Resistive load testing applies real power, measured in kW, at approximately unity power factor. Resistive-reactive testing adds reactive load so the generator can operate at a specified non-unity power factor and kVA point.
The choice in resistive vs resistive-reactive generator load testing starts with what you need to prove. Two tests can show the same kW yet place different demands on the alternator. Current and excitation can differ. Neither label proves that the whole installed system meets its requirements.
Illustrative scenario: A buyer receives an offer for a “full-load test.” Does that mean engine kW, alternator kVA, or both at the stated power factor? The offer does not say. This guide shows how to ask for a clear scope and useful proof.
Key Takeaways
- Read the duty rating, kW, kVA and power factor together.
- A resistive test exercises the alternator at the applied point; it does not leave it untested.
- Combined loading can establish a specified lagging-power-factor point when correctly selected and measured.
- Test records must identify achieved conditions and remaining exclusions.
For a project enquiry, send the nameplate and required test objective with your specification.
Read kW, kVA and power factor before comparing tests
Start with the rating basis. “100% load” needs a named reference. Is it a share of rated kW or rated kVA? Write that basis into the order so both parties know what the test must reach.
Real power, in kW, becomes work or heat. Apparent power, in kVA, describes the load based on voltage and current. Power factor is kW divided by kVA. Reactive power, in kVAR, has a direction: inductive loads draw lagging reactive power, while capacitive loads have leading reactive power.
These values describe parts of the same load point. They are not just different names for the set’s size. For this simple steady-state example, divide kW by PF to find kVA. Use measured values to review a real test.
A sourced rating example
Avtron’s resistive-reactive testing explanation uses a 500 kVA generator rated at 0.8 power factor. Its corresponding real-power rating is 400 kW: 500 multiplied by 0.8. This is an engineering illustration, not a Huali model specification.
| Quantity | Resistive point in this example | Rated-PF point in this example |
|---|---|---|
| Real power | 400 kW | 400 kW |
| Power factor | 1.0, idealized | 0.8 lagging |
| Apparent power | 400 kVA | 500 kVA |
| Share of rated kVA | 80% | 100% |
At the same rated terminal voltage, current follows the kVA ratio. The resistive point thus draws 80% of rated current in this example. That does not make the test “80% complete.” It also tells you no such percentage for heat losses.
The extra kVA is not extra rated engine output. Both columns still show 400 kW. Read them as two load points for the same example, not as two different generator sizes. A report that gives only kW leaves that difference hidden.
Keep the assumptions beside the calculation
This example assumes the stated rating and stable voltage under the agreed conditions. Do not raise kW beyond the engine’s allowed rating to chase a kVA figure. Adding reactive load is a different action. It must stay within the unit’s limits.
Other sets have different rating bases. Check the unit’s documents; don’t treat 0.8 as a rule for all sets. Put rated PF and measured PF beside each full-load percentage in the generator load-test report. This tells the reviewer what the result can prove.
What resistive generator load testing proves
A resistive bank provides a real-power load you can control. It loads the engine and exercises its fuel, cooling and exhaust systems under the test conditions. Record what happened. The kW, time, temperatures and observations give the result its meaning.
The engine responds to load over time, not a label. A point listed on a schedule is only a plan. Did the test reach and hold it? Record any stops, load cuts or abnormal conditions; don’t report the planned maximum as though it was measured.
The alternator still operates
The alternator supplies voltage and current during a resistive test. Voltage control works at that point too. So “the alternator is not tested” is misleading. Ask instead whether the test reaches the electrical point the contract requires.
In the example, rated engine kW at unity PF does not reach rated alternator kVA. That result may meet the agreed engine-load scope. A separate rated-PF point may still be missing. Approval of one point does not prove the other.
Resistive loading can also form part of a load-step test. The applied step and recorded response matter. A steady reading cannot reveal a brief voltage or frequency dip that the instruments missed. Agree dynamic criteria separately; peak kW alone cannot prove them.
The report should distinguish a held load from a changing load. Both may be useful, but they answer different questions.
Keep the result within its boundary
A factory test may use temporary connections. It cannot then prove the site’s final fuel supply, airflow or full power path. Nor does one ambient condition cover every site environment. State these limits in the report so buying and site teams read the result the same way.
Return to the quotation scenario: if the agreed objective is engine kW loading, a resistive test may be appropriate. If the specification also asks for rated kVA at lagging PF, request that additional point explicitly. Use the broader generator commissioning procedure to place each test within the project sequence.
