Three assembly lines send finished ranges to a palletizer. The palletizer puts each range onto a pallet, glues and staples a box around this arrangement, barcodes the box, then sends the result down the line to a barcode reader that will send a given unit to a preselected shipping bay.
This obviously complex system is under PLC control. To prevent finished product from piling up and getting damaged (and also damaging the production line), the palletizer uses a jam detection system that sends a signal to the PLCs of the production lines to slow down or stop. The jam detection system is designed to anticipate a jam and intervene before it happens, rather than wait for it to happen and damage to occur. It relies on a line speed loss calculation that takes into account temporary speed loss due to the weight of a range suddenly being introduced to the line.
This has worked very well for a long time. Usually the problem is excess glue at the palletizer; it drops into the conveyor and gums up the rollers, resulting in a loss of speed. The mechanics have a regular PM for cleaning glue drippings and generally maintaining the gluing system. In addition to preventing a line speed loss, these efforts reduce gumming up the stapling system that is immediately downstream of the gluing system.
But over the past couple of weeks, several slowdowns and shutdowns have occurred, and the mechanics have assured the plant engineer it’s not due to excess glue. They also checked the gear oil levels in the various gearboxes and ruled those out.
Now you have been assigned to look at this from an electrical standpoint. What should you do?
First, use a thermal imager or spot temperature gun on those gear boxes. They should not be more than a few degrees above room temperature. If they are, there is some problem such as worn bearings or the gear oil is contaminated and needs to be changed.
But before simply changing the oil, have a qualified person check the bearings with an ultrasonic instrument capable of assessing bearing noise. These particular bearings are probably spherical roller bearings, and they may be gummed up from contaminants in the oil. They may be pitted or have other damage. Ask the plant engineer to address these issues to the mechanics. And don’t forget the conveyor line itself. If there are 50 rollers on the palletizer line, how many sticky ones will it take to cause the line to slow down too much? That’s another question for the mechanics.
Tell the plant engineer this problem could be due to an accumulation of issues and you will look into the electrical ones while the mechanics go beyond merely checking the oil level in the gearboxes.
Turn your attention to the palletizer drive motor. To ensure the motor is in good shape, have a qualified person assess the motor bearing health with a suitable ultrasonic instrument. If even one bearing is on its way out, the motor is going to draw more current to produce its normal torque. Have the motor overloads blown and had to be replaced any time in the past month?
Has anyone notice a line speed loss when a range has been added to the line? There is no electrical issue that will cause the motor to turn at less than its design speed, although excess load can cause it to bog because it simply doesn’t have enough torque to handle the load. If you can monitor the motor current and look for times when the motor seems to really strain a gut, you may find another contributing factor. This goes back to the gearbox and roller issues, but also consider what is being put on that palletizer line. Has a new, heavier model of range been introduced lately?
Next, look at speed sensors. Is each one mounted securely and in the correct orientation at the correct spot? Is their cabling in good condition? Is their cabling separated from power wiring? Has one been replaced recently? If so, perhaps it is the wrong type or the installing tech routed the wiring in such a way as to induce current from a power conductor into it. If the original sensor was a magnetic one, then it was immune to interference from dirt (or even glue) because you can’t insulate against a magnetic field. But an optical one might not be suitable for this application; was a change made here? Finally, perform a calibration on each sensor loop.
In a PLC controlled system, the last thing you should look at is the PLC. Think about this. The final control element (valve, motor, etc.) is subjected to the most stress so is the most likely part of the system to fail. The input side is normally subjected to less stress than is the control element side, but sensors are still exposed to the process so they are the next likely failure area.
The PLC is not exposed to the process, and software doesn’t wear out. But PLC hardware and connections in the PLC cabinet might have gone bad. After you check the final control elements, the sensor inputs, the PLC hardware, and all connections, only the PLC logic remains. Did someone tweak things so the band around the setpoint for speed loss is tighter? If it had an 8% band, is it still at 8%? Don’t go tweaking in there to adjust anything from the normal settings, because this system worked well for a long time. Look for what has changed, and change it back (after recording what you found to have changed).
