Dynomotion

Group: DynoMotion Message: 7810 From: Steve Date: 6/28/2013
Subject: G20 and G21 seem to be reversed in MACH plugin?

MACH set for Inches. G-code and jogs work as expected with Kflop.

Any G-code containing G20 causes the axis to move many times more than they should. G21 correc ts the situation and the axis move normally.

Having identified the culprit (finally), I can edit the G-code to find and remove G20, but this needs to be corrected.

 

- Steve

 

Group: DynoMotion Message: 7814 From: Steve Date: 6/29/2013
Subject: Re: G20 and G21 seem to be reversed in MACH plugin?

Little more info; issue is present on ver 'xxx f ' plugin. NOT present in 'xxx a ' plugin.

 

 

 


--- In DynoMotion@yahoogroups.com, "Steve" wrote:
>
>
> MACH set for Inches. G-code and jogs work as expected with Kflop.
>
> Any G-code containing G20 causes the axis to move many times more than
> they should. G21 correc ts the situation and the axis move normally.
>
> Having identified the culprit (finally), I can edit the G-code to find
> and remove G20, but this needs to be corrected.
>
>
>
> - Steve
>

Group: DynoMotion Message: 7886 From: Tom Kerekes Date: 7/9/2013
Subject: Re: G20 and G21 seem to be reversed in MACH plugin?
Resend

TK

Group: DynoMotion Message: 7891 From: Steve Simpson Date: 7/9/2013
Subject: Re: G20 and G21 seem to be reversed in MACH plugin?
Tom,
 
I am not putting any logic to this at all. I am just reporting what happens. There is no way that I can know what is causing it (so far as I know anyway).
 
I'm happy to provide details, but there are none that I can think of that are pertinant to this. I further reported that the anomoly is not present with xxxa version.
 
SO, version xxxa no problem.  Version xxxf problem. I don;t know exactly where it came in, just that it is not present in xxxa and it is there in xxxf.
 
Symptom conditions are:
 
NO changes to any CNC parameters SAME MACH setup, SAME machine, SAME G-code. xxxa works fine, xxxf goes nuts as soon as it encounters a G20 or a G70 the scale goes completely bonkers. I did not measure the delta, but it very well could be 25x error and I can tell you for sure that it IS related to the metric G commands; G20 caused the problem and G21 correct it . . . which is why I speculated that they had gotten reversed somehow.
 
Is this anomoly not reproducible? Am I the only one having this issue?
 
What info can I provide that would be useful?
 
Steve


On Tue, Jul 9, 2013 at 10:47 AM, Tom Kerekes <tk@...> wrote:
 

Resend

TK

Group: DynoMotion Message: 7892 From: Tom Kerekes Date: 7/9/2013
Subject: Re: G20 and G21 seem to be reversed in MACH plugin?
Hi Steve,

I was just pointing out that if G20 and G21 were reversed you would not get the effect you are describing.

It just sounds simply like you either have the wrong "Mach3 units" or "Steps Per" set in Mach3.  I haven't heard of anyone else having the problem and I can't re-produce it.    But then I can't explain why it worked before.  Could it be that you just didn't realize it was also wrong before?  Maybe the GCode you were running at that time did not have a G20 or G21 and you didn't realize that you were running an Inch File in mm mode to correct for your error.

I'm not sure what Mach3 does to determine the default Inch/mm mode if it isn't specified.  Maybe the default mode somehow changed?

BTW everything regarding this is handled internally in Mach3.  The Plugin doesn't really know what the GCodes are or what the units are or what the "Steps Per" is.    Mach3 just tells the plugin how many "steps" to move.

Please answer the following questions:

#1 in Mach3 what is your Config | Set Native Units Setting at?

#2 in Motor Tuning what is your "Steps Per" for the X axis set to

#3 in the Mach3 MDI enter G20

#4 Record the X Destination in the KMotion Axis Screen

#5 Make a move that changes the X DRO in Mach3 by 1.0000

#6 Record the X Destination in the KMotion Axis Screen

Regards
TK

Group: DynoMotion Message: 7895 From: Steve Date: 7/10/2013
Subject: Re: G20 and G21 seem to be reversed in MACH plugin?

