Solar wye/wye transformer

Status
Not open for further replies.
Why is it needed? Well, there were some 3 phase inverters in the past that used it as a load-carrying conductor. I remember one 3 phase inverter that was connected in a WYE configuration and would turn on the phase inverters separately to match the inverter load to the PV array output. I don't know of any inverters today that do that, maybe some large central inverters still do it. That's the only load carrying situation I have seen, the other uses have to do with voltage detection. If you want to detect loss of phase voltage you have to be able to measure the line to neutral voltage. There used to be a couple of inverters that would allow a 3 wire and EGC connection as long as it was to a grounded WYE system. They just used the EGC as a substitute for the neutral and did the phase voltage measurement from line to EGC.
Now, connection to an ungrounded delta service is pretty new for most inverters. No neutral is needed there, no loss of phase detection, just monitoring the L-L voltages.
Say what you want but I believe most inverters still "want" a neutral connection. See this thread:


Granted it is 4 years old, but I doubt much has changed. Bottom line is most inverters STILL require/recommend/suggest a neutral connection when it doesn't seem to be absolutely required.
 

electro7

Senior Member
Location
Northern CA, US
Occupation
Electrician, Solar and Electrical Contractor
Yes, that is the one I was working on. I thought it was the mounting bolts hitting the case and there was continuity there. But even after I removed those I still had continuity from XO to the case ground. I couldnt see anywhere the XO would make connection with the case. I guess when I say case I mean enclosure. Is the case the enclosure? Im getting continuity between XO and the enclosure.

Sent from my SM-N960U using Tapatalk
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Say what you want but I believe most inverters still "want" a neutral connection. See this thread:


Granted it is 4 years old, but I doubt much has changed. Bottom line is most inverters STILL require/recommend/suggest a neutral connection when it doesn't seem to be absolutely required.
This is for three phase inverters, just to be clear. Single phase inverters don't need 'em.
 

pv_n00b

Senior Member
Location
CA, USA
Say what you want but I believe most inverters still "want" a neutral connection. See this thread:


Granted it is 4 years old, but I doubt much has changed. Bottom line is most inverters STILL require/recommend/suggest a neutral connection when it doesn't seem to be absolutely required.
In the last four years most major string inverter manufacturers have provided an inverter option that can connect to a delta or ungrounded WYE transformer. We use the all the time.
 

pv_n00b

Senior Member
Location
CA, USA
What does "HV is Grid-Side Voltage" mean after "Suitable for Bi-Directional Use"? Is it a problem that LV is going to be used for the grid-side?

Cheers, Wayne
Transformers are typically designed to limit inrush current on the supply side where they will be energized on startup. If a transformer is connected with the design load side facing the source that will energize it then the inrush can be significantly higher than expected. Bi-directional design simply means it is designed to be energized from the utility side and not the inverter side, but power flow will normally be from the inverter to the utility.
 

pv_n00b

Senior Member
Location
CA, USA
And I can't help but think some/most/all of this "effective grounding" stuff is hogwash or voodoo. The reason I say this is I have done multi megawatt PV systems for Georgia power that had no such requirements, but then other utilities (such as national grid in NY and new England) where they require zig zag transformers for "effective grounding" on all their utility scale PV systems. Does national grid really know something that Georgia power doesn't??? Is the luminiferous aether in Georgia different than it is in NY???
There is a lot of question about effective grounding as applied to electronic static inverters sources. Effective grounding works well on rotating sources like rotating generators. But generally, it's the only thing utility engineers have to go by so they wanted to apply the long history of limiting overvoltage on distribution systems supplied by rotating generators to PV inverters. Square peg meet round hole. But try to talk a utility engineer out of it when it's all they have known for the last 40 years.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Transformers are typically designed to limit inrush current on the supply side where they will be energized on startup. If a transformer is connected with the design load side facing the source that will energize it then the inrush can be significantly higher than expected. Bi-directional design simply means it is designed to be energized from the utility side and not the inverter side, but power flow will normally be from the inverter to the utility.
Actually, in general usage, a stipulation by the manufacturer that the transformer can be used with either side as primary specifically addresses the inrush on connection. The two directions may not have identical inrush percentates, but both should be tolerable.
The only time bidirectional (rather than suitable for reverse operation) might have that meaning you suggest is in the case of a transformer specifically branded for solar PV use.
A "standard" transformer not rated for reverse installation will still be perfectly suitable for use with a grid active inverter as long as the primary (inrush winding) is facing the grid.
 

pv_n00b

Senior Member
Location
CA, USA
Actually, in general usage, a stipulation by the manufacturer that the transformer can be used with either side as primary specifically addresses the inrush on connection. The two directions may not have identical inrush percentates, but both should be tolerable.
The only time bidirectional (rather than suitable for reverse operation) might have that meaning you suggest is in the case of a transformer specifically branded for solar PV use.
A "standard" transformer not rated for reverse installation will still be perfectly suitable for use with a grid active inverter as long as the primary (inrush winding) is facing the grid.
If the manufacturer stipulates bidirectional power flow. Before PV developed a higher demand for these types of transformers this was not a typical design.
This applies in 95% of the cases. There are some smaller dry type transformers (under about 25kVAR or so) that are designed with a nonlinear voltage response that can't be back fed without causing problems.
 

