Closed Loop Stepper Motor Control

[Rodmcm]

Perhaps not the place for this question but I cant find the answers anywhere!

Small closed loop controllers such as the CL57T and LCDA257S allow settings from 800 to 51,200 pulses/rev from the encoders they connect to. There is only two encoder connections on these modules.

Most of the small stepper motors have 1000 segment encoders and two encoders in quadrature so it is easy to derive up to 4000 positions per rev from this setup using only encoder position.

I know that 51,200 divisions/rev is the maximum figure used in current controlled stepper motors

But if a combination of current and encoder is used then there is the possibility of loosing current steps, and hence loss of accuracy. Seems to negate the idea of encoder feedback entirely.

Has anyone any idea on how this all works?

 

[itsmoked]

The encoder is the final word on position. If the move didn't get where it was supposed to another step is commanded. Some newer stepper drivers actually know when a step was skipped so something can be done about it even without an encoder. I suspect a lot of designs will go this way soon.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Rodmcm]

Thanks, yes I am aware what it should do as a closed loop controller, what I do not understand is the possibility of 51,200 divisions per rev

 

[itsmoked]

That is ridiculous in my experience. The more the steps the worse the compliance and that many is way over the top. Have you got a link to this somewhere we can look at?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Would the high count allow a motor mounted encoder to precisely indicate the position of a shaft the other side of a deep reduction gear box?
eg:

1000 segment encoders and two encoders in quadrature so it is easy to derive up to 4000 positions

using 40,000 positions would allow this encoder/motor combination to indicate the position of a shaft the other side of a 10:1 gear box.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

Bill; That would suck because of the play in the gear box. You'd want the encoder on the gear box output shaft so it could be used to deal with all the lash.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

I defer to your greater familiarity in this issue.
Yours Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Rodmcm]

https://www.omc-stepperonline.com/download/CL57T.pdf

This is the best document I can find on setup, but doesn't have any information on how it all workd

 

[itsmoked]

Hi Rodmcm; Apologies for the slow response.

Reading the manual it appears you may be laboring under a mistaken belief.

You may use a 1000ppr encoder period. Not something higher. The higher 'thing' is only the pulse train driving the stepper. Presumably these smaller steps will allow one to more accurately hit the movement endpoints.

Personally I'd start with the default 1600 steps setting and I'd be very wary of higher settings without a lot of testing.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Rodmcm]

I can understand using any microstep value to get you to a known encoder position. But that implies that you can only accurately hit 1000 divisions per rev (or maybe 4000 with quad encoders) to be held in closed loop position by the encoder

If you want to do 99.3 divisions you can microstep between the 99 and 100th division but now you are held by the microstep
current control just like a normal stepper motor control without an encoder

This is the part I don't understand, as it easy to loose stepper positions when using current steps (if it wasn't there would be no market for closed loop methinks)