I'm setting up a Predictive Maintenance Program at a company that did no PdM before now. They have conveyors handing heavy material. (Iron Ore, Coal, Limestone)
In doing some research, it's seems people are using a varity of methods to find potential failures, and are sucessful in preventing them. If you were starting a program from scratch, would you use Vibration, IR Thermography, or Ultrasonics or something else to monitor them?
FYI....There are 4 rollers per bracket housing, X's a couple of thousand bracket housings per conveyor. Some of these areas are very dirty, with dust, dibris, and water. I hate to buy some equipment and destroy it the first time out.
Also, The Rollers DO have grease fittings.
Please help!

Do you have safe and quick access to both sides of the conveyor to touch each bearing housing? If not, then you have to rule out any type of contact sensor and consider airborne sound or ultrasound. If yes, then you can consider a contact sensor with a stinger that measures ultrasound, shock pulse, or demodulated vibrations. A magnet base on a vibration sensor (accelerometer) could be too difficult to mount with the curved surface and crud on them. Traditional vibration analysis is very slow for screening or finding suspects. I have worked on the great lakes vessels and heard stories how previous attempts at conveyor monitoring were unsuccessful. Does every bearing have an identification tag that can be used to identify suspect bearings? How often does the conveyor change speed? What ever method you choose, it have to be rugged, very fast and accurate. A false positive or false negative bearing fault diagnosis can really sink the program very fast.

Walt
w_f_strong [at] msn [dot] com

I would also point you in the direction of ultrasound. I have had and heard from my Level 1 students of some excellent success using this method.
Since inappropriate lubrication is likely to be one of your biggest killers, implementing a lubrication-on-condition programme using an ultrasound system with trending software will work well for you for sure.
Implementation is quite straightforward and the learning curve is not so steep!
Best Regards,
Tom Murphy
tom@reliabilityteam.com

Tom,

What method of ultrasound was being used on the conveyor bearings; microphone sensor or contact sensor? How well did the meter and sensor continue to work in the hostile environment near a conveyor (especially coal and iron ore)? Were dB limits set or just audio in headphones? Was meter by SDT or another brand?

Walt

Newbie,

This refers to standard troughed belt conveyors with a multi-ply rubber belt being carried over troughing idlers. The most typical belt conveyor for what you describe.

If the idlers are made to CEMA standards their will be a letter in the middle of the spec number that will give you a hint as to what type of bearings you have. If your conveyors are fairly small (ie less than 200 tph). They will be CEMA C. Smaller is CEMA B. B Idlers will probably have "lubed for life" bearings and C or higher will have regreaseable tapered rollers.

Most idlers have 3 rollers with one hor. and 2 on a 35 deg angle to form the trough, so there are 6 bearings. The center bearings are almost never going to be safe to access directly because of the belt going 500 fpm 3 inches away.

Ultra-sonic is the way to go, imo.

I don't actually do this anywhere because it has to be done really cheap. Idlers don't cost much and don't usually create that big a problem when run to failure.

New2PdM,

Contact me if you would like some specific recommendations for this tough monitoring project.

Walt
w_f_strong@msn.com

Hello Walt,
my experience comes from working with the SDT170. I am aware of people using contact and/or airborne.
The contact probe is of course more desirable because if I have access with the contact probe then I will probably also have the opportunity to perform on condition lubrication.
Nevertheless, where a quick low-cost non-contact method is necessary/required ultrasound seems to deliver.

Best Regards,
Tom Murphy

I used an SDT150, with the contact probe, to check roller conveyor bearings on a bakery oven. It worked great, and was easy for the mechanics to do.