Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #1 
I have inherited a database with an AP set 3000Hz/6400L/3Av and STW (special time waveform is ticked)
this AP is used extensively throughout the database.
A particular issue I have is limited shaft revs are being captured. 4 only on a 1490 rpm shaft.
By my calculations data collection time with just 1 average should capture at least 11 or 12 revs.
So the question is. Is it the STW limiting captured revs? another question, whats the purpose of the STW? and lastly what value is there in a STW? I'm very tempted to just untick that STW box.
 
Thanks in advance for your valued comments.
Beatnik

Sr. Member / Supporter
Registered:
Posts: 183
Reply with quote  #2 
The STW is usefull when you want a TWF that have different frequencies than your AP Set. Imagine you want a low fmax spectrum to see the low frequencies on an installation but also want to see a TWF with the high frequencies to see the bearing condition all on the same measurement point.


Data collection time for a sTWF is:

T = Numbers of points / (sample rate) = Number of point / (Fmax * 2.56)  You probably thought that the maximum frequency in the TWF setup was the sample rate, it is'nt.


You can just unclick it, if you unclick it you will get the TWF used to build your spectrum. I would also increase that 3000hz to 6000hz but it's personal tastes.
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,398
Reply with quote  #3 
Hi Nok, many spectral settings will produce a tw which is not really useful.  So CSI came up with "Special time waveform" Once the spectrum has been collected it follows up immediately with the STW. So it does add time depending.  But it provides separate control over the stw.  As Beatnik pointed out using the stw also means the spectrum was not created from it so can be slightly different. I use  all the time. Others don't I think at your old office it is heavily used. rgds

Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #4 
Thanks Fellas
Reckon I'll just untick it.
Beatnik,
I was actually going to reduce the fmax, possibly even edit the (s)crap out of it and turn it into an order based set.

I saw the glint in your eye and the upturned corner of your mouth then Vibe. 
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,398
Reply with quote  #5 
Definitely!
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #6 
Nok, the number of revolutions is controlled by the number of "lines" and the Fmax you use for the Time Waveform setup.  I made a screen capture video showing the process, but it won't work here (wrong file type).  So hopefully the pics below will illustrate how it's done.  I would argue that the number lines used, the Fmax, the number of spectral lines used, should be whatever you want, without regard to the "size" of the collected data.  Makes no sense to spend so much on analyzers and other equipment, and then be hamstrung by bargain computers or laptops.
  
Lines_1.png  Lines_2.png  Lines_3.png 

When you get to this screen, the Maximum Frequency Calculator, input the Fmax (same as your spectrum is not using Special), the number of revs you want, and the shaft speed.....then click Re-calculate, then click Use Adjusted Values (it shows you how many revs you'll get, and how many lines it uses).
Lines_4.png 
Lines_5.png


__________________
"The trend is your friend"
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #7 
Nok, I use the special TWF feature to capture acceleration TWF's for all rolling-element-bearings, but I collect velocity spectra.  This is partly so I can trend Maximum Peak Waveform in acceleration, which for me, is probably the most important parameter for triggering a call to replace a bearing. I will set the Fmax for the TWF to usually twice that of the spectrum.  But it also allows me, when I collect the data in the field, to see the velocity spectrum and acceleration waveform at the same time when I've collected a point.  Invaluable to me.  As for trending the Maximum Peak Waveform, I came up with that idea on my own. Just made sense that as a bearing starts to fail, it will produce considerably more high frequency energy (acceleration), and I've found it to increase pretty much exponentially in the last stages of failure.  A few years after I started using it, I was talking to Nelson Baxter and he said he did the same thing.  If Nelson does something, I'm certain it's a really good idea.
__________________
"The trend is your friend"
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,398
Reply with quote  #8 
I second that Rusty.  Met Baxter once here at a conference in Australia. Very nice man. I remember him telling us he used to take his daughter (as baby) to industrial sites.  Guess she just slept there as he did his work. I think she might now be in this field or at least closely related. Very talented man. rgds
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #9 
Here is the video showing the process. Nok, be careful about "unticking" the Special TWF button if it specifies an acceleration TWF. Your parameter sets may use acceleration in the calculation of parameters (mine do), and your trends will be royally screwed. 

mp4 Special_TWF.mp4     



__________________
"The trend is your friend"
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #10 
Rusty, you've given me more to think about and research how the current ap set is defined.
The inherited database is getting a major overhaul with client consent,
Mainly equipment info, rpms,fault freq's etc and several ap al changes/additions
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.