Trane CGAF low pressure tripping

[chadtech]

urrrggggghhhhhh this thing!!!! Been having problems with machine for past several weeks tripping out on lpc open ckt 1 and 2 at times. It is comfort cooling, set at 48 degree chilled water, with OAT reset. It usually trips in the morning hours when OAT is around 30-45 degrees. I have checked all temp sensors, checked low ambient damper operation. I have seen it trip off and it is occuring when the 2 undampered fans have staged on and are running to lower SCT to SCTLL, but it drops the SCT way below the lower limit setting of 90, it will drop SCT to as low as 65 degrees before staging down, and this is when the suction pressure drops below 45. I have adjusted all low ambient temp adjustment as high as they will go and lowed the OA suppression temp to 2 degrees. What else can be done to prevent this from happening? I see the in the setup menu there is Control Algorithm Tuning parameters but it when trying to change the parameter is prompts for a Tuning password and I can't find this password anywhere. Anybody know the password?


MN: is CGAFC604AAA0W

[drivewizard]

I am assuming that once on and head pressure stabilizes that it works OK?
If so, change out the 45 # LPC's with 25# CO. and get on with your life.
I have had to do that on several of these units that need to run in low ambients. These units were designed more for comfort cooling, not process.
I know you said it was for comfort cooling, but most comfort chillers shouldn't run below 50-55 deg OA.

[chiller mekanik]

I am assuming that once on and head pressure stabilizes that it works OK?
If so, change out the 45 # LPC's with 25# CO. and get on with your life.
I have had to do that on several of these units that need to run in low ambients. These units were designed more for comfort cooling, not process.
I know you said it was for comfort cooling, but most comfort chillers shouldn't run below 50-55 deg OA.

I never understood why Trane sent a chiller out the door without a form of freeze protection in the barrel. I always figured that was why they used that ridiculously (high) low cutout point (45PSI) as the only real method of freeze protection.

Seems like I remember if you set it up for process cooling (even though it is a comfort cooling application) that changes some of sequence of operation. In fact, the controller bypasses the low pressure cutout when the first compressor fires in that circuit for around 3 minutes & that time may increase in process or it may increase the amount of times the compressors can fail on low pressure before it locks it out. Or, it may have reduced the minimum run times of the compressors themselves. I'm fuzzy on this so someone correct me if I'm wrong.

In my honest opinion, the real design flaw was/is in the fact that the on board processor only bypasses the safety circuit when the first compressor in that circuit fires but not when the second one does. You can add a bypass timer but thats a little tricky since the contactor coils are 120VAC & the & the actual cutout circuit is 24VAC. The first time I did this I simply bypassed the low pressure cutout across a set of NC contacts using two relays paralleled with the actual contactor coils. Therefore the coil voltage on the relays was also 120VAC. The trick is to also put a delay on make timer on each bypass relay & set them for 2 to 5 minutes, whatever you need. Now, each any compressor fires, you have a 2 to 5 minute bypass of the cutout circuit.

Otherwise, they are pretty cool machine once you get past this major obstacle.

[Tranetek]

Increase the temp. settings for your head pressure control and see if that helps. I've had to do that on several machines when the weather turns cold.

[chiller mekanik]

urrrggggghhhhhh "I have adjusted all low ambient temp adjustment as high as they will go"


MN: is CGAFC604AAA0W

I know my initial response was lengthy, but I spent a great deal of time with close to a dozen of these things when they first hit the pavement & believe me, there are some very useful & applicable comments therein.

[chadtech]

I know my initial response was lengthy, but I spent a great deal of time with close to a dozen of these things when they first hit the pavement & believe me, there are some very useful & applicable comments therein.


thanks for the info. I have been giving the whole bypass the LP circuit some thought and really likin that idea. Why did you use 2 relays paralleled? Couldn''t I just use one relay and use the NC and NO contacts?

What are your thoughts on changing the 45 psi pressure cutouts to something a bit lower?

I had also put the unit in Process cooling as you mentioned, but the compressors were just banging on and off a awful lot, so I switched back to comfort cooling and it seems happier.

