Re: Linear encoders with servos

Actually I don't think the issue here is linear vs. rotary encoder. The
issue is, "Do you monitor the table movement or the motor movement". Whether
you are using rotary or linear I think is irrelevant providing you can
obtain your reading from the electronics fast enough to match the speed
at which you're moving and doing the comparison. What is relevant
is "any" slop introduced between motor movement and table movement
(including of course leadscrew backlash) which could set up a feedback
loop. If you think about it, a system which monitors actual table movement
is the "ideal" system because, in the end, THAT is what we care about,
motor movement is simple a means to that end.

Ken (the other Ken) Jenkins
kjenkins@...

> Pete:
>
> Such machines do exist but they are very expensive and clearly out of
> the hobby price range. What you are describing is a very complex motion
> control system. The major problems to overcome are loop stability and
> velocity control. Even with very fast scales most designs still depend
> on shaft encoders or tachometers for velocity control.
>
> The two most popular software drivers, Mach2 and TurboCNC can not do
> this. Unless one has a LOT of experience in motion control system design
> (and a lot of money) this will not work. The hunting problem you mention
> is the loop stability problem I am refering to and it's a bear to solve.
> Stick with shaft encoders and save a lot of headaches.
>
> Sorry for the bad news;
>
> Ken