Supply side Interconnections

[SunFish]

I have a few questions about supply-side connections.

1) Are only fused disconnects allowed or can a breaker be installed as the first over-current protection after the supply-side connection, as long as the breaker panel/bus is rated for greater than 60 A per 230.79(D) and it is service rated & lockable per utility requirements.

Example, 6 inverter outputs are combined in an AC panel with a 200 A main breaker. If this "AC combiner" panel is service rated, lockable and grouped with the main service disconnect can the 200 A main breaker be used as the first over-current protection for the supply-side connected conductors or would an additional fused disconnect be required?

2) Grounding & bonding conductor sizing gets fuzzy between the main solar disconnect and the supply side connection. As I understand it the grounded conductor should be bonded to the pv disconnect per 250.24, i.e. the neutral is connected to ground. Am I sizing the bonding jumper to 250.66?

In a residential scenario where i am connecting to conductors inside a main service panel on the supply side of the breaker can I skip bonding the inverter neutral to ground in the PV disconnect and just land my ground on the appropriate terminals in the main service panel as the neutral bus is bonded to the ground bus at the main panel/service?

Would I still size the ground wire/egc between the PV disconnect/OCPD and the supply connection to 250.66?

Clarification is much appreciated.

 

[jaggedben]

I have a few questions about supply-side connections.

1) Are only fused disconnects allowed or can a breaker be installed as the first over-current protection after the supply-side connection ...

There's no reason you can't use a circuit breaker if the equipment otherwise meets requirements.

2) Grounding & bonding conductor sizing gets fuzzy between the main solar disconnect and the supply side connection. As I understand it the grounded conductor should be bonded to the pv disconnect per 250.24, i.e. the neutral is connected to ground.

This is all quite debateable, but that is the way I would do it.

It is arguably an either/or question. Either you bring only the neutral (no EGC) to the new PV panel, and bond the neutral to ground there... Or you bring a separate equipment ground and don't bond the neutral in the new PV panel, as you suggest later in your post. Either way could be code compliant, but it really depends what your AHJ thinks.

For the record, I'm on the side of treating it as an additional service disconnect and bonding the neutral.

Am I sizing the bonding jumper to 250.66?

Yes.

In a residential scenario where i am connecting to conductors inside a main service panel on the supply side of the breaker can I skip bonding the inverter neutral to ground in the PV disconnect and just land my ground on the appropriate terminals in the main service panel as the neutral bus is bonded to the ground bus at the main panel/service?

Perhaps yes, if your AHJ agrees. The only good reason to do this, in my opinion, is if you happen to be using an inverter(s) that doesn't require a neutral. Otherwise you are bringing the neutral to the enclosure anyway, and bonding it surely isn't that difficult.

Would I still size the ground wire/egc between the PV disconnect/OCPD and the supply connection to 250.66?

Yes. You would really have to. After all, you can't size it to 250.122 because there is no overcurrent protection ahead of it, so there's no way to use that table.

 

[c_picard]

Great answer jaggedben.

I swear the bond/don't bond debate is the PV equivalent to the global warming and evolution "debate".

 

[Zee]

Great answer jaggedben.

I swear the bond/don't bond debate is the PV equivalent to the global warming and evolution "debate".

Great response Jben. It helped me too.
Are we striking a ground rod also for every "line side tie" ? (AKA supply side connection)?
I am told I should.

 

[c_picard]

Great response Jben. It helped me too.
Are we striking a ground rod also for every "line side tie" ? (AKA supply side connection)?
I am told I should.

If a code-compliant grounding electrode system is already there, an additional ground rod is not required. A new GEC will be brought to the service disconnect for the PV, where the neutral will be bonded. That new GEC simply gets irreversibly spliced to the existing GEC.

A huge source of confusion is when this new GEC is connected to the existing GEC inside the main panel, and run via the same conduit as the hots and grounded conductor...AHJ's see four wires and forget that this is not an EGC, but a GEC. Thus mistakenly require the grounded conductor to be isolated at the new disconnect, in violation of 250.24(C).

 

[shortcircuit2]

That new GEC simply gets irreversibly spliced to the existing GEC.

