Vibration

[A_123]

we have change the shaft and impeller for FD Fan after start up vibration in Axial Direction for None Drive End Bearing 4.5 to 5.2 mm/s. and temperature of Drive End (Plummer Block Bearing) 70 to 80 Celsius while None Drive End 55-68 celsius . what do you think the reason of increase the vibration in Axial and Is the temperature acceptable?

 

[Tmoose]

Not very much to go on.

Is the "vibration" filtered to 1x rotating speed ?
Or overall, on the bearing housings, of ISO 10816 frequency range, or some other?

Why did you change the shaft and impeller?
Belt drive, or direct coupled?
Anti-friction bearings, or sleeve/journal type bearings?
If anti-friction, cylindrical or taper bore?

"None Drive End Bearing 4.5 to 5.2 mm/s" . Does this mean the (1x?) vibration before was 4.5, and now is 5.2 mm's (rms?)
Was series of axial measurements made around the face of the plummer block?
Was a survey made that includes the vertical and horizontal vibration of 1x and phase from the bearing.

10 down, 10 to go.

 

[A_123]

vibration overall

change because the shaft bent that increase vibration, Runout 0.25 mm

direct coupled

anti-friction

Cylindrical roller bearing / double-row

vibration Reading Min 4.0 mm/s and max 5.3 mm/s (peak) that means overall

Axial Measured around the face of plummer block

 

[Tmoose]

"Cylindrical roller bearing / double-row "

I'm guessing they are really spherical roller.

I was inquiring about the bearing bore/mounting method.
Some have a straight cylindrical bore and are retained be setscrews or some other hopeful method.
Much better is a taper bored bearing, with a tapered sleeve that that is drawn into the tapered bearing bore, expanding the inner race and decreasing the internal clearance. The result is the bearing is retained securely, and concentric to the shaft. It is possible to get the tapered adapter in skewed with the inner race skewed also, so the race roller path wobble axially, causing all manner of mayhem if severe enough.

I'd check the axial runout of the inner race face.

Until you can get "filtered" vibration data this is 10x more like guess work.