On Friday 02 February 2018 21:28:31 jeremy youngs wrote:

> Can serport be used to supply +-10 v pwm for servo drive control? Is
> the voltage output controllable? Wiki says 232 voltage can be +-13 or
> +- 24 so I think voltage could be an issue.
> I got the serport docs here ,
> http://www.linuxcnc.org/docs/html/man/man9/serport.9.html
> Is there anything else I need to know? Thank you

I think that means its tolerance for min and max swing at its inputs. 
Output swing will be whatever the supply puts out unless there are some 
max232 chips, in which case they might make more. But the serports real 
problem is the generally fixed baud rate, so its not possible to do a 
usable pwm thru it at any baud.

Which is why we use the parport, but contained in an interface card with 
an FPGA brain, which can run a stepper generator at many times the speed 
your computer can generate the steps at,  with many times more precise a 
timing. 

Personally, I'd use an interface card like the mesa 5i25, (or a 7i90 but 
its a 3 volt card, 72 i/o's and needs a triplet of 7i42TA's to both 
protect it from real world noise, and give you a much easier to use 
terminal strip for each of the 3 50 pin sockets on the 7i90) and a 
breakout board WITHOUT any opto-isolators, at least in the paths driving 
a pwm signal, or inputting from encoders. The opto's commonly used on 
most breakout boards are waaaayy too slow, sometimes a couple 
magnitudes. And that will for sure have bad to horrible effects on the 
linearity of your pwm controls unless the pwm is running so slow that a 
4 microsecond error is absorbed. 1000 hz is not slow enough, yet its too 
slow for good servo response. Ditto running a spindle encoder with a 
higher count per rev.  With a 1000 liner on the rear of the spindle 
motor, its a bit over a 7 to 1 at the spindle in high gear, or about 15 
to 1 in low gear, and with the opto's incircuit, my spindle tach, and 
the speed controls died at 300 rpm. Cut out and bypassed, its still 
working great at 3000 revs in high or 1500 in low.

With a 5 volt swing from a bob, and a servo or vfd that needs 10 volts, 
mesa has a SpinX-1 card about 2" square that translates a 5 volt pwm 
signal great once the opto's are excised from that signal path.  And its 
not expen$ive at all. 1% accuracy once calibrated.

For a db25 breakout board, (use with the Mesa 5i25) I found the $22 bill 
(ebay price) SainSmart to be decent, and you can pull the opto's it has 
in the 5 inputs if you need to. I am an oldtimer, and a C.E.T., so I 
have the tools and its relatively easy for me. With no opto's in the 
outputs, it has an in/out time of around 10 nanoseconds, more than good 
enough for the girls I go with. ;-)

On my biggest mill, a G0704 I've cnc'd, an old Dell with a 5i25 gives me 
two parports so I can run a 4th axis. On my biggest lathe, a Sheldon 
11x36 thats nearly 70 yo, I've cnc'd, a 7i90/7i42TA's buffered all 
driven from a raspberry pi 3b over an SPI buss running at 42 megabaud 
for writes, and 25 for reads, is doing nicely.  And I've added a whole 
bunch of do-dads to that.  And still have 20 or so gpio's unused, nice.

HTH.

-- 
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>

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to