Neutral current on an inverter with a delta transformer 2

[rockman7892]

I've recently been looking at a solar design which has an inverter interface to the utility system with the AC output of the inverter utilizing a Delta-Delta steup up transformer to output the inverter voltage at 480V. The instruction manual mentions that for a 4-wire utility system a neutral can be connected inside the inverter (4th wire) however it would only be used for neutral sensing. The inverter ties into the utility through a reverse fed feeder breaker in an upstream panel.

Obviouly with a delta connected transformer there is no neutral current flow but I'm curious if there were loads connected in the panel that had a neutral or were connected L-G would this mean that all neutral current would have to return to the utility source and none would return to the inverter? Is there an issue with this if all unbalanced current returns to utiliy although some of the supply is coming from the inverter?

Also how does this effect ground fault detection and clearing on the AC system. Does this mean that all ground fault current would have to flow to utility source and none would flow back to inverter? Again is there an issues with this as questioned above?

Thanks for all the help.

 

[davidbeach]

Wouldn't be acceptable on my system, where all customer owned generation must present itself to the system as an effectively grounded source. Just wouldn't even consider it.

 

[waross]

Think of your inverter as three single phase inverters. Any neutral current must originate from one or more of the inverters and will return to the inverters.
If you have line to neutral loads that exceed the output of the respective phase inverters you will have the equivalent of a floating neutral with both positive and negative loads. The inverters may not like the over-voltages that they may be subjected to.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rockman7892]

Thanks Guys!

Davidbeach

Why would you not allow this configuration? Simply because it is an ungrounded system that you would not want to interface with an existing system? Is this simply for ground fault reasons?

What about corner grounding the sytem? Or should i insist that a wye secondary is supplied in the inverter.

Waross

I'm not following exactly what you are trying to say. If i think of inverter as (3) single phase inverters how would any neutral current return to each of the inverters if there is no neutral point to connect to? Can you explain how this overvotlage condition would exist?

 

[waross]

Think about what happens if you use a delta winding to feed unbalanced, wye connected loads. If the neutral is for sensing only, connecting line to neutral loads to it may have interesting and possibly expensive unintended consequences.
Think "Open neutral".

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rockman7892]

Waross

If I am correct when you have a delta winding to feed an unbalanced wye connected load then the wye point of the load will not be at zero potential and will float to a voltage determined by the unbalanced loads. This floating wye point can cause higher than rated voltage across the load terminals.

If you connect L-N loads with no netural return path and you have an open neutral then the voltage across the load will be relatively small and the voltage buildup between the sensing neutral wire in the inverter and the inverter phase output will be high possibly resulting in damage inside the inverter? Am I understanding correctly?

 

[waross]

Correct rockman.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rockman7892]

Thanks waross!

How could the presence of voltage on this netural sensing wire insdie the inverter cause damage to the inverter itself?

For instance lets say that a single phase load, or unbalanced wye load in the feild caused a voltage to appear on the neutral sensing wire with respect to ground (not sure what a realistinc value might be). How could the presence of this voltage casue damage to the inverter? Wont the inverter components be rated for at least the inverer output rating of 480V L-L or 277V L-G?

It appears that this neutral sensing wire may be bonded to ground inside the inverter. If this is the case then would this voltage be present on the entire AC and DC grounding system (grounding systems bonded together) in reference to remote earth?

Thanks again for your help!

 

[waross]

Hi rockman. Without seeing the schematic???

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

Yes, a shematic or more info might help. It really doesn't make sense to cross a delta-delta transformer with a neutral wire. What would it be sensing? A bonded neutral should show no voltage. A floating neutral could indicate a ground fault on the 480V system but what exactly would a ground on the 480V system mean to the inverter? You also wrote step-up transformer which implies the inverter portion is not capable of handling 480/277V levels.

 

[controlsdude]

If I remember right the MOVs in a vfd are grounded on one side for a wye connection. I think there is a tech note warning in the vfd manuals if you do a delta connection about breaking this link. Otherwise I think you would explode the MOVs. Its true for some manufacturers, but not for others. Just something to know before you apply power and the smoke releases.

 

[rockman7892]

Found out a little more information about connections but I will try to post a schematic as well.

I am told that the neutral that would be used for sensing only would be bonded to ground inside the inverter. Esentially it is bonded to both the EGC from the inverter as well as the grounding electrode to the inverter. This AC ground system is also bonded to the DC system.

I am told that inverter will not operate in an island mode, so it will never directly serve L-N or unbalanced 3-phase wye loads. Also I am told that in the case where it is serving unbalanced loads with the utility that the inverter will shut down when it detects a certain amount of unbalanced current on its ouputs.

When I asked about how the inverter would react to a ground fault I am told that the inverter will shut down in 1/4 of a cycle based on voltage sensing during the ground fault. I was told however that the inverter used L-L voltage sensing for control. For a ground fault doesn't the L-N voltage drop to zero and not the L-L voltage? How could the L-L voltage be used to sense a ground fault?

I guess also that for the 1/4 during a ground fault all fault current would have to return to the utility source?

I'll try to get a good schematic.

 

[waross]

For a ground fault on a delta system the faulted line voltage to ground/neutral drops to zero. The line to neutral voltage on the other lines rises to line to line voltage.
The voltage across line to neutral connected inverters on the unfaulted phases will rise to line to line voltage.
If the inverters are delta connected there will be no change in the voltage across the inverters.
Ground fault currents on a delta system. It may be that your ground fault current may be very low. possibly a few amps or even measured in milliamps.
A ground fault may be limited by the capability of the inverter feeding a fault. If the inverter is delta connected, the current may be limited by the impedance of two legs of the artificial neutral circuit and may possibly be less than an amp.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

Having the neutral brought to the drive and then bonded to ground make the whole connection rather pointless. I think you could also argue that would be against code. I always understood the neutral and ground were not to be bonded at end devices in the power system. By this, I mean you bond the neutral at the incoming service but never bond the neutral at a device fed by the service.

I wouldn't think the inverter will supply any ground fault current due to the delta-delta transformer connection. But then, hte L-N system voltage would likely collapse so it might source some current due to that. Still, even if it did, it would shut-down rather quickly due to the over-current protection. I believe it would also just shut-down if the line-line voltages go out of the allowed operating range.