Half-stepping the steppers
December 09, 2007 07:44PM
I've corrected the PIC firmware so that it now does half-stepping properly. It's checked back into the Sourceforge repository in Subversion.

This means that we can definitely use the cheaper 200-steps-per-rev stepper motors. You need a bipolar one where the coils can take a voltage of 12v and each use a current of 2A or less.

To switch half stepping on (not the default, to maintain upwards compatibility) uncomment the line that says

//#define HALFSTEP

in the file stepmotor.h

It produces much smoother movement, so it may be worth enabling with the 400-step motors too - just double the number of steps/mm in the preferences and increase the speed (half way between the speed you've set now and the maximum of 255 should double it, giving the same physical speed). You may also find that you can run the motors on a lower percentage power in the preferences file too.

best wishes


Re: Half-stepping the steppers
December 25, 2007 01:19AM
(1) Is this still fixed now we've switched the trunk to autoconf-based code?

(2) It would be easier to make turning half-stepping on/off a SNAP command, and then it can be a setting in the host properties file... any real reason for *not* doing that? This way people with 200step/rev motors can use "official, released" firmware rather than having to recompile it.

Re: Half-stepping the steppers
December 27, 2007 12:59PM
well, as you've said before, modifying the SNAP protocol is a big NO-NO.

we should leave that for version 2. of course if you've changed your mind, then i have a couple changes i'd like to see as well...
Re: Half-stepping the steppers
December 27, 2007 01:19PM
Yes, (2) is probably a post-v1.0 idea, since obviously implementing it doesn't
get us any closer to v1.0. And I should have made that clear. Good catch.

Though, thinking about it a bit more, if it were added as a new SNAP command,
and the default for stepping vs half stepping was the current default, then that
enhanced firmware would remain 100% compatible with the current host software.
That's somewhat different from replacing one current command with a new and
different one, which then requires enhanced or modified host code in order for the
new firmware to function.

But overall, yes, this should probably wait until after v1.0, like quite a few
other interesting ideas. Worth recording while we think of them; better left
unimplemented until v1.0 is done, so we stay focused on the immediate task.


Sorry, only registered users may post in this forum.

Click here to login