HELP NEEDED FOR XFMR PROBLEM (Inrush?)

Status
Not open for further replies.
C

Captain Planet

Member
Location
South Carolina
Ok, so here's the back story, There is an existing setup in an electrical room in a plant I'm working on...400amp breaker at main switchgear, running 500mcm to a 225 kva XFMR (480 to 208 running in reverse) a 200amp fusible disconnect after the XFMR that powers the large air compressor for the plant.
So apparently everything has worked for more than a year...they recently had problems with the air compressor and the air compressor tech comes in to work on it and it has an error code in the compressor that says "over amperage" (on the variable drive...) Compressor tech cries bad transformer. No tripped breakers, no blown fuses, and manufacturer spec is 200amp fuse. He recorded voltage readings line to line and found 470,480,510. So before I got involved the transformer was changed to a different brand and type but same specs...once transformer was swapped they tried to turn the breaker on but it wouldn't reengage, so they swapped the breaker with a brand new one....
Then I come in. I took feeders off breaker and transformer and megged phase to ground and phase to phase 4,000+ megs on each so I'm good there. Then megged directly to conduit in case it wasnt grounded sufficiently still 4,000+megs...turned disconnect off on secondary. Megged everything on secondary side. Good there. Checked transformer wiring. Everything looks good to me there (it was tapped for 468, so I did change taps to provide 480) I tried to reengage breaker with no load on transformer and it tripped instantly twice. Since I was sure nothing was direct shorting I assumed I was dealing with Inrush (something I admittedly have a hard time understanding completely, given all the variables and misinformation)
so I cycled the breaker a 3rd time and I had an amprobe on 1 leg and the amp meter reading a different phase on switchgear (rest of plant is dead, so it's only reading our transformer setup) the breaker did not trip instantly, and instead for ~10 seconds there was transformer noise and both my amprobe (1,000 amp limit) and the amp meter (2,000amp limit) were maxed, I waited for Inrush to settle and feel as though I waited ample time, after the 10seconds I tripped the breaker manually, for fear it wasn't going to trip, and amp meters we're still maxxed.
So here is my question. What exactly is the diagnosis? It's been working for over a year with this setup and I know for a fact this plant has been through 3 outages in that year....So do I diagnose a faulty transformer? That's the only thing that's really changed...and That's what I believe....but my knowledge on inrush is limited, and I have no clue if Inrush could last that long under certain circumstances. And if it does is it normal? In my experience less than a second it's usually over, and if the breaker trips instantly, you cycle the breaker a few times hoping to hit it at the right phase angle....
Help me understand what I don't. Is it simple? Is it just a bad transformer? Or Am I missing something? Can inrush last such a long time? Where do I go from here?
 
In rush should be only a few cycles with a transformer. Motors on the other hand is dependent on rotor speed and it’s relation to impedance production.
 
208 input?
Well-balanced input voltages?
3 wires feeding primary, center tap floating?
Single core with three windings, or three independent transformers?
Transformer designed for step-up duty or 480->208 used in reverse?
I assume you already checked for obvious errors like a winding wired backwards and different voltage taps on different phases.
(a delta secondary can have large circulating currents if the phases aren't well matched)
 
1) Inform the plant folk that running a transformer in reverse causes a bunch of problems, with inrush, grounding, and input protection for VFDs. Making their current setup work is not a proper long term fix.

2) Inrush only lasts for a few cycles (unless you have something out of a lab, like superconducting coils), and can be many times the normal full load current. High current extending for 10 seconds is some other problem.

3) Megging the coils just detects faults to ground, and won't tell you if you have shorted turns or some other fault internal to the coils.

4) Usually, when a delta-wye transformer is connected in reverse (using the wye as the primary) you need to leave the neutral of that wye not connected. Connecting the neutral can cause huge circulating currents as the transformer tries to balance the supply voltages.

-Jon
 
190113-1017 EST

Captain Planet:

Paragraph your questions and comments.

Single phase transformer inrush current is mostly a 1 cycle event, and then dribbles off at a much lower level, taking many cycles. Transformer core material has a great affect on the peak current. The loading on the transformer at the time of last turn off has a substantial effect on the next turn on transient. For a single phase transformer see my plots at photos P6 and P7 of http://beta-a2.com/EE-photos.html . These are for an unloaded transformer.

As winnie mentioned you need to look at the delta circulating current.

.
 
drcampbell said:
208 input?
Well-balanced input voltages?
3 wires feeding primary, center tap floating?
Single core with three windings, or three independent transformers?
Transformer designed for step-up duty or 480->208 used in reverse?
I assume you already checked for obvious errors like a winding wired backwards and different voltage taps on different phases.
(a delta secondary can have large circulating currents if the phases aren't well matched)
Click to expand...

