AM64
Well-Known Member
- Joined
- Feb 11, 2016
- Messages
- 28,590
- Likes
- 42,430
Vibration especially flow induced vibration is really a complex issue, and strangely enough every now and then there are simple solutions. Sometimes aircraft and car manufacturers just add a weight to change the dynamics. A steam generator tube in some ways is like a string in a stringed instrument; it has a fundamental vibrating frequency and when excited it vibrates at that frequency (plus the harmonics); change mass, unsupported length, etc and you detune (or sometimes unfortunately tune) the component. Flow through or over a tube, a wing; or other object will induce vibration, and it it isn't moderated, it can be devastating. One great example was wind induced vibration in the Tacoma Narrows Bridge - the bridge broke apart.
Design engineers (at least before I retired) used computer based Finite Element calculations to design mechanical systems - to include vibration modes - bending and torsional modes and frequencies. We found over the years in using Modal Analysis to measure vibration in completed assemblies that the calculated vs the measured vibration modes and frequencies often didn't match - sometimes not well at all. Modal analysis involves adding accelerometers and then exciting the structure. The most straight forward test is to mount an accelerometer at a specified location, impact the structure with an instrumented hammer - a load cell to measure the impact amplitude and characteristic. Move the accelerometer to the next location and repeat. The data go into the modal analyzer (Fast Fourier Transform) and the magic takes place. The Hewlett Packard FFT based Modal Analyzers were fantastic. You got an animated display to demonstrate the bending modes and all the response characteristics amplitudes and especially frequencies. Always fascinating.
We actually used a large speaker and white noise to excite a reactor coolant pump impeller following an in plant failure. If you direct a noise source at an object, can cause it to vibrate at it's natural frequency. Flow noise either over a control surface or inside a piping system is often (if not generally) something like white noise (all frequencies) with some embedded deterministic noise like the blade passing frequency of a pump. You can also measure the internal noise externally with accelerometers to characterize normal behavior and abnormal behavior of systems - but that's another story. You can also expand that to look at noise in plant instrumentation (like pressure instrumentation) to determine if the transducer itself appears to be functioning correctly.
