On Monday 07 March 2016 02:21:53 Danny Miller wrote:

> Aaaaand that did it!  Well, is there any reason to hook that enable up
> to anything?  Won't it issue a stop via Modbus if we do Toggle Machine
> Power?  (I should have checked that).
>
> I found the code has a bug, it's making a "calculated frequency" by
> multiplying the freq by 100x.  But neither the x200 nor wj200 do that.
> They go by 10x (e.g. 100Hz is 100.0Hz and formatted as "1000" on both
> read and write).  So it was trying to send impossible frequencies to
> the VFD.
>
> Now when I first hooked up this system under Mach3, the Modbus didn't
> read or write and it turned out to be a problem with massive
> common-mode noise on the VFD, even with no motor output. RS485 is
> differential, but unless it has opto isolating input stages, its input
> IC can't read a signal outside its common-mode range even if the
> differential is fine.

Ouch.  The speed of the opto's may restrict performance at the higher 
baud rates by causing errors. I'd certainly look at it with a o-scope.

> I fixed that with an RS232-RS485 isolating converter.  At least I
> thought I did.
>
> There's a problem that the X200 has no ground reference to tie the
> converter's RS485 ground to- not that I could find.  But it worked
> fine.
>
> The thing is, now that I have fprintf reporting, I DO see that the
> readbits() task fails a LOT when the VFD is decelerating- and to a
> much lesser extent on acceleration.

That I'd suspect would be a direct artifact of the lack of a common mode 
reference ground.

> I suspect there's additional excessive ground noise going on during
> deceleration.  Doesn't seem like it's gonna break anything, though.
>
> Danny

I expect that I had better test this vfd & motor I've purchased by 
setting it up and driving with with a function generator. I have been 
assured by the seller that its quite common to drive its 0-5 volt input 
with a high frequency PWM logic signal, its well filtered and will 
integrate the pwm just fine.  Something the Mesa 5i25 is very good at.  
It has a well marked common ground input that should work well.

But it worries me that I'll lose rigid tapping unless I use it as a 
replacement motor so my encoder still functions, so I may wind up 
putting it on in place of the slow 1hp, fitting the gear from that 
motor, or a replacement of it, to this motor in place of its ER16 chuck.  
That will obviously require replacements of all the spindle and back 
gear bearings with top quality stuff in order to deal with that whole 
gearbox being turned at 6x the speed the existing motor can muster. 
Something in the 6x territory if the nameplate speeds are used. That 
would give about 13,500 revs for wood carving, a huge improvement over 
the 2500 or a bit more that I can get now, using the Pico pwm-servo 
driver.  Bearing life is a huge ? though.

That is unless someone can tell me if the speed tally this vfd returns is 
actual motor speed, and not just the driving frequency, a question thats 
been asked of the vendor and danced around for a reply.  So my question 
to this list is, has anyone attempted to do rigid tapping by using that 
signal as feedback into an a/b/z encoder? Its not available as 
quadrature, its a single pin which I believe is a copy of the driving 
frequency.

And I don't believe in magic.  Not even when David Copperfield does 
it. :)

> On 3/6/2016 6:01 PM, andy pugh wrote:
> > On 6 March 2016 at 23:46,  <dan...@austin.rr.com> wrote:
> >> I'm sure it will- how do I connect it??  Or just get rid of it?
> >
> > Simplest is to "setp pin-name.enable 1" in the HAL file.

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>

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://makebettercode.com/inteldaal-eval
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to