Thanks to everyone for the help.

[Healey Nut]

Thats what changing to process cooling does it eliminates the anti short cycle timer but it does alter the low pressure bypass timer circuit.
You never said if it has glycol and what % ??
Changing out the LPC controls to 25psi setting is a good idea . I believe there is also a loss of charge switch on there which is set at 7psi ,I think ????

[chadtech]

Its not glycol, just plain old H2O. Chilled water setpoint of 45 with 7 degrees of reset. Running at the minimum water flow through evap according to the Trane chart. (about 75gpm) Has multiple fan coil units on it and only several of them have 3 way valves and no bypass..... I didn't design it just trying to make it run...:-)

[jayguy]

...Running at the minimum water flow through evap according to the Trane chart. (about 75gpm) Has multiple fan coil units on it and only several of them have 3 way valves and no bypass..... I didn't design it just trying to make it run...:-)

you can't fine tune poor engineering.

get a load with some more flow and you will find all of your problems will disappear. if you are not sure, then test it. open up the pump and let the load come up and then let the unit run...if all works well...let the engineer know.

remove the chilled water reset...it just complicates things. get the unit to run well then add the reset for energy efficiency later.

[chiller mekanik]

thanks for the info. I have been giving the whole bypass the LP circuit some thought and really likin that idea. Why did you use 2 relays paralleled? Couldn''t I just use one relay and use the NC and NO contacts?

What are your thoughts on changing the 45 psi pressure cutouts to something a bit lower?

I had also put the unit in Process cooling as you mentioned, but the compressors were just banging on and off a awful lot, so I switched back to comfort cooling and it seems happier.

Thanks to everyone for the help.

The way I wired it (which seemed right at the time) required two relays, one for each compressor contactor. That way the LP control would be bypassed for a couple of minutes anytime any compressor started.

Personally, I have no issue with replacing the LP that is currently set to cut out @ 45 psi with one that cut out @ 25psi except that it is the only mechanical form of freeze protection on the machine. I'm not sure how the on board controller would respond to the barrel being in a freeze condition. I would want to be certain my flow switch worked & I would even consider adding a DP control for the chilled water & possibly wire the aux contact on the pump starter in series with the flow detection devices as well. Call me paranoid, but that's my thoughts. If my memory serves me correctly, the controller will shut the unit down if the entering water is less than the leaving water (which should be an automatic result of no flow) for a given period of time. That's okay as long as the entering water sensor is accurate.

On the matter of process cooling, are you certain the compressors were staging up & down or cycling on low pressure repeatedly? I have the engineering bulletin somewhere but I think it allows the LP to open 3 times before it locks out.

[Tranetek]

Don't put the relays on. Find the real problem and solve that. If you are running on the min. water flow and the valves close off and take away more flow you are going to continue to have problems. you may need to put a bypass at the end of the loop in order to keep the correct water flow.

[chiller mekanik]

Is none other than the fact that Trane did not program the E-PROM with bypass time for the LP cut out when ANY compressor started. They only provided this for the first compressor in the circuit but not for the second, which was a mistake. We installed several of these machines back in the late 90's & I discovered this flaw & we (me, my bosses & the customer) stayed on Trane's case until they finally programed a chip just for us with bypass time on the LP circuit anytime any compressor started.

Before that, I wired in several of the bypass timers which proved to be a complete success & served as the prelude to them re-programing the chip after the engineer from Clarksville flew down & witnessed the smooth, uninterrupted operation.

Why this never went into regular production is beyond me.

Consider this: One 15 ton compressor is on in a 30 ton circuit. The second 15 ton compressor starts. When only one compressor was on, the OAT was 40*, the entering chilled water was 48 while the leaving was 44, the suction pressure was 72 psi while the head pressure is floating between 170 & 195, now the suction pressure "dips" to 44 psi & then presto, it has failed on low pressure. With the bypass time in there, the pressure will dip to between 40 & 45 but it will climb back up into the 60's in less than a minute most of the time.

