Splicing in a meter bank

[david]

I see both sides. The pv output is 7 amps. Would it be any different if each system was back fed at the panel in each apartment? I don't think it makes any difference.

Granted there will be load diversity.

I’m not an electrical engineer but you want to add 280 amps solar single phase to this center fed three phase buss gear at 80 different points without asking an engineer?

 

[david]

edit

 

[david]

I see both sides. The pv output is 7 amps. Would it be any different if each system was back fed at the panel in each apartment? I don't think it makes any difference.

My math is a bit rusty but I think it makes a difference.

The way I see it if you supply the single phase solar source current by tapping the 2 pole feeder at the MDP you are Y connecting single phase sources to a 3 phase buss.

Are you going to balance the source input onto the three phase Y?

Is the MDP center fed 40 apartments, 20 on each side of the 1200 amp main ?

Did you calculate the bus amp rating 705.12(D)(2)(3)?

Are you not concerned that this is center fed gear?

 

[iwire]

The way I see it if you supply the single phase solar source current by tapping the 2 pole feeder at the MDP you are Y connecting single phase sources to a 3 phase buss.

Are you going to balance the source input onto the three phase Y?

I may be misunderstanding you but the way I see the OPs installation it will be balanced due to the construction of the meter center. Each feeder to each unit that the OP plans to tap would already be arranged AB, AC, BC, AB, AC, BC and on.

I do agree with you that there are code issues as far as main.

Let's be realistic though, each inverter is rated 1500 watts, they are not going to produce 1500 watts and each apartment will be using at least part of what the inverter puts out and often much more.

I don't see much reverse current heading out of the main.

 

[kwired]

I may be misunderstanding you but the way I see the OPs installation it will be balanced due to the construction of the meter center. Each feeder to each unit that the OP plans to tap would already be arranged AB, AC, BC, AB, AC, BC and on.

I do agree with you that there are code issues as far as main.

Let's be realistic though, each inverter is rated 1500 watts, they are not going to produce 1500 watts and each apartment will be using at least part of what the inverter puts out and often much more.

I don't see much reverse current heading out of the main.

Me either, if there would happen to be a time where there is absolutely no load in any of the 40 units, and all the PV was producing rated output then unless I got something wrong you are only putting about 166 amps back into grid through something designed to handle 1200 amps. I am not well versed in the rules that apply to these types of installs, some may be questionable here but rules aside I don't see much chance of overloading anything here because of the PV install. If there is any overloading it was because of poor load calculations before the PV was ever introduced.

 

[david]

I may be misunderstanding you but the way I see the OPs installation it will be balanced due to the construction of the meter center. Each feeder to each unit that the OP plans to tap would already be arranged AB, AC, BC, AB, AC, BC and on.

I do agree with you that there are code issues as far as main.

Let's be realistic though, each inverter is rated 1500 watts, they are not going to produce 1500 watts and each apartment will be using at least part of what the inverter puts out and often much more.

I don't see much reverse current heading out of the main.

I may be misunderstanding you but the way I see the OPs installation it will be balanced due to the construction of the meter center. Each feeder to each unit that the OP plans to tap would already be arranged AB, AC, BC, AB, AC, BC and on.

I don't know how to calculate inverters 19 & 20 on both sides of center without adding a third inverter to balance the 3 ph. I imagine others here can do it I just do not know how. Or if two extra inverters should be added to balance the 3 phase

 

[david]

I do agree with you that there are code issues as far as main.

Let's be realistic though, each inverter is rated 1500 watts, they are not going to produce 1500 watts and each apartment will be using at least part of what the inverter puts out and often much more.

I don't see much reverse current heading out of the main.

I agree that the buss rating is conservative and may be adjusted or even eliminated at some future point. I was just trying to apply the 2014 NEC to this and I did find it somewhat intriguing.

 

[jaggedben]

@sketchy
Okay, there's a lot going on here and a lot already been said, so hopefully I'm not missing anything important...

Main points:

1) Since the bussing between each meter and breaker is not to be modified, I believe you must be talking about tapping the feeder for each unit on the load side of it's breaker. All further comments below are based on this assumption.

2) It matters a whole lot what code cycle you are on, regarding whether your AHJ should/will allow such load side taps, period. 2014 has clear rules which require either that the feeder is rated high enough for both sources, or that you install an overcurrent device downstream. See 705.12D(2). The 2011 code is highly ambiguous about such taps, since no rules are explictly written to address them. If on that cycle you should make no assumptions about your AHJ's allowing such taps at all (but also don't give up). Best case, with only 7amps on each PV system your feeders may already rated high enough for both sources. For example #2 Cu on 100A breakers would be just fine under 2014 and you'd have a solid case for convincing the AHJ under 2011.

3) Next we have the question of whether each unit's breaker is a service disconnecting means or not. It's paramount that this question be clear between you and the AHJ. If each unit's breaker is a service disconnecting means, and they allow the load side taps, then we are all good if the taps are allowed. However if another disconnect upstream of the meters is considered the service disconnect, and the AHJ considers the meterbank a panelboard, then this is totally uncharted territory for me personally. Again, it matters hugely what code cycle you are on. 2014 has more different rules and options for ratings of panelboards, and 705.12(D)(3)(C) could probably give you the permission you need. 2011, on the other hand, could put the kabosh on the entire project, since all your sources couldn't be installed on the 'opposite end' if you're invoking the 120% rule.

In conclusion, there's a lot of ambiguity here that could really screw up this whole project if your AHJ doesn't understand/agree with what you're doing. You'd better be real clear with them in your proposals and how you're interpreting each code section that could bear on the subject.

 

[iwire]

The main ahead of the meters is the service disconnect it is required to be. The unit breakers are feeder breakers.

