New Vibezone

[ Home | Contents | Post | Reply | Next | Previous | Up ]

Re: Cracked Roll Journals

From: Arne
Remote Name: 217.210.179.50

Comments

OK, that is important information. I read your test so that you inform that the actual shaft has cracked where it is inserted in the main roll material. A crack there is due to mainly two issues adding up: One is the actual bending stress. How has the tension in the machine here been changed during the years? The other is the stress concentration due to a sharp transition from the cylindrical shaft where the inner ring sits to a shoulder of some kind. The radius of that transition is very important. A very soft (large radius) to a sharp edgy step can differ in crack starting sensitivity with several multiples. To avoid a future crack start, you can bring the roll to a lathe and turn out a nice soft transition. To find a crack growing is not impossible. As the crack is there, the roll will change its resonance. Softer and hence lower in the crack direction. So regular free bump tests can be one way. The freq. should stay the same in all radial roll directions turned to point downwards. When running, a crack that opens and closes during a turn can cause multiples (harmonics) to develop. Normally not very offensive signals. But I do have a pump cracking case recorded like this which showed good vibrations well in advance, although nobody really cared..... I would rather consider to invest in a good improvement of the design backed by a proper calculation of the stress knowing the load on the roll. It should give you a good sleep. Regards Arne

Last changed: October 16, 2007