Disclaimer: The above pressures & temperatures are an anomaly & do not represent an actual log of an actual operating condition. The readings are random to some degree while they are the result of my best effort to recreate what those cold winter nights with my CGAF'S in the late 90's looked like. :couch:

[Tranetek]

Seems funny when I have had a problem like this it is either a design issue with the system or and something cause by installation. When THE REAL PROBLEM is found and made right it all goes away. Without adding relays.

[chillerguy81]

A few years ago I modified on of these chillers the same way that chiller mekanik described, and as far as I know it is still running strong. I do agree that system design is usually the main culprit, but a bypass timer is a cheap easy fix until the rest of the system can be looked into. The bypass timer won't hurt anything and will make you customer happy that his chiller isn't tripping off every other night.

[jayguy]

i am concerned that the recommendations being made here are without finding the real problem.

just adding relays may make the problem go away and work well for the life of the unit...but is it the reason for the problem? i have several of these units that do not have anything extra added to them and they work quite well (even in low ambients...i think our high today is 11F)

if your car had a leaking gas line...would you just get a bigger gas tank? a bigger gas tank buffers the pain of the problem...but it does not solve the real problem.

the "real problem" has not been discovered. all of the solution paths have not been trekked. it sounds as if the unit is still fairly new, so we can still get the original engineers attention to consider other options (which he/she should pay for...i guess good luck on that part). if you "temporarily" fix the problem and then discover the real problem several years later...you would be lucky to get a call back from the engineer let alone pay for the solution.

do i think that Trane has perfect equipment? no. but i do think that they have most of it down pretty good. and if you can read between the Marketing Departments b.s., then you can find real solutions that work under reasonable conditions.

while 81's solution of adding the relays until the real problem is found has merit...it only works if you can stay on the problem. if this "solution" fixes the pain but not the problem...everyone will want you to go away (and then be pissed years later when YOU did not get back to it...gotta love the blame game!)

i am still concerned with the low flow, i always am. sometimes i think that marketing over-steps their bounds in order to get the sale. the relays will probably fix the low pressure trips but they will not solve the compressor over-cycling that you are most likely going to experience (although i hope that you do not).

[chillerguy81]

Well put Jayguy, I totally agree that the whole system should be looked into. You are right about not wanting to get yourself in a bind by bandaiding the problem then the customer not wanting to fix the root of the problem. Next thing you know your an a-hole when other issues arise from short cycling and what not.

[Healey Nut]

Jay . Yes you are correct but remember its ALWAYS ?????? the equipments fault and chad tech needs a fix(band-aid) to help him out of a bind with a customer .
It maybe that the chiller severs a bunch of computer server rooms or something that needs to run NOW. It maybe a system that has been built for future expansion and has very low load at the moment .
Dont get me wrong the real soloution is a bypass to maintain flow or more load or both .
The suggestions given will work but if I did it I would be very clear in writing that the modifications are a band-aid to an engineering problem which needs correcting ASAP for correct system operation and long term reliability.

[chadtech]

Here again I would like to think everyone for there opinions. (there like elbows and a-holes :-)) I realize that this is a band-aid fix but understand I have spent many hours watching, checking and adjusting this system before resorting to band-aids. BTW: it is 11 years old (98 year machine). Here again I didn't engineer it or install it and have just started working for this customer and supposedly this has been a problem in previous years during low ambient operation (but the "other" company adjusted something to make it run without tripping, another band-aid I reckon). The chiller serves about 12 fan coil units, and only a handful of these have 3 way valves. The chiller operates only on a time schedule and doesn't know if any of the fan coil are even running. I and the customer realize that this a engineering, install problem.

cheers!!

