What's new
What's new

G71... Please help! What is wrong with this program?

yardbird

Titanium
Joined
Jul 3, 2013
Location
Indiana
Hello All,

I’ve got an issue with my G71 roughing program. I have scratched my head and looked and I just don’t see what the problem is? First run of this new program, everything is going just as it should until the very last roughing pass. It comes down the angle then swings the 4” radius then comes a cross the 1”dia. but instead of swinging the .38rad it plunged right into the shoulder. WTF? No I didn’t run it through the graph or out in dry land first. :mad5:

I’ve attached some pictures of my drawing and the actual program and the end result. If someone could help me figure out what the deal is here I'd greatly appreciate it. At this point I’m going to go in and run the same program out in dry land and see if it does it again. :scratchchin:

Other than that I got nothing. I hope you can see the drawing and pictures? The exact code is at the bottom.

Thank You….

Brent





help-1.jpg
20171127_223355.jpg
help-2.jpg


N30(ROUGH OD)
G0G99G40G54X14.Z10.T0
T0303
M41
G97S150M4
G0X3.5Z.1M8
G50S700
G96S450
G71U.05R.05
G71P31Q32U.03W.003F.012
N31G1X1.4854Z.0000
G3X1.4950Z-.0063R.005
G1X1.2726Z-.4213
G2X1.0000Z-1.4566R4.
G1X1.0000Z-2.0100
G2X1.7600Z-2.3900R.38
G1X2.9400Z-2.3900
G1X3.0000Z-2.4200
G1X3.0000Z-3.3160
N32G1X3.5
M9
G0G99G40G54X14.Z10.T0
M1
 
I guess you have confirmed you have a type 2 control and can turn reducing diameters in G71?

N31 G0 X1.485 Z.1
G1 Z0.0 F.006
G3 X1.495... etc...
 
I guess you have confirmed you have a type 2 control and can turn reducing diameters in G71?

N31 G0 X1.485 Z.1
G1 Z0.0 F.006
G3 X1.495... etc...

Thanks! Fanuc 0i-td I've used type 2 roughing many times.

The thing is I was standing right there watching it run, small DOC for fist part, it roughed down to the 1 1/2" on the end, then started carving out the concave portion, everything looking good. Multiple stabs on the .38rad, all is good until the very last pass, I watched it sweep the 4" radius and come across at 1.03". It got to the start point of the .38rad and the sumbitch just plunged right into the shoulder. WTF? I'm stumped....

Brent

20171127_224809.jpg
 
Last edited:
...It got to the start point of the .38rad and the sumbitch just plunged right into the shoulder. WTF? I'm stumped....
Your start of profile looks funky to me. The first line after the P address block should be a G1 Z0.0 move at the min X dia., with the feedrate for the G70 pass, and the P block should be a G0 to the X Min with to the same Z position as the start position.

I'd make those edits and run it in air to see what it does.
 
Your start of profile looks funky to me. The first line after the P address block should be a G1 Z0.0 move at the min X dia., with the feedrate for the G70 pass, and the P block should be a G0 to the X Min with to the same Z position as the start position.

I'd make those edits and run it in air to see what it does.

First thing I'm going to do is run the exact same program again watching it on the graph. I'm not trying to be argumentative but I've run that same roughing routine a thousand times but I'm willing to give your suggestions a shot. %99.9999 it's my fault, on very few occasions over the past 29yrs have I thought the machine had a brain fart, this is one of them. IDK? Thanks again...

Brent
 
@jancollc,

Firs two pictures are the rough and finish as I had it. Changed program to like 3rd picture, results are in 4th picture. I truly don't know why but you nailed it. Thank you for your help!

:cheers:

Can anyone explain exactly what the deal is here? I believe this is the second time this has happened to me and I've written hundreds of programs using this roughing template but I'm changing that today. This could have been really ugly if I wasn't close. Brain fart? Lol....

Brent

20171128_162609.jpg
20171128_162837.jpg
20171128_164814.jpg
20171128_164729.jpg
 
Wow!!!

Although I always position my tool in Z ( or X for G72 ) in the same position as the P block, but would have never thought it to be of an issue
for the control...

As you have gotten me very curious, in a few hours I'll load your program into an OiTC and a Haas and see if they behave the same.
 
Wow!!!

Although I always position my tool in Z ( or X for G72 ) in the same position as the P block, but would have never thought it to be of an issue
for the control...

As you have gotten me very curious, in a few hours I'll load your program into an OiTC and a Haas and see if they behave the same.

That would be great! Thanks...

I had this one other time a while back using G72 but never asked about it then.

Brent

20151024_211133.jpg
 
Wow!!!

Although I always position my tool in Z ( or X for G72 ) in the same position as the P block, but would have never thought it to be of an issue for the control...
On the type 1 G71 where diameter is increasing, you don't even need a Z position on the P block. It will just use the Z start position. On the type 2 you need the Z position.

In both types, the next line has to be the G1 Z-axis feed to the start of profile. It can't be a G3.

At least, them's the rules as I understand them...
 
OK so I run the exact same program in post 8 the new improved version except I change the DOC from .15 to .05 and this is the results. I shit you not!

20171128_181619.jpg

20171128_181543.jpg

Then change the DOC to .1 and run again and then this? What's up with that?