Or connected with connectors listed as grounding and bonding equipment.

 

[c_picard]

right, sorry bout that. not a code requirement but commonly enforced by rebate programs here in the northeast.

Sent from my SCH-I535 using Tapatalk

 

[jaggedben]

If a code-compliant grounding electrode system is already there, an additional ground rod is not required. A new GEC will be brought to the service disconnect for the PV, where the neutral will be bonded. That new GEC simply gets irreversibly spliced to the existing GEC.

A huge source of confusion is when this new GEC is connected to the existing GEC inside the main panel, and run via the same conduit as the hots and grounded conductor...AHJ's see four wires and forget that this is not an EGC, but a GEC. Thus mistakenly require the grounded conductor to be isolated at the new disconnect, in violation of 250.24(C).

Alternatively, the PV GEC could be brought directly to the grounding electrode, or connected to the GEC at an accessible point outside of the existing service equipment. Might be less wire in some cases, and clarify that debate.

 

[c_picard]

In many places that's what we do jaggedben, funny because it's the same thing really. Just makes it so there are only 3 wires in the conduit between enclosures. smh.

Sent from my SCH-I535 using Tapatalk

 

[Smart $]

The GEC connection may have a bearing a bearing on the 'service disconnect or not' debate. If it is a service disconnect, then one has to comply with GEC requirements of Article 250. Affected is the requirement to be continuous per 250.64(C). A PV system is not a separately derived system, so alleviation for such do not apply. That means the only place where the GEC can connect without being continuous is per 250.68(C), which only allows water pipe, structural steel, and concrete-encased electrodes to be the locations where connection of a separate GEC for the PV system is permitted. You are not permitted to connect to any other service disconnects' neutral or grounding bus... as a service disconnecting means.

 

[c_picard]

My personal policy is to refrain from refering to this as a "debate". Anyone who does not think this is a service disconnect doesn't understand why we bond the grounded conductor in the first place, and hasn't earned a spot at the podium.

Sent from my SCH-I535 using Tapatalk

 

[Smart $]

My personal policy is to refrain from refering to this as a "debate". Anyone who does not think this is a service disconnect doesn't understand why we bond the grounded conductor in the first place, and hasn't earned a spot at the podium.

Touche!

FWIW, it's only a debate for those that are worthy of being at the podium...

 

[shortcircuit2]

A PV system is not a separately derived system, so alleviation for such do not apply.

A PV system that uses a Isolated Inverter is a separately derived system.

 

[GoldDigger]

Arguable in that a GTI does not produce any power output until after a direct wire connection to the "host" system has been made.

Tapatalk!

 

[shortcircuit2]

Arguable in that a GTI does not produce any power output until after a direct wire connection to the "host" system has been made.

Tapatalk!

Neither does the load side of any isolation transformer.

 

[GoldDigger]

But the SDS distinction depends on the fact that there is no direct wire connection between primary and secondary, or in the case of a generator that it can produce output power before the output wires are connected to anything more than a load.

Tapatalk!

 

[shortcircuit2]

The Code makes no distinction on whether an electrical power source has to have electricity flowing for implementation of its rules. If the wiring and equipment is intended to connect to a source of electricity than we follow the Code.

I don't follow your reasoning that a GTI with a isolation transformer is not separately derived until it connects to the "host". The isolation transformer in the PV System Inverter provides no direct connection of the circuit conductors from one system to the other.

 

[GoldDigger]

What about the large number of GTIs that do not have isolation transformers and use ungrounded PV arrays.


Tapatalk!

 

[shortcircuit2]

What about the large number of GTIs that do not have isolation transformers and use ungrounded PV arrays.

Grid tied PV systems that serve premises wiring without isolation transformers are not considered separately derived.

Stand-Alone PV systems that serve premises wiring with or without isolation transformers are considered separately derived.

 

[GoldDigger]

And in a standalone PV system there will be no service disconnect(s) because there is no service.
Unless you are talking about a system with a transfer switch.
I have lost track of the point of this discussion, which I thought was concerned with the nature of the PV disconnect in a grid tied system.

Tapatalk!