[jayguy]

Jay . Yes you are correct but remember its ALWAYS ?????? the equipments fault and chad tech needs a fix(band-aid) to help him out of a bind with a customer .
It maybe that the chiller severs a bunch of computer server rooms or something that needs to run NOW. It maybe a system that has been built for future expansion and has very low load at the moment .
Dont get me wrong the real soloution is a bypass to maintain flow or more load or both .
The suggestions given will work but if I did it I would be very clear in writing that the modifications are a band-aid to an engineering problem which needs correcting ASAP for correct system operation and long term reliability.

very true...i just wanted to make sure the caveat was out there. i have been in a situation like this before where several months later it became a huge problem because i got it "temporarily fixed" and other customers had no cooling..."we'll get you back there ASAP" becomes the normal response and the next thing you know...ASAP means "when the customer complains again". i do not want chad to get burned like that.

there are also several other potential fixes that have yet to be investigated...some of these are pretty quick.

one more thing...if this unit has had this many problems for this long...i would not take anything for granted. who knows how much extra charge has been added or what else was bypassed or added or rewired.

[jayguy]

...The chiller serves about 12 fan coil units, and only a handful of these have 3 way valves...

about the 75 gpm that you state you have (or is spec'd)...is that the total with everything open or with just the handful of 3 way valves open?

[txhvac]

They updated the software on units built after yours due to the same problems. The condenser fan staging can drop you out on low pressure once the LP bypass timer is up. Also the reset on the old switch was like 60#'s may not close in-time, again locking you out once timed out. Fan coil units are always another name for female dog. Valves are important, Jayguy is right. Have people been in there screwing around w/ circuit setters? 75 gpm is the bare minimum for this machine. Post a log of this chiller. Look at the big picture before just wiring components, no offense to anyone, to me something sounds wrong in your loop, which should open up more investigation. Hope this can help.

[chiller mekanik]

However, I do find it interesting (more like baffling) that no one had an issue with removing the 45 psi cutout switches & installing 25 psi cutout switches.

My suggestion was to simply build a better mouse trap. In fact, there are timers tied to these relays.

I'm always impressed with principal minded folks. It almost brings a tear to my eye. (not really)

But here in the real world, I don't know about the rest of you but I myself don't exactly have customers lining up to hand out purchase orders for me to spend a day or two or three trying to get a hold of the original design engineer, as builts, service history of all equipment & whatever other parties involved to get to the bottom of the mystery. Albeit worthwhile I'm sure, most of the time (especially if the equipment is of the age we are discussing) customers simply want the problem to go away so long as the equipment isn't compromised in terms of safety, efficiency or usability. But make no mistake, if they have the money, I've got the time & I too will leave no stone unturned in search of "the real problem".

Quite often, we're dealing with 1K or 2K dollar limits & occasionally 3K (if we're lucky) but that's usually the only opportunity we have to bring some version of a workable solution to the table before they will move on to another vendor or give us trouble paying the bill because it only consists of two days of research with no actual repairs being made because we're still gathering data. I've got no problem with the idea of offering a follow up (thorough) investigation.

:deadhorse:

[CaptJackSparrow]

I recall some of these that I upgraded the mods to the new cond fan control but in the end I installed an ambient temp control so that when the oat dropped below say 50 deg I took over fan control with pressure switchs. It was a 24hr/365 telephone equip bldg. I don't know if they ever did correct the control but i prefer increasing head vs lowering the lpc. Trane has always been good Bout correting the real problem but my opinion is they are using to few condenser fans, which makes overshooting the issue. Hard to go back and add multiple fans after the fact. Cgafs suck is the bottom line for me!

[troubled_jeff]

Chadtech had the exact problem that I am having right now. I have a 50ton chiller, 4 compressers, 6 fans(2 with dampers) average 60 low side tripping on low pressure. I have raised the supply temp to 45, set the delta T up to 20 degrees, opened my pumps wide open, set the cold water valves to 100% and let the boiler work for a livin' to increase the load, diabled the low ambient temp cutout, and watched the guages for 10 hours at a time without seeing it drop out. Setting the delta T seemed to improve it but the problem is still there. I have low press switches on the way but that is my last ditch effort. Not much hope. Ckt 2 always goes out before Ckt1. I work in an air traffic control tower, so they get kind of testy when they're uncomfortable. I've had TRANE techs come out, but it's hard to troubleshoot if you can't see when it happens. Anything you guys can tell me will be appreciated.

