I did not say solidly grounded.
Wye Inverter on a Delta Service
- Thread starter Rock-Noll
- Start date
ggunn
PE (Electrical), NABCEP certified
- Location
- Austin, TX, USA
- Occupation
- Consulting Electrical Engineer - Photovoltaic Systems
Perhaps "capacitively referenced to ground" would be a better way of putting it?
It's a hack because the EGC is not the neutral and the inverter is required to measure line to neutral voltage. The NEC is pretty careful about making sure people don't use an EGC as a neutral, because it was not uncommon in decades past and a hazard. Technically, the inverter should have a neutral run to it, even if a reduced size neutral since it's not current carrying. In the way back days, inverters used to require a neutral to be run, then some bright bulb at an inverter company looked at a single line and thought, you know the neutral is in parallel with the EGC. We can tell installers they don't have to run a neutral if they connect a jumper between the ground and neutral terminals in the inverter and they will like that and buy more of our inverters. Of course once one manufacturer allowed this as an option then the others quickly followed.
ggunn
PE (Electrical), NABCEP certified
- Location
- Austin, TX, USA
- Occupation
- Consulting Electrical Engineer - Photovoltaic Systems
Well, I guess that is one way to look at it, but it's a distinction without a difference. The NEC is made up of rules drawn up to prevent fires and injuries, and this method of voltage reference does not increase the probability of either.
electrofelon
Senior Member
- Location
- Cherry Valley NY, Seattle, WA
Well we are going to have to disagree. Having to run a neutral used to drive me absolutely crazy and. I was so glad when that ridiculousness stopped. And tell me again, what is it about the green dye in the insulation that the electrons and voltage measurement circuit don't like?
ggunn
PE (Electrical), NABCEP certified
- Location
- Austin, TX, USA
- Occupation
- Consulting Electrical Engineer - Photovoltaic Systems
Red electrons are attracted by the insulation and green electrons are repelled?
Pulling a neutral was not just annoying but a huge waste of resources, I wish that bright bulb figured out a way to do it without a ground reference at all but ill take what I can get. I always thought it was strange how VFD's (basically a type of inverter) never needed a neutral and those PV inverters did. VFD's are not without issue on delta banks, probably due to odd ground reference. I mean these things are basically computers now right ? They should be able to detect every oddball North American transformer configuration there ever was in 8 bits of memory.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
Surely the issue is that UL 1741 on grid-interactive inverters requires the inverter to confirm the grid configuration for both L-L and L-N voltages before the inverter can export power, so that's not possible.
Cheers, Wayne
GGunn, thanks for explaining that, ungrounded Wye now makes sense. Just the side that we (the customer or building electrical contractor) are seeing is ungrounded, but it is grounded at the service, right?
But I am afraid that is not what I am dealing with. The POCO meter tech was out there today, so I was able to look into the CT cabinet, no ground run from the utility transformer, just 6 sets of 3 phase conductors. He said it’s an open Delta service, ungrounded on any phase, and it looks that way to me. Only grounds are the groundING conductor bonding the steel raceway and enclosure. The utility (WE Energies) is slowly swapping factories over to Wye; they paid the customer side of a service rewire for a whole industrial park a few towns over. At the industrial park where I’ve been working on this job, they are slowly doing the same, but customers are footing the bill for new CT cabinets, wiring etc. I asked if he’d open the transformer for me, but said he didn’t have time, had to get to another job.
I got all 4 inverters working today (had communication and roof issues so failed last week,) they all show ~ 281 volts to ground, 490 phase to phase. But then turned them back off until we get inspection. Hopefully that inspection is next Monday, so am thinking I’ll just keep quiet about this til the inspection, discreetly confirm the service type with the utility engineer (the meter tech didn’t seem 100% sure) then if we need to install a Wye transformer it will be a “personal” issue. The other thing is that I’ve had them all working for a few hours now, so am tempted just to let them run and monitor the voltage (can do remotely on the inverter monitoring.)
I have one picture attached here of the groundING, going to follow up with a few posts of other pictures. I'm getting "to large of a file" message.
Thanks for your advice!
But I am afraid that is not what I am dealing with. The POCO meter tech was out there today, so I was able to look into the CT cabinet, no ground run from the utility transformer, just 6 sets of 3 phase conductors. He said it’s an open Delta service, ungrounded on any phase, and it looks that way to me. Only grounds are the groundING conductor bonding the steel raceway and enclosure. The utility (WE Energies) is slowly swapping factories over to Wye; they paid the customer side of a service rewire for a whole industrial park a few towns over. At the industrial park where I’ve been working on this job, they are slowly doing the same, but customers are footing the bill for new CT cabinets, wiring etc. I asked if he’d open the transformer for me, but said he didn’t have time, had to get to another job.
I got all 4 inverters working today (had communication and roof issues so failed last week,) they all show ~ 281 volts to ground, 490 phase to phase. But then turned them back off until we get inspection. Hopefully that inspection is next Monday, so am thinking I’ll just keep quiet about this til the inspection, discreetly confirm the service type with the utility engineer (the meter tech didn’t seem 100% sure) then if we need to install a Wye transformer it will be a “personal” issue. The other thing is that I’ve had them all working for a few hours now, so am tempted just to let them run and monitor the voltage (can do remotely on the inverter monitoring.)
I have one picture attached here of the groundING, going to follow up with a few posts of other pictures. I'm getting "to large of a file" message.
