Practical Solutions to Machinery and Maintenance Vibration Problems
Chapter 3, Detuning and Proving Resonance
Section 15, Using Braces to Temporarily Change the Resonant Frequency
You can temporarily detune a non-rotating part such as a pipe or beam by clamping temporary braces consisting of approximately three inch leg angle iron (or any conveniently available, easy to carry angle iron) or relatively lightweight square tubing. They do not have to be cut to close length, but can protrude well beyond the sections being braced. Angle iron or tubing usually enables at least one side to be in a position to be clamped to the suspect part. If the suspect part is so braced and the vibration is not drastically altered, then you can assume that you are not bracing the actual resonant part. When the resonant part is braced, the vibration should decrease (or increase) appreciably.
However, there is a symptom that is more important than the change in amplitude. It is the possible change in phase angle. Phase angle changes can be observed digitally or with a strobelight that is triggered by the actual vibration. As the use of the strobelight method can be confusing to some analysts, the instructions are as follows:
1. Put vibration pickup on the bearing housing that is vibrating at the highest amplitude, or put the pickup directly on the suspect resonant part itself, such as a pipe, pedestal, and so on.
2. Tune the vibration instrument (or if using an FFT, position the cursor) to the frequency of the excessive vibration. (As resonance magnifies only one frequency, it is possible to perform this test using an instrument that gives only "overall" vibration.)
3. While reading the vibration amplitude, prepare to watch phase as well. If using a strobelight triggered by the vibration signal, direct it toward the phase mark on the rotating part. If the dominant vibration frequency is equal to 1 x rpm, the strobelight will reveal a "single image" that makes the rotor look as if it is standing still with one reference mark. If the dominant frequency is a full integer multiple of the rpm (such as 2 x rpm, 3 x rpm, 4 x rpm, etc.), the rotor's image will look as if it is "standing still" but with the number of reference marks equal to the multiple of the rpm. The reference marks will be equally spaced. If using an FFT, obtain phase with the cursor positioned to the desired frequency.
4. While reading amplitude and phase, brace the suspected part. Watch while the second clamp is tightened.
If the part being braced is actually resonant (in the direction being stiffened), the stiffened part will then have a changed resonant frequency. Therefore, the phase will change an appreciable number of degrees, anywhere from 15° to 180°.
If the suspect part is at the very peak of resonance, bracing enough to completely get out of the resonant frequency range will most often cause a phase change of approximately 90°. However, if the temporary stiffening effect of the bracing is too small, the phase change may be much less.
If using a strobelight, the shifting of the reference mark(s) is similar whether there is only one or several that are equally spaced.
If the phase change is negligible, the part being braced is either not resonant at all or not resonant in the direction being stiffened. If the part is a pipe, also try bracing it in a direction that is 90° from the original bracing direction.
It is important to note that when applying temporary braces, the brace itself may not be rigid enough to completely detune the resonant part. Therefore, when the temporary brace is applied, the phase change may be only partial as compared to what it will be when the final detuning is accomplished. Another possibility is that relatively light bracing may alter the resonant part's resonant frequency only slightly, but to a frequency that is closer to the vibration frequency originating from the source. In this situation, the amplitude will increase rather than decrease. The fact that the amplitude appreciably decreases or increases indicates that the resonant part is being affected. The crosscheck is that phase will have changed as well.
For piping (especially reciprocating compressor piping), excessive
vibration may be due to "organ pipe" resonance rather than
"spring system" bending resonance. Pulsations in organ pipe
resonance is beyond the scope of this course. Experienced vibration
specialists can get more insight on this subject by reviewing a college
physics textbook section on sound in organ pipes. Another good and practical
source is a company that sells "pulsation dampers." Organ
pipe resonances can be cured by changing the effective pipe length or
its diameter, or through the use of a pulsation damper.
This textbook contains only part of the information in our Practical Vibration Analysis seminar.