Load test on a single-phase transformer
Direct load test on a single-phase transformer: connections, step-by-step procedure, tabular column, efficiency and regulation calculations, and viva questions with answers.
Aim
To conduct a direct load test on a single-phase transformer and determine its efficiency and voltage regulation at various loads.
Apparatus required
| Apparatus | Specification | Qty |
|---|---|---|
| Single-phase transformer | 1 kVA, 230/115 V | 1 |
| Single-phase autotransformer (variac) | 230 V, 0–270 V, 8 A | 1 |
| Resistive load bank | 115 V, 10 A, step-loaded | 1 |
| Voltmeter (MI) | 0–300 V and 0–150 V | 1 each |
| Ammeter (MI) | 0–5 A and 0–10 A | 1 each |
| Wattmeter (UPF) | 300 V, 5 A and 150 V, 10 A | 1 each |
| Connecting wires | Adequate rating | As required |
Theory
The load test measures efficiency instead of deducing it: the transformer is actually loaded, and the input and output powers are read directly from two wattmeters. Efficiency is the ratio of the two, and everything the transformer loses appears as the difference.
Voltage regulation is measured at the same time. The secondary terminal voltage is recorded on no load — that is E₂ — and again at each load step. The fall between them, as a percentage of E₂, is the regulation at that load.
The output falls short of the input by the sum of the core loss, which stays constant, and the copper loss, which grows as the square of the current. That is why efficiency rises steeply at light load, peaks where the two losses are equal, and falls slowly after it.
The method is honest and simple, but it wastes the full output power as heat in the load bank and needs a supply capable of delivering it. Beyond a few kVA it becomes impractical, which is why the OC/SC method exists.
Circuit connections
Check every point below against your board before switching on. There is no diagram here on purpose — a wrong diagram is worse than none, and this is the list a demonstrator actually walks through with you.
- Supply through the variac to the primary, via the HV-side ammeter and the current coil of wattmeter W₁. Voltmeter V₁ across the primary terminals.
- Secondary to the load bank through the LV-side ammeter and the current coil of wattmeter W₂, with voltmeter V₂ across the secondary terminals.
- Both wattmeter pressure coils connected across their respective sides, on the load side of the current coil.
- Load bank switches all OFF and variac at zero before switching on.
Procedure
- 1Complete the connections and have them checked. Keep the load bank switched off entirely.
- 2Switch on and raise the variac until V₁ reads rated primary voltage. Record V₂ with no load connected — this is E₂, and every regulation figure depends on it.
- 3Switch on the first step of the load. Record V₁, I₁, W₁, V₂, I₂ and W₂ together, as one set.
- 4Increase the load in equal steps up to rated secondary current, recording a full set at each step. Readjust the variac if V₁ sags, so that the primary voltage stays at its rated value throughout.
- 5Switch the load off step by step, bring the variac to zero, and switch off.
- 6Compute efficiency and regulation at each step, and plot both against output power.
Work out your readings
Type in the numbers off the meters. This fills the tabular column, works the calculation through step by step, plots the characteristic — and tells you when a reading cannot physically be right, which is the part a manual can't do. Everything stays on this device, and it works with the network off.
Nameplate and machine data
Measured before any load is switched on
Load readings
One row per step of the loading rheostat, from no load up to full load.
| # | V₁(V) | I₁(A) | W₁ in(W) | V₂(V) | I₂(A) | W₂ out(W) |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | ||||||
| 3 | ||||||
| 4 | ||||||
| 5 |
Fill in the machine data and at least one complete row of readings to see the results, the worked calculation and the curve.
Precautions
- Keep the primary voltage constant at its rated value at every load step. If V₁ is allowed to sag, the core loss changes and the efficiency curve is meaningless.
- Never exceed the rated secondary current — the load bank will happily draw more than the transformer can give.
- Record E₂ before any load is switched on. Taking it later, with a step still connected, is the most common cause of a wrong regulation figure.
- Read all six meters at the same instant, not one after another while the load heats up and drifts.
- The load bank gets hot. Do not touch the elements, and leave the fan running if there is one.
Sources of error
Every record asks for these, and every record gets the same three lines copied from the one before. These are the errors this particular experiment actually has.
- Wattmeter and voltmeter loading: the meters themselves draw power, which is counted in the input but never reaches the output.
- The load resistance rises as the elements heat, so a step drawn early in the experiment is not the same load ten minutes later.
- MI instruments are accurate to a class of 1.0 or 1.5 of full scale, so a reading taken at a quarter of scale carries four times that percentage error. Choose ranges that put readings in the upper half of the scale.
- Frequency and supply-voltage variation on the bench during the run.
Viva questions with answers
Why is the direct load test not used for large transformers?
Because it dissipates the whole rated output as heat in a load bank, and needs a supply able to deliver it. For a 100 kVA transformer that is 100 kW wasted for the duration of the test. The OC and SC tests need only the loss power, a few percent of that.
Which is more accurate for a large transformer — this test or the OC/SC method?
The OC/SC method, counter-intuitively. Efficiency here is the ratio of two large, nearly equal numbers, so a 1 % error in either wattmeter becomes a large error in the difference. The indirect method measures the small loss quantities directly, and the error stays small in absolute terms.
Why must the primary voltage be held constant throughout?
Core loss depends on the applied voltage, not the load. Letting V₁ sag as the load increases quietly changes the constant loss from step to step, so the points no longer lie on a single efficiency curve.
What does the efficiency curve look like, and why?
It rises steeply from zero at no load, peaks where copper loss equals core loss, and falls gently after. The steep rise is the fixed core loss being spread over a growing output; the gentle fall is copper loss growing with the square of current.
Why is regulation measured against the no-load secondary voltage rather than the rated value?
Because regulation is defined as the change the load causes. E₂ on no load is the reference the transformer starts from; the rated nameplate voltage is a design figure, and the actual no-load voltage of a particular unit at a particular supply voltage will differ from it.
What kind of load gives the worst regulation?
A lagging power factor load, and the lower the power factor the worse it gets — the reactance drop then adds almost directly to the resistance drop. A leading load gives the best regulation, and can even make it negative.