One of the first steps in vibration analysis is identifying the operating or running speed of a machine. Knowing this value is important because details about the state of the machine are usually linked to this particular frequency, as explained in a previous posting.
Identifying the running speed can be achieved using several methods, including taking tachometer reading, using speed encoders, or simply by analyzing the spectral data. Some machines, however, may have multiple rotating components that turn at different speeds due to intermediary gearboxes or belt and chain drives. In these cases it is important to know the speed ratios between the different components, and to realize there are multiple running speeds involved.
For example, let us assume we have a motor turning at 1,800 rpm and it is connected to a gearbox which drives a pump with 6 vanes. If the gearbox, has an input-to-output ratio of 9:1, then the pump is running at:1,800 rpm x 1/9 = 200 rpm
Normally we would expect the vane pass frequency to occur at 6 times the running speed; this is true if we take the pump speed, 200 rpm, to be the running speed, but not if we take the motor speed to be the running speed. The vane pass frequency in this example is:
200 rpm x 6 = 1,200 rpm
If we assume the motor speed to be the running speed, then the vane pass frequency is two-thirds of that value, and shows up as a sub-synchronous frequency on the spectral data .
This example demonstrates why we need to understand the components and operation of the system, and it illustrates why we have to correctly identify the machine’s running speed before proceeding with further analysis of the data.