Is this possible? I have been monitoring this unit for some time with bearing frequencies present, Cage and Roller. Cage is .45x rpm. Today I see .227x running speed. I have never seen this before, have you?

1/2 cage frequencies are in Velocity. Spike energy shows 1x cage and harms.

This message has been edited. Last edited by: jbalko,

jbalko,
Anything is possible

Are you seeing this in the velocity fundamental area? Is there a 1 1/2ftf and 2 1/2x or just this 1/2x and 1x harmonics?

If this is a true 1/2x ftf that would mean it is happening every other revolution of the shaft rather than every revolution. May be that a roller is only "catching" or exciting the cage every 2 revs of the shaft. Also could mean if it is showing in the velocity data at the fundamental as a true 1/2 and 1x of the cage rather than as sidebands around something, such as the roller frequency, it may be in bad shape. Is it a machined brass cage?

The 1/2 are sidebands up higher, around 40,000 cpm. Could be sidebands around roller freq. The bearing is definitely bad. Just a matter of time till it is done.

Timken 329000

jbalko,

Could you show some of your data? (g's, TWF) Also, what type of mechanism is this? Is there a belt involved, or anything else that could cause this modulation?

Pinion gearbox driven. Just bearings and gear. No belts.

"Only nearby center freq is 3x or.." This is in a text box on your velocity spectral data. Could you clarify? What is 875 cpm... shaft speed? Do you have g's spectral data with a high enough Fmax to see 3x gearmesh? Also, it would appear your TWF does not capture at least 10 revolutions of the shaft (based on guessing without enough information). How is the transducer mounted and how close to the bearing is it?

875 is running speed. GEar mesh is no existant. See new plots. Gearbox had same bearing fail twice already. The velocity waveform shows over 20 revolutions. Pickup is mounted with magent within a foot of the bearing. All activity in the g's spectrum is roller and cage sidebands.

1_2cage.doc (160 Kb, 26 downloads)

quote:

Gearbox had same bearing fail twice already.

Jbalko,

Have we done a root cause failure analysis on this gearbox. Three bearing failures seems too excessive?, huh?

What are the gearbox teeth numbers?

28 teeth. Poor design in the gearbox. Root cause unfortunately is not my job, but the poor design and excessive impacts during wrecks surely contributes. Roller spin freqs. were present since bearing was new. I think they even used all new bearings this time.(as apposed to reconditioned).

quote:

28 teeth. Poor design in the gearbox.

28 teeth on input and output?

Yes, a pinion box, splitting 1 shaft into 2.

Assuming this is an AC motor.

New theory.
Direct Coupled
Ration 1:1
No load speed 900
Speed of motor 875
Slip Frequency 25
Number of poles 8
25 X 8 = 200

200/875= .227X

Is this maybe what you are seeing rather than 1/2 ftf?

motor is variable speed, dc. would that still work?

Jbalko

quote:

motor is variable speed, dc. would that still work?

Ralph

quote:

Assuming this is an AC motor

No, not with a "real" DC. Guess that is what one gets for assuming ANYTHING in life, which I normally do not do, but once again I have slipped into the old habit of doing so::::ASSUMING, so I will exit back into reality and stop ASSUMING.

To assume is to subject one to profound circumstances that could render to be quite embarassing, to say the least. IMHO.

To "take for granted" (Definition:"To consider as true, real, or forthcoming; anticipate correctly.")is another BAD thing to get in the habit of doing.

This message has been edited. Last edited by: Ralph Stewart,

Lets not call it an assumption, but another theory. Your theory actually made me think outside the box (pun intended). Maybe it might not be cage but sidebands from a modulating force outside the gear box. You never know. So dont apolagize. All thoughts are helpful sometimes in getting us to think a little differently about the problem.