GENSET TRIPPED WHEN BROUGHT ON SECOND COMPRESSOR

[TONY2023]

Hi All, I have a question. We have a 2.5MW, 13.8KV genset to provide power to two compressors. Each compressor is rated for 600HP, 460V. There is a 1500KW step down transformer between genset and compressors. The genset started up and provided power to first compressor without any issue or tripping. As they brought up the second compressor, genset got tripped in 20 seconds with negative KVAR. I did the calculation based on the above information, everything looks okay. The genset has enough capacity to handle the two compressors load.

I would like to get any feedback on how or what to check to solve this issue?

Thank you,
Tony

 

[crshears]

Hmmm . . . more information required.

"Enough capacity to handle the handle the two compressors' load" implies once steady state operation is reached; is there extra margin available for the compressor start-up process? Is the compressor start direct on line / across the line, or soft starter, or an older reduced voltage autotransformer start, or what?

Is this a new installation [think warranty], or a new problem with an existing installation?

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[electricpete]

Some more questions

[ol 1]
[li]Is the system operated isolated from the grid?[/li]
[li]Does negative KVAR mean vars into the genset? (as if the excitation was too low).[/li]
[li]Do the protective relays indicate the "type" of trip? (maybe negative KVAR the type of trip? Or time overcurrent?, Or?)[/li]
[li]Is there any information about current or voltage prior to the trip?[/li]
[li]Is there indication of whether the 2nd machine came up to speed?[/li]
[/ol]

 

[waross]

Starting a second 600 HP motor with one already running is marginal on a 2.5 MW set.
Your KVA during starting may well be over 2.5 MW?
Did you mean 2.5 MVA?
Gen sets are normally rated in KVA or MVA rather than MW. (But not always)
Does it take the compressor 20 seconds to come up to speed?
You may have reversed CTs.

Starting on a gen set.
Your compressors may be drawing about 500 KVA running and 3000 KVA starting.Total 3.5 MVA
A 2.5 MW set will be capable of 3.15 MVA continuous.
But that 3000 KVA starting may be about 1500 kW and the balance will be KVARs.
The prime mover has the power to supply the required kW, and the AVR will supply the excitation needed to generate the KVARs or MVARs.
There will be a momentary current overload but well within the thermal capacity of the generator end.

I suspect a protection setting error or reversed CTs.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[LittleInch]

What do you mean by 1500kW transformer?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[crshears]

Hopefully 1500 kVA was intended . . .

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[TONY2023]

There was only one genset with rating 2.5KW with unity pf 1.0 or 3125KVA with pf 0.8. No grid power
Please see attached picture when one compressor was running. As they turned on the second compressor toward the end, EMCP tripped genset. Is it normal to get a negative KVAR on first compressor? My coworker said it is normal because of induction motor. Is it worth a try to swap the CT wiring?
Waross, would you please share the calculation on how you got 3000KVA for starting a compressor?

 

[TONY2023]

Sorry, I meant 1500KW or 1875KVA step down transformer.

 

[jraef]

1200HP, 1875kVA transformer... The 2nd compressor is likely putting the transformer in saturation.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[electricpete]

> The 2nd compressor is likely putting the transformer in saturation.

I didn't understand that. Can you explain how/why that would happen.

For op, the meter shows
Total KW 642
Total KVA 677
Total KVAr -215

To me the +642KW tells us the reference direction for real power flow is leaving the generator.
If we apply the same reference direction to reactive power flow, then we'd say there is reactive power flowing IN to the generator (as if there was a capacitor or other var source connected to the terminals).

That's a head scratcher to me if I understood it correctly. Two thoughts
[ul]
[li]Longshot: Are there power factor correction capacitors? (maybe way too much correction was used)[/li]
[li]I wonder if there might be an miswiring in the instrumentation. Do you know what the meter read when only one motor was running (positive or negative kvar)? [/li]
[/ul]

EDIT - I see Bill said something similar to the 2nd one. That's probably much more likely.

