Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #1 
We’ve discussed before how difficult it is to use vibration data to quantify the effects of misalignment, and conversely, the effects of good alignment. I’ve never had the opportunity to do a “close enough” alignment, collect data, and then do a “precision” alignment and collect identical vibration data.

When we “align” any machine, there are a number of things that have likely happened that would affect the before/after vibration data: rebuilt components, coupling hubs removed/replaced, coupling element replaced, hold-down bolts loosened/retightened, feet reshimmed, bearings replaced and/or relubricated. And then things affect machine operation such as load and speed changes/variations.

Have any of you ever had the chance to do a valid comparison, where nothing has changed besides the alignment?

I envision doing a “pretty good” alignment using indicators, or even straightedges. Then collect multiple sets of data (for statistical analysis) initially at 5 minute intervals, and then repeated at hourly intervals. Then shut it down and do your very best alignment with the laser, and collect identical data. As well as the normal data we collect, I would include data filtered at 1X, 2X, 3X using a tach. Sensors would be stud or magnet/pad mounted, and not removed between trials.

Anyone up to the challenge of doing this? Anyone have a machine with this kind of availability? Ideally it would be a machine where you could install different coupling types (gear, grid, torus, 3-jaw, solid spider, split spider. I’m not talking about an “idealized” machine, in a laboratory environment, but rather a real-world machine, subject to the all the normal things that can change when we ONLY do an alignment.

__________________
"The trend is your friend"
fburgos

Sr. Member
Registered:
Posts: 670
Reply with quote  #2 
I'm looking forward to this test, I've done it on laboratory like environment, first visually not aligned, straight edge and final laser.

From not aligned to straight edge there is a real difference, sound, feel, with straight edge vibration was just right under "iso tolerance"
From straight edge to laser there was a measurable difference.

On field, doing only alignment I've had mixed result between real difference and no difference at all (alignment not was the real issue)
JuddJones

Sr. Member
Registered:
Posts: 382
Reply with quote  #3 
Attached is a sheet from the lab exercise we did in a training given by reliability solutions. We had electric motors coupled to a shaft and bearings. We then proceeded to take overall vibration and gE readings with a Monarch overall vibration meter through every step of the process of aligning the setup. We also measured the amp draw, RPM of motor, and temperature during every step. I wish I would have had my data collector with us to have some waveforms and spectrums to look at. Sorry, it is a little messy. Now that I do everything paperless on an ipad my work looks much neater!

Attached Images
jpeg IMG_4565.jpeg (274.18 KB, 27 views)

RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #4 
Thanks for posting the data. Can you describe the system? Driver/driven, coupling type, machine size (hp, weight), how it was loaded, where the measurements were (1,2,3,4). Based on the final readings, it looks to be very well balanced, and that perhaps misalignment was the only issue present? It certainly illustrates the benefit of good alignment, but I’m not sure it shows the benefits of “precision” alignment vs “good” alignment.
__________________
"The trend is your friend"
JuddJones

Sr. Member
Registered:
Posts: 382
Reply with quote  #5 
It was a small tabletop setup. Motor was probably 1 or 2hp. The driven was a (2" IIRC) shaft in 2 pillowblock bearings. It's been a few years, but I believe the coupler was a 4 jaw with flexible element. It was well balanced. One of our experiments was to purposely unbalance it and take similar measurements. The measurements (1-4) started at the motor NDE and proceeded to the Shaft NDE. I agree with your statement about good/precision alignment. I just thought the data aligned with your post and might facilitate conversation.
the_rookie

Member
Registered:
Posts: 8
Reply with quote  #6 
I should be able to help with this is the not-too distant future.

If a pump (direct couple) goes down during production we just throw a new one in place and go by "feel" for alignment, then do a correct alignment during scheduled down time.
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.