[jayguy]

welcome t_j! i like my doctors and pilots happy (i guess that would go for the guys helping to land the plane too!).

1. model number?
2. serial number or at least age?
3. type of fluid (if glycol, type and concentration)?
flow rate or at least the pressure drop (we can figure the flow rate with model number, fluid type and pressure drop...and the pressure drop needs to be as close as possible to the chiller and use the same gage for both pressure measurements)
4. what does "average 60 low side" mean? pressure switches don't trip on the "average".

what is this unit doing...exactly?! this is a very technically-oriented site. more details equals a more exact diagnosis. remember, we can always throw away the details that we do not like or mean nothing be we can not invent the details we need. i tell every newbie that when you are stuck and you want to call me, you better be able to answer every question i have...even the color of the compressor paint! you don't know what it means, but it may mean something to me.

[troubled_jeff]

Hi Jay. Thanks for the reply and yes I try to keep the controlers happy. The only problem is I'm Avionics playing with HVAC. I think they for get that.
Anyway, to answer your questions:


1. model number? CGAFC50EACA
2. serial number or at least age? C99K2230OM - I was assuming that makes it a '99?
3. type of fluid (if glycol, type and concentration)? Just water. I'm not sure of the flow rate. I haven't seen anywhere to ba able to check the pressures close to the chiller. The closest place on the supply line is just before the pump which is 30 psi in and about 45 psi out. I haven't seen anything except places for temp on the return but I will check closer when I get back or when I get called in again. It has a flow switch but I'm not sure what it's rated at.
4. what does "average 60 low side" mean? pressure switches don't trip on the "average".
As for the "average 60 low side" I was just trying to say that when it's running the suction pressure is about 55-60 psi. I have not seen it fall offline so I'm not sure what the pressure is actually getting down to. I've spent up to 10 hours a day on cold days watching the temps and pressures but the thing never fails when I'm watching it. I evacuated and weighed in the refigerant to the factory specs and it was accurate.
I have set the
Saturated Condensor temp Control Band to:
Lower Limit : 90 deg
Upper Limit : 125 degrees
Temporary Low Limit Suppression : 5 deg

Saturated Condenser Temp
Efficiency Check Point : 105 Deg
Low Ambient Control Point : 100 deg
The Low Ambient Control Deadband and OAT Supression Setpoint I couldn't find where to check it.
I also have the Delta T set to 20 deg. That slows down the staging, right? And I set the Leaving Solution Setpoint to 45 deg. It usually gets down to about 40 degrees and goes into hot gas bypass. For the most part it only uses one ckt but when it does use both ckts the #2 doesn't run for long and cycles off. The only consistant thing I see is that the #2 ckt always fails first and the #1 ckt doesn't always fail. Also, when the #1 ckt is running the fan with the damper on it constantly runs and the damper will be open but the temp stays around 105 deg. But it doesn't have to hit the Saturated Condensor temp Upper Limit to turn it on. I am wondering if that is staying on too long. When I first reset it all of the fans come on and then stages back down to that one fan. The #2 ckt fan comes on at about 75 deg and cycles off with the compressers which is when the Leaving Solution gets within a degree or 2 of the 45 deg setpoint.
That is all I can think of for now. If it's not enough let me know what else I need to look for and I'll do it. The cycle of operation on the fans don't really seem right to me but maybe it's just me.
Thank you for your time and I'll be awaiting further instuctions. I have our second level techs comming out monday but you guys have the experience. Oh, and my compressers are gray.

[MrSlim]

Does anyone have any other manuals to share on the Trane CGAF Chiller?

http://www.trane.com/Commercial/Uplo...-svp01b-en.pdf

[MrSlim]

Does anyone have any other manuals to share on the Trane CGAF Chiller?

http://www.trane.com/Commercial/Uplo...-svp01b-en.pdf

http://www.trane.com/Commercial/Uploads/Pdf/1063/cg-svp01b-en.pdf