Using Motion Camming (MAPC) to couple a servo to a laser micrometer

I am trying to couple a servo axis to a non-motion signal, and am currently stuck in a Try->Fail loop.

The axis needs to follow the laser micrometer measurement at 1:1 scaling. I tried using a virtual feedback axis and then using MRP to constantly define the position of that feedback axis with the micrometer's signal, but no luck.

What's the best way to do this?

Why not just send the laser Mic measurement reading as an absolute position motion command?

I don't understand either. If the ratio is 1:1 then why not simply gear the servo axis to the laser position.

Because gearing, like camming, requires the master to be a motion axis. The laser micrometer measurement is just a value coming from an Ethernet/IP source. I was putting that value into a virtual axis via a continual MRP, but that doesn't work.
If I use a simple MAM instruction, then I have to re-initiate it every time the measurement changes, which happens constantly over a 30-60 second period. Further, I only want to track in the positive direction.

phuz said:

Because gearing, like camming, requires the master to be a motion axis.

Click to expand...

There is some trick you ware missing with your motion controller or you have the wrong motion controller. This should be simple. We gear hydraulic servo cylinders to encoders on feed chains all the time.

An L81 processor is more than capable of handling any motion instructions, which is what we have on this project.
I have used encoders as an axis before for camming instructions, but how would you take a regular analog value into an axis?
This is a Keyence micrometer which comes in via a GENERIC module in the controller tree.

Since it is setup as a generic module, I don't think you will be able to mimmick it as an axis because, at minimum, an axis has position and speed. So to make an imaginery axis track, you would have to MCP and MCD on every scan. I don't know that RSLogix will allow you to do either while a cam/gear is active on that axis.

I suspect the answer might be that you need a different module profile other than generic - one that can appear as a CIP axis.

You current solution may be the MAM on every scan - being careful with your selections for merge and merge speed in the MAM - as well as your velocity/accel etc.

I'm very very curious if some one has a solution. Thanks for posting the question

For now, I am using MAM conditioned with a XIO(MOTIONINSTRUCTION.IP) and a one-shot, which seems to work. As I may have mentioned, I was just looking for a cleaner solution, but I can live with this for now.