What resistive-reactive generator load testing adds
A combined resistive-inductive bank adds lagging reactive load to the kW load. A suitable system can then reach the required kW and kVA at non-unity PF. Check the bank’s limits first. It needs enough kW and kVAR at the test voltage and frequency.
Reactive loading changes the alternator’s current and excitation requirements. It can therefore expose electrical operating conditions that a unity-PF test at the same kW does not reach. ASCO’s technical paper on kVA testing explains the separate load components and their combined use. Its historical standards discussion is not a substitute for the project’s applicable requirements.
Record electrical and thermal conditions
At the lagging-PF point, log voltage, current, kW, kVA, kVAR and PF. Add the required temperatures and observations over the agreed time. These are actual test records. A bank’s advertised capacity alone cannot show that the set reached the point.
One full-load screenshot is not proof of thermal acceptance. It shows a moment. It cannot show how long the load stayed there or whether heat levels settled as required. Choose a recording method that can reveal limit breaches as well as the peak load.
Lagging-PF evidence also does not prove leading-PF capability. A resistive-inductive arrangement is not a capacitive test. Keep the manufacturer’s applicable operating limits and the requested load direction in view before defining a combined-load point.
Keep measured current separate from a calculated estimate. If the record omits voltage or PF, ask for the missing channel rather than filling it from the nameplate. The nameplate gives a rating. It does not prove the load reached during a test.
Separate real and reactive sharing
For parallel sets, review both kW sharing and reactive-load sharing. The bus total does not show each set’s share. Ask for per-unit readings in the agreed modes. A combined bank alone cannot prove every control setting.
Illustrative scenario: A parallel-generator test schedule asks for equal kW sharing but contains no per-unit kVAR or PF record. The buyer should ask how reactive sharing will be assessed. The missing scope is a question to resolve, not evidence that the equipment has passed or failed.
A balanced bank also does not establish acceptable operation with every phase imbalance. Refer to the three-phase generator load-balancing guide when the actual installation introduces unequal phase loads. Keep that assessment separate from the comparison between unity and lagging PF.
Choose the load bank from the question you need answered
Start with the result you need to review. Then choose the load, readings and test boundary that can provide it. More equipment does not always mean more useful proof. The scope needs to answer your question.
| Objective | Suitable loading approach | Evidence to request | What remains unproved |
|---|---|---|---|
| Engine real-power loading | Resistive kW at the agreed points | Achieved kW, duration and engine observations | A separately required non-unity-PF point |
| Rated-PF electrical loading | Combined load matched to specified kW and kVA | kW, kVA, kVAR, PF, voltage, current and relevant temperatures | Other PF directions and untested conditions |
| Parallel load sharing | Loading that exercises the specified real and reactive objectives | Per-unit and bus records under agreed arrangements | Unexercised control modes and protection functions |
| Transient response | Defined load changes with suitable recording | Time-related voltage and frequency response | Other step sizes or actual load-start profiles |
| Nonlinear or leading loads | A specifically engineered test arrangement | Agreed waveform or leading-PF evidence | Coverage inferred from an ordinary R/L bank |
| Installed-system transfer | Test scope including the required transfer path | Switching sequence and required supply observations | Components bypassed by temporary connections |
Read each row as a separate scope choice. A single project may need more than one row, but the report should say which questions each test answers. This avoids asking one static load point to stand in for a whole site acceptance programme.
Check compatibility before accepting the proposal
Ask the bank supplier to confirm:
- Available kW and kVAR at the intended voltage and frequency.
- Load increments that can reach each proposed point.
- The method for measuring voltage, current and PF.
Check each point, not just the maximum rating. A bank can have enough total capacity yet lack the steps needed for the agreed schedule.
The connection point changes the conclusion too. Testing at the terminals covers a different path from testing through the final distribution system. Ask the responsible engineer to define the boundary, connections and safe airflow clearances. This guide does not provide wiring instructions.
For equipment-setup context, Crestchic provides the following load-bank commissioning video. It illustrates a manufacturer’s equipment context; it does not establish the acceptance criteria for your generator or demonstrate every combined-load condition discussed here.
Illustrative scenario: A buyer asks whether an R/L test replicates a nonlinear UPS load. The equipment label alone cannot answer that question. Ask which current waveform and operating behavior the proposed test reproduces, and which need a separate assessment. Do not substitute PF alone for that evidence.
For a generator project discussion, share the unit rating, application and requested load points. State the evidence you need alongside the load-bank type so the enquiry remains specific.
