Original message
| niko851
|  "Cent. Chillers Water temp fluct's."
, posted Mon 6 Jun 11:01  
Hi all - especially Chiller Guy since we couldn't get this post to work in the other forum.... LOL
EQUIPMENT:
1 Trane Cent. 600 Ton (Lead), 1 400 Ton (Lag).
2 Evapco CT's, Tracer (upgraded last year) BCU.
2 HTW CB Boilers (both offline for the season now).
24 AHU's, 2 each floor.
Issue w/ the above is that my condenser water temp's are fluctuating WAYYY to much.. I started this up a few weeks ago, during a 78 degree day, and programmed Tracer for a temp of no greater than 90, no less than 80; specifically, Cooling tower fan(s) should come on at +/- 85d to maintain the specified threshold. Pump(s) always on.
Chilled water is (now) set at 43-47 since we're warm enough to justify bringing it down to NOCWT. Before, I had it at 55 +/- since pulling it down was unnecessary. Nevetheless, I still, to this very minute, keep getting alarm notifications that the condenser temp is above 90. After logging into the system remotely (going back to about a week ago), found the cooling tower fan came on @ 94, off at 88, back on in less than 2 mins. at 93, off at 88... Can't seem to pull down to the 'normal' 85... History over the last few days (and almost 2 weeks) show the lowest the temp has been is 87. Since this is an ongoing problem to date, the history still shows a higher-than-specified cond. WT..
I've thought about reprogramming the cut-in & cut-out times, but that doesn't address the root cause. As an FYI, the cooling towers were covered (NOT BY ME) during the Winter by the buidling owner. I'm wondering if condensation build-up from the winter has anything to do with it, i.e. getting into the components section of the CT. Obviously, the things are used to being in a wet location, given that's their design and intent (couldn't get that through the owner's head, but what the hell do I know - i'm just the Engineer - LOL), but what's the opinion on a bad water temp. sensor @ the CT? The sensor in the equip. room below (it has 2 sensors - one at the CT and one set at the inlet/outlet on the Chiller) gives appropriate readings wthin my specified thresholds. I have defeated the sensors being the problem pursuant to comments later in this post, BTW.
I found it gave me an 'average' temp of the combination of the two sensors (the ones at the CT & Chiller). Upon further digging, Im speculating that the fan on the CT is being controlled by the remote sensors (on the CT) instead of the sensors on the Chiller....
Anyway, I wanted to address this before we start getting warmer up here in MI. Afterall, it was 80's 2 weeks ago, then 73 next day, had to kick the lag boiler back on (due to whining - go figure), now we're in the 90's. I didn't take the chiller off-line due to humidity issues when the boiler was put back online, however now I'm only using the chiller(s), obivously. So, after reading my novel, I open the floor for discussion, questions, or comments.
Also, at first thoughts, and just brainstorming now with others' suggestions/posts, I veered towards normal startup woes. Higher cond. temperatures means higher heat transfer rates across the exchanger with resulting condenser rise. High exchange with early season tower reduction goes to the "Settles out" part of the (previous) suggestion to get the balance. Much like waiting for the residential small system to settle the startup load before readiness to read service pressures and temperatures ... but trickier balance with big-mass equipment. High condenser temperatures could mean a need to get the tower water side further down below the expected alarm number to absorb the heat sent to it (precooled water mass) so it can absorb the heat with less resulting rise to the alarm temp. Tougher to do with fan control near the summer expected temperatures set for my expected balanced load.
Nevetheless, I'm getting quite irritated with the never-ending alarms, justifiably. The builing was closed yesterday, so I decided that (since no one was there), it would be ideal to go there and troubleshoot, given it was 92. I also brought #2 online due to the temps as I know it's now that time. Temps are expected to be this way for the next week as well. Upon arrival, only #1 operating:
ChWTs: 51 (setpoint 45)
CnWTs: 92 (setpoint 85)
CtFANs: on
Div's: Fully open
I brought a friend to be a 'monitor' on the computer while I was going up to the roof and back down. All this in my 'crippled' state (disabled due to back surgery in Dec. 04, but still 'truck on' as I can't just sit back and not do anything).
Before I brought #2 online, I would tell him when the fan would cut-out and he'd give me the temp. We're still cutting out well above the set-points and not coming on until 1) Alarms, Alarms, and more alarms, 2) CnWT: 90+.. Again, bearing in mind that I've got 92 degrees outside, I figured bringing #2 online will resolve it. So, (before starting #2) I took the software offline and went into manual mode, allowing me to troubleshoot sensors per a prior post. Manual readings of the temps were dead-on with the systems' reporting temps, so I cannot forsee a sensor issue now. OF course, still up for debate with my collegues on here...
