Brownouts from our chiller

[StradivariusBone]

So a while back our building had a new chiller installed. Since then we've had periodic brownouts on the order of a couple to several per hour. We did a lot of troubleshooting and determined that the outside building transformer was fine, our county electricians put a really neat Fluke meter on our building mains to chart the voltage drops and found that they were occurring. We have a sister building that also got the same chiller upgrade but does not have the brownouts. It's enough to occasionally reboot a computer or sound board, which are all now on UPS. It seems to coincide with the compressor kicking on at the chiller. I've watched the lights dim at the exact moment the compressor motor starts up and the chiller shows a persistent alarm for the second compressor having a low current. They claimed it was a flow sensor switch that was faulty and they replaced that part, but I'm given to wonder if they were unsure of the definition of current.

I've heard of issues with loose lugs causing problems like this when a load is applied, but I'm getting ready to push the county to work on it again. Has anyone had a similar experience? Sorry if this is better suited in the electric forum. It seems to me to be more of a facility issue rather than one of lighting and electrics.

 

[SteveB]

Sounds like they undersized the feeders on the compressor system. Likely as well miscalculated the existing load on the system. Possibly the utility company never upgraded the feeds on the street and they just went with what existed ?. Our local provider is Con-Ed and they routinely underfeed supplies all over the city, is why the buried feeders are always failing catastrophically and blowing up. This might be a problem getting corrected as utility companies hate being told they are wrong.

 

[StradivariusBone]

Well the whole chiller plant is fed from the same main panel that our building is fed from. As far as I know they didn't replace the feeder when they swapped it out, but we never had this issue with the older chiller plant. I'll mention the feeder though. It's possible that it might be damaged in some way, we have had issues with our underground conduits and lightning damage causing all sorts of electrical gremlins throughout the years. A nearby school with an almost identical building got the same chiller and has not had similar issues.

 

[BillConnerFASTC]

I feel like you need an electrical engineer on your side.

 

[Ben Stiegler]

Can you compare the specs on old vs. new chiller to understand if the load has changed? If not, may point to an installer issue .. loose lug, etc. if new chiller draws more, even if it appears to still be under the panel limit, then the search scope expands ... upstream resistance at a splice, breaker, etc. or degraded conductor.

Not to be confused with an inebriated conductor, who caused me to lose interest in my college orchestra.
Good luck and keep us posted!

 

[Dionysus]

Have you had your Power Factor checked not just on the unit itself but overall?

 

[StradivariusBone]

I can't speak for the old chiller but the new one has info here-

I know a little about AC motors and how the current draw will be greater when a motor starts from the inrush of current and then stabilize as it gets up to speed. The brownouts are happening as compressor A turns on. From what I was googling, the LRA is the locked rotor amps where that current draw would be highest. It's interesting that the two compressors have different ratings, I figured they'd be matched and duty cycled, but maybe the additional one is only activated at peak use and hence why it's a smaller system? Perhaps the low current alarm on B is when that second kicks in as the other is already running and there's not enough current available, but it's quick enough where it won't trip the breaker? What I'm curious about is why there are two numbers for the LRA. 930 amps is more than the building's main switch is rated for. Granted, that inrush would be like a second I'd wager, but that's a lot of current!

Anyway, the catch is there is an identical-ish building with an identical chiller that does not have this problem about 30 minutes up the road from me. So I'm inclined to believe that it's something unique to my building and not necessarily a mismatched chiller plant, though I'm prepared to be wrong of course.

 

[StradivariusBone]

Have you had your Power Factor checked not just on the unit itself but overall?

Maybe during the study? The main disconnect has a display that is indicating around .85. It would jump to 1 randomly and float up to .90. I don't know how accurate that is, but power factor has to do with how the phases are loaded right? I think I watched an ElectroBoom video on that. Or maybe Practical Engineering.

 

[Dionysus]

Power Factor has to do with the difference between Resistive, Capacitive and Inductive loads. This effects the waveform and the lag on both each phase and all together.
Between .90 and 1 are great. The further from 1 the more out of sync the phases are with what would be expected. When Power Factor is consistent its not a problem as well, but large instantaneous swings can cause brownouts (and other problems). The lower the power factor also the more you will be billed for your electricity.

