On Tuesday 24 November 2020 23:21:08 Sam Sokolik wrote:

> Granted..  I know these are pretty basic.  You can usually find them
> for under 10 dollars.  They do work quite well though.   I was working
> on some youtube content and was goofing around with this bob and the
> printer port. Setting up a pwmgen to test the analog out unearthed
> some things..
>
> -It isn't very linear.  lincurve is your friend..  It can be off by as
> much as .5v over the range.
> -it is sloooowwwww.  It is most likely a simple low pass.  The photo
> here is of a 1khz pwm signal modulated by a 1hz (yes - hz) sin wave.  
> You can already see it is phase shifting.   10hz and it is attenuated
> to about 2v p-p.
> -the pwm frequency didn't really affect the linearness or performance.
>
> http://electronicsam.com/images/greenmachine/IMG_20201124_183913.jpg

Waiting on electronicsam.com forever.

> one of these bobs...
>
> http://electronicsam.com/images/greenmachine/IMG_20201124_183919.jpg

Ditto. So I've no clue what bob you are playing with. So from here I'm 
relating my experience.

My fav bob is a sainsmart, marked as "CNC Interface Board ST-V3" right 
behind the DB25 socket, has no analog converter and only an aux relay 
I've never used and often removed. And I use it because it has no speed 
killing OPTO's anywhere. But its 5 inputs are slowed by caps to filter 
noise that if feeding a 5i25's second encoder thru P2, must be removed. 
Those grain of sand caps are C6 thru C10 on the bottom of the board. 
They also do not have the analog facility so you must use a mesa SpinX1 
or similar if you need a truly analog signal. In some cases I have fed a 
higher frequency pwm directly into the analog input of a vfd IF the vfd 
has a 5 volt control option available, its input filters will average 
the pwm fairly well if a bit of non-linearity can be tolerated and you 
won't need a Spinx1 for adequate control. Toss in a PID for a feedback 
loop and even that non-linearity dissapears till you hit the top speed 
limit of around 380 hz.  And the SpinX1 is quite fast. Driving a cheap 
clone vfd with a pwmgen in a 7i90HD, a 1 hp 3 phase 50 yo motor and some 
hal logic to sequence the direction change, my Sheldon with a nearly 40 
pound 8" 4 jaw chuck mounted, can be reversed with only .24 of a turn of 
overshoot at 100 rpms. And not a PID in sight, nor any feedback from the 
encoder, that is only used for rigid tapping and to display the revs on 
a tach dial with pyvcp.

Propagation delay for the digital pwm signal is around 10 ns thru this 
bob connected to p2 of a 5i25, is a cmos buffered rail to rail signal 
and at 20 kilohertz, the vfd cannot tell the difference between that, 
and the SpinX1's output IF its control input can be switched from 
10volts analog, to 5 volts analog. 

> Still pretty cool what you can do with the printer port still ('I'm
> not dead yet!') and rt_preempt.  (50us base thread)  This could be why
> I wasn't able to get a very tight spindle speed control on the little
> emco lathe. (I am using this bob's analog out driven from mesa pwm) 
> And you are certainly not going to control an analog servo with it ;)

Why not? With one of Jon's pwm-servo's, I am doing that on TLM, a 7x12 
lathe, fast enough that if speed changes aren't rate filtered, it will 
strip the teeth off the drive belt and burn up the plastic drive pulley, 
and has done so many times. I put a limit3 between motion and the PID to 
slow that and haven't had a drive failure since. And it made all the 
stuff I needed to refit the Sheldon since.

Now if your web server worked... I'm a huge fan of your you-tube 
videos. :) 

> sam

> _______________________________________________
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>


_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to