Solar transformer question

G

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
Scenario:

A client requested assistance from a consulting engineer on a project connecting 480/277V inverters to a 208/120V service. The engineer recommended using a 208V delta to 480/277V wye transformer. The client instead bought and installed a 208/120V wye to 480V delta transformer. The client is lucky because he is using SE inverters which can connect to a 480V delta service, and he does not want to replace the (large, expensive, and probably not returnable) transformer with the recommended one.

The client installed the transformer with the neutral connected back to the service from the transformer primary. The engineer advised the client to remove the neutral connection to the service, making the transformer essentially delta-delta. The client did so, but in the process the engineer saw that the client had also bonded the transformer primary neutral to ground, connecting the neutral and ground conductors between the service and the transformer in parallel, which is a clear code violation.

The question:

The client has removed the neutral connection between the service and the transformer primary, but the otherwise floating primary neutral is still connected to ground; is this a problem?

Aside:

The system was inspected and green tagged by the POCO, despite that the transformer is not the same as the one on the sealed planset, there are numerous discrepancies in wiring, and the code violation.
 
The utility service connection should be the standard N-G bond.
The 208Y primary side connection at the transformer should not have any connection to ground or neutral, the X0 terminal should be floating.

Your 480V side of the transformer will not have any reference to ground unless your inverters are connected in a wye.
You need to either corner ground the 480V side, which your inverters may not like, or you need to add a ground detection device/scheme.
If this is a standard transformer which they are running in reverse, you might even run into inrush problems when you try to energize it after a utility power blip.

I hope they saved lots of money by buying the wrong transformer.
 
jim dungar said:
The utility service connection should be the standard N-G bond.
The 208Y primary side connection at the transformer should not have any connection to ground or neutral, the X0 terminal should be floating.

Your 480V side of the transformer will not have any reference to ground unless your inverters are connected in a wye.
You need to either corner ground the 480V side, which your inverters may not like, or you need to add a ground detection device/scheme.
If this is a standard transformer which they are running in reverse, you might even run into inrush problems when you try to energize it after a utility power blip.

I hope they saved lots of money by buying the wrong transformer.
Click to expand...
The secondary side of the transformer is OK, or it will be once the client has added the ground detection that is mandated by 250.21(B). Luckily for the client, the SE inverters they bought can operate on a 480V ungrounded delta system. Most inverters cannot.

I agree that the wye point of the transformer should be left floating, but if it is grounded is it a code violation or safety hazard?

The client did not save any money that I know of by buying the wrong transformer; they just screwed up.
 
Last edited:
It's funny about the inspection, innit? Usually we are complaining about inspectors failing systems that should have passed. :D
 
The wye point of the transformer is going to be a 'derived neutral' likely injecting circulating current into the neutral and EGC conductors.

The POCO may have signed off on this and been happy for 'effective grounding' reasons. But in general it is an NEC no-no.

A zig-zag grounding transformer might be used to derive a neutral on the 480V side; you could ground that and have a 277/480V wye system for the inverters. I don't know all the code details for implementing this.
 
ggunn said:
...
I agree that the wye point of the transformer should be left floating, but if it is grounded is it a code violation or safety hazard?
Click to expand...
I believe it might be an objectionable current violation (250.6) even if it's not strictly a violation of something more specific like 250.24(B) (It was definitely in violation of the latter until the neutral connection was removed.)
 
winnie said:
The wye point of the transformer is going to be a 'derived neutral' likely injecting circulating current into the neutral and EGC conductors. ...
Click to expand...

He stated that the neutral was removed. So the circulating current wouldn't go there, but would take any equivalent path formed by the GEC, GES, EGC, and/or earth.
 
I'm not really 100% up to speed with all the grid connected solar stuff, so forgive my ignorance. But the way I understand the OP, 480/277 volts comes from the solar inverter and goes into the delta side of a transformer to be stepped down to 208Y/120 volts feeding the grid. So the wye point would be left floating. That makes sense to me.

