Re: Why is the push to such high step rates (Grex and Arc100)?

AHHHHH, Thanks Les. That is making sense. So, what you are saying
then if I understand correctly, is that the staircase effect of the
range of frequencies is limitted in both "Systems". But with a
45Khz system, you have say a total of 1024 discrete different
frequencies you can work with - most of them being in the lower
range - leaving large steps in the higher frequencies. Applying the
same to these new devices, you have a broader range of discrete
frequencies within that 4Mhz range? Again, having the higher
resolution of frequencies within the lower useable range.

If that's the case, that makes good sense. I have run into this
problem working with micros too for various velocity control
issues.

Thanks!!!

Chris





--- In CAD_CAM_EDM_DRO@yahoogroups.com, Leslie Newell
<lesnewell@...> wrote:

>
> Hi Chris,
>
> One reason for high step rates is harmonic purity. Take Mach3

running at

> 45kHz. There are a limited range of frequencies it can produce:
> 45kHz
> 45kHz/2 = 22.5kHz
> 45kHz/3 = 15kHz
> 45kHz/4 = 11.25kHz
> and so on
>
> Now what happens if you need 30kHz? Mach will produce a burst of

pulses

> at 22.5kHz then a burst of pulses at 45kHz then back to 22.5 and

so on.

> The average will be 30kHz. Your poor stepper motor is trying to

rapidly

> switch between two speeds. Of course inertia means there is no

chance of

> it accurately following the rapidly changing signal. What happens

in

> practice is that a lot of torque is wasted trying to track these
> variations. If you are running close to the motor's max torque

anyway

> you will end up losing steps. The really evil thing about this

problem

> is that you cannot predict it. You can set up your machine so it

runs

> great until one day you feed at just the wrong speed and the motor

loses

> steps.
>
> Now try the same thing with a pulse generator that can produce

4MHz. The

> two frequencies are 30.075kHz and 29.851kHz. This variation is so

tiny

> that it will have virtually no effect on the motor.
>
> Servos are much less affected by this problem but it will still

affect

> performance at higher speeds.
>
> As well as the above problem, Mach has to contend with delays

generated

> by PC hardware. These delays cause the pulse train to be even more

ragged.

>
> Les
>
>
> lcdpublishing wrote:
> >
> > Both of the devices, the GREX and the ARC100 promote some very
> > intresting features and performance specs. At first glance I

thought

> > I might be able to utilize these features but after looking at

the

> > numbers a bit more closely, I couldn't see a practical

application for

> > hobby use. So, that got me thinking about commercial

applications

> > and, well, I just couldn't think of one so I had to ask.
> >
> > I know from the posts by people in the various groups that there

are a

> > number of these units out there. What is it that they are

building

> > that utilizes such amazing performance?
> >
> > Chris
> >
> >
>