Thanks for your advice!
Attachments
ggunn
PE (Electrical), NABCEP certified
- Location
- Austin, TX, USA
- Occupation
- Consulting Electrical Engineer - Photovoltaic Systems
A bit about terminology - if the service is grounded anywhere it is grounded everywhere, in the sense that the phase voltages are referenced to ground. It either is or it isn't. If the utility transformer is wye on the secondary side, the neutral is almost assuredly grounded at the transformer even if the neutral conductors are not run out of itr, which makes it a grounded wye. I hope that is what you have, but without looking at the transformer terminals, you can't tell; there certainly aren't any neutral conductors in those enclosures.
It is a pad mounted transformer; is it not? Someone else in here may know better than I, but although open delta services are indeed usually wye, I didn't think pad mounted transformers are ever open delta; all the open delta transformers I have seen have been pole mounted, and I cannot imagine having six sets of conductors coming from a pole mounted transformer. I get that that is what the tech told you but it just doesn't sound right to me. Anybody?
BTW, when he told you that it is ungrounded on any phase, that just means that it isn't corner grounded; that is something else entirely from what "grounded" means with respect to a wye service.
I realize that the horse has left the barn, but before you design an inverter into a PV system, you MUST know what the service configuration is. I don't mean to sound harsh, but it sounds to me that you are in a bit over your head. My advice to you is to turn off the inverters until you have this definitively sorted out; if the service is not a (grounded) wye you may be risking damaging the inverters, and if that happens SMA won't honor the warranty.
That's my free advice and worth every penny you paid for it.
- Location
- Wisconsin
- Occupation
- PE (Retired) - Power Systems
WE Energies specifies their padmount transformers as having tri-plex cores rather than a common core. I have no doubt they could supply an open-delta padmount transformer. Ungrounded delta services are not uncommon throughout Wisconsin but are not normal for new installations.
The three simple ground indicator lamps look like this is an old existing service.
These are probably not 480V bulbs but are likely transformer or resistor style. These should continue to function even if one lamp is burned out.
The three simple ground indicator lamps look like this is an old existing service.
These are probably not 480V bulbs but are likely transformer or resistor style. These should continue to function even if one lamp is burned out.
Thanks, I'll be sorting this out with the engineer. I appreciate your advice about leaving them off until I figure out exactly what is going on. I don't want us to buy 4 new inverters in addition to transformers!
I have definitely learned my lesson on clarifying the exact nature of the service at the beginning of the job, and over design changes! Early on, we contemplated using a smaller SMA's inverter (?30ish kW) which is fine with an open Delta!
I have definitely learned my lesson on clarifying the exact nature of the service at the beginning of the job, and over design changes! Early on, we contemplated using a smaller SMA's inverter (?30ish kW) which is fine with an open Delta!
It's just the intended purpose. EGCs are for bonding to equalize potentials and to carry fault current in grounded systems. It's not that they can't do 5 other things, it's not that the EGC would not work as a current carrying conductor in place of a neutral. It's that it's not intended to be used for anything else. I've had to spend a lot of time explaining to people why on a single line diagram we have several conductors that are running in parallel that look like they could all do the same thing, carry electrons. Each has an intended purpose in the NEC.
If this is going to turn into a, why do we follow the dumb NEC, discussion please don't.
There are a very few inverters that can operate on an ungrounded service. It's got to be hard to design one that passes UL 1741 loss of phase detection or I'm sure there would be more. Taking ungrounded services, corner grounded deltas, and split phase deltas off the table probably does not reduce 3ph inverter sales all that much.
Last edited:
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I strongly agree with the policy of not using the EGC as a circuit conductor, and not intentionally injecting current into the EGC.
However the EGC must always carry some current, as a result of capacitive coupling and insulation leakage.
IMHO, using the EGC as a voltage reference is not unreasonable, as long as the current coupled to the EGC by this use is small compared to the normal operating current expected on the EGC.
However the EGC must always carry some current, as a result of capacitive coupling and insulation leakage.
IMHO, using the EGC as a voltage reference is not unreasonable, as long as the current coupled to the EGC by this use is small compared to the normal operating current expected on the EGC.
Clearly this is an example of a ungrounded 480V service, one of many many thousands of ungrounded systems in use. Hopefully the inverter manufacturers take notice that its not just 120/240, 208/120 and 480/277 out there. An inverter is supposed to be a 'smart' thing it should figure out that there is no neutral or L-N voltage reference.
Well having to install costly transformers just because a manufacturer & UL can't match the service voltage is unacceptable. I hope the manufacturers can smell the coffee and work with the utilities to correct UL 1741 as we can clearly see its a enormous waste of resources to have idle transformer losses hours per day or any time the sun don't shine on all the thousands of solar systems that could be on some of the best roofs for solar, industrial roofs.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
OK, well the first question is if there are inverters commercially available for ungrounded 480V deltas? If there are, then the issue is just that the OP selected the wrong inverter.
If there aren't, then you have a point. So the next question would be whether UL 1741 permits such an inverter. If so, the issue is that inverter manufacturers aren't filling this market niche; take it up with them. If not, then the issue is with U 1741, and it should be expanded to cover that possibility.
Unless, of course, there is some real safety or performance issue that means a grid-interactive inverter will not work as safely or as well on an ungrounded 480V delta as it will on a 480V wye.
Cheers, Wayne