I gave the commands for constant fan, div's fully open. Since the software was bypassed, I got the CnWT down to 80; so it CAN be done and this indicates to me that I am (more than likely) having a software problem. Given the 92 temps outside, I then brought #2 online. After about an hour, give or take, also going back into auto-mode, I had the following, #1 being lead, #2 being lag:
ChWT #1: 49
ChWT #2: 50
CnWT #1: 91
CnWT #2: 91
Setpoints:
ChWT #1: 43
ChWT #2: 48
CnWT #1: 85
CnWT #2: 85
Irritating eh??
Once I got ChWT's down to <55, I started all AHU's to put a load onto the system. ChWT's 'slightly' fluctuated, as anticipated, for about 10 minutes until the ChWT's went up to >55. Now I had ChWT alarms in addition to CnWT alarms... While waiting for stabalization, since I put loads on teh system, CnWT's went above 100. NOT good at all, and in fact kicked the lag offline.
Went back into manual mode, put both CT's on constant, pulling the CnWT's down to 80 +/-, then back into auto. Brought #2 back online, CnWT's steadied at 85 (thought I finally got it), monitored for about an hour w/full loads, ChWT's were +/- 50, and started to pack up. NO SOONER did I get out the door the cell rings; I don't even need to tell you why.... For the sake of the occupants, I put the things into manual again, fans constant w/div's only. Another hour passes, building is cooling down nicely, temps within reason all around, and left it this way so they didn't walk into a sweat-box this AM. So, it is apparent that the CT fan(s) controls/programming/who the hell knows is the issue. The ONLY way to maintain my setpoints is via manual fan operation w/auto diverters.
To make matters worse, here's another kicker. We had a huge storm blast through resulting in power outages. Oh yes, got notification of that at 9:30PM as well. I couldn't get into the system remotely, so I assumed the building was (still) powerless. Got there @ 4AM, power was on. Logs indicate loss from 9:30 - 2AM (rounded), system did not restart automatically, despite having the software side on generator backup. Before restarting any of the equipment, I rebooted the PC to start 'fresh' for lack of better words. Then I went through the start-up procedures, having pulled down to 50 degrees in about a half-hour.
After doing the above, I kicked on all the AHU's for each floor, starting at the top and working my way down. Wasn't too bad in the building itself, highest temp being 78 on the 12th (and highest) floor. Lowest was 75 on floors 1-4.
So far today, now that I was/am (still) on-site in order to start it back up, the ONLY way to keep a <90 CnWT is by semi-auto operation. I have the fans on constant, only using diverters, both chillers are set at 40 (to get the building down quicker, in hopes I should say), and when daylight broke, went up to the roof again to see if there is ANYTHING noticable on the equipment that would be indicative of a problem. Both had nice blowing mists flowing out of the tops of the CT's, so I sat up there for a while, smoked 2 cigarettes, and more/less enjoyed my 'break' on a 70 degree, 90% humidity, 5:30AM rooftop with 2 CT's blaring in my ear.
It's 10AM now, building is still powerless and is becoming QUITE stifiling and people are bitching (of course). I walked into an 80+ building as they don't want it cooling, err 'occupied-cooling temps' on weekends, UNLESS requested in advance and approved by management, even though I've battled this argument of 'it won't cool down instantly in the AM's for years now'. However, after they got a taste of their own medicine a few years back (like today), I got 'approval' to keep the lag running on the weekends to maintain <80 on all floors. Once ONE floor gets above 80, the lead will come online regardless of the day, time, date, temp, year, moon, etc. (LOL)... Sorry - have to have SOME humour in order to maintain my sanity....
This is maintaing my 85 CnWT set-point, +/- 1 degree, until power went out AGAIN at around 5:45....
Again, there's about 15 MAIN roads and 30+ streets completely closed due to downed lines. The building is about 1/4 occupied since the majority of people can't get here. Generators kicked on, all MY equipment is off (not powered by generators), and now I've got the chance to sit here and figure out what to do when I have power again. In this 'office' as they call it (control room that I sit in), I at least have a window-shaker that I can use, given this room is on the generator. HA!! I'm the only cool one...
