I have a barn that is 750' from my main panel. I would like to run power down to it myself because the power company quoted some crazy high prices. I plan on burying the wire and would like to know what size wire I should go with or if it is too far to do. Any help would be appreciated. The barn is going to have lights, some outlets and maybe a 220 outlet or two in case I need to power a welder from time to time, but thats about it. I would rather go bigger in case I need to expand in the future.

  • 4
    We've had a few Q&A's about long power runs and some of the answers are quite detailed about how to do it cheaper and more effectively by using transformers at both ends and running the intervening line at a higher voltage (not kilovolts, typically 480V rather than 240V, (so standard 600V wire insulation is fine) but it helps the wire size a LOT, which helps the cost.) For Example: diy.stackexchange.com/a/164293/18078
    – Ecnerwal
    Commented Oct 2, 2019 at 15:20
  • 1
    If you need lots of power, it is indeed a good question as to "more expensive and or higher voltage (thus, special wire and techniques) transformers being worth it" or if an engine driven welder that's common and portable makes more sense for the application.
    – Ecnerwal
    Commented Oct 2, 2019 at 15:34
  • 1
    Depending on how often you'll be working there, and how much you like hi-tech stuff, consider covering the barn roof with solar panels & putting in a wall-pack storage unit (and of course a DC - AC converter). Won't be cheaper at the outset. Commented Oct 2, 2019 at 19:14
  • can you get by with solar for lighting and a generator for heavy loads? i bet that could be cheaper, and more flexible later on, or if you move.
    – dandavis
    Commented Oct 2, 2019 at 21:12
  • 1
    How open are you to plopping a couple of transformers in? Also, how big is this barn, and how big of a welder do you have (I1eff and/or I1max)? Commented Oct 2, 2019 at 22:40

2 Answers 2


Long-distance circuits are a huge opportunity to spend way too much on wire unnecessarily. Given the amount of money that is at stake here, let's debunk the usual 3% knee-jerk.

There are two huge mistakes when made when sizing circuits.

  • sizing for 3% voltage drop. 3% is simply wrong. There's no support in Code for it anywhere (it's a suggestion, one place). This myth is widely spread by the companies who sell wire, for obvious reasons -- the first place they do it is in their "free" voltage drop calculators. They default to 3%, and by that they mean literally 3.00% - if a cheaper option would come in at 3.25%, they will hide it, and "blame the computer" for being too literal.
  • Sizing for "breaker trip rating" instead of practical load. Remember, loads are already supposed to be derated 20% (literally, breaker should be 125% of load), so sizing to breaker is always wrong. Indeed, the simplest step of sizing to 80% of breaker usually results in a couple of wire size drops and thousand$ saved. But you should go farther, and size to actual load.

Voltage drop is proportional to current right now

Remember that voltage drop in a circuit is a function of current actually flowed. Here's what's not true: "The voltage drop the calculator says will always apply to all loads". Actually it will never apply to any loads.

Suppose someone puts in monster wire on their "50A" circuit and gets to 2.5% drop at 50A.

  • You draw 10A @ 120V. You actually get 1.0% voltage drop. Not what we expected, eh?

  • What happens if you have a "30A" dryer, which actually is 23A, and that actually is about 21A on the 230V side and 2A on the 120V side. So drops of 0.525%, 0.05%% and 0.575% per leg. The 240V heating element sees 1.1% voltage drop, and the 120V mechanism sees a 0.575% voltage drop.

  • A 40A (9600W) heater sees 2.0% voltage drop, which it doesn't need -- heaters will work on 30-40% voltage drop.

Getting the idea how wasteful this is?

For a welder, consult with your manufacturer, but voltage drop is pretty normal for a variety of reasons inside welders, so it's probably not going to bother them all that much.


Then there's using transformers to step up voltage. All fixed-installation wiring is rated for 600V, and there is nothing wrong with stepping up power that high for the transition. Transformers are expensive, but on a long haul, they're cheaper than wire. Often, simply "stepping up" the circuit to 240V, and using a transformer at the far end to make 120V, is all you need. I have plenty of postings about this.

Let's run some numbers in your case.

And to be clear, we'll be running aluminum wire, because running so much copper isn't even stupid. Use the goop and torque to spec.

Scenario 1: Obedient Consumer.
Let's do exactly what the voltage drop calculator and wire companies say we should do. Compute on breaker trip and stick to 3%. 3/0 wire would have 3.05% drop and the computer says "no" to that and computers are smarter than us, so we are forced to 4/0 wire. 750' of 4-wire 4/0 URD will cost $2145, trench it at 24" and we're done.

