Re: starting up EMC/STG II day 1

[rehenry <rehenry-at-up.net>]

Hey Les, sounds great.  The only thing I'd do different is set the accel 
and the axis gain much lower to start.  It is easier to see the need for 
more accel or gain than it is to see the need for less.  

With your big machine, I'd set accel about 2 and gain about 125.  If the 
axis starts to run away when you close the loop and command a small 
motion then raise it.  If it overshoots and returns at the end of a 
motion then raise it some more.  When you get a reasonably smooth stop 
then increase the accel a bit and start over.  

Thanks for the great notes.

Ray

On Monday 04 March 2002 05:22 pm, Les Watts wrote:
> Well I started in earnest running the big gantry mill with
> EMC today. I mostly worked on setting up the .ini file.
>
> I will post today's notes. COMMENTS WELCOME!
>
> Using EMC_run and emc.ini on the bdi.
>
> 1) set motion control on .ini to stg8mod.o
>
> 2) set MIN_LIMIT_SWITCH_POLARITY to 0; I am not
> using the limit switches with emc directly so the STG inputs
> should pull hi indicating no limit.
>
> 3)Fired up emc and had no encoder indications in display.
> Servo amps unhooked.
> Set task controller from minimilltask to bridgeporttask.
>
> 4) Fired up emc again; have encoders indicating on display
> on all 3 axes. Direction right but scale wrong.
>
> 5) With 25x5mm ballscrews and 500x4 encoder counts/rev,
> set INPUT_SCALE to 2000 counts/rev x .2 rev/mm x
> 25.4 mm/in =10160. Set Z to 20320 as it has 4000 counts/rev.
> All axis displays checked as correct now with dial indicator.
>
> 6) Turned estop off machine on manual mode with amps off.
> Display says amp fault axis 0. This is correct as the amps are
> unhooked. Temporarily cange all axes FAULT_POLARITY
> to 0 in order to check other functions with amps off.
>
> 7)Fire up again in manual mode with amps off and do a jog on
> x, y, and z. All return following error. Correct response.
>
> 8)Set HOMING_POLARITY to 0 (neg direction) for all axes;
> home will be upper left back when facing machine front.
>
> 9) Check HOME_SWITCH_POLARITY all axes set to 1; nc
> switches ground inputs to card. When open they should pull
> hi.
>
> 10)Check amp ENABLE_POLARITY to 0 all axes (low enable).
> Switch Copley 412 amps S1 off for low enable. This seems the most
> failsafe.
>
> 11) Set DEFAULT_VELOCITY to 0.1
> set DEFAULT_ACCELERATION to 20
> set MAX_VELOCITY to 1
> set MAX_ACCELERATION to 20
> .. Just reasonable numbers I hope.
>
> 12) Checked following error calibration  by jogging with amps off while
> watching commanded position on the display. Error
> is displayed at .01". The velocity proportional 0.01
> MIN_FERROR of 0.01 is confirmed.
>
> 13) set P gains at 1000. This should give max output to the amp input
> at 10v/.01". May drop it down some before enabling the amps.
>
> 14) set I,D,FF0,FF1,FF2 at 0 all axes. One variable at a time.
>
> 15) Check OUTPUT_SCALE set to one for now.
>
> That's today's work. It took a lot of fumbling to do this as I am
> neither familiar with Linux or EMC. The documentation isn't the
> best for EMC- but it's great for the price!
>
> I will continue with the documentation of going from a BDI disk
> in hand to a running servo machine (i hope) and will put it up on a web
> site.
>
> Les
>
> Leslie Watts
> L M Watts Furniture
> Tiger, Georgia USA
> http://www.rabun.net/~leswatts/wattsfurniturewp.html