# A Generator Sized for the Plant's Running Load Was Never Sized for the Load That Actually Starts It Author: Nitin Verma Author URL: https://insights.spans.co.in/author/nitin-verma Published: 2026-10-11 Meta Title: Standby DG Set Sizing for a WWTP | Spans Meta Description: A standby generator correctly sized against a treatment plant's total running kW can still stall the moment it tries to start the plant's largest blower or pump. URL: https://insights.spans.co.in/standby-generator-sizing-starting-load-wastewater A standby DG set for an [ETP or STP](https://spans.co.in/effluent-treatment-plant-etp/) usually gets specified off one number: add up the running load of every blower, pump, panel, and light the plant needs during a grid outage, apply a comfortable margin, and size the generator to that total in kW. It is a reasonable-looking calculation, it produces a defensible number for the purchase order, and it answers a real question. It does not answer the question that actually determines whether the generator works the first time it is needed, which is not how much power the plant needs once everything is already running, but whether the generator can get everything running in the first place. ## Running load and starting load are different questions A motor started direct-on-line does not draw its full-load current the moment it is switched on. It draws a surge well above that, commonly cited as somewhere in the range of five to eight times full-load current, for the few seconds it takes the rotor to accelerate from standstill to running speed, though the exact multiplier is specific to that motor and worth getting from the manufacturer's starting data rather than assumed from a generic figure. A generator sized against steady running kW was never asked to answer whether it can supply that transient surge without its own output voltage sagging so far that the starting motor cannot develop enough torque to actually accelerate, a failure that shows up as the engine bogging down, the motor stalling, or protection relays tripping, none of which has anything to do with whether the generator's running capacity was correctly calculated. This is not a subtle distinction once you see it, but it is an easy one to miss on a spec sheet, because [inrush current](https://en.wikipedia.org/wiki/Inrush_current) and running current both get measured in amps and both relate to the same motor, so a reviewer scanning a generator sizing calculation for "does the kW number look right" can walk straight past the fact that nobody checked whether the alternator can actually absorb the starting transient of the single largest motor on the panel. ![A bar chart comparing two scenarios on a generator: on the left, total running load sits comfortably below the generator's rated capacity line; on the right, the same running load stack is topped by a tall red starting-surge spike that blows well past the rated capacity line, labelled as the last motor starting on an already-loaded bus](https://prod.superblogcdn.com/site_cuid_ckxsj7b8y397701kn8deih6cyt/images/dg-set-starting-load-inline-1791682348303-compressed.png) ## Why the single biggest motor is the real design constraint A generator sized exactly to the sum of every running load plus a flat percentage margin can still be undersized for the plant's actual worst case, which is not everything running steadily together, it is the single largest blower or pump starting direct-on-line at the exact moment the rest of the plant's running load is already sitting on the generator's bus. That scenario, the last and biggest motor trying to start against a bus that is already carrying everything else, is the worst-case starting event the generator will ever face, and it has essentially nothing to do with the total running kW figure that drove the original sizing decision. What makes this failure mode particularly stubborn is that it stays invisible through almost every routine check a plant normally runs. A monthly DG test that starts the generator on no load, or brings up the plant's loads one at a time in a sequence the technician happens to choose, never reproduces the specific worst-case starting event. The generator can pass a dozen scheduled tests and still fail the first time a real grid outage forces every motor that was running when the grid dropped to restart in whatever order the plant's own control logic happens to bring them back, which may well include the largest blower trying to start last, against a bus already loaded with everything else that came back first. Picture a plant where the aeration [blowers](https://insights.spans.co.in/twin-lobe-vs-tri-lobe-blower) are the single largest motors on site, sized generously for the biological process and running direct-on-line because nobody expected the starting current to matter once the generator's kW rating cleared the running load with room to spare. The grid fails during a normal operating day, the plant's automatic transfer switch brings the generator online, and the control logic restarts loads in the order it was configured, smaller pumps and panels first, the blowers last, by which point the generator's bus is already carrying everything that came back ahead of them. The blower's starting surge lands on a generator with little headroom left, the voltage sags further than the blower motor can tolerate, and the motor stalls rather than accelerating to speed, tripping on overcurrent protection within seconds. The generator itself never faults. The kW rating was never wrong. The plant simply asked it a question, start this motor against this bus, that the original sizing calculation never considered. We have written before about how a [pump or blower foundation](https://insights.spans.co.in/pump-blower-foundation-mass-design-wastewater) can satisfy its footprint specification while failing the actual dynamic question the foundation exists to answer; a generator sized only against running load is the same shape of mistake wearing an electrical engineering costume. Two numbers that look like they are answering the same question, total kW required versus the transient the alternator must actually survive, are in fact answering different questions, and a spec sheet that only checks one of them can be fully compliant with itself while still failing on day one. ## What actually needs checking The fix starts with treating motor starting data as a required input alongside running load, not an afterthought: locked-rotor current or starting kVA for the largest motors on the panel, obtained from the motor manufacturer rather than assumed from a generic multiplier, because the real figure varies by motor design and the generic rules of thumb found in vendor literature disagree with each other by a wide margin. With that data in hand, the generator can be sized against the actual worst-case starting event, the largest motor starting last against an already-loaded bus, rather than against running load alone. Often the cheaper fix is not a bigger alternator at all. Reduced-voltage starting, a soft starter, or a [variable frequency drive](https://en.wikipedia.org/wiki/Variable-frequency_drive) on the largest motors cuts the starting surge dramatically, which can bring a plant's generator requirement down rather than up, and sequencing motor starts so the largest loads never have to start simultaneously against an already-loaded bus avoids the worst-case scenario entirely rather than sizing the generator to survive it. Whichever approach a plant chooses, the only way to actually verify it works is a motor-start voltage dip study run against the real starting data, not a vendor's rule-of-thumb sizing table, since the margin between a generator that handles the start and one that stalls on it can be narrower than those generic multipliers suggest. A DG spec sheet that lists only total running kW has answered how much power an [ETP or STP](https://spans.co.in/sewage-treatment-plant-stp/) needs once everything is already on. It has not answered whether the generator can actually get there, and those are different engineering questions that happen to produce numbers in the same unit, which is exactly why one so easily gets mistaken for the other. If you are reviewing a DG sizing calculation and want a second opinion on whether the worst-case starting event has actually been checked, that is worth raising before the next outage finds the gap for you: [spans.co.in/contact](https://spans.co.in/contact/). --- This blog is powered by Superblog. Visit https://superblog.ai to know more. ---