Tom,

Someone else did mention a 'scaling' problem when they used G20/G21, which prompted my comment. However, if the behavior I have described cannot be caused by the Kflop plug-in, and is not reproducable, that's as far as you need to take it.

You seem always willing to go the 'extra mile' for people, but I do not expect you to support MACH issues.

Thanks,

Steve

 

 

 


--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Steve,
>
> I was just pointing out that if G20 and G21 were reversed you would not get the effect you are describing.
>
> It just sounds simply like you either have the wrong "Mach3 units" or "Steps Per" set in Mach3.  I haven't heard of anyone else having the problem and I can't re-produce it.    But then I can't explain why it worked before.  Could it be that you just didn't realize it was also wrong before?  Maybe the GCode you were running at that time did not have a G20 or G21 and you didn't realize that you were running an Inch File in mm mode to correct for your error.
>
> I'm not sure what Mach3 does to determine the default Inch/mm mode if it isn't specified.  Maybe the default mode somehow changed?
>
>
> BTW everything regarding this is handled internally in Mach3.  The Plugin doesn't really know what the GCodes are or what the units are or what the "Steps Per" is.    Mach3 just tells the plugin how many "steps" to move.
>
> Please answer the following questions:
>
> #1 in Mach3 what is your Config | Set Native Units Setting at?
>
> #2 in Motor Tuning what is your "Steps Per" for the X axis set to
>
> #3 in the Mach3 MDI enter G20
>
> #4 Record the X Destination in the KMotion Axis Screen
>
> #5 Make a move that changes the X DRO in Mach3 by 1.0000
>
> #6 Record the X Destination in the KMotion Axis Screen
>
> Regards
> TK
>
>
>
> ________________________________
> From: Steve Simpson steve@...
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, July 9, 2013 3:10 PM
> Subject: Re: [DynoMotion] G20 and G21 seem to be reversed in MACH plugin?
>
>
>
>  
> Tom,
>  
> I am not putting any logic to this at all. I am just reporting what happens. There is no way that I can know what is causing it (so far as I know anyway).
>  
> I'm happy to provide details, but there are none that I can think of that are pertinant to this. I further reported that the anomoly is not present with xxxa version.
>  
> SO, version xxxa no problem.  Version xxxf problem. I don;t know exactly where it came in, just that it is not present in xxxa and it is there in xxxf.
>  
> Symptom conditions are:
>  
> NO changes to any CNC parameters SAME MACH setup, SAME machine, SAME G-code. xxxa works fine, xxxf goes nuts as soon as it encounters a G20 or a G70 the scale goes completely bonkers. I did not measure the delta, but it very well could be 25x error and I can tell you for sure that it IS related to the metric G commands; G20 caused the problem and G21 correct it . . . which is why I speculated that they had gotten reversed somehow.
>  
> Is this anomoly not reproducible? Am I the only one having this issue?
>  
> What info can I provide that would be useful?
>  
> Steve
>
>
>
> On Tue, Jul 9, 2013 at 10:47 AM, Tom Kerekes tk@... wrote:
>
>
> > 
> >Resend
> >
> >
> >TK
> >
> >
> >
> >________________________________
> > From: Tom Kerekes tk@...
> >To: "DynoMotion@yahoogroups.com" DynoMotion@yahoogroups.com
> >Sent: Friday, June 28, 2013 6:07 PM
> >Subject: Re: [DynoMotion] G20 and G21 seem to be reversed in MACH plugin?
> >
> >
> >
> >Hi Steve,
> >
> >
> >Your logic doesn't sound right to me.  If G20 and G21 were reversed then I would expect G20 to move tiny amounts (mm's instead of inches).
> >
> >
> >More likely your resolution is incorrect by a factor of 25.4 so that specifying G21 reduces the motion by a factor of 25.4 and corrects for the original mistake.
> >
> >
> >If you provide more details we can probably help you sort it out.
> >
> >
> >Regards
> >TK
> >
> >
> >
> >________________________________
> > From: Steve steve@...
> >To: DynoMotion@yahoogroups.com
> >Sent: Friday, June 28, 2013 5:47 PM
> >Subject: [DynoMotion] G20 and G21 seem to be reversed in MACH plugin?
> >
> >
> >
> > 
> >MACH set for Inches. G-code and jogs work as expected with Kflop.
> >Any G-code containing G20 causes the axis to move many times more than they should. G21 correc ts the situation and the axis move normally.
> >Having identified the culprit (finally), I can edit the G-code to find and remove G20, but this needs to be corrected.
> > 
> >- Steve
> > 
> >
> >
> >
> >
>

