Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment   Page 1 of 2      1   2   Next
mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #1 
Dear colleagues,

Will appreciate your comments/thoughts about my first vibration survey report. Is it a right diagnosis?

 
Attached Files
docx PM-23202B Vibration survey report_updated.docx (4.12 MB, 43 views)


__________________
_______________________
_______________________

mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #2 
update: the report has been updated with extra information.
__________________
_______________________
_______________________
fburgos

Sr. Member
Registered:
Posts: 659
Reply with quote  #3 
TLDR, presenting to much information why a columns with alert and alarm value why spectrum with cross phase use a bubble diagram.

What I would like to read first, what's failing, how to improve, how much time to take action.

Overall history, graph, are the support information can go at the end of the report
mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #4 
@fburgos

T
hank you for the notes, these details were requested by my line manager probably to see how I am working.

__________________
_______________________
_______________________
John from PA

Sr. Member
Registered:
Posts: 940
Reply with quote  #5 
Quote:
Originally Posted by mnewiraq
@fburgos

T
hank you for the notes, these details were requested by my line manager probably to see how I am working.


Absolutely nothing wrong with the details you’ve included.  Moving forward these details can be thought of as reference material.  When the boss asks you a question, hopefully that info will be readily at hand.

I can remember that Don Bently required us to document in every report the nameplate data.  Was it needed for a technical reason?  Maybe, but this was the day before cell phones and taking a picture of a page from the tech manual.  What it did was get the engineer off their backside and walk out and see the machine.
fburgos

Sr. Member
Registered:
Posts: 659
Reply with quote  #6 
im not saying a lot of data is bad, im saying there are simpler ways of presenting same data, every person is different, your boss likes big reports I like them brief, I try to present something like this.


-Machine sumary, nameplate data etc, overal alarms etc.

-findings and observations, for example...

on mm/dd/yy a vibration survey was done the overall was ## mm/s rms (se table 1) 5% from alert... etc etc,

phase in axial and vertical is 180 probably cause misalignment (Diagram 1),

spectrum on moa, and mia bla bla (Diagram 2, 3, 4)

-Recomendation.

Do... when... what.

in next planned oportunity /in the upcoming week/tomorrow/ within a month do.. coupling inspection, regrese , investigate posible 2fl by waterfall plot during shutdown, verify alignment and motor solo run


-charts.

*Table 1
*Diagram 1
*Diagram 2



mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #7 
Must say that the layout you've suggestin' is much more convenient and indicative.

Thank you for sharing it with me. Accept my regards.

__________________
_______________________
_______________________
mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #8 
@John from PA thanks for sharing your thoughts. And yeah, it seems that the adherence to the classics at the beginning of a journey will pay back for it later.
__________________
_______________________
_______________________
MarkL

Avatar / Picture

Sr. Member
Registered:
Posts: 1,108
Reply with quote  #9 
On an off topic note, you seem to be using a skf ax80 analyser from the screenshots, but the computer software derived fft shown is from a different software? Bentley Nevada or commtest?
fburgos

Sr. Member
Registered:
Posts: 659
Reply with quote  #10 
Probably it's a Bruel & Kjaer vibro vibroport 80
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 837
Reply with quote  #11 
mnewiraq

attached is my brief version of your report.
7 pages reduced to 1

 
Attached Files
docx PM-23202B Vibration survey report_brief version.docx (38.50 KB, 22 views)

mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #12 
@MarkL
@fburgos

t
he company has only one license that is reserved by other guy. so, I had to transfer the measurement files from the device memory, putting their content in new excel sheet and build a fft by an excel chart, annoying but 5 minutes task.

so what you see is an excel chart of the fft.

__________________
_______________________
_______________________
mnewiraq

Member
Registered:
Posts: 11
Reply with quote  #13 
@MarkL
@
Noknroll
@
fburgos

B
y the way, would you please explain why the phase change of 180 may shift from the radial to the axial direction?

__________________
_______________________
_______________________
MarkL

Avatar / Picture

Sr. Member
Registered:
Posts: 1,108
Reply with quote  #14 
Wow that's pain in the ass!
What about the sidebands clearly visible in the analyser and excel generated Ffts?
wopajh

Member
Registered:
Posts: 3
Reply with quote  #15 
My guide is that a report detail needs to be in line with the cost of the equipment monitored. No point spending hours on a report for a $20K motor pump combination. But for a multi miliion dollar gas turbine then you should spend more time on your investigation and the report. So in this case yes it's too detailed a report.

As for the diagnosis, having 10 years of VA before I began doing laser alignments, I was discouraged to discover that vibration is not a very good way of diagnosing misalignment - it needs to be a long way out to produce appreciable vibration and is dependant on the type of coupling. I cringe to think how much poor advice I'd given in those 10 years (I'm a slow learner). There is a study done by CSI and others which indicated 80% alignment diagnosis were incorrectly identified and I would agree - that is my experience also. Having said that, the system is probably misaligned as the tolerances for a 3000rpm machine is very fine, but misalignment is probably not the cause of the vibration. My reasoning -the higher motor vibration is at the NDE yet the coupling is at the DE, and the pump thou of lower mass and less stiff is not vibrating appreciably higher. than the motor.  The trend shows the vibs are up and down historically, and looking at the suction pressure of the pump (excellent practice that this is monitored), I'm guessing the driving vibration is from the pump - check the pump curve. I'm betting the vibs are then not being absorbed by the base frame and could possibly be in resonance. From memory, good design of a baseframe is ~3 times the mass of the motor / pump and this is often over looked. I suggest you check hold down bolts, ensure the grouting (if it exists) is good, do an ODS of the frame, and a bump check to check for resonance.  I recently had a similiar experience and the only resolution was to replace the pump with one more suitable for the application.  The client has at this point choosen to stiffen the baseplate with more steel (which helped take it out of resonance) and to live with the issue of continually replacing bearings albeit at a reduced rate due to the pump running off the curve.
Remember when measuring phase, the accels need to point the same direction or you will get a 180 degrees change.
Goods luck.
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.