But grounding the 480 side, Jim suggests making it corner grounded if the inverters will tolerate it. But my question is why wouldn't you just ground the neutral of the 480Y/277 to establish the ground path?
 
CoolWill said:
I'm not really 100% up to speed with all the grid connected solar stuff, so forgive my ignorance. But the way I understand the OP, 480/277 volts comes from the solar inverter and goes into the delta side of a transformer to be stepped down to 208Y/120 volts feeding the grid. So the wye point would be left floating. That makes sense to me.

But grounding the 480 side, Jim suggests making it corner grounded if the inverters will tolerate it. But my question is why wouldn't you just ground the neutral of the 480Y/277 to establish the ground path?
Click to expand...
Because the client erroneously bought and installed a 208/120V wye primary 480V delta secondary transformer instead of the 208V primary 480/277V wye secondary transformer specified on the sealed planset. Everything after that has been damage control.
 
ggunn said:
Because the client erroneously bought and installed a 208/120V wye primary 480V delta secondary transformer instead of the 208V primary 480/277V wye secondary transformer specified on the sealed planset. Everything after that has been damage control.
Click to expand...
I understand, but the inverters are outputting 480/277, as stated in the OP, correct? I get that you have a delta on the 480 side, but there is still a neutral coming from the inverters? Why can't that be grounded to establish a fault current path on that side?
 
CoolWill said:
I understand, but the inverters are outputting 480/277, as stated in the OP, correct? I get that you have a delta on the 480 side, but there is still a neutral coming from the inverters? Why can't that be grounded to establish a fault current path on that side?
Click to expand...
You can't do that on an ungrounded three phase service because the phase voltages are undefined relative to ground, but luckily for the client, the inverters he chose can connect to a 480V delta ungrounded service. Most (all?) others cannot; they may be able to connect without a neutral but the phase voltages must be referenced to ground. The only additional thing he needs to do to the 480V side is to add a ground detection system. The problems I have been talking about are all on the 208/120V side.
 
Last edited:
If the 208V side X0 terminal is connected to ground you will end up with circulating currents caused by having nit allowing the ptimary side voltages to be imbalanced. Along with creating an NEC violation, of objectionable currents, you can overload your transformer as it tries to maintain balanced line currents.

I say it regularly, "Do not 'ground a transformer primary side wye point unless you are a utility."
 
CoolWill said:
I understand, but the inverters are outputting 480/277, as stated in the OP, correct? I get that you have a delta on the 480 side, but there is still a neutral coming from the inverters? Why can't that be grounded to establish a fault current path on that side?
Click to expand...
The inverters can be configured to be connected to 480/277 wye-grounded or ungrounded 480 delta. They don't really output 480/277 if they're configured to interconnect to a delta.

The transformer does not have a wye point on the inverters' side, so grounded-wye is physically not an option.

The inverters won't like a corner-grounded delta.

So with the incorrectly purchased transformer, there is no grounded system on the inverter side that will function.
 
ggunn said:
I agree that the wye point of the transformer should be left floating, but if it is grounded is it a code violation or safety hazard?
Click to expand...
I have real world experience that yes that can be very bad. About 10 years ago I was doing some work on a building that had a 208Y->240 delta transformer that was feeding a bunch of HVAC units. There was a neutral ran and landed to the wye point of the primary. I decided to do a good deed and remove the neutral, however it was dark and in a bad spot with a bunch of crap around and I didnt notice there was one of those ribbon jumpers bonding the XO to the frame, so the XO was still ultimately bonded to the service neutral thru the EMT. Well a bit after that, one of the phase connections burnt off a the serving pole transformer. This caused such an imbalance and load on the transformer that one of the conductors faulted and blew a big hole in the EMT. I have a picture, Ill try to dig it up.
 
You must log in or register to reply here.
Top