Group: DynoMotion Message: 7923 From: Steve Date: 7/13/2013
Subject: Re: G20 and G21 seem to be reversed in MACH plugin?
Tom,

Reporting back on this problem as it may be useful to others.

You were correct in that the behavior was caused by MACH being set to metric units. So the behavior is explained, but the setting remains a mystery.

The steps per inch and other settings were correct for inch as these settings were calculated and entered directly on the target machine.

Metric vs Inch and a host of other stuff was never explicitly set on the target machine because the basic config for Kflop, including the MACH .XML, a ling list of macros, custom screens, the Kflop plug-in and Kflop programs, were copied from my first 'Kflop machine' which is indeed set to inch units.

Further Kflop programming (including all homing code) was done on the target machine and the plug-in on the target machine was upgraded because of some fix that you made that effected the target machine.
It was immediately following this upgrade that I discovered (coincidentally perhaps) the 'scaling' problem.

I don't have any explanation as to how the units cold be set to metric on the new machine when the entire config was copied from another machine which is set to inch. And I do not have an explanation as to how the xxxa version of the Kflop could have not displayed the issue.

Given the above scenario, the only logical conclusion is that somehow upgrading the Kflop software and/or poking around in MACH with the Flop programming reverted the MACH config to metric, but that is pure speculation.

Something else to consider in diagnosing this particular anomaly is that if all axis are zero at the beginning of a program, there is no clue that the units are bonkers until the program starts a move. Zero being the same for metric and inch. With any numbers in the DRO's, executing G20 and G21 from MDI clearly shows what the 'native' units are.

Thanks for the help, Tom, and apologies for assuming that you did not respond to the initial question. Going forward, if I do not get a rersponse to a post, I shall assume it to be a Yahoo glitch and just post a reminder.

