How many amps used to estimate the voltage drop for residential application?

[zemingduan]

First, I want clarify that I am not asking how to calculate the voltage drop. I know how to calculate the voltage drop per NEC and I use this calculator by Soutwire to calculate the voltage drop (https://www.southwire.com/calculator-vdrop). But I want to know how many amps I shall use when calculate the voltage drop and consider the effect of the voltage drop on conductors size. This problem bother me for a long time.

For example, a 830 sqft apartment unit with 4.5kva water heater, 1.5kva washer, 5kva dryer, 8kva range, 1kva dishwasher, 1kva fridge, 1.5kva microwave/hood, 0.8kva garbage disposal, 2.5kva heat pump outdoor condenser(12A @208V), 1.2kva air handler unit (5.8A @208V) with 3.6kva@208V electric heating kit, the calculated load per NEC 220. 82 is: ((0.83*3+4.5+1.5+5+8+1+1+1.5+0.8)-10)*0.4+10+2.5+1.2+3.6*0.65 kva = 22.356 kva. The calculated amps @208V is 22.356kva/208V = 107.5A. A 125A feeder is required. The apartment feeder is 200'. When you size the minimum conductor size to obtain the 3% voltage drop or estimate the voltage drop based on #1 cooper wire, do you use 107.5A calculated per NEC 220.82? But I think NEC calculated load is always larger than in reality. Is it overkill to use this amps to estimate/consider the voltage drop on the apartment feeder?

The NEC 310.75(B)(7) states the ampacity of the feeder/service doesn't need to exceed 83% of the feeder rating which is 125A *0.83 = 103.75A. So what about using 103.75A to calculate the voltage drop? But is it still overkill?

Do you use an experienced number? How do you consider the voltage drop?

What about considering the voltage drop on the service conductors that supply a multifamily dwelling building? For example the calculated load of (55) above dwelling units is around 1200A @ 208V 3phase per NEC 220.84. The service conductors in conduit under ground that go to the meter bank is 450'. I have a project that the meter bank is 440' away from the utility pad mount transformer with under ground conductors... The largest run of a unit feeder is 200' from the meter bank to unit load center. How do you consider the voltage drop? What size of cooper conductors and how many sets will you use that takes the voltage drop into account?
Thank you!

 

[tom baker]

With 2 exceptions, voltage drop calculations are not required in the NEC. NEC calculations for service size are very conservative resulting in large service wires with little load.
So do this. Calculate the voltage drop with minimum size NEC conductors. Is it large enough to be concerned with

 

[W@ttson]

Voltage drop is based on your calculated load. Just as an easy example see below:

Suppose you have a 40A Single ph Breaker feeding a 1ph 1KW unit heater from a 3ph 480/277V panel. Whereas, it is obviously a noneconomical install, it is compliant as long as the conductor size is sized per the breaker and load. Your VD would be on the 3.6A since that is your load, not on your 40A.

 

[zemingduan]

With 2 exceptions, voltage drop calculations are not required in the NEC. NEC calculations for service size are very conservative resulting in large service wires with little load.
So do this. Calculate the voltage drop with minimum size NEC conductors. Is it large enough to be concerned with

I understand voltage drop calculation is not required in NEC since it's a performance issue not a safety issue. But I am very concern with this long run service conductors.

In my example, the calculated load per NEC is 1198A and the minimum NEC required conductors for the (55) units meter bank are (3) sets of 4#600KCMIL cooper in (3) sets of 4" conduit whose ampacity is 420A*3 = 1260A. Based on 1198A amps, I get a 5.31% voltage drop based on 0.9 power factor at the (3) sets of 4#600KCMIL service conductors. And get the (208*(1-5.3%))V = 196.96V at the meter bank. There is another 3.24% drop on the 125A apt. feeder which you get (196.96V*(1-3.24%)) = 190.62V at unit load center. The voltage drop at the load center is (208-190.62)/208 = 8.36% which is larger the recommended 3% by NEC.

It's impossible to obtain 3% voltage drop at apt. load center based on the calculations. In order to obtain 5% at the apt. load center I need to increase the service conductor to (5) sets of 4#600 KCMIL or even more sets per this calculations which becomes crazy.

 

[wwhitney]

In my example, the calculated load per NEC is 1198A

You need to check the energy or green codes applicable to your project. I seem to recall seeing at least one such code that provides a mandatory maximum voltage drop requirement and also specifies to use the NEC calculated load. In that case you have no choice.

Absent that, I would say it's fair to recognize that the NEC calculated load is conservative and that a lower amperage could be used for calculating voltage drop. However, I'm not aware of any guidance on how to determine that lower value, which I think is what you are asking for.

