Hello. With the EPIC, PAC control, and RIO as digital inputs, what is the minimum pulse width I can detect on channels 0, 1, and 2? I will be working with OnLatch and OffLatch commands.
Hello. With the EPIC, PAC control, and RIO as digital inputs, what is the minimum pulse width I can detect on channels 0, 1, and 2? I will be working with OnLatch and OffLatch commands.
Only channels 0 and 1 can be used for latching on a groov RIO. The max. frequency is 10000Hz. The min. period from on-latch to the next on-latch is 0.1 msec. See below for more details.
What are the specs you need?
Thank you for the quick response. I will use the RIO's counter function for the first two sensors and the third channel for the other sensor. My question is whether the RIO's third channel, used as a digital input, can detect pulses of approximately 400 milliseconds with Node-RED or Pac control.
Yes, 400ms isn’t a problem in PAC Control when configured correctly.
Node-Red typically has a 1 second polling/inject time. There are ways around that though. For fast IO reads in Node-Red, I will use OptoMMP streaming (when there isn’t a lot of post-processing needs) or a hand-rolled OptoMMP read request loop with short delays. I use this to detect button presses on RIOs when the first two channels are in use.
In my opinion, it is silly that the RIO doesn’t support latching on all digital inputs, even if the other channels are lower performance. It would be much easier to have a start and stop button on a RIO when you also have a flow meter if this was built-in to the firmware. These work arounds are much more resource intensive.
I totally agree Phil.
Hi Philip
Is there a way to filter out bounces on the RIO's channel 0 and 1 counters? My sensors are mechanical (axis counting pads) and I need to filter out the bounces.
Not built-in that I am aware of. To use the built-in counter, you will need a hardware debounce circuit tuned to your process. Otherwise, you will need to build a software implementation. If 400ms is the longest period, a sample rate shorter than 200ms, yet longer than the debounce time, will work for you to track state and then count the changes from 0 to 1. If your bounce is longer than 200ms then you will have trouble.
Since it sounds like you have a controller with PAC Control, that would be the simplest place for a software implementation.
Thanks for the help. It would be interesting if the RIO counters had filters to eliminate bounces; I’ve seen them in small PLCs.