Trigger camera off encoder counts in Rslogix 5000

I am running into some problems trying to trigger a camera to take a picture every 15 degrees when I spin the part I am looking at one full rotation. Here is what I am working with:

PLC-AB 1769-L18ERM-BB1B
Motor-AB TLY-A130T-HJ62AA TL-Series Servo 502.200/EA 502.20 Motor
Driver: AB 2097-V31PR0-LM
Rslogix 5000

Both the motor driver and camera connect to the PLC through ethernet IP. The problem I am having is that when I increase my speed and try to trigger the camera based on the encoder count, it will often not take the picture. I believe that the cause is that although the motor driver sees every encoder count, it does not send every encoder count to the PLC. I think it does not send data fast enough to the PLC for it to recognize the the encoder position was reached.

I have tried slowing the motor down, and looking for ranges of encoder counts instead of exact encoder counts. This helped but I am not able to get precise enough positions in the time I want. I think I need to find a way to increase the speed at which the PLC receives the encoder counts. Any tips on how to do this would be appreciated or please me know if there is a better way to approach this.

Thanks

This is where a PLC isn't going to help you at all. By the time that the servo gets around to sending the signal out, the PLC responding to it, and then sending a command to the camera, and then the camera finally taking a shot, you could be anywhere.

If at all possible, you want to go with a separate encoder, or better yet, a switch that makes 24 times per revolution. You could put a gear on the rotary part and then gear it down 24:1. Then a simple cam switch closing a contact would be all that you need (assuming the camera responds to trigger signals of course).

this might have ZERO to do with your problem - but it might be a good idea to take a look at this post ...

http://forums.mrplc.com/index.php?showtopic=26162&p=125399

be sure to listen to the WAV file - you can actually hear the "I/O task" problem being discussed with your own ears ...

Thanks for the responses. I figured out where I was going wrong. I was monitoring the encoder position and using compare logic to find when it is at the position I wanted. The MAW looks like it was designed for just that and works correctly.

Motion Axis Watch and Motion Axis Output Cam are both excellent for making discrete triggers based on a precise position during a motion move.

I'm doing a project right now with a ControlLogix and the super-fast 1756-OB16IEFS scheduled output module. My servo axis is also a Kinetix 350.

While my motion profile isn't challenging (it's moving at a constant speed with a constant load), the precision I'm getting is impressive. The duration of the output is crucial in my application, and I can adjust it with a precision of about 20 microseconds.

There's an Ethernet-based output module that can participate in these sorts of motion output cams, the 1732E-OB8M8SR.

If MAW works for you with the hardware and speed you want, that's the easier way to go. I just wanted to post about that Scheduled module because I'm really excited about the performance I'm getting from mine.

Hey Ken,

I've looked at those scheduled output modules before and though I've never used one, I'm quite interested in their performance because I've done some high speed camera triggering before (I don't have the speed specs, I'll have to check when I get to the office in the morning) . I believe they are limited to a resolution of 1/16 of the coarse update period of the motion group, so I'm curious which controllogix you are using and how many axes you have and what the CUP for your motion group is?

Thanks!

In my case I have a 1756-L73 controller running just two Kinetix 350 drives, with a 2 millisecond coarse update period. I'm just testing a single 1756-OB16IEFS module now, but will eventually have 5 of them, so I'll be using up 5 of the 8 available output cam targets for the axis.

It's an L73 because we are doing some massive array handling elsewhere in the program.

The 2 millisecond coarse update rate for this motion group gives me (0.002/16 =) 125 microsecond precision on firing the output.

In my application, the precision of the start position isn't as important as the precision of the duration, and from what I can tell the module processes the Left and Right positions and the Duration all onboard.

I'll know more once I roll it out to all 80 of my high-speed outputs.