Published on: 

While completing yet another experience list for compressors, we had the following thoughts about the value of these lists: Although fleet information about centrifugal compressors can be informative if the right questions are asked, many important facts are often overlooked. If the question is, “How many compressors of this model have you built that operate at over 2,000 psi discharge pressure?” the answer will be next to worthless.

This is mainly due to the fact that the model name a manufacturer uses for its compressor is often a summary designator that includes individual compressors of vastly different designs, bearings, seal systems, shaft lengths and thicknesses, and possibly custom-designed impellers. There are far more “Serial Number One” compressors than most people expect.

Every year, there are machines that, despite having met all the scrutiny of experience list requirements and supposedly have large fleets, fail to perform as predicted once installed. This can lead to significant impacts on project profitability due to increased fuel consumption, inability to meet production goals due to high power consumption, and reduced flexibility due to early surge or premature choke. In even more severe cases, projects are delayed due to rotordynamic, aero instability or other mechanical problems with machines.

Compressor uniqueness

The reality is that operating conditions lead to compressors that are often unique in one or more aspects, regardless of the size of the fleet of similar models. Let’s have a look at a few of them:

Compressor aerodynamics, among other things, are determined by the impeller and the subsequent diffuser. It is common practice in the industry to start out with a base impeller and then to modify the diffuser either by changing the diffuser length or adding vanes to the diffuser. Impellers often get modified by cropping, changes in the hub diameter, or modifications to the shroud contour. Obviously, factory or field test data of the base configuration only gives a limited amount of information on the performance of the modified configuration. Also, any experience data must encompass the Mach number levels encountered or it is mostly irrelevant.

Similarly, changes in individual rotor components, such as changing bearing pre-load or modifying seals can have an impact on the overall rotordynamic behavior. Merely looking at experience at a certain pressure level is misleading, since it does not address the gas density in relation to rotor characteristics (such as operating speed in relation to the first critical speed of the configuration).

Another basic fleet issue is the capability of manufacturers to have reproducible fabrication and assembly processes. Especially with welded impellers, it is difficult to control critical flow path parameters.

Three-piece welded impellers pose the largest challenge to repeatability. Changes in impeller materials may impact the manufacturing methods required, and thus, the outcome in terms of performance and strength.


Essentially, the evaluation of compressor fleet experience levels will require knowledge in compressor design. Answers to the question, “How many of these have you built previously?” will not lead to thorough, reliable assessments.

The seemingly obvious conclusion would be to insist on more testing, but it should be noted that testing is often too late in the project cycle to allow for remedies without impact on schedules.

On the other hand, experience lists for gas turbines are far easier to evaluate than compressors, because industrial gas turbines are usually standardized products. Successful operating hours are a good indicator of the validity of a given design.

It should be noted, however, that in some instances, so called “equivalent operating hours” are used, where each start or shutdown adds disproportional operating hours. Some manufacturers use this practice to define their maintenance intervals. For the purpose of demonstrating experience, however, only actual running hours should be used.

Proven designs

Where does this leave the user of turbomachinery? Using proven designs and engineering knowledge to evaluate designs regarding the capability of the manufacturer can significantly reduce the amount of problems in the field. Evaluating designs requires engineering knowledge that goes far beyond the completion of simple fleet experience lists.

Obviously, we want progress in the industry, so new designs are necessary. What should not be required is that a manufacturer’s development work is done at the users’ expense unless this was specifically agreed upon. Good communication with the manufacturer as well as a good description on the intended use of the machine will improve the chances of successful operation after installation.


Klaus Brun is theMachinery 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.