Cheers, Wayne

 

[Hv&Lv]

Find a set of similar apartments and ask the utility for demand data for those apartments.
I see houses that have 320 amp service that have a peak demand of 22kW, and these are total electric.
I dare say the apartments with 125 amp feeders will ever see amperage as high as 75A.
That range kVA is with every eye and the oven on. 55 apartments will never have everything on at once.

 

[zemingduan]

You need to check the energy or green codes applicable to your project. I seem to recall seeing at least one such code that provides a mandatory maximum voltage drop requirement and also specifies to use the NEC calculated load. In that case you have no choice.

Absent that, I would say it's fair to recognize that the NEC calculated load is conservative and that a lower amperage could be used for calculating voltage drop. However, I'm not aware of any guidance on how to determine that lower value, which I think is what you are asking for.

Cheers, Wayne

Wayne, thanks for the response! That is exactly what I am asking for. I am seeking a guidance to determine the lower amps value ( used for voltage drop calculation) than the calculated amps per NEC or how you guys take the voltage drop into account to determine the conductors size experience for similar situation.

I will check the IECC energy code to see if I can find any clue. Thank you.

 

[Charged]

Has someone mentioned, get demand data from utility for something similar, or even come up with what you think the realistic load is. See what that gets you and then think about what equipment ( if any) would be effected or sensitive voltage tolerance. I came across a sheet that ONCOR had that had a bunch of project types and the connected vs actual demand data. I think it was an internal training thing. Nothing was over 50%. Sometimes you just have to come up with your own diversified number if your trying to make sure you don’t have any issues and leave the NEC calcs out of it. Same thing you would do if sizing a whole building generator.

 

[electrofelon]

I usually use 60-70 amps for a typical single family. For feeders or services supplying multiple units in a multi-family, I might add something like 20 amps for each additional unit. But it all depends on the situation too.

 

[zemingduan]

You need to check the energy or green codes applicable to your project. I seem to recall seeing at least one such code that provides a mandatory maximum voltage drop requirement and also specifies to use the NEC calculated load. In that case you have no choice.

Absent that, I would say it's fair to recognize that the NEC calculated load is conservative and that a lower amperage could be used for calculating voltage drop. However, I'm not aware of any guidance on how to determine that lower value, which I think is what you are asking for.

Cheers, Wayne

I check the energy code. Philadelphia adopt the IECC 2018. The voltage drop requirements are only found in the commercial provisions. (https://up.codes/viewer/philadelphia/iecc-2018/chapter/CE_4/ce-commercial-energy-efficiency#C405.9)

"C405.9 Voltage Drop in Feeders and Branch Circuits

The total voltage drop across the combination of feeders and branch circuits shall not exceed 5 percent."

This means it only apply to commercial building. And it doesn't require the voltage drop calculations or specify the calculation based on NEC calculated code. I think inspector will measure the voltage on site to check if it is code compliant.

The residential provisions of IECC 2018 regarding the electrical power and lighting system is very short. It only has below requirements. (https://up.codes/viewer/philadelphia/iecc-2018/chapter/RE_4/re-residential-energy-efficiency#R404)

"R404.1 Lighting Equipment (Mandatory)

Not less than 90 percent of the permanently installed lighting fixtures shall contain only high-efficacy lamps.

R404.1.1 Lighting Equipment (Mandatory)

Fuel gas lighting systems shall not have continuously burning pilot lights."

 

[zemingduan]

Has someone mentioned, get demand data from utility for something similar, or even come up with what you think the realistic load is. See what that gets you and then think about what equipment ( if any) would be effected or sensitive voltage tolerance. I came across a sheet that ONCOR had that had a bunch of project types and the connected vs actual demand data. I think it was an internal training thing. Nothing was over 50%. Sometimes you just have to come up with your own diversified number if your trying to make sure you don’t have any issues and leave the NEC calcs out of it. Same thing you would do if sizing a whole building generator.

I think get the peak kw demand from utility company can work for each apartment. For the whole building, I guess it doesn't have a utility meter for the whole building? I think asking utility company for their sizing is a good idea since utility usually get a utility transformer KVA rating much smaller than sized per NEC.

 

[zemingduan]

I usually use 60-70 amps for a typical single family. For feeders or services supplying multiple units in a multi-family, I might add something like 20 amps for each additional unit. But it all depends on the situation too.

Do you try to obtain the 5% voltage drop or 3% (which recommended by NEC) at apartment load center by increasing the size of service conductors and apartment feeder?

