A plain-language walkthrough of NEC Article 220 — general lighting loads, demand factors, and how a load calculation turns a building's connected loads into a required service size.
What Article 220 Actually Does
NEC Article 220 is the code section governing how to calculate branch circuit, feeder, and service loads — the procedure that turns "what's connected to this building" into "how big does the electrical service need to be." It exists because a building's total connected load (every outlet, fixture, and piece of equipment added up at face value) is almost always far larger than what actually gets used simultaneously — sizing a service to the full connected load would be enormously wasteful. Article 220's demand factors are the code-sanctioned way of accounting for that: statistically reasonable assumptions about how much of the connected load actually operates at once, applied so the calculated service size reflects realistic demand rather than a worst-case sum of everything installed.
The Basic Structure: Connected Load, Then Demand Factors
A load calculation generally works in two stages. First, tally the connected load by category — general lighting (calculated by a unit load per square foot, not by counting individual fixtures), receptacle outlets, fixed appliances, HVAC equipment, and any other significant loads. Second, apply the demand factors Article 220 specifies for each category — a demand factor less than 100% reduces a load category's contribution to the total, reflecting that not everything runs at once; some categories (like the first several kVA of general lighting in certain occupancies) are calculated at 100% before demand factors reduce the remainder.
General Lighting Load — Why It's Calculated Per Square Foot
Rather than counting and summing every light fixture individually, Article 220 specifies a general lighting unit load in volt-amperes per square foot, varying by occupancy type (dwelling units, offices, retail stores, warehouses, and other categories each have their own specified value). This reflects that lighting density is reasonably predictable by building use, and calculating by area is both simpler and more consistent than trying to anticipate every future fixture change over a building's life — a load calculation done at design time doesn't need to know exactly which fixtures will eventually be installed.
Demand Factors — Why Not Everything Counts at 100%
Demand factors are the mechanism that keeps calculated loads realistic. For dwelling unit general lighting and receptacle loads, for example, Article 220 applies a declining demand factor as the load quantity increases — the first several thousand volt-amperes count at full value, and the demand factor decreases for load beyond that threshold, reflecting the reality that a house with more square footage and more circuits doesn't use every single circuit simultaneously at full capacity. Appliance loads (dishwashers, garbage disposals, and similar fixed equipment) have their own demand factor tables when several are present in the same dwelling, again reflecting that not everything runs at once.
Feeder and Service Calculation — Putting It Together
Once each load category's calculated (post-demand-factor) value is determined, the calculation sums them to a total, then converts that to the minimum required ampacity at the building's service voltage. This final number — not the raw connected load — is what actually determines minimum service size, and it's typically substantially smaller than a simple sum of every connected device's nameplate rating would suggest, which is exactly the point of the demand factor structure.
Optional vs. Standard Calculation Methods
Article 220 provides more than one calculation method for certain occupancy types — a standard method (the full category-by-category process described above) and, for dwelling units specifically, an optional method that uses a simplified, more conservative demand factor structure in exchange for less calculation complexity. Both are code-compliant; the choice is generally a matter of which produces a more favorable (or more straightforward to calculate) result for a specific project, not one being inherently more "correct" than the other.