A head scratcher for me

More
23 Aug 2018 00:32 #116466 by scottsdalejohn
I'm new to all of this so the answer to by problem may be obvious to you, but not to me.

I have a 4 axis hot wire cutter to make wings for RC gliders. I'm using a TB6560 controller and Nema 23 motors and Jedicut to generate the G-Code. All has been going well and I'm learning the entire process but I have one panel that will cut perfectly on the left side, but the right side goes into a high speed mode as soon as the actual cut begins. I have looked at the code for for each side and they look identical to me.

Here is a snip it of the code in question:

( Plan XY )
G17
( SET LENGTH UNITS )
G21
( SET DISTANCE MODE )
G91
( SET CUTTER COMPENSATION )
G40
( SET TOOL LENGTH OFFSET )
G49
( SET PATH CONTROL MODE )
G64
( SET FEED RATE MODE )
G94

( Move max speed )
F300
G1 X0 Y10 U0 V10
( Move cut speed )
F150
G1 X177.907 Y0 U0 V0
G1 X10 Y0 U10 V0
( Move cut speed )


Everything up until here appears to be work fine ... the feed rate is what I expect, but as soon as it hits the next line the feed rate appears to double ans stays at that speed until it come to this bit of code:( Move cut speed ) and it slows back down the anticipated feed rate and continues to the second part of the code:
G1 X70 Y0 U70 V0
G1 X0 Y90 U0 V90
( Move max speed )
F300
G1 X-277 Y0 U-346 V0
G1 X0 Y-91 U0 V-93
( Move cut speed )
F150
G1 X177.907 Y0 U0 V0
G1 X10 Y0 U10 V0

Any thoughts on what is causing this problem will be appreciated.

Please Log in or Create an account to join the conversation.

More
23 Aug 2018 01:59 #116468 by curtisa
Replied by curtisa on topic A head scratcher for me
You have two feed rates being called alternately throughout your Gcode ('F150' and 'F300'). Are you saying that the program is not following the requested feed rates every time they change?

Don't forget that feed rates are modal. Ie, specifying a feed rate will result in every subsequent G1 move to be at that feed rate *until* a different feed rate is specified.

Please Log in or Create an account to join the conversation.

More
23 Aug 2018 04:16 #116473 by scottsdalejohn
I believe that is what is happening. The strange thing is it will cut fine until I get to a certain amount of sweep back and then it seems to have a mind of its own. In the morning I'm going to set all of the feed rates to F150 and see if that resolves my issue.

Thanks for the response!
John

Please Log in or Create an account to join the conversation.

More
24 Aug 2018 10:58 - 24 Aug 2018 11:05 #116526 by andypugh
Replied by andypugh on topic A head scratcher for me
I think that the issue here might be that feed rate is calculated _purely_ in the XY plane, and if the UV plane distance is larger then that side will move faster.

Ideally your CAM should use inverse time ( G93 ) mode where the time required to make the move is specified.

Out of interest, do you have Axis configured with FOAM=1 ? That gives a foam-cutter specific preview mode. It seems to be present in 2.7 but only documented in 2.8 (the dev docs)
Last edit: 24 Aug 2018 11:05 by andypugh.

Please Log in or Create an account to join the conversation.

More
02 Nov 2018 23:11 #119858 by rcKeith
Replied by rcKeith on topic A head scratcher for me
Hi
I hit exactly the same problem today and was puzzled. I have foam =1 set in the ini file. Right wing cut perfect then the left wing went crazy. I thought I had a problem with the code but nothing wrong with that. I then installed my Mach3 disk and ran the gcode and it worked perfectly.
Did you find a solution ?
LinuxCNC for foam cutting is much nicer than Mach3 with the foam option set.

Keith

Please Log in or Create an account to join the conversation.

More
05 Nov 2018 11:28 #119982 by andypugh
Replied by andypugh on topic A head scratcher for me
Does the-code for both wings have the big end at the XY end?

When you say "went crazy" what do you mean?

Please Log in or Create an account to join the conversation.

More
05 Nov 2018 12:16 #119986 by rcKeith
Replied by rcKeith on topic A head scratcher for me
Hi Andy
The right wing has the big end at XY but the left at UV. The UV iside started ok then went really fast. I tried it on Mach 3 and it worked perfectly, I used Profili2 Pro to generate the code which is quite old now. I contacted Stefano at devcad.com asking if it supported G93 and it doesn't but his new software DevWing Foam does.
I think I can get around it by generating the gcode for right wing inverted which the software will allow and the left normal , but I'm not 100% certain if I'll end up with a wonky wing , need to try that,.
I prefer to use LinuxCNC because AXIS shows the profiles correcly, Mach3 doesn't do that.
Thanks
Keith

Please Log in or Create an account to join the conversation.

More
05 Nov 2018 12:42 #119987 by andypugh
Replied by andypugh on topic A head scratcher for me
Are you in a position to test an air cut?

It is possible (but not likely) that setting the machine to the UV plane will solve the problem in cases where the UV side is the big one.

linuxcnc.org/docs/2.7/html/gcode/g-code.html#gcode:g17-g19.1

Please Log in or Create an account to join the conversation.

More
05 Nov 2018 12:52 #119988 by rcKeith
Replied by rcKeith on topic A head scratcher for me
Hi Andy
Thanks for that, sounds like that may be the answer, I give it a try and let you know.

Please Log in or Create an account to join the conversation.

More
05 Nov 2018 12:58 #119989 by andypugh
Replied by andypugh on topic A head scratcher for me
Well, it feels like it should be but a cursory search in github didn't show the selected plane as being used in the velocity calcs.

Please Log in or Create an account to join the conversation.

Time to create page: 0.083 seconds
Powered by Kunena Forum