Re:Re: PWM stepper driver with external bridge

> > Hi, it's my first post here, hope I don't offend
> > anyone but I want to dispute a couple of points.
>
> Speak up .. we tend to fight in a "friendly" manner in this group.

Ha ha! No fighting, just discussing! :o)

> > having designed power control hardware for many
> > years I can tell you 5A to 10A drivers are as
> > different as 2A to 5A drivers are...
>
> I've run that design at 8 amps with no problems and have no reason

to

> believe that 10 amps won't work as well .. although I do admit that

10

> amps IS NEAR, if not ON, on the edge.

Cool.

>
> > Wrong! Yes microstepping CAN be "at all" accurate,
> > the question is "how accurate"?
>
> Forgive me the lack of precision in my previous post. I was

assuming

> 1/16 or 1/32 step as "microstep" .. to me, 1/4 step is some kind of
> "middle ground". Also, I was assuming the current ratios to be
> uncompensated.

Sure, I see this attitude a lot lately, seems there
is a very definite "microstepping bad - gearing
good" attitude with CNC guys in particular. In
many other areas microstepping is quite accepted.
I suppose current compensation stuff is quite new
and not as widely understood. :o)

> Moreover, as far as static friction goes, some quick testing has

shown

> that the behaviour of a motor on a bench and connected to a lead

screw

> moving a table are quite different.

We agree there. But in most cases the motor static
friction is small compared to the machine static
friction. But in most cases for 1/4 and 1/8th
steps the motor will assume a final rest position
dependant on the final microstep selected.
Things like backlash and static friction will
always be there, but people often assume that
steps less than full steps will have NO effect
on positioning which is completely false. In a
torque compensated microstep system the motor
WILL turn to a position relative to the microstep
selected, with the same torque and load angle
deflection that is used to turn it to a full step
position.

> > The a3955 will do as Lawrence requires, it will
> > drive ONE PHASE perfectly as it was designed.
> > Obviously you need two for both phases. It will
> > drive a h-bridge fine with a few resistors.
> > Remember a stepper motor requires TWO h-bridges
> > for bipolar operation.
>
> I think he wants the chip as a driver for a higher current output

stage

> AND to maintain the various current decay modes. Not sure how this

can

> be accomplished with just resistors.

Good point, the different current decay modes
make it more compliacted, but from what I know
about the 3955 it turns on both the bottom
bridge semis for fast decay, this will still
work when it is driving a h-bridge I think,
but you could be right. :o)

> > This sounds somewhat similar to the commercial
> > project I am currently working on. :o)
>
> My goal was not commercial, just an exercise in "improving" what I
> already had. Just about the time I completed the design, I talked
> myself into a servo system.

Nice!

> > I am curious how you generate vref with the PIC,
> > by hardwired resistor ladder? Or pwm??
>
> First step was PWM. Some quick math convinced me that I couldn't

change

> the PWM widths quickly enough for the step rates I had in mind. ( I
> could be wrong about this, but it was too close for comfort )
>
> Next step was an R2R ladder ( right out of the Microchip App note ).
> Great idea, just ran out of I/O pins with the processor I had in

mind.

>
> Last step was to use discrete resistors of JUST the right value,
> because, as you rightly point out, there's not that many values
> involved.

Yeah, it gets hard to get resistors just right,
you either have to go to 1%'ers in non standard
values or stat using a number of R in combination
to get the right values.

-Roman