Just curious what you guys think about this. I've been asked to help try to predict a bearing failure on a rotary feeder. The feeder turns at 17 rpm and is failing due to what appears to be lack of lubrication. This feeder remains at a pretty constant 350 or so degrees Farenheit. The lube currently being used is only rated for 350 and is being basically burnt away leaving only a powder after a while. There appears to be no pattern to the failure as far as timing (it can go a month and fail or only go 5 days and fail) and the engineer assigned to this project is working with the feeder vendor to get another type of grease specified for the bearings. Bearings installed are stacked 6217 C3 shielded bearings on both DE and NDE. Vendor specifies that shims be used to shim the inner race to remove a certain specified amount of clearance to maintain proper bearing clearance due to the high heat. Bearings are placed side by side and clamped by the outer race. The clearance is measured by fitting feeler guages between inner races. Clearance measured then has a vendor specified clearance subtracted and that is the amount of shim placed between inner races as the bearings are installed. Failure mode doesn't seem consistent with a periodic event so conventional analysis seems worthless. I do, however, see a use for some type of high frequency detection being of use for the lubrication degradation. I'd like to be able to have a solution such as a 4-20 ma output sensor of some type that could feed a temporary signal back to our DCS and have it trended to watch for an impending failure until the lubrication issue can be squared away. What do you folks think?

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

Billy,

I am not aware of a gauge that can do in-line measurement of grease degradation but daily ultrasound or high frequency/demodulated vibration measurements could be effectively used and trended.

As a side note the whole procedure of setting up the bearings is little unclear to me. Do they insert shims/washers in two tandem mounted bearings to eliminate axial thrust/loading at high temperatures?

David

David, I was actually in the market of a good sensor to use to measure high frequency datat at the feeder by way of an accel or ?ultrasonic tranducer?. I would probably want it to feed back to DCS by way of a 4-20 ma signal. The only kicker is it has to be able to withstand upwards of 300 degrees Farenheit. I've got a vendor scheduled to come into the plant after lunch to see the application and see if she has sensor for the job. Good question on the shims David. I'm not real sure and we haven't got a good explanation from the vendor yet (German manufacturer). More to come hopefully.

Billy,

I know this wasn't your question, but 350 degrees is really hot and unless there is something special about the 6317 C3 bearing, you may be closing up the clearances.

Have they tried a looser clearance bearing?

In such a high temperatures you'll gave lubrication issues. Why don't you evaluate ceramic bearings or some kind of external cooling system?

Danny, higher clearance bearings are being discussed. As is doing away with the shims and leaving the C3 clearance as is.
Alex, we're hoping to get a remedy to the solution by way of the new lubrication being spec'ed out by the vendor. This is a fairly new install so we need to go through them first. I'd much rather call a bearing and lube vendor and get them in the picture, but it has been decided to go with manufacturer first.

Met with my vendor this afternoon and we are considering going with a high temp accelerometer connected to a 4-20 ma transmitter that will feed either an RMS acceleration value (can't remember the band width) or a "Peakvue like" value that is calculated in the transmitter. This will be passed on to DCS for constant trending and will hopefully be a good indicator of bearing degradation.

Good day, just a thought, you mentioned the setup for your "stacked bearing" as being adjusted by shimming the inner races for clearence. That seems to me to be the procedure for setting up an Angular Contact Bearing such as a 73xx series bearing. I have never heard of it being done on a Deep Groove Ball Bearing such as 62xx series. I guess the Bearing types have been checked before they were installed. I could be way off base here but sometimes the simple things are the ones overlooked.

If these bearings are running 350F running 17 rpm, then they either have a very tight clearance (why?), or a very heavy load (why use a ball bearing with minimal contact area for that)? Sounds like a misapplication to me. Don't assume that the "manufacturer" knows what they are doing. I'm with Sean on this one.... I'd question them about the bearings they are using.

I agree completely. My first question was why we weren't conversing with a bearing and lube rep to get to the nitty gritty of the situation. I was "informed" that we would be giving the manufacturer a chance to work things out first. I had to step back and reevaluate my position in the big picture. I was asked to try to find a way to predict the failure. My hands are sort of tied right now with getting to the root cause.

Jon Ellerbrake 402-467-5221 uses and oven 60' square @ 600F with 5-10 rpm shaft IN the hot air for 1-3 hours. It then goes through a cool down cycle. Give him a call to discuss what grease he is using 20+ yr old company. I also came across a company called www.Graphalloy.com.
Run hot, cold, wet or dry
Excel at -450F to 1000F
914-968-8400

Something to ask your vendor is what heat treat specification are the installed bearings. At those temperatures, you could be significantly reducing the hardness of the races and contributing to the failures. Bearing manufacturers have different HT processes for these high temperature applications.

I would also look closely at the grease, even though the grease is rated for 350 degrees, it may not be able to handle that temperature for continuous duty.