Re: getting the 57cnc to work with G540

[DanL]

hay Art I have got it down to the spindle control not working, I am going to get it all going through M3 so it gives me the pin numbers, then get it working in Auggie, I am very sure it's the spindle.
I worked it through from nothing ticked, bringing up the scripting window each time, it showed what was need to be turned on or off, I did that and repeated to it got to the spindle control.
using the scripting window to test makes it quite fast.

[ArtF]

Dan:

If you turn on a spindle library it should run... relayonly maybe..

Art

[DanL]

I got going the 57CNC or computer needed to be powered down, or pin 17 needed set it was one of these 3 I did all 3

[ArtF]

:)

  Next version will have some safeties as Ive added ways for a script to call
a routine name which may or may not yet exist, and do so with no error. This
will keep you from erroring out..

Art

[DanL]

thanks art, it's still a problem with the G540 the other pin does not show up so I may have to do it through the 57CNC only more testing will tale I should have 2 pins active, I only have pin 17 output 1,

[ArtF]

Hi Dan:

You shouldn't really get any errors in run-time based on what the G540 is doing.. I don't think I'm following
whats happening, yell if you need a hand or it isn't making sense..

Art

[GlennD]

Dan
I am trying to following along on the G540 stuff, incase I don't like these chinese drives.
From the way I understand it you are not seeing a signal on pin 17 out of the 57CNC?
What about the pin20 PWM(2) which looks like pin 14 on the Gecko parellel cable?

What were the other two pin numbers you turned on out of curiosity?

Thank you
Glenn

[DanL]

I have a signal on 17 it only came up after I fired up M3 I will turn on pin 20 and see if I am getting anything on pin 14 PWM output in the G540, its a bit of not knowing the corresponding pins, there is 3 things I have to look at it to get the info I need it get's a bit confusing.
I know pin 17 is working it's turning the superpid on from pot to pwm control, its just the PWM output that needs done.

Art the G540 is still unstable with 4 motors turned on, the problem with the net being on is gone, I will test the Ethernet connection soon.

I ran the roadrunner code it started good as it got to half way through the code and crashed, I will run the 2 examples to see if it's more than just random
most times when I home the machine it does not zero out the work concordats.

[DanL]

Art the Ethernet it still run's ruff.
the 2 example file's run fine, that as far as I have got 2 day I will bring up the scripting window and run it again just to see if anything poops up.

[ArtF]

Dan:

Homing will not zero the work coordinates, you need to manually do that. Sometimes one would want to, other times one wouldn't.. its nice to see losses before cancelling them out. But in the end a personal homing script could do it for you.

  I'm still stumped on the roughness, I'm beginning to think its CPU speed or graphics speed, I use a pretty good machine. ENet should be the smoothest if that were the case, but I still cant find a spot where that may happen as yet, but I'm sure it will be found eventually.  The weird thing is, Bob find his bad on 3 axis, not 4, and I find mine much smoother on 4, than on 3. Makes sense, the 8 axis engine has 1/2 pulse width..

  More testing I guess, we'll figure it out over time..

Art

[DanL]

homing is fine like that, it does Zero out sometimes by it self 1 in 5 times. I will try my fast machine to see if that helps

[DanL]

Changing computers did nothing it still ran ruff with 4 motors turned on, everything else was the same

[GlennD]

Dan
I added these lines to my System\SystemCallBacks  OnHome script and it will Zero them out every time.

Code: Select all

  if (axis ==1){ GlobalSet("Axis1CurPos",0.0);} 
  if (axis ==2){ GlobalSet("Axis2CurPos",0.0);} 
  if (axis ==3){ GlobalSet("Axis3CurPos",0.0);} 

Glenn

[DanL]

Ta I am offr on holiday for a week starting tomorrow I will put it in when I get back, with YaNvrNo homing script it must be a bug that it happens sometimes by it self, more for Art to do.

I tried to get pin 20 PWM pin 14 on the G540 to work the 57CNC and Auggie did not like it one bit, I am going to use the 57CNC to do the spindle control, it did not have a problem with pin 17 on the G540 being used what is output 1 to turn the spindle on and off that was working.

will have to see if bob get's a 57CNC if he can work it out or someone else it's the same with the limits. but that depends on if the new 57CNC is V 1.2 or V 1.3, there is meant to be a certified version coming out next week or week after

in Mach 3 it all works fine with Art not having a G540 it make's it a bit hard to find the problem, the X, Y and Z stepper driver on the G540 work so it's not that much of a problem

[ArtF]

Dan:

Enjoy your vacation, Auggie will be here when you get back. :)

  Im still puzzled by Mach3 vs the CNC57...when you think of it,
its my code in both cases, mach3 and Auggie,lol, so if I'm a bit stumped,
you can imagine how curious the results are.

    I had a G540, but that's the one Bob is using. I wanted to see it put to use,
but its one of the first ones ever made, Marris sent it to me to play with
in my retirement, and I suspect its near serial #2. I think he'd be glad its being
used to troubleshoot new software. :) (Marriss is nothing if not a curious
individual..)

    So while testing on other drivers reveals smooth stepping on Bobs system,
and you'd think that would prove the G540 is doing something or expecting
something odd, the fact MAch3 runs it fine is very puzzling. I mean, I wrote the
pulse code for both, and I cant see much of a difference, other than pulse width.
  Mach3 is about 3us, Auggie is 20us on 3axis, and 1/2F on the CNC57.

  The final thing I can think of, and its not likely, is that pins 18-25 on a printer port are
usually grounded together. Perhaps only 1 of them is really grounded in the G540,
and we're not grounding the right one to the CNC57's ground?

  Anyway, it will likely show up eventually.. enjoy your vacation..

Art