The conductor size is the easy part

When to use this: wiring a new spa, checking an installation before an inspection, or working out why a spa heats slowly.

Sizing the circuit is ordinary arithmetic. What makes a spa different is the cluster of Article 680 requirements around it — and those, not the wire, are what fail inspections.

The circuit

A spa heater runs for hours, so the circuit is continuous and takes 125% of the nameplate current under 210.19(A)(1). Sizing on the nameplate alone is the most common way these come out one size short.

Where the code and the trade disagree

For the common 40 A spa, the table permits 8 AWG copper — 8 AWG is exactly 50 A at 75 °C. Almost every spa in the country is wired in 6 AWG. Both facts are true and the reason is worth knowing:

• 8 AWG lands on 50 A with no margin whatsoever.
• Spas sit a long way from the panel, and voltage drop over 60 to 150 ft settles the argument on its own.
• Manufacturers frequently specify 6 AWG in the installation instructions, and 110.3(B) makes those instructions enforceable — they override the table.

Read the equipment pack label before the code book.

GFCI — the whole circuit, once

680.44 requires GFCI protection for the entire circuit, not just a receptacle. Two arrangements work:

• A GFCI breaker at the panel feeding a plain disconnect at the spa.
• A plain breaker at the panel feeding a GFCI spa panel.

What does not work is both. Two GFCI devices in series race each other on every fault, and whichever wins is unpredictable. The result is nuisance tripping that is genuinely hard to diagnose, because each device tests fine on its own.

The disconnect: 5 to 50 feet, in sight

The two bounds do different jobs and inspectors check both. “Within sight” is a defined term — visible and within 50 ft, not one or the other.

Equipotential bonding

The most commonly omitted requirement, and the least understood. 680.26(B) requires 8 AWG solid copper tying together every conductive part within 5 ft of the spa: the shell, the pump housing, metal handrails, ladders, and the perimeter surface on a permanently installed unit.

It is not a ground. It carries no fault current and trips nothing. Its job is to hold everything a person can touch at the same potential, so that someone bridging the handrail and the water sees no voltage across their body. That is a completely different failure mode from the one a breaker addresses, which is why a properly grounded spa can still be dangerous without bonding.

It must be solid, not stranded, and it is a single continuous conductor rather than a daisy chain of short pieces. Self-contained portable spas on a deck are treated differently — check the listing.

Why a spa heats slowly

Almost always voltage drop. A heater is resistive, so output falls with the square of the voltage: 10% low is 19% less heat. Nothing trips, nothing looks wrong, and the spa simply takes longer to reach temperature and struggles in winter.

A 60 ft run in the code-minimum conductor is usually fine. A 150 ft run usually is not. Our voltage drop calculator will tell you where the line falls for your run, and wire size handles the upsizing.

Things that catch people out

• The ground is not the bond. A spa needs both: an equipment grounding conductor in the circuit, and a separate 8 AWG bonding conductor. Our ground wire size calculator covers the first.
• Aluminium is permitted and rarely wise here. Two sizes larger, in a damp outdoor enclosure where aluminium terminations are least forgiving, for a small saving on a short run.
• Underground runs need burial depth. Table 300.5 — 18 in for direct-buried UF at 240 V, 6 in for rigid metal conduit, and a spa circuit crossing a garden almost always goes underground.
• The panel may not have room. A 50 A two-pole breaker needs two adjacent full-size positions and 50 A of capacity. Our panel capacity calculator checks both.

How this calculator is checked

Technical reviewer

Salamot Hok

Electrician · 10+ years of installation work in Bangladesh and the wider South Asian region

He reads the result the way an installer would: are the defaults values people actually meet, does the warning fire where you would stop and think, and is the answer something you could buy and fit? The code figures themselves come from the published standards cited below, not from him — that boundary is set out on his profile.

• The maths lives in a pure function with its own test suite, asserted against worked examples from published references and standards. A calculator does not ship until those tests pass.
• 2 sources cited by name and linked, so any figure on the page can be traced back to the document it came from.
• Last reviewed 17 August 2026. Review dates are advanced only when the page is actually re-read, never to look fresh.
• Unusable input returns no answer. Where the inputs do not describe a real design, the calculator says so and withholds the number rather than printing a plausible-looking wrong one.

The full process is written up in the methodology and editorial policy. Results are engineering guidance, not a code sign-off — see the disclaimer. If a result looks wrong, tell us; corrections are answered before anything else.