by Lev Nelik, Pumping Machinery, LLC

For the rest of the world (95 percent of cases), vibration is first sensed by the hand of the maintenance operator, who happened to come by the pump to check or change the oil. These operators do not have instrumentation but use simpler, but still reliable, vibration measurement instruments—such as accelerometers attached to their neck-held readout, reading overall (root mean square—RMS) vibration values and (if needed) spectral analysis capabilities—to get the fast Fourier transform (FFT) signature, to troubleshoot the unit if the unit appears to be at the overall level of vibrations that require troubleshooting. Therefore, the HI publication essentially reflects the value of a new installation, which design engineers, contractors and consulting firms usually use. If you take a closer look at the HI publications, note the scale values on the graphs—0.04, 0.08, 0.12, 0.16 … etc. Such plot scales would not be expected of a seasoned engineer. You attached ISO 10616-1, which also reveals the same type of authorship by well-qualified folks, but with “interesting” practical understanding of the world.

In practice and in the field, such academic methods are not used. To differentiate between a 0.17 inch per second at the pump and to allow 0.21 inch per second at the top of the motor is practically impossible. What they need to know is one number, which they can easily remember and which would be reasonably applicable to a wide range of pump types.
With this approach in mind, the typical field value for the warning is usually 0.30 inch per second and alarm at 0.50 inch per second, although most pumps, almost regardless of type, would run at about 0.1 inch per second to no higher than 0.2 inch per second. Spending most of my time these days in the field and seeing daily a different type of pump, this appears to be a mainstream of data.

Having set the field limit warning at 0.30 inch per second, one would be comfortable with a pump at 0.05 inch per second, its vertical motor at 0.12 inch per second. If the motor is tall, a comfortable number would be 0.16 inch per second, as an example. Therefore, the conservatism (practically speaking) of the selected value would dismiss the need of differentiating between a tall or short motor setting.

As you may know, the pump repair standard for vertical pumps is in the process of being drafted. If you are interested, please join our committee, representing Australia. We have representatives from the U.S., Middle East and Europe also on the committee. We are also working on a standard for horizontally split pumps. You may find related articles, available at the Pump Magazine ( or at, where you originally saw the article.

Speaking specifically to your pump, I have seen double-suction split-case pumps mounted similar to what you have in Figure 1, in a vertical arrangement, and they usually have problems. What type bearing do you use? Do you have a sectional drawing you can email me? I hope these thoughts and ideas are helpful. Let me know if you want to join our committee.