Newsletter Sign-Up
January 2015 · Energy-Tech Magazine
April 2005 Go to Page 1 2
Determining Crankshaft Distortion in Large Reciprocating Compressors, Engines
Web Deflection & Measurement Techniques

Measuring web deflections is a common technique to determine crankshaft distortion in large reciprocating compressors and engines. Crankshaft distortion may be caused by frame misalignment, driver misalignment, or bearing damage. Excessive distortion can lead to crankshaft damage, bearing failures, and other problems. The purpose of this article is to define what web deflections are, describe typical measurement techniques, list some typical causes, and explain how to interpret web deflection readings.

Web Deflection

A typical two-throw crankshaft is shown in Figure 1. As shown, the webs are the parts of the crankshaft that extend outward from the centerline to support the connecting rod bearings. The distance "X" in this figure is the web opening. If the crankshaft is straight, the web opening will not change as the crankshaft rotates. If the crankshaft is not straight, the web opening will change as the crankshaft rotates. For example, let's assume that as a result of a faulty condition, the crankshaft has taken on a simple bow shape. As shown in Figure 2, at 0º crank angle, the web opening, "X1" is greater than the web opening at the 90º crank angle, "X2." Also, "X2" is greater than the web opening at the 180º crank angle, "X3." The flexing of the webs cause cyclical stresses, which if excessive, can lead to crankshaft damage, bearing failures, and other problems.

Web Deflection Measurement

Web deflections can be measured with both dial indicators and electronic devices. Figures 3 and 4 illustrate how web deflections are measured with a dial indicator. As shown, the indicator is set up between the compressor webs. The gauge is normally set at zero at some reference position. The crankshaft is then rotated, and any movement of the webs can be observed. Although some technicians only record the maximum deflection, it is better to record deflections at various crankshaft angles. This will aid in the evaluation. Ideally, web deflections should be recorded at 0º, 90º, 180º, and 360º of crankshaft rotation. However, because the connecting rod will eventually interfere with the dial indicator, web deflections are normally taken at five locations as shown in Figure 3.

An electronic device does the same thing as the indicator, but it can be safer, easier to use, and more accurate. Once an electronic device is set up, web deflections can be read outside of the crankcase. It eliminates the need of using lights and mirrors to read an indicator and eliminates ambiguous readings that can result from incorrect indicator installation.

In many cases, web deflections are taken with the compressor in a cold condition. Sometimes this can't be avoided, but it does introduce some unknowns into the interpretation of the web deflection readings. It is best to measure web deflections when the machine is hot since this will represent the actual running condition. In general, the compressor should run for about three to four days to allow adequate time for the entire compressor, piping, and foundation temperatures and any thermal movement to stabilize. Web deflections should then be checked shortly after the compressor is shut down.


Assuming that when the compressor was installed, the web deflections were minimal, something would have to change to cause the web deflections to increase. Some typical causes are as follows:

  • Bent crankshaft.
  • Foundation degradation that causes the crankcase to distort.
  • Misalignment between the driver and the compressor.
  • Main bearing failure.
  • Cylinder support failure that causes the crankcase to distort.
  • Bent connecting rod.
  • Cylinder misalignment.

Go to Page 1 2 Next Page