Instability in toolpath, retracts added? Rotary axis windup issue???

[Bruce Caulley]

Hello all,

I have managed to work my way around some of the problems I was having with 5 axis surfacing recently, but now that I am posting I have run into a couple of issues.

 

First: The posted code has several of the following

code:

---

 105  L X-24.014 Y+17.651 Z-2.028 B+1.152 C+311.892

106 L X-24.037 Y+17.625 Z-2.023 B+.951 C+312.438

107 L X-24.07 Y+17.585 Z-2.017 B+.683 C+312.727

108 L X-24.103 Y+17.545 Z-2.01 B+.415 C+313.51

109 L X-24.137 Y+17.503 Z-2.004 B+.157 C+313.042

110 L X-24.071 Y+17.432 Z+47.996

111 M129

112 L Z-1 F MAX M91

113 L X-495 Y-350 R0 F MAX M91

114 ; INSTABILITY IN TOOLPATH

115 M00

116 TOOL CALL 238 Z S5000

117 CYCL DEF 19.0 WORKING PLANE

118 CYCL DEF 19.1

119 M128

120 L X-24.216 Y+17.506 F MAX ; CLEARANCE FOR TABLE ROTATION

121 L Z+48.002 F MAX ; CLEARANCE FOR TABLE ROTATION

122 L B+.102 C+134.56 F MAX

123 M5

124 L X-24.172 Y+17.461 Z-1.997 F2

125 L X-24.208 Y+17.416 Z-1.991 B+.353 C+135.615

---

Secondly, the code doesn't appear to be starting in the same C axis position each time it adds a retract. The machine is a DMU 50e and the c-axis limit is set at 30,000 degrees in the control. I was hoping to be able to finish this whole path in one pass, but it would need a c-axis windup of nearly 600,000 degees.

 

My questions are: What is causing the instability in the post and can I set the post to retract and reset the c-axis once it gets to 30,000???

 

I will try and do this with a 4 axis path and see if I can play around with the limits on the machine, but any help I can get would be great.

 

Bruce

[Bruce Caulley]

bump

[Doug Gabbey]

Bruce,

My experience on 5 axis programming is up to three programs now, but someone, Dave Thomson I think, told me that if the tool is cutting very close to the center of rotation it causes instability. Possibly very small numbers involved in the Math.

We just got a DMU80T with renishaw, laser etc.

Best regards,

Doug

[deckerc]

The instability is when your doing five axis work and the tool passes back thru vertical. The math equations then try to divide by zero which is impossible. I usually try changing my paths to prevent the tool from going perfectly vertical.

 

I'm not sure about the wind up.

[Allan Brown]

If you go vertical in a 5-Axis toolpath, it may result in large C-Axis motion. Vertical has infinite C solutions, but minor tilts in B have only 1 C-Axis solution (there are 2 B solutions +/-, with 1 C solution each).

 

We check for C motion (inst_lim) in your post. If the C-Axis motion exceeds your inst_lim variable, you get a retract. We also don't feed back down at the same point - we skip the motion from start point to end point of the move that causes the large C-Axis motion. On the Heidenhain, moves are interpolated. If you want the tool in contact during the C-Axis swing, loosen off the inst_lim setting.