On Thursday 22 November 2012 09:59:53 andy pugh did opine:

> On 22 November 2012 11:26, Dave Caroline <dave.thearchiv...@gmail.com> 
wrote:
> > Current encoder inputs need 50% ratio
> > Some cleaning/retiming required
> 
> I can't decide if it is a special case of counter mode, or a whole new
> component.

Because the pulse width is the direction signal, and the rep rate is the 
speed, that does sound like a whole new component to me.

We are, generally speaking used to the quadrature mode, where the condition 
of one pulse on the given edge of the other, is the direction.

This would seem to need a timer counter, that for software convenience 
sake, would probably need to be married to the BASE_PERIOD.  On slower 
boards, this could be a problem because its pulse widths are a bit noisy if 
you look at the pdf sheets. For fwd rotation, the tolerance is 38-52 u-
secs, while the reverse is 76-104 in duration.

The speed part is relatively easy, how many pulses in a SERVO_THREAD 
interval, which should probably be averaged over several intervals with the 
lowpass component for slower speeds.

For those of us using the D525MW boards which seem to be pretty happy 
running at 23 to 25 u-secs BASE_THREAD we can probably say that a 1 count 
or 2 is fwd, and a 4 count is reverse, leaving a don't change deadband at a 
3 counts.  With finer timing of coarse we can likely get pickier, primarily 
because if its turning even at .25 rps, it sure isn't going to reverse in 
one BASE_PERIOD.  So I think I would nail the dir to whichever rail, until 
a very slow speed has been detected.  A speed at which its possible it 
could actually have been told to reverse.  Consequently I see the direction 
detector essentially frozen in place as long as the speed precludes its 
reversal without using hoover dams output for 20 microseconds to achieve 
the reversal.

In any event, it will need a separate index pulse, used to make sure there 
have been no missing or extranious(sp) pulses since LinuxCNC tracks windup 
with very wide counters.

I can't code it, not on these machines (I could on a TRS-80 Color Computer, 
but its clock is 1.79Mhz=max speeds in the 1 rps area, but first I'd have 
to cobble in a parport & write a driver, it doesn't have one), but I think 
that about covers it.

Humm, wierd thought about the reversals.  Is anyone here using a big 
honking frame 56 or bigger stepper to turn a lathe spindle?

Cheers, Gene
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
My web page: <http://coyoteden.dyndns-free.com:85/gene> is up!
Behind every great man, there is a woman -- urging him on.
                -- Harry Mudd, "I, Mudd", stardate 4513.3

------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to