pv_n00b

Senior Member
Location
CA, USA
The SMA, Solectria, and Chint 3 phase 480V inverters I have used recently all show the neutral as optional.
There is a very fine point here that people need to understand. There are inverters that use the EGC in place of a neutral if the neutral is not run to the inverter (SMA for instance) but they require a grounded WYE service. The EGC connected to the neutral at the service entrance allows the inverter to still monitor the phase to neutral voltage by monitoring the phase to EGC voltage at the inverter.
Then there are inverters that don't need any neutral reference and can be connected to a delta service. Do not confuse the two types of inverters even though both say they don't need a neutral. If an inverter that uses the EGC for phase to neutral voltage monitoring is connected to a delta it will probably run but will have intermittent shutdowns as the phantom phase to neutral voltage bounces around.
 

electro7

Senior Member
Location
Northern CA, US
Occupation
Electrician, Solar and Electrical Contractor
The tranaformer manufacturer again recommended I leave the XO neutral disconnected and said the imbalanced current on the high side would not be significant enough to cause problems. He said leaving the nuetral from the service connected on the XO terminal could possibly cause problems because of the nuetral ground bond at the service. Thats what they have seen in other installations.

Sent from my SM-N960U using Tapatalk
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
The transformer manufacturer again recommended I leave the XO neutral disconnected and said the imbalanced current on the high side would not be significant enough to cause problems. He said leaving the nuetral from the service connected on the XO terminal could possibly cause problems because of the nuetral ground bond at the service. Thats what they have seen in other installations.
Unless I'm missing something, the transformer manufacturer's recommendations are a violation of the NEC. With a wye-wye transformer, the neutral terminal on the grid side wye must not be bonded to the case. If the transformer you have can't do that, it's not suitable for your application.

Cheers, Wayne
 

pv_n00b

Senior Member
Location
CA, USA
The tranaformer manufacturer again recommended I leave the XO neutral disconnected and said the imbalanced current on the high side would not be significant enough to cause problems. He said leaving the nuetral from the service connected on the XO terminal could possibly cause problems because of the nuetral ground bond at the service. Thats what they have seen in other installations.

Sent from my SM-N960U using Tapatalk
The X0 should be connected to the utility neutral but not connected to ground. Maybe that's where the confusion is coming from, those are two different connections. The NEC does not require that the X0 be connected to the utility but you are going to have problems with voltage stability if it is not, particularly if your inverters monitor the line to neutral voltage. X0 can't be connected to ground if it's connected to the service on the load side of the existing service disconnect.
On the 480V side since this is a separately derived LV system H0 has to be grounded. There's no requirement that you run a neutral from the 480V side if you are not using it for anything, but it has to be grounded, that's an NEC requirement.
On a side note, the fact that the transformer came with the X0 bonded to ground would indicate that it was designed to be energized from the 480V side. Check with the manufacturer to find out what the inrush will be if energized from the 208V side. It might be much higher than you think.
 

electro7

Senior Member
Location
Northern CA, US
Occupation
Electrician, Solar and Electrical Contractor
They are SMA Core 1 inverters. Do you know if they use line to neutral for voltage stability?

The nuetral from the utility is ran to the transformer but not landed on XO, which then the nuetral is not bonded at the transformer. Hammond said they have seen problems with the loop when the XO nuetral is landed at the transfomer going back to the service through the EGC and connected to the HO, since the HO is bonded at the transformer.

I do have a GE and GEC at the transformer bonded to the enclosure and the HO.

I do have a nuetral conductor ran to the inverters on the high side.

Sent from my SM-N960U using Tapatalk
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
The neutral from the utility is ran to the transformer but not landed on XO, which then the nuetral is not bonded at the transformer. Hammond said they have seen problems with the loop when the XO nuetral is landed at the transfomer going back to the service through the EGC and connected to the HO, since the HO is bonded at the transformer.
As long as X0 is not bonded to the transformer case or the EGC, there is no loop. So again, (1) 208V neutral to X0 and (2) no bond of X0 to case or EGC. If (2) is not possible, you have the wrong transformer.

Cheers, Wayne
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
As I mentioned in my first post on this subject, the reason you _may_ need to wire the neutral to X0 is that single phase current on the secondary (H) side must have a corresponding path on the primary (X) side.

If you don't have a neutral connected to the X0, then there must be some sort of magnetic coupling from each secondary phase individually to all of the primary phases. A tertiary delta provides this coupling, but the transformer core itself can also provide the necessary coupling. This is why you might not need anything connected to X0.

IMHO if the manufacturer knowing the installation conditions recommends not connecting X0, then I would follow the manufacturer. I would simply keep in my back pocket the knowledge that you might need X0 connected.

In no case should X0 of this transformer be connected to the case directly. (repeating what others have said)

-Jon
 

electro7

Senior Member
Location
Northern CA, US
Occupation
Electrician, Solar and Electrical Contractor
Got it Jon, thank you! Yes I explained it in detail to Hammond.

I think the only question left is if the SMA Core 1 inverters need the neutral for voltage stability? It sounds like according to pvn00b's post #72 they do not and can use the EGC for that.

So then would I disconnect the nuetral conductor running to the inverters on the HO side?

Code:
 As long as X0 is not bonded to the transformer case or the EGC, there is no loop. So again, (1) 208V neutral to X0 and (2) no bond of X0 to case or EGC. If (2) is not possible, you have the wrong transformer.

these are made and have a label stating "solar duty". And the EE approved them (so at least I have that to fall back on). I'm 99.5% sure they are okay to use.

Sent from my SM-N960U using Tapatalk
 
Status
Not open for further replies.
Top