Yes 208 input.

Stable voltage (~214 steady)

Yes 3 wires feeding primary center floating.

Single core. 3 windings

It's a step down wired in reverse.

And yes windings wired correctly and all tapped same (for 480)
 
winnie said:
1) Inform the plant folk that running a transformer in reverse causes a bunch of problems, with inrush, grounding, and input protection for VFDs. Making their current setup work is not a proper long term fix.

2) Inrush only lasts for a few cycles (unless you have something out of a lab, like superconducting coils), and can be many times the normal full load current. High current extending for 10 seconds is some other problem.

3) Megging the coils just detects faults to ground, and won't tell you if you have shorted turns or some other fault internal to the coils.

4) Usually, when a delta-wye transformer is connected in reverse (using the wye as the primary) you need to leave the neutral of that wye not connected. Connecting the neutral can cause huge circulating currents as the transformer tries to balance the supply voltages.

-Jon
Click to expand...

1) I will definitely inform them of this.

2)that's great to know for sure. I was not entirely sure before

3) didn't think about the coils internally shorting

4)x0 was left untapped, so I assume we are good there
 
gar said:
190113-1017 EST

Captain Planet:

Paragraph your questions and comments.

Single phase transformer inrush current is mostly a 1 cycle event, and then dribbles off at a much lower level, taking many cycles. Transformer core material has a great affect on the peak current. The loading on the transformer at the time of last turn off has a substantial effect on the next turn on transient. For a single phase transformer see my plots at photos P6 and P7 of http://beta-a2.com/EE-photos.html . These are for an unloaded transformer.

As winnie mentioned you need to look at the delta circulating current.

.
Click to expand...

Sorry! First post on here, I'll be sure to paragraph properly from now on. Love this site, I've used it for years!
 
I'm a slow poster - could be all this has been covered
Here is what I translated:

  • System is a 225kva, 208/480 fed from the 208 side, powering a 480V compressor. Feeder is 400A CB. Secondary disconnect (480V) is 200A fused disconnect.
  • A/C faults on "Over Amperage". No tripped CB, no blown fuses.
  • Transformer is changed. 208 feeder CB is changed
  • I not sure exactly what was megged.
Transformer primary to ground (208V side) ?
Transformer secondary to ground (480V side) ?
Transformer primary conductors ?
Transformer secondary conductors (transformer to 200A disconnect) ?

I tried to reengage breaker with no load on transformer and it tripped instantly twice.
Click to expand...
So secondary disconnect (200A fused disconnect) was open?

so I cycled the breaker a 3rd time and I had an amprobe on 1 leg and the amp meter reading a different phase on switchgear (rest of plant is dead, so it's only reading our transformer setup) the breaker did not trip instantly, and instead for ~10 seconds there was transformer noise and both my amprobe (1,000 amp limit) and the amp meter (2,000amp limit) were maxed, I waited for Inrush to settle and feel as though I waited ample time, after the 10seconds I tripped the breaker manually, for fear it wasn't going to trip, and amp meters we're still maxxed.
Click to expand...

I'd say there are two problems:
The first one is with the compressor/drive, down stream of the 200A fused disconnect. Changing the transformer/CB didn't fix that.

Second issue is with the new transformer.

  1. Different brand/type that has different inrush characteristics,
  2. wired wrong,
  3. Not the same transformer.

Some notes on transformer inrush, xfm feeder design:
According to a GE paper, the inrush current for a back fed 208Y/480D transformer can be 37X the FLA. For a 225kva xfm, the 208V side FLA is 625A. Inrush current as much as 23KA. Yep, 23,000A. Expected duration is 100 milliseconds.

The transformer that should have been speced is 208D/480Y. The 208V side FLA is still 625A, however, the inrush is 11X FLA = 6700A.

Normal xfm feeder CB for a 225KVA, 208V primary is FLA x 1.25 = 625 x 1.25 = 781A, next size up, 800A.

I'm a bit surprised it ever worked with a 400A feeder CB.

Now for a bunch of wild guesses:
The old xfm was a 208D/480Y.
The new transformer is 208Y/480Y

The new transformer is wired wrong. Two instantaneous trips of the 400A CB is normal (what I would have expected) The CB holding to ~10 seconds, with the measured currents above 2000A, before being manually opened, is bad news (X 2). The inrush current won't last longer than 100 ms. And, the CB should have tripped as it did the first two times.

So, what to do?

