Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
RickP330

Member
Registered:
Posts: 30
Reply with quote  #1 
Hello all,
Curious for your take on this data, this is a 17HP positive displacement oil pump 1765 RPM.  A lot of activity way out there at 80X, but only in one direction (My Accelerometer is not oriented correctly, X axis is axis of rotation).  .7 g's.  Bearing issue?
Much TIA,
Rick

Attached Images
jpeg VibrationChartControl350.jpg (183.60 KB, 57 views)

Beatnik

Sr. Member / Supporter
Registered:
Posts: 183
Reply with quote  #2 
I'd take this measurement at a couple of different places to make sure the haysack is not simply the resonnance of the accelerometer.

I'm curious about the sidebands though, what are they?
RickP330

Member
Registered:
Posts: 30
Reply with quote  #3 
VibrationChartControl3502.jpg  IMG_9807.JPG 

I installed a larger magnet on the my accelerometer and took another reading set.  The newest reading set is in the background.  Much better I presume, but I wonder if my magnet set up is not appropriate?




RickP330

Member
Registered:
Posts: 30
Reply with quote  #4 
Guys,
I'm getting the hang of this I think.  I believe those first indications I was seeing was due to a poor accelerometer set up.  I now put an even larger magnet on my accelerometer and you can see those .7g spikes are now gone.
RickP

Attached Images
jpeg VibrationChartControl35016.jpg (81.51 KB, 21 views)

John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #5 
You might want to review the document at https://www.ctconline.com/pdf/pubTechPapers/29-Mounted%20Resonance.pdf where about 2000 Hz is the stated upper limit of measurement of a particualr accelerometer mounted on a "curved surface magnet".  See Figure 2.

Your 80X, using the motor speed of 1765 RPM is 2380 Hz (142800 cpm) so you may very well be pushing the limit of your hardware.  

As an aside 1765 RPM means a fair amount of slip (35 RPM) for an induction motor.  Are their other issues with this motor?
RickP330

Member
Registered:
Posts: 30
Reply with quote  #6 
John,
Thank you for your reply.  So for using the "Curved Surface Magnet" as pictured, I am really just wasting storage space by recording to over 100X.  Does it make sense to reduce my Fmax to the limit of the transducer configuration?  What would you set Fmax up to for this case?

The 1765 is the nameplate RPM of the motor.  Guys, I appreciate this - you are holding my data to a higher standard.  Next time I'll haul out the strobe light and actually measure the RPM of the motor.  These motors only run when the machine is starting up or shutting down.  We stop every two weeks or so for maintenance so on the next shot I'll measure RPM as well.

This has not been easy for us.  I've had to beg borrow and scrape up hardware as no one is willing to budget money for CBM.  In addition to trying to learn all of this on the side.  So your comments and assistance are really approached.  Thank you!

Oh, and another quick question.  Orientation of the accelerometer, these are vertical motors - I've been aligning the X axis as along the shaft pointing down or in the direction of the free end to the pump end.  Is there a better convention to follow here?

RP IMG_9969.JPG  IMG_9963.JPG  IMG_9968.JPG

John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #7 
I'm not a walk around guy so others have to answer on the Fmax question.  I do know there is an online calculator for Fmax at http://www.mobiusinstitute.com/site2/detail.asp?LinkID=54.  See what you can get from that.

An alternative site is http://www.maintenance.org/topic/how-to-set-fmax-and-lor.
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,398
Reply with quote  #8 
For a motor like this 100x is plenty. Most likely 63** series DGBB's. With a BPFO or around 3.08x you will see more that 30 harmonics and more than 20 harmonics of BPFI.  I would normally go out to 120x but that is neither here nor there. 1600 lines is good. I also have a few motors set up at 150x at which point I would go to 3200 lines. 


JB1

Member
Registered:
Posts: 18
Reply with quote  #9 
Rick,

Based on your first plots, 1X = 1782 and the 11X peak appears to be the oil pump gear teeth count (19602cpm)?  And the high freq peak is the 6/7X harmonic of gearmesh w/1X sidebands?  Trending, mounting and always taking measurements at the same discharge psi should help.  It does not look like a bearing based on the limited data presented.  What is the GMF?
OLi

Sr. Member
Registered:
Posts: 1,918
Reply with quote  #10 
Just to make it complicated, most tests on accel mounting are with single axis accels, this is a 3-axis so only one of those axis compare to the tests. Other directions have even worse mounting IMHO and may be the reason why more magnet is better. We use a 3-axis and the magnet with most pull we can find and it works. It is also tricky to find good solid places to place the sensor. Keep doing as you do, replacing sensor on the same spot every time is important. Put a marker ring to get back to the same spot next time. I put a water level bubble on ours to get it in the same angle every time, that is also important. Keep struggling. You should find up to 10% bad bearings in a plant where no data been collected before, have you found some? Happy hunting.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
RickP330

Member
Registered:
Posts: 30
Reply with quote  #11 

Vibe-Rater,

                Okay, I reduced my readings to 80X but now wondering if I really needed to.  I’ll take a reading set anyway and see what it looks like.  I have one motor with 20X and 30X peaks 1/4G.  Could this be a bearing indication?

JB1,

                Yes, I agree, the 11X peak must be the oil pump.  I tried to get some drawings / data on the pumps, but from 1969 there is nothing available.  It is a common signature.  I have 2 of these pumps and they both have 11X peaks just like this.  Plus on the free end of the motor, farther away from the pump, I see a much more damped 11X response.   If I ever get a chance to take a pump apart I’ll document it and check if the forcing frequency matches – but I suspect it will.  I also have 4 other smaller pumps with 4X and 5X peaks which also appear to be the oil pump.

Oli,

                I have not found any bearings yet that I know of.  I am still trying to get comfortable taking and reading the data.  I’m sure we will find something.

                I might have a maintenance day tomorrow – I’ll take more measurements and bring along my strobe light as well.  I’ll keep you posted and thank you for commenting.

Kind regards,

Rick

Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.