VD?

[Alwayslearningelec]

So when doing voltage drop calc and you have 3P 4W feeder from 277/480 panel to another 277/480 panel is the feeder 277 or 480? And why? Thanks

 

[winnie]

If you have a dominant single phase 277V load, then you need to do the calculation for that single 277V load.

But the common assumption is that the panel will be roughly balanced and thus treated as a 480V three phase load. So common practice (and IMHO a very reasonable assumption unless the panel is small) is to treat the feeder as 480V.

Where might it be more appropriate to treat it as 277V? Imagine you have a very small panel with a few circuits, such as a 30A 3 phase feeder supplying 6 20A lighting circuits in a warehouse. You could easily imagine a situation where only 2 circuits on a single phase were on, in which case your voltage drop is set by the loading of that single hot and neutral, and the fact that other hots are present is irrelevant.

If you really want to get into the weeds on VD calculations, you need to consider both transient (eg. motor starting) and steady state loading, but common practice is to calculate voltage drop in the steady state.

-Jon

 

[wwhitney]

If you really want to get into the weeds on VD calculations, you need to consider both transient (eg. motor starting) and steady state loading, but common practice is to calculate voltage drop in the steady state.

Is there a common example where the transient state loading will end up controlling, and it won't be sufficient to just look at the steady state?

Cheers, Wayne

 

[winnie]

Is there a common example where the transient state loading will end up controlling, and it won't be sufficient to just look at the steady state?

Cheers, Wayne

I'm thinking oversized motor loads, or loads where common usage is well below the motors rated capacity. Things such as saws or compressors at construction sites.

-Jon

 

[Carultch]

So when doing voltage drop calc and you have 3P 4W feeder from 277/480 panel to another 277/480 panel is the feeder 277 or 480? And why? Thanks

Given a three phase load, or a collection of single phase loads that add up to a balanced 3-phase load on the feeder to a subpanel on a WYE grid, you can use either of the following two formulas:

%VD = 100% * I*r*L/(N*Vpn)
or:
%VD = 100% * I*r*L*sqrt(3)/(N*Vpp)

where:
r is the effective resistance per unit length for the size, raceway type, temperature, and whether it is AC or DC.
L is the one-way length of the circuit
I is the operating current
N is the number of sets in parallel, and 1 if not applicable
Vpp is the phase-to-phase voltage
Vpn is the phase-to-neutral voltage

For unbalanced 3-phase loads, consider the worst case scenario total line current among the entire circuit, ans use that as the basis of the calculation.

As for why this formula works:
Vpp / Vpn = sqrt(3) by the vector math of the WYE grid, therefore the two formulas should get you the same result either way. To no more than 3 significant figures of course, because that is as accurately as we know the voltage in the first place.

The term L*sqrt(3) is the effective round trip length. For single phase and DC circuits, this is 2*L, because it is the round trip length. All current from the source comes back to the source using the other wire.

For a 3-phase AC circuit that is completely balanced, the circuit behaves as if it is 3 individual sources at the phase-to-neutral voltage, each of which sends the power from source to the load, and takes credit for the other two phases being conveniently timed to simultaneously carry back the return current. No line needs to use the wire twice to get current to travel the full round trip path. So using the phase-to-neutral voltage as the basis of this calculation, the effective round trip length is the one-way length.

To replace Vpn in the equation with Vpp, we multiply by 1 in a fancy way, which is sqrt(3)/sqrt(3). The term sqrt(3)*Vpn gets replaced with Vpp, and we are left with a sqrt(3) in the numerator.

 

[drktmplr12]

For unbalanced 3-phase loads, consider the worst case scenario total line current among the entire circuit, ans use that as the basis of the calculation.

No reason to not do it this way, unless there is a physical constraint such as installed cable or conduit that isn't being replaced.

If you do the calc assuming the highest phase current, and it doesn't comply with state building code, then sharpen your pencil. In the vast majority of cases, the need for the level of precision being discussed is an academic exercise at best and usually unwarranted. If it's close and you can control the conduit and cable size, just go 1 size bigger and be done with it.