Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Rang

Member
Registered:
Posts: 57
Reply with quote  #1 
I'm looking for information on triax's and attenuation due to transmissibility when magnet mounted. I'm assuming there is signal attenuation at the two axis perpendicular to the magnet's orientation?

i.e. If you mount your triax in the Horiz. direction, is the signal acquired in the Vert. and Axial axis attenuated? As the signal is effectively crossing the magnet/machine interface in a perpendicular manner.

I've heard of some consultants (much brighter than me) refusing to use a triax for this reason for magnet based data collection. But I've not data to back it up....
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #2 
Short answer:
As long as you have a decent magnet on a decent surface and it feels well stuck, the signal is not attenuated/amplified at low frequencies (at least < 500 Hz). The signal will never be attenuated, it will only be amplified, so at least it is conservative.

Long answer:
Your uniaxial accelerometer (or primary axis of the triaxial sensor) has a mounted resonant frequency (MRF) that determines what the upper limit of the frequency range is. This is often in the range of 12-36kHz. As you approach this limit, your signal is not attenuated, it is amplified. Very common are these figures that show how different kinds of mounts affect the MRF:

  2018-11-06 10_08_57-10.104.001b_Sensor_Placement_-_Accelerometer.docx .png 

What I've never seen, is the same MRF data for how mounts affect triaxial's secondary/transverse axes. It is clear that their maximum MRFs will almost always be lower than that of the primary axis, but how they react to a softening of the mounting method is not well documented. I imagine that if you have the 8kHz primary MRF, my guess would be a secondary MRF about 1/4 of that (pulling that outta my ass). Your upper bound of the frequency range is again 1/4-1/2 of your MRF depending on your tolerance (3%-3dB), which gives you a max frequency range for the secondary axes as 500-1000 Hz. For most, this is good enough, but I can see why many would shy away from it.

Worst, is that you can completely mess up your measurement if you are not familiar with what a good magnetic connection feels like. If you have any kind of rocking on your triax, the primary will show some distortion on the order of 10%, but the secondaries may  be distorted 300%.

Rang

Member
Registered:
Posts: 57
Reply with quote  #3 
The official word from a sensor manufacturer is;

"Up to 2k Hz there is no distortion or amplification, and only a slight possibility of attentuation."

So beyond that it seems the signal is affected. Something to watch out for if you are getting PeakVue data from all three axis (NB: Only applicable to magnet mounting).
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #4 
Quote:
Originally Posted by Rang
The official word from a sensor manufacturer is;

"Up to 2k Hz there is no distortion or amplification, and only a slight possibility of attentuation."

So beyond that it seems the signal is affected. Something to watch out for if you are getting PeakVue data from all three axis (NB: Only applicable to magnet mounting).


Definitely a universal rule of thumb that you should only consider high-frequency signals and features like peakvue from the primary axis of a triax. I think this is even restricted on the CSI 2140 if I remember correctly from another thread.

Just keep in mind that the vendor gives you 2kHz as the cutoff probably in 'typical' conditions,  which for them is really ideal conditions. That means, a flat magnet on a machined surface. If you have a dual-rail magnet on a curved, painted surface (more what I would consider 'typical', then that number is going to fall quite low.
Rang

Member
Registered:
Posts: 57
Reply with quote  #5 
I found a white paper with some testing that suggests using a flat rare earth magnet is sufficient to capture PeakVue data, assuming it is collected in the Z-axis, vertical direction.

https://www.emerson.com/documents/automation/triax-accelerometer-for-route-based-vibration-analysis-en-39110.pdf
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.