Dl-05

I use the high speed inputs on these little guys for counting, have half the plant converted so far with great success. I was thinking of using one for speed, with a simple gear tooth proxy. Was going to approach it by perhaps taking count samples with a timer. Maybe take 1/4 or 1/2 second samples of the count and do the math and indicate speed. Not too sure how accurate this would be, there is the other option of using interrupts, but haven't used them yet. Just some random thoughts, any suggestions?

Have a great day all!

Assuming a constant speed your indication variance (as a percent) would be ((1/Number of scans in a timing period) + (1/Typical number of pulses in a timing period)) * 100

So the greater the time period or the greater the number of pulses the less variance as a percent. How long do you want to wait for a new update? What's the highest speed you can bring in pulses?

If the pulses are infrequent it may be better to measure the time between two pulses (they don't have to be adjacent, possibly the first and the tenth or whatever works) and derive the speed from that.

Thank Bernie, the highest frequency these can measure is 5000hz if I recall. It was just an idea we where throwing around a few weeks ago, to put speed indicators on some equipment to ensure the operators were in the target range. For that, time samples would work just fine, even 1/2 second, but I would rather put the effort into something that is not just 'slapped' together, and I like the idea of measuring the time between pulses. Then there is virtually no time lapse, and I don't have to worry about a small time sample missing a pulse and indicating 0 when that may not be true. Not sure how or if the dl-05's can do that, but will keep you posted.