Hello Guest it is March 28, 2024, 03:53:20 PM

Author Topic: corners getting radiused at higher speed on plasma table  (Read 18175 times)

0 Members and 1 Guest are viewing this topic.

Re: corners getting radiused at higher speed on plasma table
« Reply #10 on: January 29, 2012, 01:43:11 PM »

Short of rebuilding the machine so you can achieve triple digit acceleration, the CV setting are the next best thing.
Which is more or less what I said.

[/quote]

Actually, you equated using the CV setting as "hitting the proverbial CV nut with a sledge hammer."

My point was that if you don't have to power to accelerate and decelerate the load at an acceptable rate, you either use the CV settings (which is a workaround to the real issue) or you rebuild your drive system to achieve the speed and acceleration you need.
Re: corners getting radiused at higher speed on plasma table
« Reply #11 on: January 29, 2012, 02:01:27 PM »
Thanks guys;

Instead of describing the machine, I'll post a picture. It's not an industrial machine, it's one I built.
I made the gantry from aluminum to keep the weight down. Aproximate weight is probably 50Lbs .
I run dual 960in/lb steppers for the Y axis and one for the X.
The Cad/CAM program I'm using is Dolphin Cad. All my electronics are from CandCNC.

Here is the g-code. I changed the speed to 200IPM from the original 350IPM.
You'll notice that the Post Processor did not program radiuses in the corners.

Thanks for having a look.
Dan


N0000 (Filename: Flight box side.tap)
N0010 (Post processor: MP1000-THCdan.scpost)
N0020 (Date: 29/01/2012)
N0030 G20 (Units: Inches)
N0040 G53 G90 G40
N0050 F1
N0060 (Part: Flight box side)
N0070 (Process: Outside Offset, Layer0, T2: .060 ALUMINUM Jet tool)
N0080 M06 T2 F200  (.060 ALUMINUM Jet tool)
N0090 G00 Z0.5000
N0100 X6.2300 Y2.9800
N0110 Z0.1500
N0120 G28.1 Z0.12
N0130 G92 Z0.0
N0140 G00 Z0.0940
N0150 G92 Z0.0
N0160 G00 Z0.1500
N0170 M03
N0180 G01 Z0.0600 F360.0
N0190 X4.2700 F200.0
N0200 Y0.5200
N0210 X6.2300
N0220 Y2.9800
N0230 M05
N0240 G00 Z0.5000
N0250 X10.5000 Y8.2700
N0260 Z0.1500
N0270 G28.1 Z0.12
N0280 G92 Z0.0
N0290 G00 Z0.0940
N0300 G92 Z0.0
N0310 G00 Z0.1500
N0320 M03
N0330 G01 Z0.0600 F360.0
N0340 G02 X10.5200 Y8.2500 I0.0000 J-0.0200 F200.0
N0350 G01 Y0.0000
N0360 G02 X10.5000 Y-0.0200 I-0.0200 J0.0000
N0370 G01 X0.0000
N0380 G02 X-0.0200 Y0.0000 I0.0000 J0.0200
N0390 G01 Y8.2500
N0400 G02 X0.0000 Y8.2700 I0.0200 J0.0000
N0410 G01 X10.5000
N0420 M05
N0430 G00 Z0.5000
N0440 X0.0000 Y0.0000
N0450 M05 M30


Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: corners getting radiused at higher speed on plasma table
« Reply #12 on: January 29, 2012, 02:58:25 PM »
Actually, you equated using the CV setting as "hitting the proverbial CV nut with a sledge hammer."
That is NOT what I said at all. If you're going to quote me quote me correctly.

It was ME who said:

You either have to do one or more of these until you're happy. Increase your accel (if you can), lower your feed, tweak CV params (see CV angle, CV distance, CV feed etc.), use exact stop (not good with plasma) or live with it I'm afraid. This is why the often forgotten quest for good accel is every bit as (if not more) important as the more poular quest for speed.

What you said was:

You can either turn on "stop CV on angles" and enter 89 degrees....or turn on "CV distance" and set it to something like .01".

That is the sledge hammer to crack a nut. Get it right.

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: corners getting radiused at higher speed on plasma table
« Reply #13 on: January 29, 2012, 03:01:24 PM »
Hi hemi43 - sorry about this. My first post was my best shot and I see there are no outside arcs so I'll have to have a think about it.
Re: corners getting radiused at higher speed on plasma table
« Reply #14 on: January 29, 2012, 03:36:34 PM »
Thank-you Stirling !
Any help is greatly appreciated.
I will try and mess around with some of the points you made earlier.
Dan

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: corners getting radiused at higher speed on plasma table
« Reply #15 on: January 30, 2012, 05:34:53 AM »
The outside corners ARE arcs as I suggested in my first post.