- Steve

--- In DynoMotion@yahoogroups.com, "Steve" <steve@...> wrote:
>
>
> Tom,
>
> Someone else did mention a 'scaling' problem when they used G20/G21,
> which prompted my comment. However, if the behavior I have described
> cannot be caused by the Kflop plug-in, and is not reproducable, that's
> as far as you need to take it.
>
> You seem always willing to go the 'extra mile' for people, but I do not
> expect you to support MACH issues.
>
> Thanks,
>
> Steve
>
>
>
>
>
>
>
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Steve,
> >
> > I was just pointing out that if G20 and G21 were reversed you would
> not get the effect you are describing.
> >
> > It just sounds simply like you either have the wrong "Mach3 units" or
> "Steps Per" set in Mach3. I haven't heard of anyone else having the
> problem and I can't re-produce it.   But then I can't
> explain why it worked before. Could it be that you just didn't
> realize it was also wrong before? Maybe the GCode you were running
> at that time did not have a G20 or G21 and you didn't realize that you
> were running an Inch File in mm mode to correct for your error.
> >
> > I'm not sure what Mach3 does to determine the default Inch/mm mode if
> it isn't specified. Maybe the default mode somehow changed?
> >
> >
> > BTW everything regarding this is handled internally in Mach3. The
> Plugin doesn't really know what the GCodes are or what the units are or
> what the "Steps Per" is.   Mach3 just tells the plugin how
> many "steps" to move.
> >
> > Please answer the following questions:
> >
> > #1 in Mach3 what is your Config | Set Native Units Setting at?
> >
> > #2 in Motor Tuning what is your "Steps Per" for the X axis set to
> >
> > #3 in the Mach3 MDI enter G20
> >
> > #4 Record the X Destination in the KMotion Axis Screen
> >
> > #5 Make a move that changes the X DRO in Mach3 by 1.0000
> >
> > #6 Record the X Destination in the KMotion Axis Screen
> >
> > Regards
> > TK
> >
> >
> >
> > ________________________________
> > From: Steve Simpson steve@
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, July 9, 2013 3:10 PM
> > Subject: Re: [DynoMotion] G20 and G21 seem to be reversed in MACH
> plugin?
> >
> >
> >
> > Â
> > Tom,
> > Â
> > I am not putting any logic to this at all. I am just reporting what
> happens. There is no way that I can know what is causing it (so far as I
> know anyway).
> > Â
> > I'm happy to provide details, but there are none that I can think of
> that are pertinant to this. I further reported that the anomoly is not
> present with xxxa version.
> > Â
> > SO, version xxxa no problem. Version xxxf problem. I don;t know
> exactly where it came in, just that it is not present in xxxa and it is
> there in xxxf.
> > Â
> > Symptom conditions are:
> > Â
> > NO changes to any CNC parameters SAME MACH setup, SAME machine, SAME
> G-code. xxxa works fine, xxxf goes nuts as soon as it encounters a G20
> or a G70 the scale goes completely bonkers. I did not measure the delta,
> but it very well could be 25x error and I can tell you for sure that
> it IS related to the metric G commands; G20 caused the problem and
> G21 correct it . . . which is why I speculated that they had gotten
> reversed somehow.
> > Â
> > Is this anomoly not reproducible? Am I the only one having this issue?
> > Â
> > What info can I provide that would be useful?
> > Â
> > Steve
> >
> >
> >
> > On Tue, Jul 9, 2013 at 10:47 AM, Tom Kerekes tk@ wrote:
> >
> >
> > >Â
> > >Resend
> > >
> > >
> > >TK
> > >
> > >
> > >
> > >________________________________
> > > From: Tom Kerekes tk@
> > >To: "DynoMotion@yahoogroups.com" DynoMotion@yahoogroups.com
> > >Sent: Friday, June 28, 2013 6:07 PM
> > >Subject: Re: [DynoMotion] G20 and G21 seem to be reversed in MACH
> plugin?
> > >
> > >
> > >
> > >Hi Steve,
> > >
> > >
> > >Your logic doesn't sound right to me. If G20 and G21 were
> reversed then I would expect G20 to move tiny amounts (mm's instead of
> inches).
> > >
> > >
> > >More likely your resolution is incorrect by a factor of 25.4 so that
> specifying G21 reduces the motion by a factor of 25.4 and corrects for
> the original mistake.
> > >
> > >
> > >If you provide more details we can probably help you sort it out.
> > >
> > >
> > >Regards
> > >TK
> > >
> > >
> > >
> > >________________________________
> > > From: Steve steve@
> > >To: DynoMotion@yahoogroups.com
> > >Sent: Friday, June 28, 2013 5:47 PM
> > >Subject: [DynoMotion] G20 and G21 seem to be reversed in MACH plugin?
> > >
> > >
> > >
> > >Â
> > >MACH set for Inches. G-code and jogs work as expected with Kflop.
> > >Any G-code containing G20 causes the axis to move many times more
> than they should. G21 correc ts the situation and the axis move
> normally.
> > >Having identified the culprit (finally), I can edit the G-code to
> find and remove G20, but this needs to be corrected.
> > >Â
> > >- Steve
> > >Â
> > >
> > >
> > >
> > >
> >
>