 

[electrofelon]

Do you try to obtain the 5% voltage drop or 3% (which recommended by NEC) at apartment load center by increasing the size of service conductors and apartment feeder?

It is so rarely a problem, I don't remember if I have ever had to increase conductor size for an apartment. It would probably have to be 150 ft or more before I would even worry about up sizing wire.

 

[jaggedben]

If energy loss is the concern, rather than load performance under high draw, then you don't need peak load figures since that may occur as little as once a year (we're talking resi here). It could be more appropriate to use average annual draw, or average draw in the highest usage month to account for seasonal variation.

I get that the OP is doing a new service and that there may be no meter data yet. There is no correlation between NEC calcs and actual energy consumption, which depends highly on the residents behavior and can change when nothing about the installed load changes (kid moves in/out, someone starts/stops working from home). Most data I've looked at have a peak load (hourly interval) around 1/2 to 1/4 of the NEC calc load, and an average load of much less. But there's always exceptions.

I don't understand green energy codes that limit voltage drop. At the user end, higher voltage means more energy consumption or some cancelling out. I suspect that, load performance quality aside, the extra energy that goes into the production, transportation and installation of larger size conductors for reduced voltage drop is never recovered in energy savings. But it'd take a PhD to prove or disprove that.

 

[zemingduan]

If energy loss is the concern, rather than load performance under high draw, then you don't need peak load figures since that may occur as little as once a year (we're talking resi here). It could be more appropriate to use average annual draw, or average draw in the highest usage month to account for seasonal variation.

My major concern is the performance wise. For example, if you obtain a voltage less than 190V at the A/C condensers on the roof, I think that is not good for the compressor long term.

Most data I've looked at have a peak load (hourly interval) around 1/2 to 1/4 of the NEC calc load, and an average load of much less. But there's always exceptions.

Is this data for single dwelling unit or a multifamily dwelling apartment?

Thank you!

 

[kwired]

In addition to what some have already mentioned, there is voltage drop associated with steady loading, usually @ calculated load and there is short time VD during starting of inductive loads.

You can have things sized for the usual 3 and 5 percent but a AC compressor that starts across the line may still draw enough for really short duration to be noticeable by lights momentarily dimming. Some of this might be because of transformer impedance or length/size of utility conductors that you have little control over as well.

 

[zemingduan]

For my project the 440' service conductors from utility owned pad mount transformer to the utility approved meter bank are in scope of electrical contractor instead of utility company. But I heard some one say that utility company will adjust the transformer taps to provide 208V at the service entrance disconnect which is at the meter bank. Can someone share some knowledge and information regarding this? Will the utility company do this?

For example, if the utility company owned pad mount transformer have tap setting (2) +2.5% like typical transformer do, we can get 208V * (1+5%) = 218.V at the secondary of the utility transformer. In that case, the voltage drop at the meter bank which is 440' away from the utility transformer is not gonna be a big problem in order to obtain 208V without increasing the service conductors size. What are your thought?

Thank you!

 

[petersonra]

I don't know how you can make VD calculations for any electrical system and have them be realistic. Are you supposed to average the VD out over a whole year? Or is it the worst case which would seem to be when large motors start.

 

[kwired]

For my project the 440' service conductors from utility owned pad mount transformer to the utility approved meter bank are in scope of electrical contractor instead of utility company. But I heard some one say that utility company will adjust the transformer taps to provide 208V at the service entrance disconnect which is at the meter bank. Can someone share some knowledge and information regarding this? Will the utility company do this?

For example, if the utility company owned pad mount transformer have tap setting (2) +2.5% like typical transformer do, we can get 208V * (1+5%) = 218.V at the secondary of the utility transformer. In that case, the voltage drop at the meter bank which is 440' away from the utility transformer is not gonna be a big problem in order to obtain 208V without increasing the service conductors size. What are your thought?

Thank you!

They can do that somewhat reliably for a stead load, but changes in load will throw it off. If you have "down days" with little or no load the voltage can possibly be higher then desired during those periods. Times with even more load than designed for will result in lower voltage.

 

[mivey]

roughly every 100 ft will require a wire size upgrade. for long runs, the utility may change the taps for you but with limits. they can usually go 2 up or 2 down at 2.5% per step. however, they are usually compensating for primary drop, not secondary drop. with a dedicated transformer, they may consider changing the tap but should look at voltage regulation and voltage at minimum load.

utilities usually compensate for secondary drop with bigger wire or sometimes a bigger transformer and may expect you to do the same. the bigger transformer will help secondary regulation but they probably will ask you to pay for the upsizing.