In short. In a chemical or petrochemical plant, power is part of the process. A generator system there must be sized from a critical-load register built on the plant’s process-safety studies, must handle large motor starts without upsetting control systems, and must be located and protected with hazardous areas in mind. The aim is a supply that keeps the process safe and controlled during an outage, not simply one that matches connected load.
Why process plants are different
When power fails in an office, people wait. When it fails in a reactor building, cooling, agitation, inerting or scrubbing may stop, and the process can move towards an unsafe condition within minutes. A generator failure can therefore become a process failure, and an oversimplified generator solution can create new risk instead of removing it.
The supply chain to consider runs from the energy source through the generator and switchgear to motors, pumps, compressors and process utilities, and then to instrumentation and the PLC, DCS or SCADA system that keeps the process within limits. Weakness anywhere in that chain shows up as a process upset.
Identify the application first
Chemical sites in Telangana and Andhra Pradesh range from bulk drug intermediates and speciality chemicals to solvent recovery, paints, fertiliser blending and petroleum product storage. Each creates different electrical requirements.
- Batch reactors need cooling and agitation continuity, often with tight time limits set by the reaction chemistry.
- Continuous plants need orderly shutdown or ride-through, depending on what restarting costs in time, product and safety.
- Tank farms and terminals need power for pumps, fire water, gas detection and control rooms.
- Utilities such as chillers, compressed air and nitrogen generation support everything else and are often underestimated.
Start with process criticality
The first question is what happens to the process when electrical power disappears. Consequences can include loss of cooling, runaway reaction risk, loss of inert blanketing, release through vents, product loss in lines and vessels, or inability to execute an orderly shutdown. The plant’s approved process-safety and electrical studies determine criticality. A generator supplier should work from those studies, not invent them.
Build the critical-load register
Once criticality is known, every load is assigned to a category with its interruption tolerance. The register then drives generator capacity, transfer method and load-shedding logic.
| Category | Typical loads | Interruption tolerance | Usual supply arrangement |
|---|---|---|---|
| Safety and emergency | Only loads designated by the approved safety architecture, such as gas detection, emergency lighting, fire water controls | Zero to a few seconds | UPS or dedicated emergency source, backed by generator |
| Process critical | Reactor cooling pumps, agitators, scrubber fans, critical instrumentation | Seconds to minutes, as defined by the process | Generator via automatic transfer, priority in load sequence |
| Process supporting | Chillers, air compressors, nitrogen plant, cooling towers | Minutes | Generator, sequenced after critical loads |
| Business and operational | Packing, warehouses, offices, non-critical HVAC | Can wait or be shed | Shed during constrained generation |
Connected load is not generator demand
A plant may show several megawatts of connected load while its critical demand during an outage is a fraction of that. The sizing path runs from connected load to utilisation and diversity, then to critical demand, the largest motor or transient event, the load sequence, future expansion, redundancy and site derating. Our generator sizing guide covers the general method.
Sgen ≥ (Pcrit / PF) × kmargin
Here Sgen is the required generator rating in kVA, Pcrit is the coincident critical demand in kW from the load register, PF is the expected load power factor and kmargin is a design margin for growth and ageing, commonly in the range of 1.1 to 1.25. The result is then checked against motor starting, which often governs.
Example (illustrative): a critical demand of 400 kW at 0.8 power factor needs 500 kVA of running capacity. With a margin of 1.2, the starting point is 600 kVA, which is then tested against the largest motor start and the step loads in the sequence.
Motor loads often dominate
Process plants contain large pumps, compressors, blowers and agitators. A direct-on-line start can draw six to eight times rated current at low power factor, and the voltage dip that follows can trip contactors, drop out control relays or reset instruments. Selection must consider running demand, starting method, starting current, acceleration time, the driven equipment and the permissible voltage and frequency disturbance for sensitive loads.
Variable-speed drives and harmonics
Many modern plants run pumps and fans on variable-speed drives. Drives reduce starting stress but inject harmonic current, which a generator tolerates less well than a stiff grid transformer. Alternator selection, AVR type and harmonic mitigation should be reviewed together.
Hazardous areas and location
Where flammable vapours or gases may be present, the plant will have a hazardous-area classification under the IS/IEC 60079 series. Generators, their exhaust and their fuel tanks should normally be located outside classified zones, with exhaust outlets and air intakes positioned away from vents and release points. Fuel storage above certain quantities needs statutory approval, and the installation must coordinate with the plant’s fire protection. Our generator fire and site safety page sets out the general precautions.
Testing, maintenance and temporary power
A process plant generator that is never tested on real load cannot be trusted. Periodic on-load tests, load-bank testing when real load is unavailable, and documented transfer tests during planned shutdowns are part of the design. During turnarounds, rental power can keep critical services running while permanent equipment is overhauled, and our generator services team can plan maintenance around the production calendar.
Common mistakes
- Sizing from connected load. The set ends up oversized, runs lightly loaded and suffers wet stacking in an application that rarely tests it properly.
- Letting the generator supplier decide which loads are safety critical. Criticality belongs to the plant’s process-safety studies, and guessing can leave a hazard unprotected.
- Ignoring the largest motor start. The voltage dip trips control systems at the exact moment the plant needs them.
- Placing the set or its exhaust near a classified area. A hot surface or spark becomes an ignition source.
- Restoring all loads at once after transfer. The block load stalls the set or trips its protection.
What to send us
Send the critical-load register or a load list with ratings, starting methods and interruption tolerances, the single-line diagram, the hazardous-area drawing for the proposed generator location and site photos. If an existing set is involved, include its nameplate photo and recent run hours. Use the application engineering enquiry so an engineer reviews it from the start.
Questions engineers ask
Should the generator back up the whole chemical plant?
Rarely. Most plants back up safety, process-critical and selected supporting loads, and shed the rest. The critical-load register decides the split, and it usually produces a smaller, better-loaded generator.
Can a generator be installed inside a hazardous zone?
Standard generator sets are not designed for classified areas and should be placed outside them. If no alternative location exists, the design needs specialist review under the IS/IEC 60079 series. Moving the set is almost always the safer and cheaper answer.
How do we stop motor starts tripping the DCS?
Put the DCS and critical instruments on UPS, sequence large motors after transfer, and use soft starters or drives where the voltage dip is too large. The generator and alternator should also be checked for the transient they must carry.
Do drives cause problems on generator supply?
They can, because harmonic current distorts voltage more on a generator than on the grid. The alternator rating, AVR type and any filtering should be selected with the drive load known. Tell your supplier how much drive load will run on the set.
How often should a process plant generator be load tested?
Follow the manufacturer’s guidance and plant procedures, with regular runs on real load and periodic load-bank tests where real load is not available. Transfer tests are best done during planned shutdowns. Records of each test belong with the plant’s safety documentation.
Technical content reviewed by Srinivas Reddy, Managing Partner, Reddy Generators. Last updated 19 October 2026.