On Saturday 02 February 2019 10:19:48 Andy Pugh wrote:

> > On 2 Feb 2019, at 13:37, Gene Heskett <[email protected]> wrote:
> >
> > Can someone please explain to me, who is used to using a spinx1 for
> > the pwmgen output to analog voltage conversion, just how in tuncket
> > the 7i76 works?
>
> The 7i76 has a (serial controlled) digital potentiometer.  You just
> pass it a the spindle speed you want and the two resistances change.
> You wire it exactly like a potentiometer and send it a number.
>
It would be very, very, very, nice if that was mentioned in the docs, 
Andy. Not a dot for an i or the crossbar of at. So I automaticly assumed 
it was a spinx1 in function.

But whats the range of spindle speed requests that are valid in that 
case.  I currently, for testing, have the motors idle speed set to 2, 
faster than the normal .1 that my hal file uses when neither gear shift 
switch is closed because they aren't wired up yet. This results in 
nearly full scale for a value of 49 sent to the pwm, giving a 99.9% duty 
cycle at the pwmgen output because the pid.s.Pgain is 20.  And on the 
GO704 using one of Jons pwm-servo's, I will/am (not wired yet) using the 
pwmgen.1.output sent to p2-gpio-26 as the pwm for that servo amp. But I 
still need functioning ENA- and DIR- so I've got to feed spinout 
something too else they don't work. Whats currently coming out of 
pwmgen.1 has already been calibrated for the GO704 and its pwm-servo 
driving the spindle.

But on the 6040, with its bastard 120 volt vfd, I'll need this 7i76's 
spindle setup from tb4-1-2-3.  But surely it can't deal with being sent 
a value of 24,000, or even half that at 12,000 for half speed of that 
spindle. For that to work there has to be a scale converting the rpms 
requested to something that ranges from 0 top 10 volts at the equ to the 
arm of the pot at tb4-2, where 1 and 3 are the com and 10+ or 12+ volt 
lines from the vfd.

So how is that "scaleing" between 24k rpms when the motor is to be wide 
open, and a full scale output of 10 volts to the vfd to be accomplished 
when I build this same interface for the 6040?  I rx'd the 2nd 7i76d 
yesterday Peter, but haven't unpacked it yet. One track mind strikes 
again. Oldtimers?

==================

Much of the same hal file will be used, with only the gearshift stuff 
excised since that spindle has no gearshift. Nor will it have an encoder 
unless I can rig an IR optical off the wrench flats of the motors chuck.  
That will have to co-exist with the tool changer wrench if I build the 
hack-a-day tool changer. But I've already found the needed carousel spin 
is at least a full turn just to break a collet loose with the collets I 
have and only one nut so far, so thats discouraging. But its still a 
cute idea. Perhaps a ratcheting motion could be programmed, turn the nut 
1/3rd turn, lift it out of the socket, rewind the carousel and xy, drop 
it back into the gator socket, turn it another 3rd turn, wash rinse 
repeat as needed. Reverse to tighten but spring load the holding wrench 
away from an "its tight switch" so we don't loose the homes from 
slipping the motors a step or 70. Perhaps 2 switches in case its too 
tight, in which case leave the locking wrench in place, but lift it 
clear of the socket so a hand wrench can be used on the nut to break it 
free.  Use an M30 pause for that.  On a 6040, its big enough the 
carousel could have 8 pockets, and the larger diameter would give the 
motors a better chance of getting it done with less stress on the motors 
and screws.

Thinking out loud...  And all dependant on getting the vfd to respond to 
the DIR- to work, so the motor could be used to drop the nut and pick up 
the next one. In fact, it runs in reverse just fine if its 
miss-configured. I found that quite by accident while trying to make it 
work from the pc with only a 5 volt pwm, but it seems to need a 10 volt 
pwm, applied to a different input terminal.  But thats another subject 
line for Later. :)

Thank you Andy.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>



_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to