 

[iwire]

I don't know how to calculate inverters 19 & 20 on both sides of center without adding a third inverter to balance the 3 ph. I imagine others here can do it I just do not know how. Or if two extra inverters should be added to balance the 3 phase

There are 40 units, 40 is not divisable by three so it's going to be a bit unbalanced with or without PV. I don't understand what you are driving at about the ballance.

 

[jaggedben]

All that and I didn't even touch on the original question about space requirements.

I think the relevant code section would be 312.8, not 376. (But again, make sure AHJ is with you.)

You take any cross section of the wiring space and make sure the cross sectional areas of the relevant components do not add up to more than the allowed percentage.

 

[iwire]

I agree that the buss rating is conservative and may be adjusted or even eliminated at some future point. I was just trying to apply the 2014 NEC to this and I did find it somewhat intriguing.

Understood, it is most unusual at least to me to have 40 different systems and trying to tie them in this way.

 

[david]

@sketchy
3) Next we have the question of whether each unit's breaker is a service disconnecting means or not. It's paramount that this question be clear between you and the AHJ. If each unit's breaker is a service disconnecting means, and they allow the load side taps, then we are all good if the taps are allowed. However if another disconnect upstream of the meters is considered the service disconnect, and the AHJ considers the meterbank a panelboard, then this is totally uncharted territory for me personally. Again, it matters hugely what code cycle you are on. 2014 has more different rules and options for ratings of panelboards, and 705.12(D)(3)(C) could probably give you the permission you need. 2011, on the other hand, could put the kabosh on the entire project, since all your sources couldn't be installed on the 'opposite end' if you're invoking the 120% rule.

The max buss rating for this 1200 amp center feed MDP is the same as the service over current protection 1200 amp

i think compliance could be given by 705.12 (2) (3) (d) 2014 NEC where designed under engineering supervision

 

[david]

There are 40 units, 40 is not divisable by three so it's going to be a bit unbalanced with or without PV. I don't understand what you are driving at about the ballance.

I think we are talking about two different things.

I wasn't too concerned with the three phase loads being unbalanced.

I was trying to determine if the source making up the three phase bank (20) inverters would be a concern if they where unbalanced. I thought by perhaps adding two inverters would help keep the supply bank balanced

 

[iwire]

I think we are talking about two different things.

I wasn't too concerned with the three phase loads being unbalanced.

I was trying to determine if the source making up the three phase bank (20) inverters would be a concern if they where unbalanced. I thought by perhaps adding two inverters would help keep the supply bank balanced

Maybe I am still missing something but it would never be perfectly balanced as each inverter is going to be outputting different wattage.

30 amps out of balance on a large service is not going to be a problem is it?:?

 

[sketchy]

I'm dealing with the 14 code. The reason for feeding 40 meters instead of just the house meter has to do with rebate money. These will be considered taps as they are on the load side of each units 125 amp main. There will also be a 1200 amp service main disconnect. I've wired jobs like this before and never had an issue. Doesn't mean there won't be one in the future... Very interesting to hear different opinions on this. I think you should be able to back feed the full buss rating. You'd be using the DG source instead of the utility source. I don't agree with the theory that BOTH sources will be flowing through a given bus at the same time. It's either one or the other IMO.

 

[jaggedben]

I'm dealing with the 14 code. The reason for feeding 40 meters instead of just the house meter has to do with rebate money. These will be considered taps as they are on the load side of each units 125 amp main. There will also be a 1200 amp service main disconnect. I've wired jobs like this before and never had an issue. Doesn't mean there won't be one in the future... Very interesting to hear different opinions on this. I think you should be able to back feed the full buss rating. You'd be using the DG source instead of the utility source. I don't agree with the theory that BOTH sources will be flowing through a given bus at the same time. It's either one or the other IMO.

Depending on how overcurrent protection is arranged, currents can add together to go to the loads. It's not likely (since a main breaker would trip at some time when the sun wasn't shining, if that much load were needed repeatedly). But that's the reasoning behind the 120% rule and the opposite end rule.

 

[kwired]

I'm dealing with the 14 code. The reason for feeding 40 meters instead of just the house meter has to do with rebate money. These will be considered taps as they are on the load side of each units 125 amp main. There will also be a 1200 amp service main disconnect. I've wired jobs like this before and never had an issue. Doesn't mean there won't be one in the future... Very interesting to hear different opinions on this. I think you should be able to back feed the full buss rating. You'd be using the DG source instead of the utility source. I don't agree with the theory that BOTH sources will be flowing through a given bus at the same time. It's either one or the other IMO.

If you can supply via the utility with 125 amps and can supply via the onsite production an additional amount, then if the load is there the common conductor/bus potentially could carry the sum of both input abilities. reality is that kind of load is seldom there to connect, also in your case the PV input for each unit is only something like 7 amps or so (1500VA/208=7.21 was what I based that on correct me if I have the wrong size of PV unit).

 

[david]

I'm dealing with the 14 code. Very interesting to hear different opinions on this. I think you should be able to back feed the full buss rating. You'd be using the DG source instead of the utility source. I don't agree with the theory that BOTH sources will be flowing through a given bus at the same time. It's either one or the other IMO.

I thought from discussion on this the concern wasn't just normal load current but also a short time fault current.
That both source currents would contribute to a fault condition.

 

[jaggedben]

I thought from discussion on this the concern wasn't just normal load current but also a short time fault current.
That both source currents would contribute to a fault condition.

Not sure why you thought that, but the inverters really can't deliver fault current above their normal operating current. So they are basically a rounding error compared to the available fault current from the utility. With very large systems this might not be the case but even with 60kW all put together from the multiple units that's pretty much going to be the case here.