Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #1 
I know these levels are low but the trend has me concerned.
Is there a reason why the waveform has this appearance, is it a sort of ring down?
Motor has an impeller directly mounted to shaft, in a dust collector application
Capture.png

diederib

Member
Registered:
Posts: 13
Reply with quote  #2 
That's an interesting waveform. What does it look like in the circular format? Autocorrelated?

Whenever I see the pkvue spectrum like that I suspect slight looseness of the component on the shaft. In my case it would be a loosening taperlock on a drum, but the same theory.

Otherwise maybe something rubbing lightly on the impeller. The randomness of the waveform doesn't seem like the case though.
Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #3 
It could be loose as stated. It could be an impacting rub where there is an energy decay (damping) after each impact. Look for dust buildup inside fan around blade tip perimeter and inlet shroud. If you used a magnet base, then there could be some mounting looseness, so try measurement again with firm contact.

Walt
MVS

Avatar / Picture

Member
Registered:
Posts: 15
Reply with quote  #4 
That is a pretty wild TW for sure - I would agree as to impacting - however, to me it seems more like a result of a cracked inner/outer bearing race/. Especially since the trend shows a steady increase (not something I would expect to see from a rubbing condition (as in over time a rub may show less of an impact). Further to the activity shown, is the "crest factor" value at over 8.5. Never consider since values are low - there is no issue (good analysis with your trending concern Noknroll).
__________________
"1xrpm is not always unbalance - but, unbalance is always 1xrpm"
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #5 
Thanks for the replies fellas, that gives me some good direction where to focus attention.
Walt, yes its a magnet mount but I think the problem is real as my measurement process hasn't changed.
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #6 
Do you have “real” data? I know PV is useful, but is it the first, or primary, thing you looked at? Wouldn’t the acceleration TWF give a more realistic view of the periodicity or randomness of the “event” in question?

__________________
"The trend is your friend"
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #7 
Rusty
Nothing to see in the acc TWF, trend actually went down
Cant ignore the PKV trend though
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #8 
Yes, established trends - whatever their origin - are usually real.  Just seems hard to determine severity and urgency  when the 'problem' doesn't show up in the basic spectra or TWF.  Could that be because of the limitations of our collection parameters for normal data, that are not inherent in PeakVue and similar manipulations? 
__________________
"The trend is your friend"
ssimmon1

Member
Registered:
Posts: 46
Reply with quote  #9 
Quote:
Originally Posted by Noknroll
Rusty
Nothing to see in the acc TWF, trend actually went down
Cant ignore the PKV trend though


Take a very high frequency regular reading, say 20-40kHz Fmax with at least 6400 lines and report what the waveform looks like.

I've had similar things show in Peakvue, but couldn't figure out what it was (still haven't actually), but knowing that Peakvue works in the "filter-40kHz" range, I decided to try to find where it was coming from, and I did. Whatever is causing the Peakvue readings below is coming from above 10kHz. This is a 20kHz Fmax/6400 line waveform that shows the impacting events that only appear above 10kHz (which are actually oversaturating my 500mV sensor, lucky me [rolleyes])


This is on a paper machine press section felt roll bearing.

twf.png 

Capture.jpg 


That said, yours looks like a blade coming into contact with something to me.


Lastly, I'm curious as to why you're using the 2k HP filter instead of the 1k filter.

Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #10 
Lastly, I'm curious as to why you're using the 2k HP filter instead of the 1k filter

The 2000Hz/80X Fmax dictates that I use the 2k hp filter
ssimmon1

Member
Registered:
Posts: 46
Reply with quote  #11 
Quote:
Originally Posted by Noknroll
Lastly, I'm curious as to why you're using the 2k HP filter instead of the 1k filter

The 2000Hz/80X Fmax dictates that I use the 2k hp filter


That makes sense lol.
Jim Crowe

Member
Registered:
Posts: 53
Reply with quote  #12 
Usually 40X rpm is sufficient for Peakvue unless you are dealing with gears or a bearing with a lot of rollers. Post the autocorrelated waveform .
Ralph Stewart

Avatar / Picture

Sr. Member
Registered:
Posts: 235
Reply with quote  #13 
Quote:
Originally Posted by Noknroll
Lastly, I'm curious as to why you're using the 2k HP filter instead of the 1k filter

The 2000Hz/80X Fmax dictates that I use the 2k hp filter


IMO, This is a good example of "NOT USING ORDER BASED SETUP FOR PEAKVUE".

One loses control of the HP Filter setting one wants or needs to use. Even an Fmax of, for example, 1000 Hz and a 1K HP Filter, if the order based setup causes just a 1 Hz increase in the Fmax (1001 Hz), Peakvue defaults to a 2K HP Filter.
Reason for the default change of HP Filter, is that it (the 1Hz increase to 1001 Hz) breaks the assignment rule of HP Filter vs Fmax.

Just my opinion and I could be totally wrong.

Thanks and Have a Great Day,
Ralph

__________________
Ralph Stewart
AlertAnalytical.com
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #14 
Ralph
no need to use big shouting upper case letters
My setup is in hz i simply added the 80x as andicator of what my motor speed is
Jim
I like to go to 80x to see rotor ba/slot frequency
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,397
Reply with quote  #15 
Hey Nok, what about the normal acceleration spectrum?  Order based sets do play havoc with filters but if you understand the speed range of a machine - say a paper machine roll - then that is easy to avoid.  Usually not an issue with fixed speed machines. rgds
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.