Chiller overvoltage problem

[Ivan001]

Hello,
I have a problem with chiller. The chiller is connected on 400V AC (+-10%) substation. But, at the end of transmission line, on my chiller terminals, I have 410V, my chiller has Un=390V, so the chiller has often breakdown, probably the electronics are so sensitive...do you have any idea, how I can reuglate this voltage to 390V? The classic transformer sounds like not good idea, do you have any idea or you think that transformer 410/390V should be enough?

 

[Asymptotic]

Is the chiller line voltage tolerance also +/-10%. If so, 410V is within specification.
390V +/-10% is a range of 451 to 429 volts.

What sort of breakdown? Motor winding failure, or a failure of control electronics?

What is the chiller KVA and/or full load current rating?
How far is the chiller from the substation?
Have you measured how low the line voltage dips during a chiller startup?

If the problem is caused by marginally high line voltage, and the substation transformer has variable taps (most larger ones do; often in several 2.5% increments) the easiest thing to do may be to switch the substation transformer tap changer to one setting lower. -2.5% of 410V is 399.8V.
Note: Transformer must be de-energized. A qualified electrician must perform this adjustment.

However, my suspicion is something other than line voltage alone is implicated.

 

[Ivan001]

Is the chiller line voltage tolerance also +/-10%. If so, 410V is within specification.
390V +/-10% is a range of 451 to 429 volts.

What sort of breakdown? Motor winding failure, or a failure of control electronics?

What is the chiller KVA and/or full load current rating?
How far is the chiller from the substation?
Have you measured how low the line voltage dips during a chiller startup?

If the problem is caused by marginally high line voltage, and the substation transformer has variable taps (most larger ones do; often in several 2.5% increments) the easiest thing to do may be to switch the substation transformer tap changer to one setting lower. -2.5% of 410V is 399.8V.
Note: Transformer must be de-energized. A qualified electrician must perform this adjustment.

However, my suspicion is something other than line voltage alone is implicated.

Thanks for reply. The chiller doesn't have tollerance +-10%, his voltage is fixed to 390V. Chiller has 15kVA. The substation is near, so I think that voltage drop isn't enough for chiller. The tap change in substation isn't allowed. Is there any possibilities to regulate this voltage with some device?

 

[DaveE]

If the chiller has any US, Canadian, or European safety certifications, then the +/-10% (ok +6% for some European standards) voltage range is implied. Also true if it is a good design. No one should (does?) design this sort of equipment without some tolerance for input voltage variations, since that is what you will find at real world installations. How designers communicate this on the ratings label can be confusing, or non-existent.
I suspect that the issue is more likely transient voltage spikes, particularly since you said it was at the end of a transmission line. Perhaps you could connect a monitor to it?
Insulation failures (arcs) are more often due to high voltage transients. Overheating is more common with slightly high voltages.

 

[Ivan001]

If the chiller has any US, Canadian, or European safety certifications, then the +/-10% (ok +6% for some European standards) voltage range is implied. Also true if it is a good design. No one should (does?) design this sort of equipment without some tolerance for input voltage variations, since that is what you will find at real world installations. How designers communicate this on the ratings label can be confusing, or non-existent.
I suspect that the issue is more likely transient voltage spikes, particularly since you said it was at the end of a transmission line. Perhaps you could connect a monitor to it?
Insulation failures (arcs) are more often due to high voltage transients. Overheating is more common with slightly high voltages.

Any idea how to regulate this spikes or slighty high voltages?

 

[Asymptotic]

Thanks for reply. The chiller doesn't have tollerance +-10%, his voltage is fixed to 390V. Chiller has 15kVA. The substation is near, so I think that voltage drop isn't enough for chiller. The tap change in substation isn't allowed. Is there any possibilities to regulate this voltage with some device?

@DaveE is right. Everything has a tolerance . If it isn't on the chiller data plate it should be in the operations manual, and at worst the manufacturer can be called for this information.

A more thorough investigation is necessary. "I think that voltage drop isn't enough" is similar to answers I've received from less experienced colleagues, and I'll tell you what I've told them, "Stop thinking, measure!".

Voltage sag during start-up probably has nothing to do with these failures (especially in this case: 25 KVA isn't a very large chiller, and ought not be capable of causing sags large enough to matter) but substituting preconceived ideas of what is happening in place of close observation and actual measurement doesn't further the troubleshooting process. One tends to either chase ghosts down one garden path, or ignore fruitful lines of inquiry on another by thinking "it must be X", when in actuality "X" has little or nothing to do with the failure.

But to answer your question, yes, it is possible to regulate and condition line voltage.
Search on CVT (Constant Voltage Transformer) and AVR (Automatic Voltage Regulator). There are several ways to go about automatic voltage regulation, and Staco, one of the manufacturers that specialize in this equipment, has a good overview. Follow the 'Product Brochure|Application and Engineering Data' menu path at:
http://www.stacoenergy.com/products...-mechanical/stacoavr#literature_documentation

All approaches to AVR are relatively costly, come with disadvantages of their own, and don't solve all voltage-related problems. If 410V is indeed too high for this 390V rated chiller and overvoltage is causing the failures, then a properly rated buck-boost transformer wired to "buck" would be one of the less expensive fixes.

What is failing (control electronics? power electronics? motor windings?) and how (overheating? winding insulation punch-through?). Answering these questions will point you towards a solution.