Voltage Neutral to Grnd - Transformer WYE-Grnd 120/208 step up to WYE-Grnd 480/277

Status
Not open for further replies.
E

electron_ranger

Member
Location
Lilburn, GA
Hello,
I'm installing a 100kW PV array on a building with a 120/208 service. It's less expensive to use inverters with a 480/277 output and a transformer than it is to use 208v inverters. So we energized the system today and all the voltage and resistance readings on the primary (building/208 side) of the transformer are correct. However, I got some strange voltage readings on the secondary (solar/480 side): Phase to neutral is 277v, phase to phase is 480v, Phase to ground is 180V, neutral to ground is 15v. I've attached an image of the line diagram.

The transformer doesn't say that it's a neutral isolating type, but I think it may be based on the voltage readings. If I understand correctly, an isolating transformer is treated as a separately derived system, which would call for a new neutral-ground bond on the secondary side of the transformer.

SLD.jpg
 
Unless it is designed as an autotransformer or shows a factory jumper between H0 and X0, you should assume that a transformer is an isolation transformer and that you *must* bond the neutral on the inverter side.


Sent from my XT1585 using Tapatalk
 
electron_ranger said:
Hello,
...However, I got some strange voltage readings on the secondary (solar/480 side): Phase to neutral is 277v, phase to phase is 480v, Phase to ground is 180V, neutral to ground is 15v. I've attached an image of the line diagram.
Click to expand...

I cannot read any details on your picture; too low resolution so [guessing now on]...

I assume NOT an autoxfmr or 480v side would act just like the 208v neutral to ground side. But not of concern.

I again assume the PV panels are somehow grounded - their frames or something else. I cannot tell from your pix how that PV inverter works, but we can be sure there are high freq, high voltage, high power switching IGBTs firing in it.

Now when we have a MOTOR on one side of the VFD inverter (same concept as yours), the 3 phases create one heck of a high voltage CAPACITIVE coupled voltage from each phase to ground (motor case). If one leaves the motor case ungrounded, and measures voltage case to ground, it is typical to see - ready for this? 180v or so... Ground the case and it is gone - generally it is pretty low current.

So here you are with a high freq switching inverter at 480v and you are no doubt getting the same capacitively coupled voltage to your PV system ground.

[\guessing now off]

If I were you I would read the inverter manual again, and if still unclear, call the manufacturer ASAP and ask - if you should tie the 480V neutral to ground! I cannot imagine this not being a requirement- for safety as well as life of the system; let it float like that and every lightening strike within 20 miles of the system will likely be slowly killing your IGBTs, a few molecules at a time, until the hit 2 miles away blows the inverter off the wall since it does not have a ground reference.
 
Last edited:
Mike, sorry for not giving you all the facts, your guessing was a reasonable thought process though. Solar inverter technology no longer uses modified sine wave technology of the 90's and early 2000's. All of the major brands today use capacitors and electronics to precisely match the voltage, frequency, and phase of the grid. And because they use electronic/solid-state inversion, the power factor can be adjusted in either direction. So adding solar to a building with a lot of motor loads could also provide additional cost savings by programming it to have capacitive reactance. And these definitely weren't inducing current onto the ground because the inverters we're all off when I checked the voltage readings from the transformer.

Golddigger, what you said does make sense, its the only thing that would make sense as far as I can tell. I'm going to confirm this with Eaton (transformer manu.) engineers, just to be sure.

Thanks guys!

:eek:
mike_kilroy said:
If I were you I would read the inverter manual again, and if still unclear, call the manufacturer ASAP and ask - if you should tie the 480V neutral to ground! I cannot imagine this not being a requirement- for safety as well as life of the system; let it float like that and every lightening strike within 20 miles of the system will likely be slowly killing your IGBTs, a few molecules at a time, until the hit 2 miles away blows the inverter off the wall since it does not have a ground reference.
Click to expand...
:jawdrop:
 
Yeah it seems pretty clear to me that the 480 side is not grounded and is just floating so you are getting some capacitive coupling thus the odd to ground readings. You should be able to tell by looking if there is a N-G bond or not. Is there a seperate HO and XO terminal?

Mike, I admit I don't share your belief that you can so reliably generalize about lightning and that grounding the system will significantly reduce the risk.
 
There is not a N-G bond, nor does the transformer have a removable buss or any labels indicating an optional N-G bond. HO and XO terminals are separate with no physical connection, and the HO and XO transformer conductors go into the transformer coils.

electrofelon said:
Yeah it seems pretty clear to me that the 480 side is not grounded and is just floating so you are getting some capacitive coupling thus the odd to ground readings. You should be able to tell by looking if there is a N-G bond or not. Is there a seperate HO and XO terminal?
Click to expand...
 
electron_ranger said:
There is not a N-G bond, nor does the transformer have a removable buss or any labels indicating an optional N-G bond. HO and XO terminals are separate with no physical connection, and the HO and XO transformer conductors go into the transformer coils.
Click to expand...
That settles it then.
You can either bond H0 to X0 and make it non-SDS or keep them separate and bond the center tap on the inverter side to the GES or a local ground depending on physical layout.

Sent from my XT1585 using Tapatalk
 
GoldDigger said:
That settles it then.
You can either bond H0 to X0 and make it non-SDS or keep them separate and bond the center tap on the inverter side to the GES or a local ground depending on physical layout.

Sent from my XT1585 using Tapatalk
Click to expand...

yep
your H side has no H0-G bonding provision
you must make one

a yg-yg is unstable under some conditions but your xfmr has a tertiary winding
in mining we require by law a tertiary winding to make this xmfr config
 
GoldDigger said:
That settles it then.
You can either bond H0 to X0 and make it non-SDS or keep them separate and bond the center tap on the inverter side to the GES or a local ground depending on physical layout.

Sent from my XT1585 using Tapatalk
Click to expand...

Wonderful. Thank you for helping to confirm this. What does non-SDS stand for?
 
Greetings,

I have an installation with a related issue.

Is a PV installation where the System is 208V and im using a Y-Y transformer for getting 480VY for the solar PV inverters.

The issue is that when i connect the 480VY/208VY transformer i get a circulating current from the neutral of this transformer to the neutral of the distributión transformer of the client 12470Y/208Y. Is there a particulary reason for this? What could be a solution?

Please help me with this topic.
 
abrahamespinal9 said:
Greetings,

I have an installation with a related issue.

Is a PV installation where the System is 208V and im using a Y-Y transformer for getting 480VY for the solar PV inverters.

The issue is that when i connect the 480VY/208VY transformer i get a circulating current from the neutral of this transformer to the neutral of the distributión transformer of the client 12470Y/208Y. Is there a particulary reason for this? What could be a solution?

Please help me with this topic.
Click to expand...

Do not connect the Y-point on the 208V side to neutral. Treat the 208V side as a delta connection (no neutral). You're getting circulating currents because the 208V side of the transformer is trying to balance out any voltage differences between the 208V phases.

The Y-point on the 480V side must be grounded to ensure proper phase-to-ground voltages on the 480V side.



SceneryDriver
 
LarryFine said:
I thought that was only for Y-D transformers. :unsure:
Click to expand...

OP should leave the Y-point on the 208V side floating, otherwise, he will get the circulating current described as the trafo attempts to force all three phases into voltage balance. In particularly bad scenarios, you can smoke the transformer and/or trip OCPD. As described, the 208V (grid side) of the transformer doesn't need (and shouldn't have) a neutral to operate. Like I said, he should treat the 208V side as a delta winding be leaving the Y-point floating.


SceneryDriver
 
Status
Not open for further replies.
Top