The dividing line is duty, not size. How load profile, loss economics and impedance decide which type a job actually needs.
Photograph: Generated (Gemini) · Pexels
The terms get used loosely, and the usual explanation — that power transformers are simply the big ones — is not quite right. It is close enough to sound reasonable and wrong often enough to cause a bad specification.
The real distinction is duty: where the unit sits in the network, and how it is loaded across a day.
A distribution transformer is the last transformation before the load. It sits on a pole, a plinth or in a small substation, and it feeds final consumers — houses, shops, a small works. Its load varies continuously and unpredictably, and for a large part of every day it is lightly loaded.
A power transformer sits between voltage levels further up: at a substation, at a large plant intake, or between two parts of a network. It feeds other transformers or a substantial single load rather than end consumers, and it typically runs at a higher and steadier fraction of its rating.
The consequence runs through loss economics.
Every transformer has two kinds of loss. No-load loss is the core loss — it exists whenever the unit is energised, regardless of load, and it is set by core material and geometry. Load loss is the copper loss, which rises with the square of the current.
A distribution transformer spends most of its life lightly loaded, so no-load loss dominates. It is energised 8,760 hours a year and it is paid for every one of them, whether or not anybody is drawing power. This is why distribution transformer design concentrates on the core.
A power transformer runs closer to its rating for longer, so load loss dominates and the economics shift to the windings.
Specify one as if it were the other and you get a unit that is technically functional and economically wrong.
Percentage impedance does two jobs. It sets the fault level that downstream switchgear must interrupt, and it determines how load shares between units running in parallel.
Lower impedance gives better voltage regulation but a higher through-fault current. Higher impedance limits fault current at the cost of regulation.
If a transformer will run alongside an existing one, its impedance must match closely. A mismatch means the two circulate current between themselves instead of sharing load — both run hot, neither delivers its rating, and the cause is rarely obvious from the outside.
Because a power transformer sits closer to the source, the fault current it can see is higher, and the mechanical forces on the windings during a through-fault scale with the square of that current. Clamping and bracing carry more of the design effort than they do on a distribution unit.
Distribution units are almost always ONAN — Oil Natural, Air Natural. The oil circulates by thermosiphon, the radiators lose heat to still air, and there is nothing to fail.
Power transformers may be specified ONAN/ONAF, where fans raise the continuous rating of the same tank when needed. That gives a dual rating, and it is worth having where load is expected to grow, but it also introduces fans, controls and a maintenance obligation that did not exist before.
Ask what the transformer is feeding:
Then check three things before you commit:
Whichever type the job needs, the routine test certificate is what turns the specification into something verifiable. Ratio, winding resistance, insulation resistance, no-load loss and load loss should all arrive with the unit. Keep them — they are the commissioning baseline that every later maintenance reading gets compared against, and without them you are guessing at whether a change is real.
Related: Oil cooled distribution transformers · Power transformers · Understanding transformer losses
Enquiries
Send us the rating, voltage class and site conditions. We would rather answer a technical question than send a brochure.