N0320 M03
N0330 G01 Z0.0600 F360.0
N0340 G02 X10.5200 Y8.2500 I0.0000 J-0.0200 F200.0
N0350 G01 Y0.0000
N0360 G02 X10.5000 Y-0.0200 I-0.0200 J0.0000
N0370 G01 X0.0000
N0380 G02 X-0.0200 Y0.0000 I0.0000 J0.0200
N0390 G01 Y8.2500
N0400 G02 X0.0000 Y8.2700 I0.0200 J0.0000
N0410 G01 X10.5000
N0420 M05

So I stand by everything I said in my first reply. So back to what you can do about your rounding. Apart from what I said about speed and acceleration, and with specific regards to what I said about CV TUNING you can....

1) Use "Stop CV on angles >" but as ALL your angles are 90 degrees using this with ANY value UP TO and including 89 will do exactly what the label name says i.e. it will turn CV OFF for ALL your angles i.e. you will get the same result as if you use Exact-Stop. Which is what I've been trying to say.

2) Turn on CV Distance and set it to SOME value. Now - you CAN set it to 0.01 if you want but this is saying - don't allow a blend until you're within 0.01 of an inch of each corner. This is really quite harsh and wil again effectively turn CV OFF. I would suggest YOU decide how much rounding is acceptable to YOU and have a play. So for example you might say I can live with 0.25 or maybe 0.1. If so, at least you're allowing CV a chance to work to some extent and you MAY find the results acceptable - you MAY not - it's YOUR call. All I have been trying to suggest is that you attempt to TUNE rather than just "hammer" it from the get go.

3) Take a look at turning on CV Feedrate and entering SOME value. This is a setting that's very rarely mentioned for some reason and when it is it's usually recommended to "hammer" this one as well rather than TUNE it gently to get results that are acceptable to YOU. See http://www.machsupport.com/forum/index.php/topic,20251.msg141302.html#msg141302 where I've attempted to explain its use.

CV TUNING is a bit of an art form - you can tweak it gently or you can smack it so hard you effectively kill it - and that's all I've been trying to say.

In the end though - you can't beat the laws of physics. No machine can change direction *instantly* but *good* acceleration can't be beaten for minimizing rounding.

Good luck and happy TUNING ;D
MODIFIED:ADDED BACKPLOT OF CODE      RICH

EDIT: Thanks RICH
« Last Edit: January 30, 2012, 11:34:55 AM by stirling »
Re: corners getting radiused at higher speed on plasma table
« Reply #16 on: January 30, 2012, 09:00:16 AM »
Thanks so much Stirlling !!
I checked the G code for arcs in the small cut out, but never thought of checking the outside corners.  ::)
I understand clearly what you're saying, and I don't have an issue having a small radius on my parts, because this is a plasma machine afterall and not a waterjet.
The radius I have been getting (about 3/8") was just too large and would ruin the part. I would be happy with a 1/8 radius.
I will draw a small sample part today and try some of your recomendations.
I will post back my results.
Thanks again for your time, it's greatly appreciated.
Dan
« Last Edit: January 30, 2012, 09:15:27 AM by hemi43 »

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: corners getting radiused at higher speed on plasma table
« Reply #17 on: January 30, 2012, 09:18:43 AM »
Hi Dan - There is actually a clue here for you perhaps which may help you in your tuning. If you're happy with the sharpness of the outside corners, that shows that CV is pretty much holding round those small arcs. You could try putting similar size arcs on the inside path. They won't cut as sharp as the outside of course but they may be better than what you have at the moment. One more thing to try anyway.

Ian
Re: corners getting radiused at higher speed on plasma table
« Reply #18 on: January 30, 2012, 09:23:11 AM »
Those arc were put there by my CAM, which I forgot to mention is from Sheetcam.
I will check to see if it's possible to create arcs for inside contours.
I'm sure I could manually modify the g-code, but I don't really want to go down that road. ;D
Dan

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: corners getting radiused at higher speed on plasma table
« Reply #19 on: January 30, 2012, 09:40:47 AM »
Yes - most CAMs will "roll" round reflex angles - or at least they should AFAIK. They will not do this for any other angle - they can't because you can't "roll" around those. You will have to create those other arcs yourself in your CAD.

Ian