Since it has been a half-hour since I started writing this, building is around 75 on all floors except for 12, which is still at 78. This is where the owner sits on his behind all day as well. So I'm still hearing it from him, but there's nthing I can do about it, given the power outage. So at least there's not as much chewing out right now; they're more worried about getting computers, etc. back up & half-way running due to the outage. They also believe I have some magic wand that I can wave to transfer power to the cooling equipment... Yeah; Sure....LOL
SO - to anyone who reads this long, but detailed novel, please give me any feedback whatsoever. I cannot and will not babysit this system much longer - even if power outages were not happening at the moment. If i had power, I'd still be in teh same predicament. One last comment - Trane's rep(s) cannot be out (or anyone else familiar with Summit for that matter) this week due to eveyrone else going through similar problems. I was told I am not 'high priority' since I'm an on-site engineer and the Owner ONLY contracted half-ass (sorry) support with them to save $$, thinking that I am a complete systems programmer on top of it.... This means I"m stuck w/the grief and BS until the reps get here next week - and given this power problem throughout the area, again, I'm low priority on the call list.. I hate like hell to bypass things and then hope for the best, but again, I'm not going to sit here all day & night, whether it be me or the other two employed operators (I'm the Engineer, they follow my lead - but had to lock out the software since they feel the need to play instead of just do their jobs, etc.).
Thanks in advance, of course, as any input will be greatly appreciated.
Niko
Check your circuit breaker or fuses before calling for service - it could save you $$!!
|
|
Replies:
|
| chiller guy
|  "Re(1):Cent. Chillers Water temp fluct's."
, posted Fri 17 Jun 19:23
niko:
How are you making out with the condenser water temp issue?
All refrigerants are safe - All refrigerant are dangerous. The difference is YOU !!!
|
| chiller guy
| "Re(1):Cent. Chillers Water temp fluct's." , posted Wed 8 Jun 20:23
For starters, I still can't get through on the other site.
OK, you can control manually but not through the BAS system, correct? Whose system is it?
It sounds like you may be trying to over control the tower loop. Try using setpoints for stop/start of fans and eliminate any timing function for the time being. The fan timing would change as the load changes.
What are you trying to accomplish other than control of condenser water temps at this time? Also, you mentioned two sensors. Which one actually signals the tower fan motors to stop/start(chiller inlet or tower outlet)and how far apart are they?
This is enough for you to work on for now. This will work! My present job has 6000 tons of chillers and the tower water is as steady as a rock.
All refrigerants are safe - All refrigerant are dangerous. The difference is YOU !!!
|
| | niko851
| "Re(2):Cent. Chillers Water temp fluct's." , posted Thu 9 Jun 07:18
I'm copying your post and answering in caps:
OK, you can control manually but not through the BAS system, correct?
I PLACE THE BAC INTO SEMI-AUTO OR MANUAL MODE TO WHERE ALL THE BAC DOES IS GIVE ME STATS. IN SEMI-AUTO, I CAN ACTUATE (OR DE-ACTUATE) COMPONENTS, BUT THE BAC WILL STILL (E.G.) SHUT DOWN THE CT FANS ONCE 'SATISFIED'. FOR SIMPLICITY SAKE OVER THE ISSUE, I TOOK THE BAC OFF COMPLETELY, CONTROLLING VIA THE CHILLERS THEMSELVES TO AVOID THE BAC SHUTTING THEM DOWN/PARIALLY CONROLLING THE EQUIPMENT.
It sounds like you may be trying to over control the tower loop. Try using setpoints for stop/start of fans and eliminate any timing function for the time being. IF I UNDERSTAND THIS CORRECTLY, AND IF I UNDERSTAND THE CURRENT SET-UP CORRECTLY, THIS IS/HAS BEEN DONE. I GIVE A 60 SECOND 'LAG TIME' SINCE THE COND. WATER HAS TO GO FROM THE BASEMNT UP TO THE '13TH' (NOT REALLY A FLOOR, AS IT IS HOUSES THE BOILERS, PUMPS, SOME AHU'S, AND THE CT'S), UNLIKE THE BOILERS WHICH ARE RIGHT THERE NEXT TO ME. I BELIEVE, AND CORRECT ME IF I AM WRONG, THAT I 'RIDDED' THE TIMING FUNCTION VIA THE SEMI-AUTO OPERATION/MANUAL OPERATION. WITH THE BAC NOT CONTROLLING, THEN IT'S UP TO ME/AN OPERATOR TO CONTROL, OR LET THE BAC PULL USE JUST THE SET-POINTS, BYPASSING ANY TIMING/ANTICPATION/DIFFERNTIAL/ETC... LOWERING THE CUT-IN TEMP TO 80 ACHIEVES +/- 84-86 UNTIL I PUT THE BAC INTO AUTO.
