Can I interconnection on the high voltage side of the existing transformer?

F

Frank6172

Member
Location
Maryland
Occupation
Engineer
Hi all,

The site got 208V, 1200A utility service and there's an existing 208/120V to 480V step-up transformer to feed some 480V machine. Also there's a 480V, 400A AC disconnect on high voltage side of the transformer. We want to avoid 208V inverters as much as possible in this project so can we do the interconnection on the 480V disconnect of the transformer?

The transformer nameplate is below and our intended inverter is 480V, 120kW.
1747057338386.png
 
The questions you will need to answer:

1) What does the utility say about the interconnection?

2) What sort of grounding can your inverters tolerate?

It appears that the transformer is a delta primary wye secondary transformer being run in reverse (the wye _secondary_ is on the utility supply side). This means that the 480V side is either ungrounded or corner grounded, something that is a problem for some inverters.

The utility may have requirements for 'effective grounding' which may force a specific transformer and primary connection.
 
winnie said:
The questions you will need to answer:

1) What does the utility say about the interconnection?

2) What sort of grounding can your inverters tolerate?

It appears that the transformer is a delta primary wye secondary transformer being run in reverse (the wye _secondary_ is on the utility supply side). This means that the 480V side is either ungrounded or corner grounded, something that is a problem for some inverters.

The utility may have requirements for 'effective grounding' which may force a specific transformer and primary connection.
Click to expand...
We are using the SolarEdge 480V transformer which says supporting Delta: IT. Does it mean it's compatible with the high voltage side winding? Thanks, 1747063996907.png
 
zbang said:
You're also going to waste some power in the transformer; not a lot, but more than zero.
Click to expand...
Yes we understood that. If the the 480V side machine is pulling enough power we don't anticipate a lot of backfeed current will be pushed to the utility side.
 
Frank6172 said:
Thanks! Do you know how to verify if the transformer is ungrounded or corner grounded?
Click to expand...
That's all in the design of the wiring connected to the transformer. Is there an intentional connection to ground on at any of the transformer terminals or the disconnect after the transformer.

You said that the transformer is feeding some existing machines. Do you know anything about the 480V distribution system currently in use?
 
Frank6172 said:
Thanks! Do you know how to verify if the transformer is ungrounded or corner grounded?
Click to expand...
Two of the lines will be 277V to ground and one will be 0V to ground if it is corner grounded. If it is ungrounded all the lines will have undefined voltage to ground.
 
ggunn said:
Two of the lines will be 277V to ground and one will be 0V to ground if it is corner grounded. If it is ungrounded all the lines will have undefined voltage to ground.
Click to expand...

I think that would be 480V to ground.

If the system is corner grounded, then you will measure 480V to ground on 2 lines, and 0 on the 3rd line. However this will only tell you that there is a ground, not if it is intentional or a fault.

If the system is ungrounded, then the voltage to ground is poorly defined. Often it will be roughly 277V to ground on all phases. If you measure with a very low impedance meter, you might even measure 0V to ground on all phases. One clue that the system is ungrounded will be some sort of ground fault detection system.
 
winnie said:
I think that would be 480V to ground.

If the system is corner grounded, then you will measure 480V to ground on 2 lines, and 0 on the 3rd line. However this will only tell you that there is a ground, not if it is intentional or a fault.

If the system is ungrounded, then the voltage to ground is poorly defined. Often it will be roughly 277V to ground on all phases. If you measure with a very low impedance meter, you might even measure 0V to ground on all phases. One clue that the system is ungrounded will be some sort of ground fault detection system.
Click to expand...
Hi Winnie, we went back and measured out all Line-to-Line voltage which is about 482V and all Line-to-G voltage between 272-282V. Do you think this is a good indication that the high voltage side side of the transformer is ungrounded?
1747332849423.png
 
Frank6172 said:
Hi ggunn: We went back and measured out all Line-to-Line voltage which is about 482V and all Line-to-G voltage between 272V~282V. Do you think this is a good indication that the high voltage side side of the transformer is ungrounded?
Click to expand...
No, it's what the voltages should be if it were a grounded 480/277Y service, though it isn't absolute proof that's what it is. If the voltages phase to ground were all over the place it would be good evidence that it is ungrounded. Look for a N-G bond somewhere else, like maybe at the transformer secondary.
 
You have a delta coil, which means the _source_ only has 480V L-L available.

You measure L-G voltage that is approximately the L-N voltage. Remember that the 480V supply doesn't have a neutral.

That is pretty good evidence that you have an ungrounded system. The approximately correct neutral voltage is either because of balanced capacitive coupling or because there are ground fault detectors that are creating an incidental high impedance ground. (The old school light bulb ground detector happens to create a high impedance wye grounding.)

The only other possibility is that you have a _separate_ grounding transformer creating a wye grounded system, but this is unlikely. I suppose another possibility is that the 480V system has some other transformer creating a neutral.

You might look around to see if there are any ground fault detectors.
 
ggunn said:
The oldest NEC codebook I have is the 2008 version; the 250.21(B) requirement for ground detectors on ungrounded AC systems is in it.
Click to expand...
2008 isnt that long ago 😉. This Just recently came up in another thread, and I was surprised that the ground detector requirement came in in the 90s, IIRC. Keep in mind, also, some states like New York are typically 8 years behind on their code adoption. Perhaps that transformer has the date of manufacture on it which would help pinpoint whether the installation was compliant or not at the time.
 
You must log in or register to reply here.
Top