I had a very interesting discussion this past week with a customer about Lube Systems and how they’re always the first thing to be blamed for any breakdown. It seems that with any failure in a mill the easiest thing to blame is not having enough lube, but that isn’t always the case.
Most systems have (or should have) some sort of monitoring. Depending on how complex the system is, or how critical the application, the more things need to be monitored. I half joke with my customers all the time by telling them “There is no problem that either time or money won’t solve.” Taking that adage and applying it to what we do can solve a lot of potential issues.
Here’s an example.
We had a customer that had a lube system on a bio-sizer, which is basically a high speed hog, and over the course of three weeks they had eight bearing failures. We were asked to come on site because they felt it was the system we supplied that caused the failures. We checked the system out and confirmed every possible alarm condition and that the PLC was receiving the alarm signals, which it was. The time to failure was approximately 8-10 hours of run time. The system had been in place for about 2 years without any issues prior to the breakdowns. So what changed?
The customer had a major bearing manufacturer’s engineer on site and he was thinking it was lube related as well, which didn’t help our cause. They had no condition monitoring on the bearings, other than the small magnetic heat sensors on a bearing housing that was about 6” thick, solid steel. By the time any heat was observed on the outside, the bearing was already in failure.
I told them that if it was actually the lube equipment failing, it would have happened regardless of what they ran in the unit. I also told them that increasing the viscosity of the lubricant would give them a bit more protection, but the downside would be a slightly higher running temperature because of the viscous drag. After changing the oil to a 68 from a 32, the equipment ran 70 hours before failing. That proved to me that there were other factors causing the failure.
After the seventh failure I successfully made my case to the customer that they needed a more sophisticated form of condition monitoring. They installed RTD’s (temperature and vibration probes) that went through the housing and rested up against the bearing.
The application had an 800hp motor, started by a small 100hp motor that used a variable frequency drive to slowly ramp the speed up over twenty minutes. At that point, the 800hp drive was brought in at about 85% of the running speed. That’s when things got interesting.
The following graph shows a lot of other information, but if you look closely at the green line at the top you can see 3 distinct and intertwined sine waves.
Down in the bottom left, right by the -2, the green line is the vibration signature. At the red vertical line, the green line starts to go up. This is because the smaller pony motor is starting to spin the hog. It goes through its first critical frequency, then when the blue line (motor amperage) goes straight up, the big motor is brought online and that’s when things start to shake.
I looked at the graph and told them they had an electrical issue with the motor. One of the three phases has gone bad, as proven by the three distinct wave forms. The uneven pulling was generating a pulsation frequency on the drive that was literally shaking the rolling elements in the bearing, causing the cages to be worm away from the inside. Once that rolling element support is gone, the bearing fails. That was proven with the increased oil viscosity which allowed the bearings to run longer. They ran longer because of the added film strength of the thicker oil, which helped to cushion the rattling of the rolling elements.
Yes, the failure was lubrication related, but it had nothing to do with the system or even the lube they were using. The situation took the loading outside the normal operating condition and that’s why it failed.
I will probably never see a failure like this again in my life but I was really glad to be there for this. To be able to experience a failure like this and get the graph with data on exactly what happened is invaluable. The learning opportunity is tremendous and I share this one wherever I go because knowledge is power. Instead of struggling through a failure like this again, we can apply this insight and look to see if this same signature is showing up earlier, rather than fixing eight bearings before considering this possibility. As Einstein famously said, “Doing the same thing but expecting different results is the definition of insanity.”
I have experienced many situations in my career where outside factors took the normal condition and turned it upside down. In each one of those, the lube system was at fault, but was not the root cause. It was interesting to investigate this case and cross off different possible causes with a systematic approach, sound mechanical theory and an understanding of how outside factors can influence and cause lubrication system issues.
You can learn to troubleshoot any piece of equipment, but sometimes it takes a different approach. Follow Einstein’s guidance and stop the insanity!
Author: Dean Maier
Dean has been in the automated lubrication business for 25 years, with 85% of that time spent in forestry applications. We are delighted to have his insight with this latest blog.
Industrial Autolube International Inc.
Do you want the next article delivered to your Inbox? Subscribe here for ideas and tips on getting the most out of your mill.
