Hello Guest it is November 12, 2019, 05:25:04 AM

Author Topic: Erratic Feedrate When Cutting Spline Lines  (Read 4480 times)

0 Members and 1 Guest are viewing this topic.

Offline Chaoticone

*
  • *
  •  5,628 5,628
  • Precision Chaos
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #30 on: September 29, 2017, 08:58:39 AM »
Good deal. thanks!
;D If you could see the things I have in my head, you would be laughing too. ;D

My guard dog is not what you need to worry about!

Offline Chaoticone

*
  • *
  •  5,628 5,628
  • Precision Chaos
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #31 on: September 29, 2017, 10:58:26 AM »
Hopefully this will be the last update.......... no functional difference. I just added the blend radius for each angle to the ini and the CV_Feedrate table so 2 files here.
« Last Edit: September 29, 2017, 11:02:31 AM by Chaoticone »
;D If you could see the things I have in my head, you would be laughing too. ;D

My guard dog is not what you need to worry about!

Offline dq828

*
  •  61 61
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #32 on: September 29, 2017, 08:36:52 PM »
Thats excellent thank you, I've learned and gained a lot from this thread.

Last weekend I was trying to cut some timber clamp knobs and the cutting was very slow and jerky, I have just run a test with the knowledge and Wizards I have got from this thread and it runs as smooth as silk.

Once again thankyou.

Offline Mauri

*
  •  306 306
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #33 on: September 29, 2017, 08:38:40 PM »
Chaoticone,
I did a lot of testing last night and this morning on lettering 2mm upper and lower case engraving a standards on brass engravings.
2 Axis selected using CV program vs all angles the same.
The Doc attachment below contains all my findings.
The first main issue that I have found is that the Mach4 program has a flaw in the CV area, in that when you go from G1 to G0 it goes off line.
If you make all the code G1, then it does not, but the time increases significantly.
The second issue on lettering is that no matter the tolerance setting on your CV program on curves it does not follow the line.
Two items affect this:
1) CV being to high on some Angles
2) The change from higher CV to lower CV.
I have provided all combinations in my DOC file with screen capture examples.
On 3D engravings and continuous line engraving it has less of an issue and the CV program works well.
Regards,
Mauri.

Offline Chaoticone

*
  • *
  •  5,628 5,628
  • Precision Chaos
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #34 on: September 29, 2017, 08:48:07 PM »
Thats excellent thank you, I've learned and gained a lot from this thread.

Last weekend I was trying to cut some timber clamp knobs and the cutting was very slow and jerky, I have just run a test with the knowledge and Wizards I have got from this thread and it runs as smooth as silk.

Once again thankyou.

Good deal!
;D If you could see the things I have in my head, you would be laughing too. ;D

My guard dog is not what you need to worry about!

Offline Chaoticone

*
  • *
  •  5,628 5,628
  • Precision Chaos
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #35 on: September 29, 2017, 09:10:57 PM »
Quote
The first main issue that I have found is that the Mach4 program has a flaw in the CV area, in that when you go from G1 to G0 it goes off line.

This is no surprise. There is no CV in Rapids (G0). Rapids are well rapid, no throttling back. Get there as fast as your motor tuning will allow.

Quote
If you make all the code G1, then it does not, but the time increases significantly.

I would imagine it does. This would be equal to starting a round track race and telling car number G0 he just has to get to the other side and can drive through the infield (shortest path) while telling car G1 he has to stay on the track.

Quote
The second issue on lettering is that no matter the tolerance setting on your CV program on curves it does not follow the line.

I fully expect that. The question is, does it stay within the tolerance you entered? If so, that's the best that can be done with CV. If you have to stay exactly on path then CV is not an option. That is what exact stop is for.
;D If you could see the things I have in my head, you would be laughing too. ;D

My guard dog is not what you need to worry about!

Offline Mauri

*
  •  306 306
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #36 on: September 29, 2017, 11:36:05 PM »
Chaoticone,
Thanks for the reply.
I decided to make one more test.
Instead of calculating an optimized G-Code on all the lines at once, I decided to do each line separately in Rhino/madCAM and optimize each line separately, but output it as one G-Code.
I tested one with CV .001mm Tolerance and the other all Angles at 100 keeping the G1 and G0 as is.
This method no longer has the off line movements on going from G1 to G0, but does takes a bit more time, however a lot less than all G1.
So for lettering this is the way to go.
The result in the DOC.
Regards,
Mauri.

Offline Mauri

*
  •  306 306
    • View Profile
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #37 on: September 30, 2017, 06:11:59 PM »
CV in CNC applications is an interesting subject and there are many out there that have been studying this issue for some time.
Here are some more mathematical ways of making a angle to a tolerance at maximum speed.
Regards,
Mauri.
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #38 on: September 30, 2017, 07:50:31 PM »
Had a look at those papers but it makes my head hurt.
Re: Erratic Feedrate When Cutting Spline Lines
« Reply #39 on: October 01, 2017, 01:26:46 AM »
Hi,
yeah, but hurt in a good way LOL. That last paper that uses a cascade of FIR filters looks like 'just the dogs dangly bits'

Overlapping two motions with a constrained error term...just brill...there are some very sick little puppies out there!

Craig
My wife left with my best friend...
     and I miss him!