Scenario 2: Compromises.
On a 50A breaker, actual draw shouldn't be more than 80% or 40A. We go for 4% voltage drop at 40A. This comes up as 1/0 wire. 750' of 4-wire 1/0 URD will cost $1297.

Scenario 3: 480V Transformers.
In this case we use 15 KVA transformers and breaker for 60A (which is more power). The transformer halves the current, which also halves the voltage drop, and voltage drop is only half as important anyway, since it's coming off 480V instead of 240V. So now we can happily use much smaller wire. 240V at 40A becomes 480V@20A. With #4 wire, voltage drop @ 20A draw happens at 2.45%. Even if we max out the circuit to 60A@240V (30A@240V), voltage drop is only 3.67%. Further, we only need 2 wires between the transformers, because neutral is created locally by the transformer.

  • The two 15KVA transformers cost $2000.
  • 3-wire #4 SEU is only $555.

Scenario 4: Mini-transformer.
In this case, we only aspire to provision 20A (16A) practical, but we use a (smaller) transformer so we can transmit at 240V on 2-wire, even though we are only using 120V. Fortunately, these smaller 5 KVA transformers are readily available on Craigslist for about $100. If we go for only one 20A circuit, then we base voltage drop on 8A@240V and the calc says we're just fine with 6 AWG aluminum at 3.58% drop.

However, instead, let's go for two 120V circuits at 20A breaker (that's table saw and dust collector). That will realistically be about 24A together, or about 12A@240V, but let's assume circuit continuous max of 16A x2, so 16A@240V. Voltage drop calc says that happens at 4.57% with 4 AWG Al.

  • one used 5 KVA transformer for $100
  • 3-wire #4 SEU for $555.

And hey, that's the same #4 wire we use on our 480 scenario, and we know we can pump that to 60A or even 80A with better transformers.

  • So for this distance, the transformers are the most expensive option (but obviously get cheaper as the distance increases - there is a reason the electricity co uses high voltage for long distance transmission lines!) Commented Oct 3, 2019 at 12:17
  • What about my suggestion of "run less power, and buy a generator to run the welder"? Commented Oct 3, 2019 at 12:17
  • 1
    @MartinBonner That's a good answer. It's just not mine. I would advise turning it into a proper answer. Commented Oct 3, 2019 at 15:43

Not a lot of information to go on here but this resource:

Wire Size Calculator

May be helpful. Assuming you want 240V and a 50A capacity (based on what my welder needs) I plugged those in with 750' distance and came up with 4/0 AWG Aluminum. At this large size I think you'll find that Al is a much better value than copper.

Lots of other variables and I'll recommend that rather than direct burial you'll want to bury a conduit that is large enough to handle the 4/0 wire or something larger if you need more power in the future.

I'm going to guess that this is going to cost $5K or more. The cost of the wire alone is going to be something like $3000.

  • Add a few load bearing splice boxes
    – JACK
    Commented Oct 2, 2019 at 14:33
  • Thanks, I know its not a lot of information. But its going to be a pretty basic set up.
    – Tony
    Commented Oct 2, 2019 at 14:49
  • 5
    @peinal you DO NOT run cat 5 in the conduit with power. Low and high voltage shall not mix. You run Cat 5e in its very own conduit, though there's little point at 750 feet unless you want analog phone or expensive (you'll need line extenders) slow data. You can run all-dielectric fiber optic with the power, but it's stupid to do so, since conduit is cheap and damaging your fiber when pulling the power wires and fiber together is all too possible. So you run a communications conduit, cap it, and open it up and run fiber or low voltage wire when you actually want that.
    – Ecnerwal
    Commented Oct 2, 2019 at 15:55
  • 2
    @peinal, those are individual wires, so you need 3 of them (+ ground). From Lowes, that would then be $4860, though buying from an electrical supply house may well be (much) cheaper, so the 3k estimate is reasonable.
    – Nate S.
    Commented Oct 2, 2019 at 16:14
  • 4
    I see the problem. That voltage drop calculator is lying up and down about what NEC requirements are. There is no 3% requirement for voltage drop, that is balderdash. Further it advises calculating drop on 125% of load, which is wrong. Also their calcs are optimized for irrigation motors so they're off by about 8% compared to my general purpose calc. The 3% is "common knowledge" but I've called it balderdash on this site hundreds of times, and nobody's ever provided a code cite. I think I see where they're getting it, and it doesn't say that. Commented Oct 2, 2019 at 16:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.