dear friends ,

can any body tell me what is the reason for mounting two probes at 90 deg for measuring bearing vibration

You can see orbits shaft movement from stand still plus many other diagnostic purposes.

Check this thread for a good understanding of orbits.

http://maintenanceforums.com/eve/forums/a/tpc/f/3751089011/m/8031086423

John from PA

Of interest may be established conventions for X,Y vibration mesurement:
http://www.reliabilitydirect.com/appnotes/convent.html

dear all,

thanks for the reply

can any one tell me why the probes are mounted as per case 1 and case 2

regards
kns

Doc1.doc (51 Kb, 35 downloads)

If you are using cap reading; i.e. an accel the 90 degrees or Horz & Vert are a good analyses tool. Example: from the direction of rotation or CW for this example the Horz would be on the right below the split.

It seems there are a number of opinions based on what you've given but X-Y may mean (and usually does) eddy current probes and they may be mounted at 90 degrees apart but rolled 45 off top dead center.

You may want to further clearify to obtain additional info. Or, that's my take.

KNS the general convention for prox probes is that the machine rotates from the X probe to the Y probe. When viewing the orbit it does not matter about the orientation of the probes as long as they are at 90° to each other. However there are some problems associated with prox probes if only looking at the numerical data. If the machine is horizontal then it is likely that the orbit will be elliptical because of the gravity loading. There fore if you use prox probes mounted as per case 2 they will give quite different readings. If mounted as per case 1 then they will give similar readings but neither will be the true indication of the shaft motion. When trending prox probe readings it is sometimes best to trend the Smax value which is the max value across the orbit. The othe rgood reason for mounting the probes as per case one is thet if mounted through a bearing housing both probes will be in the removed top half whereas mounting as per case 2 is difficult as one probe will be on the half joint.

Regards

MDE

quote:

KNS the general convention for prox probes is that the machine rotates from the X probe to the Y probe

I would have to take exception to that statement. Some people have and do use this as a 'convention,' but the more general convention is and has been to set x and y up to agree with the input to an oscilloscope. When so done, this presents a picture of the shaft centerline that agrees with physical reality.

To accomplish this one establishes a viewing direction for the equipment. For horizontal x, it would be to the right and vertical would be on top from the viewing point or in the viewing direction. Keep this relationship between x and y if they are rotated from true horizontal and true vertical.

Yes, if one transducer is on the bottom and the other on the left from the viewing direction, x is on bottom and y is to the left.

quote:

If the machine is horizontal then it is likely that the orbit will be elliptical because of the gravity loading. There fore if you use prox probes mounted as per case 2 they will give quite different readings. If mounted as per case 1 then they will give similar readings but neither will be the true indication of the shaft motion.

I’m not sure if the conclusion of the above statement is that the two methods of probe mounting will generate different orbits. In fact, when the software is configured correctly, the orbit will be generated correctly regardless of the angels at which the probes are mounted, mounting of the machine and the vibration contents. Same is true also for the shaft average centerline plot.

In many cases the relative probe cannot be installed in the true horizontal due to the split line between the bearing halves. And that is why other direction is selected.

That is just my humble opinion and I could be totally wrong.

API calls out mounting the probes at 45 degrees from TDC.

One section of API 670 states the following.

quote:

Note: The convention for X and Y probes when making piston rod
drop measurements is to view the probes from the crankshaft looking
towards the cylinder. The probe referred to as “Y” is always
located 90 degrees counterclockwise from the probe referred to as
“X,” regardless of what vertical or horizontal orientation they may
have.

Another part of 670 indicates,

quote:

a. The position of each probe in relation to the machine bearing.

Note: The direction of shaft rotation does not affect the X and Y
probe location.

The reason for mounting probes 90 degrees apart is that vibration is directional. Mounting the probes in this manner generally allows one to see at least 70% of the highest direction of vibration (Smax).

quote:

I’m not sure if the conclusion of the above statement is that the two methods of probe mounting will generate different orbits. In fact, when the software is configured correctly, the orbit will be generated correctly regardless of the angels at which the probes are mounted, mounting of the machine and the vibration contents. Same is true also for the shaft average centerline plot.

Orbit and shaft centre line plots will be correct, but some control systems only display trends of a single point values. In this case the trend value would be more accurate using the Smax derived from both transducers.