 

[waross]

600 HP = 450 kW
450 kW at combined efficiency and PF (0.90) = 500 KVA
Starting KVA = 6 times running KVA = 3000
With your reading of 677 KVA that would be 4062 Starting KVA.
For starting on a generator, use three times running KVA = 2031 plus 677 KVA for the running compressor = 2708 total KVA set capacity to start the second compressor.
If there are no other loads on the gen sets, you may be able to start with 2.5 times instead of 3 times.
At 2.5 times you may need to make provisions to prevent the controls from dropping out on low voltage.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[waross]

By the way, if you use sizing software from Cat or Onan, they will tell you that you need a factor greater than 3:1
However, 3:1 is generally acceptable.
There may be noticeable voltage dip.
I have found that most customers will accept the voltage dip when told the cost of a larger set.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[waross]

KVARs, positive or negative?
There tends to be miscommunication and confusion when looking at VARs and generators.
You won't see capacitors putting VARs into a generator during motor starting.

Much of the starting current is reactive and may be supplied by capacitors.
However, enough capacitor VARs to cancel starting reactive current is much to much for a running motor.
In very old text books you may find a reduced current starting scheme that used capacitors to supply the reactive current.
The capacitors were cut out in steps as the motor accelerated.
While the scheme does work, it is not economically feasible and never came into common use.
With that in mind, power factor correction capacitors will never overcome the starting reactive current.

The transformer may act as a de-facto series impedance reduced voltage starter.

The overload with the significantly reactive starting current will cause a voltage drop in the transformer.
For illustration, imagine a transformer with 3% regulation and 5% mpedance voltage.
Subjected with a 200% mostly resistive load, the voltage drop will be 200% x 3% or 6% voltage drop.
Subjected with a 200% mostly reactive load, the voltage drop will be 200% x 5% or 10% voltage drop.
Motor starting will be more than the regulation but less than the %imp
And the overload will be more than 200%.
Saturation depends on Volts per Hertz.
If the starting load pulls the frequency down, the UFRO feature of the AVR will prevent V/Hz saturation.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[wcaseyharman]

A couple of thoughts:
- are the motors equipped with power factor correction capacitors?
-are the motors fed from really long underground cables?
- what tripped the GEN set? The gen set controller or a separate relay?

The normal convention for generators is positive power is power out, positive VARs is vars out of the machine, so it’s surprising to see -kvar feeding a motor as others have noted. Hence my question about PF correction. My experience with smaller gensets like this has been the generators have very limited var absorption. If that truly is -vars (leading power factor, machine absorbing vars) then I would look at the system to see where all the capacitance is coming from.

 

[crshears]

This question remains unanswered: Is this a new installation [think warranty], or a new problem with an existing installation?

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[waross]

As VARs are imported into a generator rise, the voltage tends to rise.
When that happens, the AVR cuts back the excitation.
Eventually the imported VARs are providing all of the excitation and a further increase in imported VARs causes the voltage to rise.
As I mentioned, due the the reactive nature of the starting current, imported VARs are unlikely during motor starting.
Are your motors over corrected for PF?


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[TONY2023]

Hi All,

The genset EMCP displayed a negative KVARS when first compressor turned on. EMCP would shut down the genset when second compressor turned on. So, I decided to beef up the Reactive power set point from 10% to 30% on EMCP pe generator capability curve. We tested it again, and the genset was able to handle both compressors without issue or tripping. I know this was a temporary fix.

We also have another two gensets were running with load banks for unity power factor and 0.8. There was no issue. However, when we tried to parallel both gensets to bus with other bigger gensets. The EMCP shut down my gensets with Negative KVARS or reverse KVARS. So, we decided to increase reactive power set points from 10% to 30%. My gensets were able to parallel with other gensets without being shutdown. However, my gensets still displayed negative KVARS. I suspected it was the CT wiring issue to EMCP. So, I switched CT wiring X1 to X2 to change its polarity. The gensets ran and displayed with positive KVARS and negative KW. So, I rewired CT wiring to its original wiring.

Note, all loads were inductive load, and there wasn't any power factor correction capacitor. I know there were two VFDs, but I don't think they generated that much capacitive.

Does anyone have any more thought why gensets displayed Negative KVARS or consuming KVARS? I know this issue became more complicated to solve.

 

[edison123]

If your compressor motors are synchronous motors, they could be providing the leading Vars.

Muthu

http://www.edison.co.in

 

[iop95]

What VFD models?
There are input and/or output VFD filters?

 

[electricpete]

You mentioned that you swapped polarity on one ct and the reactive and real power both changed sign. It doesn't convince me there is not a wiring error. I assume there 3 current and 3 voltage inputs? Can you tell us anything about how the voltage inputs are derived and wired? (ideally a drawing showing what is the pt primary and secondary configuration and is the pt secondary signals phase to phase or phase to neutral from the secondary)