Practical Solutions to Machinery and Maintenance Vibration Problems
Chapter 3, Detuning and Proving Resonance
Section 16, Further Considerations when using Jacks and Braces
Whether or not a part is actually resonant can also be determined by only slightly changing the part's natural frequency. Sometimes there is no readily available place to clamp in order to stiffen the suspect part. For non-rotating parts, such as a floor beam supporting the machine or an attached pipe, the suspect part can have its natural frequency changed by applying a load via a hydraulic or screw jack or a wedge similar to that shown in Fig. 1.
Although the addition of a pipe or column looks like the beam's length between supports has been changed, thereby changing its natural frequency considerably, it doesn't really respond that way! Instead, the load applied by the jack puts the suspect beam under further tension. This tension increases the natural frequency only slightly. Yet, if the load is applied to a beam or part that is actually resonant, or nearly so, then the vibration amplitude and phase will decrease or increase noticeably. Usually expect at least 20 to 30 percent change.
If the source vibration as shown in Fig. 2 is in the part's
natural frequency range but at speed A, lower than the resonant frequency,
then applying the load will decrease the vibration. If the source vibration
frequency is slightly higher than the resonant frequency, the vibration
will increase. If the source vibration frequency is not in the natural
frequency range (non-resonant) such as at running speed B, then there
will be almost no change when the load is applied.
For certain situations, it may be more practical to apply strain via a cable and turnbuckle or a rope and pulleys. However, as this method is used only to determine whether the part is resonant, the final correction should be made by bracing, increasing or reducing the span, and so on. Often bracing is used. When the final bracing is completed, the part's natural frequency should be increased enough so that its natural frequency range would be well above that of the source vibration. The vibration should then decrease by as much as 80 or 90 percent.
This technique once saved a considerable amount of money in a refinery. Readings indicated that the vibration frequency of a large steam turbine driving three centrifugal compressors was the same as the turbine's rpm. Other symptoms pointed to the turbine's unbalance as the source of damaging vibration. However, the vibration amplitude at each bearing was within tolerance for its 8,000 rpm operating speed. Due to the extreme vibration of the speed control mechanism mounted on the end of the turbine, various internal parts had to be replaced three times during the past year. Each time they were replaced, approximately one-fourth of the plant had to be shut down.
As the vibration's source produced relatively low vibration at its bearings and exceptionally high vibration on the speed control enclosure, it appeared that the enclosure was resonating. To substantiate this analysis quickly and without shutting down the turbine, a load was applied to the overhung speed control enclosure by pushing it upward via a pipe and jack, in such a way as to force a strain on the case in order to raise its natural frequency in the vertical direction. As the force was applied, the amplitude of the vibration surprisingly remained the same.
Next, the enclosure was braced horizontally against a large globe valve
on an adjacent large-diameter pipe. The horizontally applied strain
raised the amplitude of the vibration several mils, indicating that
it was resonant. Phase changed about 40°. This only revealed that
the part was resonant in the horizontal direction. The next step was
to determine how to permanently change the resonance frequency.
This textbook contains only part of the information in our Practical Vibration Analysis seminar.