Neutral to Ground bonding jumper for Commercial system

[solarken]

I am finishing up an installation at a commercial site that has 3-phase 480V service, but with a 480 to 208Y120 transformer, so the PV system has 208Y120 Inverters. There is a 60A 3-pole disconnect for the system right near the transformer, and I connected the line side of the disconnect to the secondary lugs of the 480 to 208 transformer.
After I powered down the primary feed to the transformer and made the connections on the lugs, I did not see the neutral to ground bond inside the transformer that was shown on the single line dwg given to me by the customer. I have attached the single line to this post, with a red arrow showing the grounded neutral at the interconnection location. The PV disconnect is not shown on this dwg, but it is connected as I stated to the secondary lugs, with 6AWG copper line and neutral conductors and an 8AWG supply side bonding jumper to ground lug in the transformer.
I investigated while I had the power down, by measuring the resistance from secondary neutral lug to ground in the transformer with the two parallel neutral conductors that feed an existing 3-phase panelboard disconnected, and it was open, not grounded. With the neutrals from the panelboard connected, the measurement was 0.2 ohms. I opened the Eaton panelboard and did not see where/how the neutral was grounded but it must be grounded in that panelboard somewhere. GThe panelboard was pretty packed and I didn't have a lot of time.
My question is, do I need to adjust the bonding in the disconnect and install the neutral to ground bonding screw, given that the actual existing system appears to have the neutral to ground bonding jumper installed in a panelboard that is fed by the transformer rather than inside the transformer where I connected the PV system?

 

[ggunn]

My question is, do I need to adjust the bonding in the disconnect and install the neutral to ground bonding screw, given that the actual existing system appears to have the neutral to ground bonding jumper installed in a panelboard that is fed by the transformer rather than inside the transformer where I connected the PV system?

The neutral to ground bond can either be at the transformer secondary or in the fused PV AC disconnect, but not both. It cannot be in a panelboard on the other side of the disco from the transformer.

 

[jaggedben]

You can put a separate system bonding jumper in your PV disconnect. See 250.28(D)(3). It would be prudent to truly confirm the system bonding jumper in the existing panel.

Alternatively you could remove the bonding in the existing panel and install a single System Bonding Jumper at the transformer itself.

 

[solarken]

You can put a separate system bonding jumper in your PV disconnect. See 250.28(D)(3). It would be prudent to truly confirm the system bonding jumper in the existing panel.

Alternatively you could remove the bonding in the existing panel and install a single System Bonding Jumper at the transformer itself.

The neutral to ground bond can either be at the transformer secondary or in the fused PV AC disconnect, but not both. It cannot be in a panelboard on the other side of the disco from the transformer.

The neutral-gnd bond looks to be in panelboard labeled W1 on the dwg. I didn't see it but N-Gnd of the feeder to that panelboard measures 0 ohms when I had N lifted from the transformer X0 lug.

 

[jaggedben]

Personally I'd want to find the proper system bonding jumper in the panel and make sure that it's not, say, a bare EGC contacting a receptacle's neutral screw somewhere on a branch circuit.

 

[Carultch]

My question is, do I need to adjust the bonding in the disconnect and install the neutral to ground bonding screw, given that the actual existing system appears to have the neutral to ground bonding jumper installed in a panelboard that is fed by the transformer rather than inside the transformer where I connected the PV system?

NEC-wise, for customer-owned transformers, it can either be in the transformer, or in the 240.21(C) disconnect, but not both.

If it's my choice, I'd perform the connection between neutral & ground in the customer-owned transformer, and then treat all other secondary equipment as subpanels.

This way, if there are multiple 240.21(C) disconnects, I don't need to think about which one should host it, and I also don't need to think about undermining its purpose, by doing it in each one of them. The transformer initiates the separately-derived-system (SDS), and it is logical to me, to make this happen at the first point we derive the SDS.

 

[solarken]

So I investigated further and have more detail, and not sure I have a great approach to fixing it. I created a dwg showing the grounding and bonding. As I see it, the person that installed the 480 to 208 xformer and the 400A panelboard ran an AL #1 EGC from the panelboard to the transformer ground, and installed a 2/0 Cu system bonding jumper in panelboard. But there is no GEC from the panelboard to an electrode. The dwg I was given and that I uploaded before showed the system bonding jumper was to be in the transformer. I have an 8AWG Cu EGC/SSBJ from each of two PV system disconnects to the transformer ground. There is also a 4AWG AL EGC from the Transformer ground to the 480V Main Panelboard. I cannot see the main bonding jumper or the GEC inside the Main panelboard because I have not been able to power the whole site down to open it up. I did trace a 250kcm AL secondary GEC from the Main Panelboard to a water pipe because that is accessible outside the panelboard.
I need to figure out the best way to finish this job without adding too much work and cost and complexity, and make sure it is to code. I am thinking I may need to power down and open up the main panelboard to look/verify that it does indeed have a GEC to a ground rod, and a bonding jumper.
I am scratching my head about what I am seeing. Without a GEC to either the transformer or the 400A secondary panelboard, this seems kind of dangerous. In the dwg, the dashed system bonding jumper in the xfrmer is not installed yet.
Any suggestions on the best way to resolve this?

 

[solarken]

Remove sustem bonding jumper in 400A secondary panelboard, and add groundbar in transformer and add system bonding jumper from X0 to ground bar in xfrmr, and add GEC from the ground bar to a tap off the GEC in the main panelbpard "SEP"?

Right now, all the grounds in the xfrmr are on individual lugs mounted to bottom sheet metal.

 

[jaggedben]

Remove sustem bonding jumper in 400A secondary panelboard, and add groundbar in transformer and add system bonding jumper from X0 to ground bar in xfrmr, and add GEC from the ground bar to a tap off the GEC in the main panelbpard "SEP"?

Right now, all the grounds in the xfrmr are on individual lugs mounted to bottom sheet metal.

Agree with the general idea. Yes to the groundbar in the transformer, because the GEC cannot be connected to the neutral through the enclosure. See 250.68(C)(2) if the building has a steel frame.