We are interested in employing a low-profile surface-mounted temperature sensor that can detect spindle housing surface temperature changes for the purpose of spindle bearing condition monitoring. Is anyone out there using anything along these lines? Pros? Cons?

Yogi,

Try film sensor at: http://www.omega.com/ppt/pptsc.asp?ref=SRTD-1_SRTD-2&Nav=temc06
It is generally accepted that temperature is a poor indicator of bearing condition in many situations. Does this apply for your application? Bearing temperature for machine tool spindle can be affected by ambient temperature, coolant temperature, tool wear, shaft load, lubricant, and actual bearing condition. Have you considered other measurement methods like vibration and ultrasound?

As a matter of fact, we monitor bearings using both vibration and ultrasound. But there are certain spindles where we cannot effectively measure rear bearing vibration due to clearance issues. As far as the other factors that can influence spindle housing temperature, we are looking to find a delta range that would negate these, and yet get our attention before the "meltdown phase" of bearing failure. Have you used these for this purpose, or know anyone who has?

This message has been edited. Last edited by: Andrew Calcagno,

I have to agree with Walt as to temp being an indicator however, have you thought about Infrared. I use it on hard to reach conveyor bearings. Suspect bearings stick out like a sore thumb.
Kind regards

Walt,
How long have you been monitoring high speed spindle bearings? What type of equipment do you use? I would be interested in talking with you on this matter.

Yes, we use infrared, but guarding and other obstructions prevent this option. Also, we would certainly prefer continuous monitoring during the run mode as opposed to data gathering during off hours.

Yogi,

I really don't understand everything you didn't say! Why the interest in a permanent temperature sensor instead of a permanent vibration sensor (accelerometer). With either type of sensor, there is still the choice of periodic (occasional) measurements with a portable meter/analyzer or permanent monitoring with adequate capability to measure under same operating conditions. What can you "not say" more about the application? I know that the former great ball player now sells AFLAC insurance, but I didn't know that he was doing Spindles in his retirement!

Well, Walt, "you know a nickel ain't worth a dime anymore" so a guy's gotta do what a guy's gotta do...

So, anyway, yes, we have permanently mounted accelerometers to monitor the front bearings (class 7 angular contact, greased rotating @ 5000 rpm). The problem is the rear bearing set. In the past, we've had at least one get hot and seize due to coolant contaminated grease. (this was most, most likely caused by a failed high pressure coolant rotary coupling) We were unable to anticipate this failure using the front transducer alone. The reason why we don't have a transducer mounted on the rear bearing set is because of the clearance issue. (no room) The the sensor in the link that you posted earlier looks excellent. Thanks for the info. By the way, what is your experience with this sort of thing?

The future will come whether you predict it or not!
What type of measurements are you making at the outboard bearing and how far apart are the bearings? My first concern would be to maximize the usefull data from the existing accelerometer to detect increasing friction as the lubricant deteriorates. Parameters like high frequency acceleration, demodulated/PeakVue spectrum/waveform are good friction indicators, if the spindle is unloaded so that background vibrations are consistently low. Have you investigated alternative accelerometers to fit in the available space on the inboard bearing or fixed stinger-type mount or rigid housing link for conventional accelerometer mounting? Its easier to spend a dime for a nickel's worth of data, and sometimes possible to get a dimes worth of data for a nickel!