  1. Check the old transformer nameplate against the new transformer nameplate. If the old one s a 208D/480Y and the new one is 208Y/480D. That would explain the high inrush, but not the sustained high current. If so, then .....
  2. There might not be anything wrong with the old CB or the old xfm.
  3. Check the transformer internal wiring and external wiring. As Winnie said look for a 208V WYE neutral being grounded. It could be even internal to the xfm.
  4. Borrow/ rent some additional test gear - might even have to rent a tech, in fact you may want to rent outside expertise. Check transformer turns ratio (both transformers). Check the CB contact resistance. Check the CB trip unit. (Both CBs) The new CB could be bad since it could have opened on instantaneous at 20KA+ twice.
  5. After you determine the cb is good, the transformer is the right one (hopefully a 208D/480Y, the xfm is wired correctly, then ...
  6. Get a Big CT, say something in the 5000:5 range, one CT on one 208V phase is sufficient. Just put a loop on the CT secondary. Put your amprobe on the loop, and set for peak hold.Open the 480V, 200A fused disconnect. Energize.

Check the xfm nameplates and post back - we can go on from there.

the worm
 

Attachments

  • XformerWhitepaper2.pdf
    68.1 KB · Views: 0
From past experience I'd put my money on:
Check the transformer internal wiring and external wiring. As Winnie said look for a 208V WYE neutral being grounded. It could be even internal to the xfm.
 
OP states no external connection to X0, but it is certainly possible that it is bonded to case/EGC internally.

Sent from my XT1585 using Tapatalk
 
iceworm said:
I'm a slow poster - could be all this has been covered
Here is what I translated:

  • System is a 225kva, 208/480 fed from the 208 side, powering a 480V compressor. Feeder is 400A CB. Secondary disconnect (480V) is 200A fused disconnect.
  • A/C faults on "Over Amperage". No tripped CB, no blown fuses.
  • Transformer is changed. 208 feeder CB is changed
  • I not sure exactly what was megged.
Transformer primary to ground (208V side) ?
Transformer secondary to ground (480V side) ?
Transformer primary conductors ?
Transformer secondary conductors (transformer to 200A disconnect) ?


So secondary disconnect (200A fused disconnect) was open?



I'd say there are two problems:
The first one is with the compressor/drive, down stream of the 200A fused disconnect. Changing the transformer/CB didn't fix that.

Second issue is with the new transformer.

  1. Different brand/type that has different inrush characteristics,
  2. wired wrong,
  3. Not the same transformer.

Some notes on transformer inrush, xfm feeder design:
According to a GE paper, the inrush current for a back fed 208Y/480D transformer can be 37X the FLA. For a 225kva xfm, the 208V side FLA is 625A. Inrush current as much as 23KA. Yep, 23,000A. Expected duration is 100 milliseconds.

The transformer that should have been speced is 208D/480Y. The 208V side FLA is still 625A, however, the inrush is 11X FLA = 6700A.

Normal xfm feeder CB for a 225KVA, 208V primary is FLA x 1.25 = 625 x 1.25 = 781A, next size up, 800A.

I'm a bit surprised it ever worked with a 400A feeder CB.

Now for a bunch of wild guesses:
The old xfm was a 208D/480Y.
The new transformer is 208Y/480Y

The new transformer is wired wrong. Two instantaneous trips of the 400A CB is normal (what I would have expected) The CB holding to ~10 seconds, with the measured currents above 2000A, before being manually opened, is bad news (X 2). The inrush current won't last longer than 100 ms. And, the CB should have tripped as it did the first two times.

So, what to do?

  1. Check the old transformer nameplate against the new transformer nameplate. If the old one s a 208D/480Y and the new one is 208Y/480D. That would explain the high inrush, but not the sustained high current. If so, then .....
  2. There might not be anything wrong with the old CB or the old xfm.
  3. Check the transformer internal wiring and external wiring. As Winnie said look for a 208V WYE neutral being grounded. It could be even internal to the xfm.
  4. Borrow/ rent some additional test gear - might even have to rent a tech, in fact you may want to rent outside expertise. Check transformer turns ratio (both transformers). Check the CB contact resistance. Check the CB trip unit. (Both CBs) The new CB could be bad since it could have opened on instantaneous at 20KA+ twice.
  5. After you determine the cb is good, the transformer is the right one (hopefully a 208D/480Y, the xfm is wired correctly, then ...
  6. Get a Big CT, say something in the 5000:5 range, one CT on one 208V phase is sufficient. Just put a loop on the CT secondary. Put your amprobe on the loop, and set for peak hold.Open the 480V, 200A fused disconnect. Energize.

Check the xfm nameplates and post back - we can go on from there.

the worm
Click to expand...

Will do. I will post back in morning. Thanks so much everyone!!!
 
GoldDigger said:
OP states no external connection to X0, but it is certainly possible that it is bonded to case/EGC internally.

Sent from my XT1585 using Tapatalk
Click to expand...

Have you even seen an internally bonded X0 on a standard 225kva transformer?
 
Status
Not open for further replies.
Top