Write the test boundary and evidence into the acceptance scope
Link each objective to a test point and an agreed record. “Resistive-reactive test included” only names equipment use. What result would earn approval? Agree the limits and who will review them before the test starts.
Use the generator inspection and test plan to assign review, witness and hold responsibilities. Keep the detailed results in the test record rather than turning the equipment-selection note into a second commissioning procedure.
Define the requested conditions
- Identify the generator, engine, alternator and relevant build configuration.
- State the applicable duty rating, voltage, frequency and rating conditions.
- Specify load points in kW and kVA, with PF magnitude and direction.
- Agree steps, durations and any separate dynamic-recording requirements.
- Name acceptance limits and their approved document or manufacturer source.
- Identify the connection boundary, ambient conditions and measurement equipment.
This is a scope guide, not a list of legal limits. If the contract names a standard, agree which edition applies and how the project will use it. Involve the responsible parties. Don’t copy a limit from another rating, site or old white paper.
Ask for the load-point schedule before agreeing the test. It lets the buyer check the bank proposal against the unit rating while changes are still practical. A test cannot settle a limit that the parties never agreed.
Separate requested, recorded and accepted
Keep planned setpoints separate from measured values. Add required phase readings, instrument IDs, calibration evidence, temperatures, alarms and exceptions. Does measured PF match the request? Explain any mismatch before claiming the required electrical point.
| Record issue | Review action |
|---|---|
| kW present, PF absent | Request PF and kVA evidence before judging the rated-PF point. |
| Load point reached only briefly | Compare the recorded time with the agreed duration. |
| Temporary connection bypasses equipment | Mark that equipment outside the proven boundary. |
Match each conclusion to its test record. State what was not tested. If a result falls outside an agreed limit, record the issue and its disposition. Don’t redefine “full load” after the test to make the result fit.
In the parallel-sharing scenario, the scope should name the per-unit reactive measurements and review method. A kW-only table cannot answer the missing question. That conclusion concerns the adequacy of the evidence, not an assumed defect in the generator.
Use the load-test report review guide for the fuller record checklist. The practical distinction is simple: the proposal states what will be tested, the report states what happened, and the acceptance review decides whether that evidence meets the agreed requirement.
Questions buyers ask about resistive and combined load tests
Can a resistive test prove rated generator kVA?
It proves the kVA reached at the measured voltage, current and PF. Does the set have a non-unity rated PF? If so, unity-PF loading at rated kW does not reach rated kVA. Check the rating basis before claiming that this test proves the unit’s full electrical rating.
Is a 0.8 power factor mandatory for every generator test?
No: use the unit’s rating and approved test scope to choose the required point. The 0.8 PF here is part of the sourced example, not a rule for all sets. State PF magnitude and direction. Then compare the readings with that point instead of assuming one value always applies.
Does adding reactive load replace the required engine kW?
No: adding reactive load changes the electrical point; it does not replace required kW. A test could reach the requested kVA at a different kW from the agreed point. Read kW, kVA and PF together. One headline figure can hide a load point that the test did not reach.
Can a combined load bank run a resistive-only test?
Check the bank’s design and controls. Ask whether the resistive and reactive sections can run as required, then confirm the load steps at the test voltage and frequency. The phrase “combined load bank” is not enough. It does not prove that a particular unit has a resistive-only mode or the PF range you need.
Does R/L testing prove motor starting or UPS harmonics?
No, not by the label alone. A steady R/L point does not mimic every motor start or nonlinear current waveform. Define the required load behavior, readings and limits for each question. A combined bank may help with a planned test, but the load profile and recorded response provide the proof.
Which evidence belongs in the purchase specification?
Name the unit, rating basis, load points, PF direction, steps and test times; state the readings, limits and their sources. Define the test boundary and who provides the report. Ask for actual results and open issues beside the planned points. The bank type supports that scope; it cannot replace it.
Agree what the result will prove before testing
Choose resistive loading when its real-power operating points answer the agreed question. Add the required reactive capability when the scope needs a specified non-unity-PF kVA point or reactive-sharing evidence. In either case, judge the result by achieved conditions, measurements and approved limits.
Compare clear scopes side by side. One may cover engine loading; another may add rated-PF loading. Keep the remaining limits visible. Untested motor starts, nonlinear loads, transfer paths and site conditions still need their own proof.
In the opening illustrative scenario, the next step is to replace “full-load test” with the rating basis, requested points and evidence list. That gives the buyer a reviewable proposal without assuming a test outcome.
For your generator enquiry, send the specification, nameplate and required test evidence. A clear acceptance question makes the discussion more useful than a load-bank label alone.
