help in understanding the HSC and PLS in a ML1100

[mracer 2]

I have a project, that I have been working with, that I think may work better with a high speed counter and the programmable limit switch instructions. The project is a motor controlled by a powerflex 4, a 40:1 gear box with a sick photo eye looking at shaft rotation. I have a disk with 26 flag's on the output shaft that the eye is watching. It will need to wind the output shaft up a number of counts, slow down, creep to a number of counts and stop. Then the drive will rev. and repeat the process, unwinding and stopping in the same manner. My program is set up the slow the drive speed down in 2 steps, ( high speed 60, slow down-15, creep-5) I have it working but I keep loosing counts when the speed gets to the upper end of the operating range. I have a SCP that changes the counter Pre based on the drive's high speed set point. One thing that could be going on is the shaft keeps moving after the motor has been told to stop. I have the Decel set vary low that seemed to help a little. I have read the Instruction description for the HSC and PLS but I still don't quite understand if this will work for my application. Can some one share a little incite on these? Racer Edited 10 Dec 2007 by mracer

[Ken Roach 194]

First, a little response-time math. 60 hz is 1800 RPM on most motors, and you have a 40:1 reduction so at 60 Hz you're going about 1800/40 = 45 RPM. With 26 flags on the gearbox output shaft, the photoeye is sensing 45 * 26 = 1170 pulses per minute. 1170 pulses per minute is 19.5 pulses per second. If your sensor disc has 50/50 spacing, the ON-time of the pulse is 1/39 of a second = 25.64 milliseconds. In order to accurately count each pulse, you need to be able to sample at least twice in each ON period. Your PLC input and scantime must therefore be less than 12.82 milliseconds. That should be very easy for the MicroLogix 1100. Check the filtering on the inputs you are using; the default value on DC inputs is 8 milliseconds. They are DC inputs, right ? Make sure the photoeye doesn't have a debounce delay. Try uncoupling the load and doing a check for how fast the output shaft of the gearbox stops after you send a STOP command. The momentum of the load may be generating enough "back EMF" to the drive that the drive is automatically extending the deceleration time to avoid tripping on a DC Bus Overvoltage fault. I'm not a PF40 application expert so I'm not sure which parameters to check. Post your code if you can.

[mracer 2]

Ken, I still have it just setting in the office/workspace for now. I did play with the filter settings but it didn't seem to change anything. I did the math before talking to Sick about I ordered the eye and got the same cycle counts for the input. It will have vary little load on it when it is running. It will be used in our R&D area to do some cycle testing on our products. I have a short video of it running that I tried to post but couldn't get it to work. Then this laptop crashed in the middle of doing the post. I should have checked to see if the first one had posted. Sorry for the double up. I really think it has something to do with the counters and there presets.

[mracer 2]

OK, Here is what I have working to this point. It is doing a good job of counting and changing direction, but I still have a problem with the repeatability. The counter will count sometimes 1 or 2 counts after the stop. I think there should be a way to move the ACC. count from "forward" to the /Pre for "Reverse". At this point I just See "trees" lol Is there an easier way? Point me in the right direction. Thanks Racer RD_Motion_RSS.zip.zip