Dynomotion

Group: DynoMotion Message: 14305 From: engnerdan Date: 12/27/2016
Subject: KMotionCNC Error on GCODE edit

Last night I was running a batch of parts, I ran 4 from the same gcode file without issue. On the 5th I decided to change a feed speed for a plunge with an end mill. I did this while KMotionCNC was running the code, I did a feedhold and changed the value from 15 ipm to 5 ipm and pressed save. I figured this was not an issue because KMotionCNC let me do it. The change did not take effect, as the next plunge still happened at 15 ipm, at which point I assumed it would not take effect until the next running. But then shortly after I get an error about the geometry not matching up, I think it was a radius error. Upon checking my part I find the mill had gone off its path and now ruined a $100 part that was nearly done.


I think the part was running at line 6683 while I made the edit, and the line I edited was 6695 the F15. was changed to F5.


Attached is the original gcode, the edited gcode and a picture of what the controller did to my part. If only it had gone off course the opposite direction I could have saved it.


I am running 4.34e at this time.


-Dan

   @@attachment@@
Group: DynoMotion Message: 14306 From: Tom Kerekes Date: 12/27/2016
Subject: Re: KMotionCNC Error on GCODE edit [3 Attachments]

Hi Dan,

Sorry to hear that.

You shouldn't really be editing the GCode while the GCode is running.  We should make a change to prevent that.

But it is strange that it caused a problem.  Are you sure you did a FeedHold and not a Halt/CycleStart?  While a Job is running the displayed GCode has been loaded into the screen memory before the Job started.  Editing the screen should have no effect on the Disk File that the Interpreter is reading/executing.  Did you push "Save" after editing the code?

Regards

TK


On 12/27/2016 5:57 AM, engnerdan@... [DynoMotion] wrote:
 

Last night I was running a batch of parts, I ran 4 from the same gcode file without issue. On the 5th I decided to change a feed speed for a plunge with an end mill. I did this while KMotionCNC was running the code, I did a feedhold and changed the value from 15 ipm to 5 ipm and pressed save. I figured this was not an issue because KMotionCNC let me do it. The change did not take effect, as the next plunge still happened at 15 ipm, at which point I assumed it would not take effect until the next running. But then shortly after I get an error about the geometry not matching up, I think it was a radius error. Upon checking my part I find the mill had gone off its path and now ruined a $100 part that was nearly done.


I think the part was running at line 6683 while I made the edit, and the line I edited was 6695 the F15. was changed to F5.


Attached is the original gcode, the edited gcode and a picture of what the controller did to my part. If only it had gone off course the opposite direction I could have saved it.


I am running 4.34e at this time.


-Dan
Group: DynoMotion Message: 14308 From: Dan W Date: 12/27/2016
Subject: Re: KMotionCNC Error on GCODE edit
I know I didn't hit halt.  I always feed hold unless I know I am going to restart the code from the beginning.  As I have not had the greatest luck figuring out how to restart mid program (almost always does something I don't want it to do).

I am fairly sure I hit feed hold but I may have done it while it was actively running. 

I know I hit save when I was done, the file log from Dropbox showed it too, which is where I pulled the example file from. 

One thing I thought of today was my read ahead setting on the trajectory planner screen.  I am not sure where it is set.  I have had anywhere from 1 second to 15 seconds set while testing how the motion controller responds to adaptive clearing gcode routines. Which appear to load up the Kflop a bit and cause small hesitations. But I am not sure what is actually happening yet, which is why I was testing different values to see if they had an impact. I want to say I was set to 3 or 5 seconds of read ahead. 

Dan



Sent via the Samsung Galaxy S®6 active, an AT&T 4G LTE smartphone