The fan timing would change as the load changes.
THAT'S THE CONCEPT I LEARNED AND WENT ON, BUT NO DICE WHEN THERE'S A LOAD... AS I TYPE THIS, THERE'S A FULL LOAD ON THE SYSTEM AND ONLY MEANS OF ACHIEVING 85 OR THERABOUTS IS SEMI-AUTO AND/OR MANAUL OP...
What are you trying to accomplish other than control of condenser water temps at this time? OTHER THAN GET THE BUILDING COOL, AND STAY THAT WAY, JUST KEEP MY COND. WATER TEMPS WITHIN REASON VIA THE BAC, AND NOT HAVING TO MANUALLY ACTUATE COMPONENTS TO DO SUCH....I SUPPOSE MY OTHER OBJECTIVE IS TO LET THE BAC TAKE OVER COMPLETLEY, AS IT SHOULD, SO ALL WE HAVE TO DO IS MONITOR, NOT CONTROL... IT'S TOO RISKY WHEN THOSE TEMPS GET HIGHER SINCE IT AFFECTS THE CHWT, EVEN KICKS OUT EITHER THE LEAD OR LAG WHEN THE CNWT'S GET TOO HIGH..
Also, you mentioned two sensors. Which one actually signals the tower fan motors to stop/start(chiller inlet or tower outlet)and how far apart are they? I FIGURED THIS OUT TWO DAYS AGO - LOL... IT'S THE INLET (ACCOURDING TO MY SPECS IN THE 20+ BOOKS HERE, UNLESS THIS IS BOGUS TOO), AND THE CT SENSORS (I DO BELIEVE) PRODUCE THE 'ANTICIPATION' FOR THE FANS AS WELL AS ANOTHER 'FYI' TEMP. THE LOOP IS FROM A 2-STORY DOWN BASEMENT TO THE 13TH FLOOR. THERE'S ONLY 12 FLOORS FOR OCCUPANCY, 13TH IS THE EQUIPMENT FLOOR, HOUSES THE BOILERS AND THE CT'S...
This is enough for you to work on for now. This will work!
THAT'S THE OBJECTIVE... AFTER TALKING W/ THE VENDOR, I'M NOW WAITING FOR THE PROGRAMMER TO COME OUT NEXT WEEK AS I/WE DISCUSSED THE EXCHANGER SLOPE PROCESS AND THE CONCEPT OF MASS/TIME IN A WATER BASED SYSTEM. THE PROGRAMMER AFFIRMS THAT HE HAS NO CLUE WHAT THAT IS, AND AFTER THE MEETING TUESDAY, WITH THE OTHER 3 ENGINEERS FOR THE OTHER BUILDINGS, HE NOW KNOWS - FOR THE MOST PART - WHAT HE NEEDS TO DO W/THE PROGRAMMING LOGIC TO TAKE INTO CONSIDERATION THE AFOREMENTIONED.
So, that's where we stand right now... I appreciate the response, throw some more my way as you feel fit.... Thanks again,
Niko
Check your circuit breaker or fuses before calling for service - it could save you $$!!
|
| | chiller guy
| "Re(3):Cent. Chillers Water temp fluct's." , posted Thu 9 Jun 20:40
When you have some spare time on your hands (Ha) try monitoring the system in the 'manual" mode and then the "auto" mode and see what is different. Do the fans run longer/shorter times? Why does water temp drop closer to set point in manual mode? By watching the action of the components at given conditions you may be able to figure which part of the program is altering the control sequence.You should have gotten a printed sequence of operation for the system to help guide you. It will not provide the programming details but tell you what controls which.
You have several varing conditions here that have to fall in line (ie: the mass of the tower water, it's temp, the distance it travels, time it takes, set points, off sets, control loop speed and a couple more no doubt). When in manual you bypass all this "stuff" and just supply stop/start set points and adjust them a little for the "flywheel" effect of the water. That may be your best bet to keep the natives quiet until help arrives.
Good Luck and stay in touch
All refrigerants are safe - All refrigerant are dangerous. The difference is YOU !!!
|
|
|
