Load bank testing
A load bank test puts a controlled, measured load on a UPS so you can see how it holds its output at known load steps and through transfers, while the protected equipment stays on a separate path. This section explains what the test proves, how it runs, what the report should show and where its limits are.
It proves that the UPS power train carries a known, steady load at each step and keeps its output in order while it does. Vertiv describes load bank tests as the usual way to confirm that a UPS meets its own specification and operates correctly as a system. Rectifier, inverter, static switch, contactors, busbars and connections all run warm at full load, which is when a weak joint or a tired component shows itself.
ASCO's load bank paper puts the reason plainly: the major reason a generator or UPS fails to perform is inadequate or improper maintenance, and after maintenance or repairs a load bank verifies the design and functional capability of the power system. A test is evidence you can file, not an opinion that the unit looked fine.
The critical load is moved off the UPS first, then the load bank is connected to the UPS output and the load is applied in steps. In Eaton's published method for UPS load bank testing, the load goes to maintenance bypass, the UPS modules are shut down and locked out, an electrician connects the load bank to the output, and the UPS is restarted before any test load is applied. The load then runs at 50% for one hour and at 100% for 3 hours, with voltage, current and regulation watched throughout.
A short battery discharge follows (Eaton recommends up to 4 minutes, by opening the rectifier input), the batteries recharge, and the UPS is transferred to bypass and back several times before the load bank is removed and the critical load returns to normal. The full sequence, with what is recorded at each stage, is on the procedure and requirements page.
A load bank lets you prove the system without putting the protected equipment at risk, which is why ASCO calls it the best practical means of proving a power system without interrupting critical loads.
A good report lets someone who was not on site see what was tested, how, and what was found. Eaton's scope says plainly that a report is provided documenting the results; its performance-test scope adds thermal and power-quality readings. Look for:
It does not prove battery runtime. The battery discharge inside a UPS load bank test lasts minutes, enough to show that the batteries pick up the load and that the UPS behaves while they do, but not how long they would carry it. Runtime is proven by a battery capacity test, for which battery makers point to the IEEE procedure, and tracked between tests with internal ohmic readings. See UPS and battery load bank testing for how the two fit together.
It also does not test the distribution downstream of the UPS, or the generator. During the test the critical load runs on bypass, and on some UPS the batteries are not available while the unit is in maintenance bypass, as Schneider Electric notes for the Galaxy VS. The test window is therefore a period of reduced protection, and it should be planned like one.
Some large UPS can. Vertiv's safe capacity test on the Liebert Trinergy Cube and Liebert EXL S1 circulates power from the inverter back to the rectifier through the static bypass, so the power train is loaded without a load bank on the output. Vertiv states that with double-conversion efficiency above 96%, less than 4% of the circulating power is dissipated, and the test can run up to an overload of 120%. Its battery discharge mode back-feeds the utility, which the site must accept first.
Every test follows a written method agreed before anyone arrives. The test plan sets whether the site's electrical infrastructure can carry the critical load and the test load at the same time, the bypass sequence, where the load bank and its cables go and where the heat is rejected, who signs off each step, and what happens if a reading is out of range. The report includes the load steps, the readings and the findings, written for whoever owns the UPS.
In Eaton's published UPS method the load runs at 50% for one hour and at 100% for 3 hours, followed by a short battery discharge, recharge and repeated transfers. Add the time to move the load to bypass, connect and remove the load bank, so plan a working day.
Yes, on the bypass path. The UPS is taken out of the load path and the load bank is connected to its output, so the test load never reaches the protected equipment. While on bypass the load does not have UPS protection, and on some UPS the batteries are not available either, so the window is planned.
No. The load bank test proves the UPS power train at known load steps; its battery discharge lasts minutes. Battery runtime needs a capacity test, run to the battery maker's procedure.
On some large UPS, yes. Vertiv's safe capacity test on the Liebert Trinergy Cube and EXL S1 circulates power inside the UPS, so no load bank is connected to the output.
Sources, checked 2026-10-04:
Tell us the site, the UPS and the timing. We reply and quote within 4 business hours, or call 1 (438) 800-0382
Brand names are trademarks of their owners and are used only to identify the equipment we service. We are not the manufacturer and are not affiliated with it.