40 and 82 so the lift is about 8psig currently, assuming i'm calc'ing it correctly.

I know the chiller will tell what the reasoning for the trip is but that info is still on its way for the time being so trying to think as to why this would happen.

Also, if we were to install VFD's, would there still be the same issue of the lift being too low to properly cool the motor? Or would the decrease in speed mean that the motor cooling can be more easily accomplished at those low loads?

Thanks again and as always if i'm leaving out any data just let me know

[chillrdude]

That particular chiller has a refrigerant pump that pumps liquid refrigerant to cool the motor as long as you have the proper column of liquid stacked in the condenser. You have to control the lift, as long as you maintain a minimum lift of say 5psid and a max of whatever design is plus a little you should be fine.

The event log will tell you the reason the chiller is tripping off, until you get that we're just guessing.

[WAYNE3298]

As a design engineer I specified optimum head pressure control. I specified for the contractor to control the head pressure per the recommendation of the chiller manufacturer to obtain optimum performance. Optimum performance includes economy of operation as well as nuisance trips as well as chiller protection and stabile control. I had to deal with the manufacturer's standard flow permissive for start-up but got the factory on board before the fact instead of after.

[Boiler Guy]

@chillrdude

What do I need to do to set up a class and pay you to dump your knowledge? 😆😆😆

[chillrdude]

It'd be a short class that's for sure

[pecmsg]

It'd be a short class that's for sure

If Beer is served?

[Healey Nut]

If Beer is served?

Have you considered how much beer chiller guys can put away ?

[lzenglish]

Have you considered how much beer chiller guys can put away ?

The Local UA Says !