What a load calculation does not tell you about your panel

The calculation is the right first check on electrical panel capacity. It is also arithmetic done on paper, and there are five things about a panel it never looks at.

October 11, 2026 · 4 minute read

A load calculation answers one question: does the demand of the house, with the new equipment added, fit under the rating of the service. The main guide to electrical panel capacity for home upgrades argues that this question comes first, and it does. A passing number is still not a clean bill for the panel. This guide covers where the calculation stops, so a homeowner knows what else to ask before an EV charger, a generator or a heat pump goes in.

Does a passing load calculation mean the panel is in good condition?

No. A load calculation is arithmetic about demand, and nobody has to open the panel to do it. It does not see a scorched connection where a breaker meets the bus, a corroded lug, a loose termination on an aluminum feeder, or water staining inside the enclosure. A 100 amp panel that calculates at 70 amps and shows heat damage is not a panel with 30 amps to spare. It is a panel with a repair to make before anything is added.

Condition is a separate finding, and it comes from an inspection with the cover off. If the quote for a new circuit rests on a calculation alone, ask whether anyone looked inside.

Does the calculation check the rating it is compared against?

The calculation ends in a number of amps, and that number is only as useful as the rating it is set beside. Most people read the rating off the handle of the main breaker. The label inside the panel door lists a separate rating for the bus, the metal bars the breakers attach to. The service conductors and the meter socket carry ratings of their own. The capacity of the service is the smallest of those, and a calculation that takes the main breaker at its word has assumed the rest.

Solar adds a second limit the load calculation never touches. Section 705.12 of the electrical code restricts how much a solar or battery breaker can feed back into a panel's bus. A home can pass the load calculation and still have no room on the bus for a new solar breaker, or the reverse.

Why can two correct methods give two different answers?

The electrical code gives two routes for an existing home, and they measure different things. Section 220.83 builds the load from square footage and appliance nameplates, then applies fixed demand factors. Section 220.87 starts from the highest demand actually recorded at the home, multiplies it by 125 percent and adds the new load. The heat pump permit guide works a 220.83 example line by line.

The same house can fail one method and pass the other, and neither result is an error. The first is an assumption about how homes in general use power. The second is a record of how one household used it last year, and a record has blind spots: a mild summer, an empty bedroom, a previous owner who never ran the oven and the dryer in the same hour. If the house changed hands recently, the recorded peak belongs to somebody else's habits.

When does the measured method not fit the house?

The measured method gets weaker in three situations, and a homeowner can check for each one.

1. Short records. Where a full year of demand data is not available, the exception in Section 220.87 allows a continuous recording of at least 30 days. Thirty days in April does not show what a house in Lodi draws in July, which is why the same exception requires the larger of the heating or cooling load to be included by measurement or by calculation.

2. Coarse intervals. That exception defines demand as the highest average held over a 15 minute interval. Usage data recorded in one hour intervals averages a short peak down, so the hourly figure can only match or understate the 15 minute one. Ask what interval the data was recorded at.

3. Solar and load shifting. The 30 day recording route is not permitted where the service has a renewable energy system or any form of peak load shaving. The reasoning carries over to utility data too: a meter on a solar home records what the house took from the grid, not what the house used, so a sunny afternoon reads lower than the real load.

Does a passing number mean the new circuit will work where it is planned?

A load calculation is done at the service. It does not say whether the subpanel in the garage, or the feeder running to it, can carry the new load, and it does not say whether there is a physical space for the breaker. It also does not size the circuit. EV charging is treated as a continuous load, so the circuit has to be rated at 125 percent of what the charger draws: a charger that delivers 48 amps needs a 60 amp circuit.

The utility side sits outside the calculation as well. The service drop and the transformer belong to the utility, and whether they can support a larger service is the utility's determination, not the electrician's.

What should the written calculation include?

A calculation with its limits stated is more useful than a bare number. Four lines make it one:

1. The method used, 220.83 or 220.87.

2. The rating the result was compared against, and where that rating was read from.

3. Whether the panel was opened and what condition it was in.

4. For measured data, the period covered, the recording interval and whether the home has solar.

The calculation says whether the load fits the rating. It does not say whether the rating is right, or whether the panel is still fit to carry it.

None of this is an argument against the calculation. It is still the first step and the cheapest one. It is an argument for reading it as one finding among three: the number, the rating and the condition of the equipment.