Wiring garage: Circuit run length problem

[jbs]

I started on the wiring for my garage (see build thread here)and ran into an issue with the run length of my circuits. The garage is 32x40 and I'd like to have 3 20A circuits with alternating receptacles around the perimeter.

The panel will be in the front left corner inside the overhead doors (the overhead doors being on the 32ft side). Even with no outlets on the overhead door wall, that is 100+ ft going down one wall (40ft), across the back (32ft), and down the other side (40ft). Even 10-gauge would be pushing it since the route is not perfectly direct.

I am assuming a rule-of-thumb max 5% voltage drop at full 20A (giving max one-way length of 79ft for 12-gauge/120V/20A using a voltage calculator like this one). Is this typical/acceptable?

How do others tackle this problem? Is this impossible without ridiculously heavy cable?

 

[AussieDan]

If you have the space in the panel you could run 2 circuits for each wall, and feed the wall opposite the panel by running across the top of the garage doors or over the ceiling.

I ran 5x twin duplex receptacles in my 20x20 on a 12/3 MWBC with one duplex per phase at each box. The dual-pole GFCI breaker was the biggest expense with that setup but I'm happy with the results.

With the distances you're talking it might be worth looking into how the dual-pole gfci breaker cost stacks up against the extra wire and gfci breakers/receptacles for running multiple separate circuits.

 

[jbs]

If you have the space in the panel you could run 2 circuits for each wall, and feed the wall opposite the panel by running across the top of the garage doors or over the ceiling.

I ran 5x twin duplex receptacles in my 20x20 on a 12/3 MWBC with one duplex per phase at each box. The dual-pole GFCI breaker was the biggest expense with that setup but I'm happy with the results.

With the distances you're talking it might be worth looking into how the dual-pole gfci breaker cost stacks up against the extra wire and gfci breakers/receptacles for running multiple separate circuits.

Thanks AussieDan. I did think about using a MWBC (or two- I will have the space in the box), but by code we have to tie the breakers (which is a good idea anyway of course). Then I lose the advantage of still having live outlets available if I accidentally trip one. I guess since the panel is only a few steps away, this isn't a big deal.

I know a balanced MWBC can be used in longer runs than a normal circuit. But given that you can't guarantee the loads on each phase will be balanced, what is the best practice for run length for a MWBC? Code requirements?

 

[sberry]

Like you said, at full 20A draw the v drop is X but how many times is this circuit actually going to be loaded this heavy?

 

[AussieDan]

by code we have to tie the breakers (which is a good idea anyway of course). Then I lose the advantage of still having live outlets available if I accidentally trip one. I guess since the panel is only a few steps away, this isn't a big deal.

That's actually part of the reason I did the MWBC, if something trips the gfci I wanted to shut down everything. I can see arguments either way though.

I know a balanced MWBC can be used in longer runs than a normal circuit. But given that you can't guarantee the loads on each phase will be balanced, what is the best practice for run length for a MWBC? Code requirements?

I'm not aware of any special provisions for run length of MWBCs in NEC 2008.

 

[nate379]

Would it not make more sense to run a row of outlets on one circuit? I don't think I have ever seen them done the way you are talking about.

 

[Aceman]

Realistically, I'd wire it with #12 and not look back. You don't want to wrap the building with one circuit, you want two circuits going each direction meeting in the middle.

You're most likely not going to load the receps to 20 amps, maybe 15 or so at the most. What's the voltage at your subpanel on each leg? If it's 120+ volts, it's not going to matter if it drops 5-6 volts at the receptacle. Yes, that's more than the 3%(3.6 volts) typically calculated for voltage drop, but if you still have 114-115 volts at the device it doesn't matter.

Since the loads aren't balanced, MWBC's will make no difference.

I happen to be wiring an 1100' long dairy barn right now, being able to balance the lighting loads using MWBC's will save me about 2 wire sizes, not to mention 2 less conductors in the conduit. That's some serious money.

 

[jbs]

Like you said, at full 20A draw the v drop is X but how many times is this circuit actually going to be loaded this heavy?

Probably rarely if ever, but it seems like good practice to design it to operate at its rated capacity.

That's actually part of the reason I did the MWBC, if something trips the gfci I wanted to shut down everything. I can see arguments either way though.

I'm not aware of any special provisions for run length of MWBCs in NEC 2008.

As I understand it is just a "recommendation" that the voltage drop for a branch circuit or feeder be limited to 3% max each and 5% max for both together. Still seems like a good idea. I'd be at 8.5% on each branch circuit (at full 20A load).

Would it not make more sense to run a row of outlets on one circuit? I don't think I have ever seen them done the way you are talking about.

It makes sense from a wiring logistics standpoint, but not from a use standpoint (for my use at least). When I am working in one area of the shop, I'd like to be able to have receptacles from a couple different circuits available to plug saws, grinders, fans, et. al. into without worrying about overloading the single circuit whose receptacles are available in that area of the shop. If I were going to have multiple people working in different areas, I would look at it more sectionally.

Realistically, I'd wire it with #12 and not look back. You don't want to wrap the building with one circuit, you want two circuits going each direction meeting in the middle.

You're most likely not going to load the receps to 20 amps, maybe 15 or so at the most. What's the voltage at your subpanel on each leg? If it's 120+ volts, it's not going to matter if it drops 5-6 volts at the receptacle. Yes, that's more than the 3%(3.6 volts) typically calculated for voltage drop, but if you still have 114-115 volts at the device it doesn't matter.

Since the loads aren't balanced, MWBC's will make no difference.

I happen to be wiring an 1100' long dairy barn right now, being able to balance the lighting loads using MWBC's will save me about 2 wire sizes, not to mention 2 less conductors in the conduit. That's some serious money.

If I run 2 circuits, one in each direction, I won't have access to multiple circuits in any one area. Since I will work in one area of the shop at a time, I'd like to be able to plug a light, fan, radio into one circuit, and when I fire up my grinder which can draw 15A, plug it into a different circuit. Hence my desire to alternate receptacles.

Since I would sort of do ad hoc balancing anyway, the MWBC could make a difference in some cases. Not like in you lighting scenario, and maybe not reliably balanced enough that I could depend on it to extend the length of the circuit.

Don't know the voltage at the panel yet since it's not hooked up. 75 feet from the pole, so hopefully it will be 120V.

 

[jbs]

Another thought that occurs to me: Is it allowable to upsize just the homerun (i.e. the run from the first receptacle to the panel)? This could help one of the circuits since this first run could be 50ft. I could do that with 10/2, then drop to the more workable 12/2 between boxes.

That way I could have a separate circuit for the far corner (call it "B"), with some overlap with circuit "A" on one side and "C" on the other.

Another option would be to split the circuit at the first box and run in each direction. This would limit the max run to about half of what it would be if I wrapped only in one direction.

Combining the two might look something like this.

 

[hevnbnd]

So what method did you end up using?