Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment   Page 3 of 3      Prev   1   2   3
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #31 
Quote:
Originally Posted by MarkL
John,
I think hes using Erbessed Phantom sensoes which are wireless Piezo triaxial..if I recall.
[1a-300x271]

Meet Phantom Vibration Sensor:

  • 10kHz FMax
  • Full waveform and spectrum data
  • Programmable Internal RMS Trigger
  • Sample Rate: 25600 Hz
  • Lines of resolution: 6400
  • 3 years battery life
  • Internal temperature sensor
  • Noise: 630 G/√Hz


Correct me if I am wrong MarkSL?

Specs are not really detailed but they do state an Fmax of 10 kHz (https://www.erbessd-instruments.com/vibration-monitoring/#Sensors.  So, I would anticipate a mounted resonance under optimum conditions (drilled, tapped, properly torqued into a test fixture) to be around 25 kHz or higher.  Since it is designed for continuous monitoring, hence trending, I could be wrong on that number.

So as others have suggested, I would verify proper mounting torque  




Specs are not really detailed but they do state an Fmax of 10 kHz (https://www.erbessd-instruments.com/vibration-monitoring/#Sensors.  So, I would anticipate a mounted resonance under optimum conditions (drilled, tapped, properly torqued into a test fixture) to be around 25 kHz or higher.  Since it is designed for continuous monitoring, hence trending, I could be wrong on that number.

So as others have suggested, I would verify proper mounting torque and cable anchoring and perhaps swap transducers.  I think initially it is important to rule out the transducer before you go running off looking for a machine fault.  

marksl

Sr. Member
Registered:
Posts: 155
Reply with quote  #32 
Quote:
Originally Posted by Danny Harvey
mark,

The sensor manufacturer will probably have the sensor natural frequency. It could also be the mounting natural frequency, I guess. 

That's far beyond anything that I would consider to be linear for anything other than bolted connections, though.  I'd check to make sure the sensor is fully tightened if applicable.

If there are a whole bunch that aren't giving that signal, you might try switching sensors if you are able. Some sensors aren't glued together on the inside as well as others. 


I'm waiting to hear back from the manufacturer regarding the sensor natural frequency. Client checked the sensor today to make sure it was screwed in tight on the mounting pad - all OK. The customer will inspect the housing tomorrow and remove any excessive grease from the cavity
Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #33 
"The customer will inspect the housing tomorrow and remove any excessive grease from the cavity"

What is the basis for stating the bearing has excessive grease based on the OP vibration spectrum? Has there been any confirming measurements such as
Temperature to indicate excessive grease, Vibration with another sensor and portable analyzer, or Vibration by switching permanent sensor with a spare?

If the accelerometer is a triaxial MEMS type, then which axis had the dominant vibrations? Does the monitor offer demodulated waveform and spectrum?

If this is a relatively new installation for the permanent monitor, then eliminating false positives or false negatives would add to the credibility for fault detection.  

Walt
Edwin

Avatar / Picture

Sr. Member
Registered:
Posts: 131
Reply with quote  #34 
I found a more detailed datasheet of this sensor:
https://www.erbessd-instruments.com/wp-content/uploads/2018/08/Phantom-EN.pdf

You can see the 10 kHz range is only in x and y direction. In z direction it is only 5.1 kHz. So if this measurement is from z direction, it is really plausible that you capture a natural frequency of the sensors housing or mounting.


__________________
= Cost Effective Asset Management =
Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,403
Reply with quote  #35 
A totally failed crystal bond gives a very ugly twf that won't translate into a spectrum. That is to say in an accelerometer that has an audible rattle when you shake it.  I had a couple of late 1990's era CSI accels fail like that.  

Maybe a slightly loose bond or a loose wire or bolt would respond to high harmonics of something like running speed.
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.