Wire Size Calculator: How to Choose the Right Gauge

I was standing in the electrical aisle doing math on my phone… and it still felt wrong

I was that person blocking the wire rack, squinting at tiny spool labels, and tapping numbers into my phone like I was about to launch a rocket. I had a run that was “only” about 85 feet, a load that was “only” 18 amps, and a buddy telling me “just grab 12 gauge and you’re fine.”

But the thing is, I’ve worked around enough engineers (and enough inspectors) to know that “you’re fine” isn’t a calculation. It’s a vibe.

So I did what I always do when the vibe starts costing money: I backed up, wrote down what actually matters, and treated it like an engineering problem… except with real-world messiness like long runs, warm spaces, and wire that’s not living in a perfect textbook.

And yeah, that’s basically why we built ProCalc.ai in the first place.

What you’re really choosing when you pick a wire gauge

You’re not picking a number off a chart. You’re managing three things that like to fight each other: ampacity (heat), voltage drop (performance), and practicality (cost, termination size, what you can actually pull through conduit without inventing new swear words).

A wire can be “legal” on ampacity and still be a bad pick because the voltage drop is excessive. And a wire can have gorgeous voltage drop numbers and still be wrong because the insulation temperature rating or installation conditions knock the ampacity down. That’s the part people skip, and then they act surprised when the motor won’t start cleanly or the lights look a little… tired.

So if you want the short version: choose gauge for ampacity, then verify voltage drop, then sanity-check the install conditions.

That’s it.

Here’s a quick cheat table I keep in my head for what changes the answer. It’s not “code,” it’s just reality.

So why does everyone get this wrong? Because charts are comforting. They look final. Real installs aren’t final.

The math engineers expect (and the field reality you actually need)

I’m going to give you the core voltage drop relationship, because once you see it, you stop treating wire gauge like a superstition. But I’m also going to say this out loud: you don’t need to do this by hand every time. You just need to understand what the calculator is doing so you can smell a bad answer.

And here’s where people quietly mess it up: they use one-way length for resistance. But your electrons don’t teleport back to the panel. If it’s a typical two-conductor circuit, the current goes out on one conductor and returns on another, so the resistance is based on the round-trip length (yes, even if the return is a neutral, it’s still a conductor with resistance).

So, a worked example. Nothing fancy, just real numbers.

Scenario: You’ve got a 120 V single-phase load pulling about 18 A on an 85 ft one-way run. Copper conductors. You’re trying to keep voltage drop around 3% (a common design target, not a magic law of physics).

Step 1 — Round-trip length:
85 ft one-way becomes about 170 ft round-trip.

Step 2 — Pick a candidate gauge:
Say you try 12 AWG copper.

Step 3 — Estimate resistance:
12 AWG copper is roughly 1.6 ohms per 1000 ft (this varies slightly by strand count and temperature, so don’t treat it like a sacred constant).

Step 4 — Compute circuit resistance:
R_total ≈ 1.6 Ω/1000 ft × 170 ft ≈ 0.272 Ω

Step 5 — Voltage drop:
V_drop ≈ 18 A × 0.272 Ω ≈ 4.9 V

Step 6 — Percent drop:
4.9 V / 120 V ≈ 4.1%

So 12 AWG might be totally fine on ampacity, but you’re over that 3%-ish target. If this is lighting, you might not care. If it’s a motor starting under load, you might care a lot. And if it’s sensitive electronics, you’ll care the first time it reboots and you pretend you didn’t notice.

Try 10 AWG next and you’ll see the drop come down. That’s the whole game.

Also: temperature and bundling change resistance a bit, and they can hammer ampacity. That’s why I keep saying “about” and “roughly.” Engineering is precise; inputs in the field are… aspirational.

If you’d rather not do napkin math in an aisle (I don’t blame you), use the calculator and then sanity-check the result:

• Wire Size Calculator (this is the one you actually want for gauge selection)
• Voltage Drop Calculator if you already know the gauge and you’re verifying performance

And if you want the embedded version right here:

How I actually choose wire size (the quick workflow)

I don’t start with gauge. I start with the story of the circuit. What is it feeding, how far is it going, and what’s the “bad day” condition? Hot attic? Bundled tray? Long conduit with three bends and a pull that’s going to make you question your life choices?

Then I do this:

1. Get the load current (nameplate, calc, or measured). If it’s continuous, treat it like it’s continuous. Don’t get cute.
2. Pick insulation and installation assumptions you can defend later. If you don’t know ambient temperature, at least don’t assume a perfect 20°C lab.
3. Choose a wire size that meets ampacity after any reasonable derating. This is where engineers and electricians sometimes talk past each other: the engineer thinks in steady-state current; the installer lives in the world of “these six conductors are bundled and it’s hot.”
4. Check voltage drop using the actual one-way length and the correct circuit type (single-phase vs three-phase). If voltage drop is high, go up a size. Sometimes two sizes.
5. Check termination and conduit reality. Lugs, breaker terminals, bend radius, conduit fill. I’ve seen beautiful designs that died at the panel because the chosen conductor wouldn’t land cleanly without gymnastics.

And yeah, sometimes you end up upsizing “just because.” Not because you love spending money, but because you love not getting a callback.

One more thing: aluminum is totally workable, but it changes the details. Different sizes, different terminations, and you need to be honest about workmanship. If your crew treats terminations like an afterthought, don’t spec aluminum and then act shocked later.

FAQ (the stuff people ask me after they already pulled the wire)

If you want to double-check your answer from a different angle, I’ll often run the numbers through the voltage drop tool after I pick a gauge, and if it’s three-phase I sanity-check power with the three-phase calculator. Redundancy is underrated (and cheaper than rework).

And if you’re sitting there thinking “this feels like too much effort for a piece of wire,” I mean… sure. Until the run is 220 feet, the load is a compressor, the ambient is hot, and the inspector asks you why the numbers don’t pencil out.

That’s a lot of wire. And it’s a lot of regret if you guess!