Brian wrote:
> Jon,
>
>   If I had encoders on the screws, the method described on the wiki
> would work.  This is basically what Dave suggested a few replies back,
> and what my config already does to some degree.  This method will also
> work with steppers, until the motors are disabled, and you move the
> machine by hand.  Then the wiki method no longer holds water.  Trust
> me, I have tried lots of different things to come up with a usable
> solution without modifying anything.  I have thought this problem up
> and down, to no avail.
>   
Well, I simply don't understand why this doesn't work.  If your only 
position feedback is from the linear encoder, and you have significant 
backlash, and you set the system up as a stepper-servo, then you would 
likely have servo hunting behavior, with the motor running to the end of 
the backlash until it bumps the table, the PID senses the move and then 
commands it to run to the other end and bump it back.  The only solution 
to this is to set the deadband equal to the backlash, which is probably 
not desirable. 
>   With respect to backlash being a significant hindrance to my CNC
> system, it is a problem, but with a full screw map, it really isn't a
> huge deal.
Well, I (and many other CNC users and machine tool builders) believe it 
IS a big problem, and central to any accurate machining.  If the motors 
are not able to control axis position, then all else is hopeless.  The 
problem is the cutting forces can move the table over the length of the 
backlash, suddenly and without warning.  It takes time for the motor to 
zip to the other end of the backlash and try to compensate.
>   It is plenty accurate surprisingly so as a matter of fact.
>  Because I still use the machine manually, I am not planning on
> changing over to ball screws.  Moreover, the fact that I use a machine
> that has acme screws with lash is not a good reason not to allow EMC2
> to correctly utilize the data available to it.
>
>   Why is there so much resistance to this?  I must not be effectively
> explaining the situation.
Yes, I think this must be so.  I think I understand how this is supposed 
to work in EMC2.
Although I don't use the screw error compensation, I think I know how 
this is designed to work, and
that it does work on other machines where it is used.  I most certainly 
switch my machine to E-stop
and move manually, and can pick up where I left off without any 
problems, just by hitting F1-F2.
Now, the screw mapping with backlash may not work at all for manual 
moves.  I believe the code uses the COMMANDED change in direction to 
decide which value to use from the table, so it clearly will not 
compensate correctly for manual moves.

The "resistance" you sense is historical.  This battle has been fought 
on rec.crafts.metalworking for some 15 years now, and on various 
EMC-related lists for a decade.  Those who have fought the battle know 
the answer, and are tired of the fight.  I'm one of those who believe 
that backlash is a terrible problem, and makes everything else SO much 
harder.  I do practically ALL machining in the climb-milling direction, 
and this alone would be a disaster if I had a lot of backlash.

If you can't change to ballscrews (some problems with back-driving the 
handles as well as the COST!) then it is often possible to adjust the 
nuts to reduce backlash.  .001" is quite achievable, and will pretty 
much solve all these problems.

Jon


------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to