Re: [CAD_CAM_EDM_DRO] A crazy idea / maybe great for PWM

mariss92705 wrote:

>
> Hi,
>
> Parallel port output CNC programs are limited to 3 or 4 axies because
> of the limited number of port bits. Each axis requires a STEP and
> DIRECTION bit pair. What if it could be done with a single bit per
> axis where STEP and DIRECTION were combined on the same line?

Last Jan 1, 2002 I annouced the SiMAD (Simultaneous Multi-Axis Direction)
algorithm which allows you to control N axes with N+1 bits
(ie control 6 Step & Direction axes with 7 bits):
http://groups.yahoo.com/group/CAD_CAM_EDM_DRO/message/36258

I didn't explain it very well, but in a nutshell, you'd apply
only one direction bit to all the axes simultaneously, and step
the axes you wanted. Of course you can't step two axes simultaneously
in opposite directions, but in practice stepping them a microsecond
apart should be fine.

However I was told that supporting Geckos using this scheme is
impossible, due to the timing delay's required by the Gecko hardware.

What is your opinion Mariss? If solvable, the gains are great!

--------------------------------------------------------------


PWM:

A number of us have been discussing something similar, although
it is vectored more towards PWM.

It is a simple scheme, but part of the problem is that the
motor driver has to know about the scheme. That is no problem
for you Mariss, as the manufacturer!


____________|------|______|-------|_|-----|_|---------------------

signal low equal=freewheeling |--pwm CW-| signal high=brake
emit braking
behavior
(logic or power)


Comments:
- the signal is comprised of the (times) of the last two
transitions (ie time of the previous and 2nd previous signal).
If they are valid as a pair, the motor controller emits an
output based on the PWM duty cycle.

- signal is low for a long time, or high for a long time
(maybe the PC is frozen, or a cable has become undone)
emit braking behavior (maybe bring a braking output bit high
on the motor controller). The brake being on, for example
prevents the vertical Z axis from falling under gravity when
you expect it to hold its position.

Hopefully, the length of time that constitutes 'a long time'
can be controlled by a trim potentiometer.

- PWM duty cycle approx 50% emits a freewheeling behavior
(brake is off, but no power applied to motor.... this allows
the motor to coast to a stop if it is already moving....)

The possible deviation from 50% either side constitutes your
deadband, which hopefully can be controlled by a trim potentiometer.

- A PWM that is mostly 'logic high' would be ClockWise, and a PWM
that is mostly 'logic low' would be CounterClockWise.

The really neat thing about PWM control is that the width of the
PWM <high and low grouping> can be narrow for a rapidly changing
velocity, but can be wide (and thus less intensive on resources)
for a constant velocity.

Doug Fortune
http://www.cncKITS.com




















-