In many gas compression applications (e.g., a pipeline station or an upstream application), equipment operators have to determine whether to perform a required compression duty with a single large unit, or several smaller units in parallel or series. In many cases, conventional wisdom concludes that for performance and capital expense reasons, a single, larger unit will have a higher efficiency, will cost less and will operate cleaner.

There are certainly applications where this assumption holds true. LNG liquefaction projects, for example, seem to become bigger and bigger every year with gas turbine drivers now exceeding 100 MW per unit. For this type of application, the economy of scale actually works.

But LNG liquefaction is essentially a refrigeration application and operates virtually at constant load. Even in this case, users usually install multiple turbomachinery trains, but this is often caused by the maximum available size of gas turbines.

On the other hand, we find many compression facilities, especially in oil and gas upstream and midstream applications, where the load and operating duty cycle is divided among multiple units.

The basics

Many installations in pipeline compression, gas storage and gas plants, operate at widely varying loads. This is primarily due to flexible demand or supply. Also, the power capability of a gas turbine increases with lower ambient temperatures. Even at constant power demand, the relative load of the gas turbine changes. On the other hand, a gas turbine operates at its maximum efficiency when it runs at full load.

Multiple smaller gas turbine units allow a strategy of shutting down one of the units and operating the remaining units near full load if the load drops or the ambient temperature becomes lower. This reduces fuel consumption.

The advantage holds even if larger units have a higher full load efficiency than smaller units which is not necessarily the case. This is one of the reasons why pipeline optimizations have to take into account equipment efficiency at different loads and at different ambient conditions.

Many operators enjoy the increased flexibility that comes with the capability to fine tune equipment to operating scenarios by shutting units down. Small changes are often easier to accomplish with properly controlled, multiple driver units on small gas turbines. For modern control systems, the control of multiple units — even differently configured multiple units — is not a challenge.

Besides the advantage of having lower fuel consumption from the flexibility of operating smaller units, there is also the maintenance aspect: maintenance cost is determined by operating hours (fired hours) of individual compressor packages.

If we can shut down one of the units, either because the pipeline sees a lower demand or the ambient temperature has dropped, this unit will not accrue fired hours. Therefore, maintenance cost for the station is reduced. By selectively shutting down units for required maintenance and overhaul activities, one can also reduce the impact of scheduled outages on facility availability.

Further, a large gas turbine unit operating in deep part-load will often have higher emissions for both NOx and CO2 per power produced when compared with smaller units running at full load.

And then there is the issue of spare units. To reach near 100% station availability, many users install spare units. If they have to install spare units (N+1), having multiple small units requires less installed power while maintaining a given availability goal. This also holds if the concept is to install spare power, rather than spare units (that is, all installed units are oversized, so they can pick up the load if one unit fails).

Similar arguments can be made if the compressors are installed in phases, either because the demand for gas is ramped up as the supply of gas increases over time or there is uncertainty about future conditions. Smaller units allow better distribution of expenses over time as additional power can be purchased in smaller blocks. Considering overall expenses, there may be little incentive to do this; but based on net present value calculations, cost advantages can be significant.

Cost structures

The argument is often that the installation cost is higher for multiple small units versus one bigger unit. This depends largely on the cost structures in the region where the units are installed. Certain cost components, such as station coolers are independent of the number of units. Some cost factors favor smaller units: multiple smaller valves are often less expensive than one big valve.

Transportation is another issue, especially for remote sites: it might be easier to transport smaller units to remote sites. Man hours necessary for the installation are probably higher for multiple smaller units, however.

Overall, the details of an application significantly impact which gas turbine size fits best. Experience shows, however, that bigger is not always better.


Klaus Brun is the Machinery Program Director at Southwest Research Institute in San Antonio, Texas. He is also the past Chair of the Board of Directors of the ASME International Gas Turbine Institute and the IGTI Oil & Gas applications committee.

Rainer Kurz is the manager of systems analysis for Solar Turbines Incorporated in San Diego, CA. He is an ASME Fellow since 2003 and past chair of the IGTI Oil & Gas applications committee.