 

[MNicolai]

They claimed it was a flow sensor switch that was faulty and they replaced that part, but I'm given to wonder if they were unsure of the definition of current.

Happened to see this thread while on the phone with my director, a mechanical engineer who works on CEP projects regularly. "Sound like they didn't account for inrush. A flow sensor switch wouldn't do anything like that."

The 2 numbers for LRA have to do with the starter. That nameplate looks like it's from Carrier, so in their specs it usually shows up at LRA Star, LRA Delta. For everyone else, LRA "Star" means "Wye". The motor starts up in a wye configuration, at about 1/3 of the LRA in a delta configuration, and then one the motor is almost up to speed it transitions over to the delta configuration. This is known as "Wye Start, Delta Run", and reduces the amount of inrush you encounter.

Trane's documentation is a little better, so here's a snippet from their guides.

Ultimately you will likely need an electrical engineer to investigate. It's probably cheapest (free) if you can get the electrical engineer who did the upgrade involved. Failing that, you will have to go third party.

 

[Ben Stiegler]

Power Factor has to do with the difference between Resistive, Capacitive and Inductive loads. This effects the waveform and the lag on both each phase and all together.
Between .90 and 1 are great. The further from 1 the more out of sync the phases are with what would be expected. When Power Factor is consistent its not a problem as well, but large instantaneous swings can cause brownouts (and other problems). The lower the power factor also the more you will be billed for your electricity.

can you explain the billing impact, pleae?

 

[BillConnerFASTC]

Ultimately you will likely need an electrical engineer to investigate.

Yeah - what a good idea.

 

[tjrobb]

Most likely, undersized conductors. Loose lugs are possible, as is a bad capacitor. You'll need an EE most likely. Or, maybe, both circuits are starting at a cooling call instead of being staged. There are other, much strange, options but those are freak occurrences.

LRA only happens if the rotor stalls. 3-phase motors are self-starting so the breaker doesn't see that current unless something goes sideways. Which is what the breaker is for anyway.

 

[MNicolai]

Few questions:

Are the brownouts always independent of the chiller? If not, then it could be a utility supply problem.

Typically you can request the utility to come out and do a power quality analysis and they will provide the results to you.

Who is your utility? In talking with my colleague, most of our CEP electrical issues come from co-ops. They may have an underlying problem in their distribution that is being exacerbated by the load from the chiller.

For that matter, is your other building without issues on a different utility or a fundamentally different part of the power grid in your area?

 

[StradivariusBone]

So far as I know it's only when the compressor motor starts up. We already had the utility meter their side and they said nothing is wrong.

Here's the video I was thinking of that did some explanation of power factor-

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[StradivariusBone]

Who is your utility? In talking with my colleague, most of our CEP electrical issues come from co-ops. They may have an underlying problem in their distribution that is being exacerbated by the load from the chiller.

For that matter, is your other building without issues on a different utility or a fundamentally different part of the power grid in your area?

Sorry missed the last two-

FPL is the utility and the other building is a 30 minute drive away and on a barrier island. Very isolated from us.

 

[avkid]

This sounds so much like undersized feeder.
The utility has no problem on their side and EC will never admit it.
You need to get a EE firm with a logging power quality analysis system.

 

[StradivariusBone]

So when you say undersized feeder, are you talking about the cabling from the panel to the chiller, the cable from the outside transformer to the main panel, or the cable from the pole to the outside transformer? The county did put a logging Fluke meter at the main disconnect for about 2 weeks and noted the voltage drops. They also metered our sister building with the same setup. FPL metered at the outside transformer as well and reported no issue.

I'm friends with the head electrician and he is determined to help solve it, but it's been a back burner issue.

 

[avkid]

It could be at any of those points, first 2 are most likely.

 

[SteveB]

My first thought was feeder undersized. This was due to its a new chiller, possibly requiring a different feeder capacity then previous, did the engineer/architect do the math and require a new feed capacity ?, or try to save some money and go with existing.

My building had to redo the load on a linked section between 2 buildings, one of which was getting replaced. They specified a new 600A panel to be added to the remaining building, I reminded them that the remaining building had undersized feeders and should get looked at. That service entrance had been fed with about 2400 amp capacity on a panel stated at 3000 and they were now adding 600. Oops.... they replaced the feeders.