Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Shurafa

Sr. Member / Supporter
Registered:
Posts: 310
Reply with quote  #1 
Temporary Mounting for Shaft Relative Vibration Probe

What is your experience with temporary relative probes installed for diagnostic purposes? To what extent magnets gave you with reliable signals?

Many years ago, I witnessed a case of externally added probes on a boiler feed water pump. The reading was noisy. I'm not sure if many of you tried this setup with real plant conditions but it would be very interesting to hear from you about your experience/thoughts.

Regards- Ali M. Al-Shurafa
OLi

Sr. Member
Registered:
Posts: 1,916
Reply with quote  #2 
We have used the standard dial gauge magnet and holder kit for many years on many different items, latest on a gasturbine driven generator shaft. Noise may be from the surface that normally is not treated in these cases specific hydroturbine shafts may be rough otherwise it works fine. If you need the data it is good to have. I can't really say that it has been a real problem solver on the other hand. Maybe mostly the other way around a way to exclude possible reasons for trouble but that is another difficulty.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #3 
Quote:
Originally Posted by Shurafa
Temporary Mounting for Shaft Relative Vibration Probe

Many years ago, I witnessed a case of externally added probes on a boiler feed water pump. The reading was noisy.


Keep in mind that when adding external probes, noisy signals are not necessarily related to the mount, but rather the surface finish and geometry of the target area. I have encountered many feed pumps that have had rusted surfaces immediately adjacent to the seal area. Even if rust isn't visible, then someone may have cleaned up the area which can also affect the signal (flat spot. High frequency noise, etc.).

I have only used a magnet to hold a proximity probe on only few occasions and then it was a situation where I had few if any alternatives. However, given a good target area, I have applied small drilled and tapped metal blocks with a suitable adhesive very frequently. These blocks were a standard part of my field kit.

vogel

Sr. Member
Registered:
Posts: 76
Reply with quote  #4 
Hi,

probably John and Oli have a broader experience on this, however my 2 cents:

I've used prox probes with temporary magnet mounts in several occasions with mixed results. Noise is one issue. As they've already said, noise comes usually from rough surfaces. From my experience too, feed water pumps are horrible in this sense. I've seen data of fwp retrofitted with permanent mount prox probes in which the target area was not treated and the noise was there.

Another concern is the support stiffness. If the bracket/probe holder or whatever you use is not stiff then it will vibrate and the measurement will be distorted. Try to make the probes mount as stiff as possible. I think that mounting brackets like these should work: https://reliabilityweb.com/articles/entry/temporary_mounting_of_proximity_probes

When evaluating the data, I would take into account that the measurement is not on the bearing section. Depending on how far the probes are from the bearing and the shaft mode shape, this may have a big impact in amplitudes.

Generally speaking, I would be very careful to make a call based only on this type of data, but it usually rules out some possibilities. Sometimes, mounting accelerometers on the same measurement points may be helpful.
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #5 
Quote:
Originally Posted by vogel
From my experience too, feed water pumps are horrible in this sense. I've seen data of fwp retrofitted with permanent mount prox probes in which the target area was not treated and the noise was there.


"Vogel" brings up a good point.

Over the years I have seen many a feed water pump have its bearing journals repaired by chrome plating.  A knowledgeable shop can do this well and the result is a base metal probe target area with proper mechanical and electrical runout.  But it is also very easy to get a shop that doesn't know how to properly treat the probe target area.  You end up with a chrome/steel boundary area that is very near useless for diagnostics.  To find good mechanical and electrical runout in a probe track outside the casing of the machine would be unlikely.
OLi

Sr. Member
Registered:
Posts: 1,916
Reply with quote  #6 

On the other hand, don't make the mounting so stiff so you get a resonance at 50 or 60 Hz or whatever speed the machine is running, that will be seriously bad.

Here most remaining large nuke turbines have/had eddy probes on purpose mounted as far as possible from the bearing to get a better signal to noise ratio so everything is not always as API describe in this world. Some also had a cube of different material sunk in to the shaft to verify calibration in operation, funny things people do.


__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
ivibr8

Sr. Member
Registered:
Posts: 163
Reply with quote  #7 
On several occasions I used prox probes mounted on "L" brackets, 45 deg from vertical. It is easy to make and install (I used 2 part epoxy to hold onto the casing). I used them on our SSTG units (NAVY) that are supported in journal bearings. The Navy does not utilize prox probes but they proved VITAL in figuring out the root-cause problem (vs. accel casing readings).

As others have warned, look closely at the shaft surface as it adds what B/N calls "glitch". In our case, the shaft was in good shape so I simply needed to get appropriate "slow roll" vectors to obtain good information at speed.

Regards
Jim P

 
Shurafa

Sr. Member / Supporter
Registered:
Posts: 310
Reply with quote  #8 
Really interesting comments. Thanks a lot for every one.

Regards- Ali M. Al-Shurafa
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #9 
Quote:
Originally Posted by OLi
Here most remaining large nuke turbines have/had eddy probes on purpose mounted as far as possible from the bearing to get a better signal to noise ratio so everything is not always as API describe in this world.


How exactly do you achieve a higher signal to noise ratio being far from the bearing?

On a related note, some other designers in my department were working on a highly engineered pump that required mechanically quite a long distance from the bearing to the probe. As a result, the probe displacement was not equal to the shaft displacement at the bearing, so they convinced the end user to add 12% to the limit for first mode displacement and 5% for second mode displacement to account for the higher relative displacement at the probe. If I remember, its explicitly forbidden by API, but when you have such novel equipment, you gotta take what you can get, I guess.
OLi

Sr. Member
Registered:
Posts: 1,916
Reply with quote  #10 
Well, if your aim is to monitor bending of the shaft and not the futile oil film thickness related things you get a higher signal further from the bearing, normally and as your example indicate sometimes accepted by the enduser.... So if the signals are mainly used for balancing you get a higher signal level and the noise is about the same or at least not depending on axial position so signal is higher and noise the same so signal/noise relation is higher. There are more than 10 larger than 500MW turbines following that principle around here and they made it the last 30+ years.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #11 
Quote:
Originally Posted by OLi
Well, if your aim is to monitor bending of the shaft and not the futile oil film thickness related things you get a higher signal further from the bearing, normally and as your example indicate sometimes accepted by the enduser.... So if the signals are mainly used for balancing you get a higher signal level and the noise is about the same or at least not depending on axial position so signal is higher and noise the same so signal/noise relation is higher. There are more than 10 larger than 500MW turbines following that principle around here and they made it the last 30+ years.


[rolleyes] whoops, I was expecting something more than just measuring as close to an antinode as possible. I don't know why, haha.
OLi

Sr. Member
Registered:
Posts: 1,916
Reply with quote  #12 
Nop nothing fancy, just mechanics..... Otherwise it could be something to sell and it would be used everywhere....
By the way, anybody tried capacitive probes as they don't add the surface problem by not being eddy current?

__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.