In short. Multi-megawatt temporary power is normally built from several generator modules synchronised onto a common bus, then transformed and distributed to the load. Modular plant offers staged capacity, maintenance flexibility, redundancy and load following, but only when the synchronising controls, bus architecture, protection and reserve strategy are engineered. Three generators are not redundancy if all three are needed to carry the load. Treat it as a power plant with a temporary address.
Why modular rather than one very large source
Above about 2 MW, a single rental machine is rarely the right answer. Large single units are hard to transport on Indian roads, a single failure removes all capacity, and a big set running at 30 percent overnight load burns fuel inefficiently and builds carbon. A plant of, say, four 1,000 kVA modules can follow the load by starting and stopping modules, allow one to be serviced while the others carry the site, and survive the loss of one unit if the reserve was designed for it. The typical arrangement is: generator modules, synchronising and load-sharing controls, common bus switchgear, step-up transformation where distance or site voltage requires it, then distribution to the load. Below this scale, the questions of cabling, fuel and site layout are covered under large generator rental.
Sizing modules and reserve
The number of modules follows the electrical duty, not a preference. There is no universal best number. Size the plant against the actual operating envelope: minimum overnight load, average load, peak load and the largest single step, plus the reserve needed to survive losing one module.
Modules required = ceiling(peak kW / (module kW x loading limit)) + spare modules for N+1
Peak kW is the highest sustained demand, module kW is the continuous rating at site ambient after derating, the loading limit is the fraction of rating the plant should normally run at (often 0.8 to 0.85 to keep transient headroom), and the spare count reflects the redundancy target. As an illustration (illustrative), a 2,400 kW peak with 800 kW modules and a 0.85 limit gives ceiling(2,400 / 680) = 4 modules for capacity, and N+1 makes it 5. A plant designed only around an 800 kW average would have looked adequate with three modules and failed at every peak. The N+1 power engineering guide sets out the reserve logic.
The common bus is a real power plant
Synchronising and sharing
An incoming module may not simply close onto a live bus. Its voltage must match, its frequency must be suitably aligned, the phase angle must be inside the closing window and phase sequence must be correct; the synchronising system checks these and permits closure only within the approved limits. Once on the bus, kW is shared through the governors and kVAr through the AVRs, and both need tuning so that no module carries more than its share or drifts into reverse power. The synchronisation guide covers the closing conditions.
Bus architecture
| Arrangement | How it works | Strength | Weakness |
|---|---|---|---|
| Single common bus | All modules feed one bus section | Simple, full sharing of capacity | A bus fault removes the whole plant |
| Split bus with coupler | Two sections joined by a bus-tie breaker | A fault on one section leaves the other running | More switchgear and protection logic |
| Separate islands | Module groups feed separate loads, no coupling | Independent failure domains | Capacity cannot be shared between islands |
A critical project should identify which infrastructure remains shared. Five modules on one bus, one fuel manifold and one master controller have common failure points that redundancy at the module level does not remove.
Fault level
Each module adds fault current. Five 1,000 kVA sets in parallel can contribute a prospective fault far above what a single set produces, and the bus switchgear, outgoing breakers and cables must be rated for that combined level. Conversely, when only two modules run overnight, the fault current is much lower, and protection must still trip reliably at minimum generation.
Defined response to the loss of a module
Every multi-megawatt temporary plant should state in its operating philosophy what happens when one module trips. With a properly designed reserve, the remaining modules pick up the load, frequency dips briefly and recovers, and a standby module starts and synchronises. Without that reserve, the right response is to shed lower-priority load fast enough that the remaining modules are not overloaded into a cascade trip. Shedding priorities are written before the event, and restoration is staged so that shed loads do not all return together once capacity recovers. A plant that sheds 20 percent of its load in a second can preserve the other 80 percent; a plant without shedding often loses everything. See the load shedding guide.
Distribution, cables and earthing at scale
Several megawatts at 415 V means thousands of amperes and many parallel cables, so most plants of this size generate at LV and step up to 11 kV through rental transformers, or use HV modules directly. Transformer energisation inrush must be within what the running modules can supply, which usually means energising transformers one at a time with downstream load off. Earthing is designed for the plant: which neutral points are earthed, how the HV side is earthed, and how earth-fault protection discriminates between modules. Copying an earth-electrode rule from a smaller job is not adequate.
Fuel, footprint and maintenance access
A 4 MW plant running continuously at average load can consume in the region of 20,000 to 25,000 litres of diesel a day, depending on engines and load factor. That requires bulk storage, transfer pumps with backup, day tanks at each module and filtration, plus a delivery schedule that survives monsoon road conditions. The layout must fit the site while retaining service access between modules, crane access for a module exchange, and radiator discharge that does not recirculate into the next module. Gas-fuelled alternatives for long-duration plants are discussed under temporary gas generation.
Availability is confirmed per project
We do not publish a standing multi-megawatt fleet figure. Plants of this size are configured for each requirement, and equipment availability, synchronising controls and any partner supply are confirmed against that project before commitment.
Common mistakes
- Counting modules against the average load. The plant cannot carry the peak and trips modules in sequence at the busiest hour.
- Calling a plant N+1 when every module is needed for normal load. The first module failure overloads the rest and the whole bus collapses.
- Rating bus switchgear for one module’s fault level. A bus fault with all modules running exceeds the breaker’s breaking capacity.
- Restoring all shed loads at once when capacity returns. The combined inrush trips the plant a second time.
- Feeding every module from one fuel manifold with one pump. A single pump failure stops all modules together.
What to send us
Send the load profile with minimum, average and peak demand and the largest steps, the required voltage at the point of connection, the project duration and operating pattern, and the consequence of losing part or all of the supply. Include a site plan showing available space, access and the connection point, and any existing HV switchgear or transformers. Use the generator rental enquiry to start the configuration review.
Questions engineers ask
How long does it take to set up a multi-megawatt rental plant?
It depends on transport, site preparation, cabling and commissioning rather than on the generators themselves. A straightforward LV plant on a prepared site can be assembled in days, while a plant with step-up transformers, HV cabling and a bulk fuel system usually needs weeks of planning and installation. Commissioning of synchronising and protection is never rushed.
Can a rental plant run in parallel with the grid?
Technically yes, with appropriate synchronising, interface protection such as reverse power and loss-of-mains detection, and DISCOM approval. Most temporary plants run as an island because grid paralleling adds approval time and protection requirements that a short project rarely justifies.
Why not hire one 3 MW set instead of several smaller ones?
One large set is a single point of failure, runs inefficiently at light load and is harder to transport and position. Modules give redundancy and allow units to be stopped when demand falls. A single large unit can still be right for a steady, non-critical load with good site access.
How is load shared between modules of different sizes?
Load-sharing controllers divide kW and kVAr in proportion to each module’s rating, so a 1,000 kVA and a 500 kVA module carry load in a two to one ratio. Mixing sizes is workable but complicates reserve calculations, so matched modules are preferred where available.
Generators by rating
- Under 15 kVA generators
- 15–50 kVA generators
- 50–125 kVA generators
- 125–250 kVA generators
- 250–500 kVA generators
- 500–1000 kVA generators
- 1–2 MVA generators
- 2 MVA and above generators
Technical content reviewed by Srinivas Reddy, Managing Partner, Reddy Generators. Last updated 19 October 2026.