Update on my double counting problem

Thanks For All The Help

After spending almost every moment since last posting working on my problem a solution presented itself yesterday.

I added a rung under the original (the one that counted the sensor closures), on that run I added a timer that was enabled by the sensor closure with a timebase of .50, then placed the done bit after the original XIO (which was a XIC) now it appears that if the closure is less than .50 that it doesn't get counted (the bird makes a closure of about .75)

I did many of the other things that were mentioned in the original post also, and today had a log that looked like something.

The timer idea came from of all places a sensor saleman that said he had a sensor that had a built in timer that could hide the false counts, (the sensor was 400.00 apiece) We didn't get any of the sensors but it made my project work.

Now what do I do Tommorow?

Thanks Again

Clint

Just incase anyone else did not see/read the part one of the story...

http://www.plctalk.net/qanda/showthread.php?t=33957&page=1&pp=15

Input debounce is an extremely common use of PLC timers.

Hook up an oscilloscope to a DC pushbutton input some time and take a look at the trace; you'll be amazed at how erratic the voltage is when you focus in at a small timebase. The same effect comes from any physical contact device like a limit switch, and can certainly come from solid-state devices like photoeyes or proximity switches.

Some PLC's have built-in filtering, usually configurable, on their inputs but typically it is only meant to address contact bounce or electrical noise and isn't more than about 100 milliseconds maximum.

Sounds like it was a very good learning experience about the way things in the real world outside the logic program affect your coding. Thanks for posting the followup.