Help with IJK Rads

[New Job]

Did you upgrade you post to X2?

[Lee.]

yes I did.

 

I don't think the prob is with the post, somewhere in the definitions I think, but just not sure.

 

The same post worked for ever in V9

 

?????

[New Job]

Have you posted successfully from this post since the upgrade?

[Lee.]

I ran a program that didn't screw up, if thats what you mean.

 

Not sure why or why not tho'

[gcode]

N2570 X-16.0196 Y3.121 I-4.4367 J-.3351

N2580 X-16.0195 Y3.1149 I-4.4493 J0.<<<<<<<<

 

Lee,

there is .0001 difference bewteen these 2 arc moves

Your machine cannot do it so its traveling a full circle to get to that point.

Look on your tolerance page of the Control Def

and check the settings for

"Min distance bewteen arc endpoints"

default is .002

and

"Min Arc Length"

mine is defaulted at .0005

 

I suspect one of these settings is the culprit

[New Job]

I'm kinda stumped here, without anything to look at I'm not sure where to go next.

[Lee.]

I'll check G-code and thanks for the input.

[?Mark]

I noticed that gcode, but there is .006 difference in Y movement...

[?Mark]

Shot in the dark but try to uncheck "optimize toolpath" in op parameters page (under compensation type or in cd under cutter compensation)and see if it fixes it.

[Lee.]

G-Code.

 

I changed the settings as you said, then saved the settings etc.

 

Started a new Group, copied the op down in to the new group, then posted it.

 

Same results.

 

I'm really screwed here, any ideas?

 

Lee.

[gcode]

quote:

---

but there is .006 difference in Y movement

---

There is nothing wrong with the gcode, Lee's machine just doesn't like it.

 

I'm wonder what the code would look like posted with a stock Fanuc X2 post??

[88Matt]

We have run into this on our wire on occasion. It has always been a rounding issue. Usually 2 tangent arcs or splines. Tightening up the chordal deviation and using the filter does the trick usually for us. HTH

[gcode]

quote:

---

I'm really screwed here, any ideas?

---

Go to Help/ Zip2go utility, make a zip2go file

and upload it to the ftp sight. I'll check it out and see if I can find out whats going on.

 

Have you tried posting this with a stock post???

[Matt Berube at Ferron Mold]

quote:

---

There is nothing wrong with the gcode, Lee's machine just doesn't like it.

---

exactly.

 

Lee,

 

This sounds a lot like something that happened to me a while back.

 

Let me guess... You have a Fanuc O-M control ?

 

When I had my problem I tried the exact same .nc file in a better control and it worked perfectly.

[Lee.]

I'll upload a file called g-cODE-fOR-lEE.z2g

 

I hope you find something.

 

Lee

[New Job]

There is parameter that tells the control how much deviation to allow on your arcs. you may need to tighten or loosen this setting. I don't have an OM so I can't tell you where it is.

 

HTH

[Lee.]

Matt,

 

I'm not sure how that helps me though?

 

As much as I know how old the control is, it doesn't change the fact that it was fine with V9 MR***05

 

G-Code, it's in X2 files.

[gcode]

Lee upload the X2 file, there are no tool paths in the one on the FTP

[?Mark]

quote:

---

but there is .006 difference in Y movement

---

I meant to point out that code is good

 

Lee, have You switched to R's and break at quadrants yet?

I think that's what worked for me on a yasnac control several years ago.

[Brandon Walker]

Hi Lee,

 

I just emailed you,

 

My suggestion is to try using R value radius output for arc motion as well if it is at all possible. As i understand the problem (arc is taking the opposite arc motion, as if using a signed -R value), it seems to me that the IJK values may not be right therefor the machine cant interpolate/calculate that motion therefor it is moving in the opposite direction with a much larger sweep angle to try and get to the end position of the arc.

 

Let me know if this works for you.

[Lee.]

G-Code.

 

It's called Model-Large-Shrink

 

in MCX2 files.

 

R-Rads don't work well for this control guys.

 

I have had absolutely no probs with MR0105

[Matt Berube at Ferron Mold]

quote:

---

I'm not sure how that helps me though?

 

As much as I know how old the control is, it doesn't change the fact that it was fine with V9 MR***05

---

Lee, I regret that my post was not helpful.

 

The actual problem I had WAS on MR0105.

 

old thread if you're interested

 

I hope that with the suggestions made in that thread or this one you will find a solution.

[Lee.]

Matt,

Thanks, I'm just not convinced it's the control.

 

Although it could be a coincidence.

 

I'm pulling towards either a loose or tight tolerance in X2 that I haven't found yet.

 

I hope G-Code finds something.

[Gilberto]

we had this problems a few years ago.

editing post variable "arccheck" fixed it.

 

METRIC VALUES

arccheck : 1 #Check for small arcs, convert to linear

ltol : 0.1 #Length tolerance for arccheck

[Lee.]

Gilberto,

 

Could you explain this a little more for me?

 

Also, is it now relevant since X looks for some values in the Control Definition.