Using 1734-SSI encoder signal as a position reference for PF527 axis

Hi all,

I have an application coming up where I have a motor that runs forward or backward to adjust the linear position of a machine component. Position feedback is via an encoder wired to a 1734-SSI module. Previous iterations of this machine have just used a standard VSD, done a simple position compare, given it a speed and a direction and stopped when it's within a specific tolerance window. The adjustment is very slow, and the precision requirements are not unusually high, so this approach works just fine.

However for reasons unimportant to this question, I'm using Powerflex 527's on all motors for this project, which are motion class drives, and I had the thought that if I could turn this into a position loop, the positioning would become remarkably simple. Basically I'd just be using an MAM to send the required position to the axis, and the motion planner would do all the work of determining speed/direction/start/stop and tolerance windows, instead of me coding it in logic.

I have already determined that a PF527 can't accept an SSI encoder, and it's unfortunately too late in the process to change the type of encoder. So I can't just wire the encoder to the drive and do it the "right" way.

What I'm wondering is, is there a "creative" way to have the position from a 1734-SSI module effectively become the motor feedback for a position loop axis using a powerflex 527?

Bump?

Let's confirm what you're asking for.

A PowerFlex 527 functions as a CIP Motion axis in the Logix environment, and you want to use it in Position mode. Of course, that requires some kind of feedback to close the velocity and position loops.

The PowerFlex 527 has an encoder option board (25-ENC-2) that lets you connect ordinary quadrature encoders, in a couple of flavors of voltage and single-end/differential modes.

And that's... pretty much it. There are no options for setting the feedback method to be anything but the onboard encoder input. You can't set it to an encoder attached to a Feedback-Only axis, or a 1756 analog/encoder Motion card.

Forgive me for what I am about to post

There is a method you might be able to try to make a Virtual Axis follow the positions provided by a Counter type module. There's an example in the Knowledgebase using a 1756-HSC.

But that just gives you a motion axis to follow. It doesn't do anything to make your PowerFlex 527 able to close its own Position loop.

If you go to the RA Sample Code library and use "Virtual Axis" as your search term, the first two documents are methods to approximate a virtual Position axis using only a speed-control VFD.

That method is intended to be used with VFDs that take their speed reference as a command over EtherNet/IP. I am not sure how you might be able to use it with the PowerFlex 527 in Frequency Control mode (which doesn't use an encoder) and sending speed control CIP Motion instructions to it (MAJ, I figure).

It would be Rube Goldberg-esque in the extreme to gear together two Virtual Axes each of which is lashed to the wrong kind of device with zip-ties and wood glue, instead of buying a quadrature encoder to fit the drive.

I happen to be working tonight with a Motrona encoder signal converter. I figure if anyone had an SSI/Quadrature converter it would be them, but I don't see one in the catalog.

A-B used to have an absolute encoder to incremental encoder converter, and it's got a third-party successor:

http://www.rapidcontrols.com/products/output-converters/sab-q2/

I think part of the idea with those is that they would send a burst of quadrature pulses on command in order to "catch up" the position of an axis, then send pulses proportional to speed.

Ken Roach said:

I happen to be working tonight with a Motrona encoder signal converter. I figure if anyone had an SSI/Quadrature converter it would be them, but I don't see one in the catalog.

A-B used to have an absolute encoder to incremental encoder converter, and it's got a third-party successor:

http://www.rapidcontrols.com/products/output-converters/sab-q2/

I think part of the idea with those is that they would send a burst of quadrature pulses on command in order to "catch up" the position of an axis, then send pulses proportional to speed.

Click to expand...

The rapid controls converter was used way back in the early 2000s so that SSI encoders could be used with M02AEs by converting SSI to encoder counts. This didn't work well. The M02AE customers quickly switched to the M02AS when that came out.

Ken mentioned why the SSI to incremental encoder didn't work but not why. If the pulses don't come out a steady stream, it is impossible to use the derivative gain. If you can get by with PI control only you may be OK.

Ken Roach said:

Let's confirm what you're asking for...

Click to expand...

You've described it correctly, yes.

Ken Roach said:

Forgive me for what I am about to post

Click to expand...

Well, that gives me a pretty good idea of where this is headed

Ken Roach said:

It would be Rube Goldberg-esque in the extreme to gear together two Virtual Axes each of which is lashed to the wrong kind of device with zip-ties and wood glue

Click to expand...

That is an excellent description!

I think in summary, the answer is "well, it might be possible but it would be a very bad idea". Given that the existing control method works just fine, it doesn't seem like breaking out the zip ties and wood glue just for the sake of being able to use a position loop is in any way the right solution to the problem. The whole idea came about as an idea that maybe I could make this more simple, but as I kind of suspected from the beginning, any solution that involves a position loop* is going to be significantly less simple than just doing a ladder-logic compare-position-and-start-motor. Even an encoder converter, if I could find one that would fit the application, would make it more complicated than just sticking with what we have.

Thanks Ken (and Peter) for the discussion!

*Other than using the right encoder, of course, which is unfortunately not possible at this stage in the project.