20171128_182345.jpg

OK now just for shits and giggles I changed the program back to my original program as it is listed in the OP but increase the DOC from .05 to .1 and got this.

20171128_183838.jpg

If anyone else besides SeymourDumore has the time and could test this on their machine it'd be interesting to see if yours acts the same as mine.

Something funky is going on and appears to have to do with the DOC and not necessarily the start block of the cycle? IDK I got nothing...

Brent
 
Yard

I've just ran this in the Haas:

%
O00100
G28
G50 S700
G00 G99 G40 G54
G00 G97 T202 S150 M03
G00 X3.5 Z0.1
G96 S450
G71 D0.05 P31 Q32 U0.03 W0.003 F0.012
N31 G01 X1.4854 Z0.
G03 X1.495 Z-0.0063 R0.005
G01 X1.2726 Z-0.4213
G02 X1. Z-1.4566 R4.
G01 X1. Z-2.01
G02 X1.76 Z-2.39 R0.38
G01 X2.94 Z-2.39
G01 X3. Z-2.42
G01 X3. Z-3.316
N32 G01 X3.5
M09
M05
G28
M30
%

And it ran perfectly!
Note that Haas uses the single line G71 so that's the only difference but it should not matter.
I've tried a few different DOC-s, all was well.
It looks like that when your control roughs out the tapered section, it wants to ( for some odd reason ) rapid to the end of the path
without making a final smoothing pass.
On the Haas, after the taper is done, it rapids up-to start-X ( X3.5 ), then to start Z ( Z.1 ) and then skims the whole profile from there leaving .03 in X, .003 in Z.
Your control looks like wants to rapid to StartX and End Z instead ....

The Fanuc is still tied up with a job due tomorrow, but I'll plop it into it as soon as I can because I do want to know!
 
OK so I run the exact same program in post 8 the new improved version except I change the DOC from .15 to .05 and this is the results. I shit you not!

Dude, there IS something funky going on there!

On the Haas I get something like this:


The only thing changes with the DOC variation is where it rapids up to Start-X before the smoothing.
 
Yard

I've just ran this in the Haas:

%
O00100
G28
G50 S700
G00 G99 G40 G54
G00 G97 T202 S150 M03
G00 X3.5 Z0.1
G96 S450
G71 D0.05 P31 Q32 U0.03 W0.003 F0.012
N31 G01 X1.4854 Z0.
G03 X1.495 Z-0.0063 R0.005
G01 X1.2726 Z-0.4213
G02 X1. Z-1.4566 R4.
G01 X1. Z-2.01
G02 X1.76 Z-2.39 R0.38
G01 X2.94 Z-2.39
G01 X3. Z-2.42
G01 X3. Z-3.316
N32 G01 X3.5
M09
M05
G28
M30
%

And it ran perfectly!
Note that Haas uses the single line G71 so that's the only difference but it should not matter.
I've tried a few different DOC-s, all was well.
It looks like that when your control roughs out the tapered section, it wants to ( for some odd reason ) rapid to the end of the path
without making a final smoothing pass.
On the Haas, after the taper is done, it rapids up-to start-X ( X3.5 ), then to start Z ( Z.1 ) and then skims the whole profile from there leaving .03 in X, .003 in Z.
Your control looks like wants to rapid to StartX and End Z instead ....

The Fanuc is still tied up with a job due tomorrow, but I'll plop it into it as soon as I can because I do want to know!

Thank you! It's crazy to me that regardless of how the start P block is defined the DOC seems to be the determining factor on weather it screws up or not. In my mind this should not matter. I'd be interested in how your test come out on your Fanuc control? If you should get the same results try playing with the DOC and see if that changes the nasty move on the last pass?

Thanks again I appreciate it!

Brent
 
Brent

Now that I look at your graphs, it doesn't even rapid to the end, rather into oblivion in Z!
Almost looks like it wants to finish that 4" radius circle in full!!

Interesting that it doesn't alarm out with a Z-overtravel.
Don't know how Fanuc does the backplot, but Haas takes all travel limits into consideration and it throws a fit if the move
would go beyond the limit, BEFORE!!! it even starts the path.
Ditto in real running mode. If the programmed path goes beyond the travel limit, it will stop a few blocks ahead!
 
Dude, there IS something funky going on there!

On the Haas I get something like this:



The only thing changes with the DOC variation is where it rapids up to Start-X before the smoothing.

Not only funky but dangerous too, this fucked my stuff up and could have been a lot worse, I do hope you test this on your Fanuc 0i control and you get to see this shit with your own eyes. It appears by looking at the graph and seeing it running that the last jacked up move is a radius?

I somehow feel if I was to zoom out and let it finish the roughing cycle complete you'd see an 8 inch circle on the graph? Tomorrow I'm going to rerun the cycle with the apparently wrong DOC and let it run complete, I have a feeling there will be an 8 inch circle when it finishes?

BTW I increased the DOC so as to not have this problem on this particular run and I'm now using the tailstock to support the end.

Brent
 
Brent

Now that I look at your graphs, it doesn't even rapid to the end, rather into oblivion in Z!
Almost looks like it wants to finish that 4" radius circle in full!!

That is exactly how it appears. If you look at graphs you can see the changes to the P start block. I feel this is a something totally unrelated? Hell IDK?

Brent
 








 
Back
Top