Output Transformer connection for PV inverter
We have a 100KW, 480V, PV inverter that we need to load side tap to an existing 208V, 3 phase switchboard. We specced a 480V wye: 208V delta transformer to be wired between the inverter and the 3 pole breaker used as our PV tap point in the existing swithcboard. The electrician ordered a 480V Delta:208 Wye transformer.
Can we use the ordered transformer? The inverter does not require a neutral connection and the inverter manufacturer says the delta configured transformer is acceptable to them.
We are not sure if we will have voltage imbalances with this configuration. Also we're not sure of how to handle the GEC connections.
The short answer is yes. This is typically the setup I would use for your situation. NEC 250.20(B) does not require that a 480V 3ph 3W circuit be grounded.
The DC GEC from the inverter is routed to the premises grounding electrode system. The EGC will run from the inverter to the transformer ground bus.
I'm getting some push back from our 3rd party PE reviewer. He feels a short to ground on the inverter (480V) side of the transformer will not trip any overcurrent protection. He is requesting we use the 480V Wye primary: 208V delta secondary transformer. Is there a way to persuade him the 480V delta primary: 208V Wye secondary is safe besides quoting NEC250.20(B)?
He is correct, a ground fault will probably not cause an OCPD trip, but then again a ground fault on a WYE will not cause an OCPD trip either in this circuit configuration. If you did use a WYE on the 480V side exactly what OCPD would you be tripping on a ground fault? The only OCPD on the 480V side is probably in the inverter and since it is part of a listed assembly it has to be installed in accordance with the listing and the installation manual, which probably allows an ungrounded Delta connection.
I would start by asking your 3rd party engineer exactly what OCPD would be tripped by using a WYE on the 480V side and how it would be tripped. Once they untie themselves from that knot they will probably be more open to using the transformer you have. I'm going to guess that they are just used to having everything grounded but they don't really know why it would not be required in this situation.
If the engineer says something about the OCPD on the 208V side tripping due to a ground fault on the 480V side you can point them to NEC 240.21(C)(1) which does not allow secondary conductors to be protected by primary side OCPD in a WYE-Delta transformer.
Last edited by Marvin_Hamon; 01-04-12 at 12:33 PM.
Thanks for your valuable input. I was trying to wrap my head around what the fault conditions would be given a ground fault on the inverter side of the transformer. In all likelihood the inverter would trip off but you still have the service as a source. Is personnel safety an issue with a faulted cable between the inverter and transformer? I don't think so. The cable would fault to the conduit that is bonded. The inverter is grounded through the DC GEC to the service grounding electrode system. There is an equipment ground between the inverter and transformer. There shouldn't be a significant potential difference if someone touched the conduit and case, right?
Whenever you work with PV systems you have at least two possible electrical sources, the inverter and the utility. The utility is the one with the huge available fault current. The available fault current from a static inverter is only 2 to 3 times the rated current for half a cycle or so. When designing fault protection we look mainly at the elephant and not so much the mouse.
For a fault on the 480V side, the inverter will probably trip off line fairly quickly due to the voltage drop on the faulted leg and now you have to look to the utility side to see what is protecting you from the utility fault current. If you are installing the transformer secondary feeders in accordance with NEC 240.12(C) then there really is no protection for the conductors, and according to the NEC that's acceptable if you go by 240.12(C). Like the tap rules, NEC 240.12(C) allows the installation of conductors that would not be considered to be protected from fault current because they are installed in such a way that a fault is less likely, a fault will have limited ability to cause collateral damage, and there is a readily accessible way to disconnect the faulted conductor.
If it were me I would be asking the engineer two things, where does my design not conform to the NEC and if I changed the design, exactly how would the protection operate during a fault that is different from my design. Assuming you are complying with the NEC requirements for transformers I don't see how the engineer can provide answers to these two questions that would require you to use a different transformer. As I said before I have done many designs that use an ungrounded delta transformer on the inverter side. So far no one has questioned this circuit topology.
Change all my references to 240.12 to 240.21.
Thank you again! The code does not specify in 240.21(C) the transformer configuration, only the proper way to provide overcurrent protection. And our design is per code. I don't know if I will win over the engineer as it comes down to the mentality that we have always done it this way. Having two sources requires letting go of that mindset.
That being said is there any advantage to Wyeelta or Delta:Wye configurations when connecting to an inverter that doesn't require a neutral? With a loadside PV tap, connecting to a circuit breaker, I can see a slight advantage of the delta toward the breaker. But with a new service, connecting to a utility transformer, should the Wye be toward the utility transformer? We would carry the service neutral to the transformer and terminate the transformer GEC back at the service transformer.
There are reasons why you might want to use a WYE-Delta transformer and not a WYE-WYE or Delta-Delta but they have to do with distribution system voltage problems that are pretty advanced. There are a number of distribution system design books that cover transformer configurations pretty well.
This is an issue with MV ready inverters that have to be connected ungrounded to MV transformers. In that instance using a WYE connected transformer with a grounded neutral on the inverter side would not be allowed, and these are done all the time. We really have this "everything must be grounded" mindset in the US and once we get outside of that people get very uncomfortable. But sometimes that is the only way things work and it is allowed in the NEC.
Could you please recommend a couple of sources that cover the proper selection of transformer configurations.
Originally Posted by Marvin_Hamon