Sign up Calendar Latest Topics Donate

  Author   Comment   Page 4 of 4      Prev   1   2   3   4

Posts: 10
Reply with quote  #46 
A couple of things to keep in mind when doing any type of torsional measurement...

If you are actually measuring torsional vibration (angular displacement) you will want to take the measurement at a location along the shaft where the torsional vibration is high.  In many cases, the only location that can be accessed safely is at (or very near) a coupling.  Where torsional resonance is involved, the coupling will typically be at (or very near) a node point for lower torsional natural frequencies so the torsional vibration might actually be low.  However, the strain will be quite high.  This is part of the reason we typically use strain gages and a telemetry system for any type of torsional measurement.  Occasionally we will install encoders if we are interested in torsional displacement but that is less common for us. 

It all comes down to what you are trying to get out of the test:
Trying to document torsional natural frequencies?
  • Strain gages work well for this.  You can sample during a machine startup or a trip and usually identify the torsional natural frequency (first mode and sometimes second mode) from the time waveform.
  • Torsional vibration (angular displacement) is also valid for this as long as you can access the shaft at a location not near a node point for the mode(s) you are trying to measure.

Was there a failure that indicates torsional vibration was a contributor? 
  • Again, strain gages are a good choice here.  If you measure the strain, you can calculate the shear stress and look at material properties to evaluate fatigue life.
  • The problem with measuring the torsional vibration (angular displacement) is that it is difficult to determine how much is too much.  The only effective way I've done that in the past is to measure the torsional vibration amplitude and then compare to a torsional model of the machine train.  There are still challenges with that method since the model accuracy and forcing functions are big players.  
Are you trying to monitor torsional vibration continuously?
  • I've done this with strain gage systems in the past and it works well with the right hardware and careful installation of the gages.
  • This seems like a good application for an optical sensor if you can install at a suitable location.
When you do regular vibration measurement with an accel, you usually have the option of measuring at multiple locations/directions.  If you do it for a while you get a feel for where the best locations/directions will be for a given machine to look for certain problems but it isn't a huge problem to take multiple measurements.  With torsional measurement (vibration or strain) you often only get one location due to the cost of hardware and installation time so it is important to have a feel for what the torsional mode shapes look like and how that will influence your measurement type.

Hope that adds to this great discussion.

Previous Topic | Next Topic

Quick Navigation:

Easily create a Forum Website with Website Toolbox.