Need Help!- Coordinate Rotation G68

[dougtyler]

We are using a Fanuc 31I control

The G68 command is limited to .001 precision.

Is it possible to make the control calculate a precision of .0001?

Optimally, we would like to get the precision to .00001

Anyone run into this problem before?

[underthetire]

.00001? You would have to switch the entire control over, and it would still be useless without a high accuracy machine behind it. Scales, hydrostatic spindle and ways, thermal to .01 Deg or better, on and on.....

[dougtyler]

Yea, I know.

The control is on a machine that has a positioning accuracy of .00001 inch.

The accuracy of the machine is there, but this feature (g68) of the control is not.

[christinandavid]

Do you mean that 0.001 of a degree is not accurate enough (a common issue), or does the control only work out co-ordinates to three decimal places?

DP

[dougtyler]

.001 of a degree is not accurate enough.

We have a part that has bores that are 36 inches apart.

We want to be able to load the part on the machine, probe one bore and set a work offset, probe the other bore and find the angle the part is setting at.

The control will calculate the angle out to .00000000001 degrees. However, when I program the G68 code this angle gets rounded off to the nearest .001

At 36 inches this equals .00064 inches radial movement.

.0001 degrees would work for what we need. This would be .000064 at 36 inches.

But to realize the full accuracy of the machine, it needs to work at .00001 degrees.

Hope this clears up the question.

[christinandavid]

How much complexity is involved in the machining operation ie is it straight forward milling/bolt hole patterns - can you get around the problem using macro b?

DP

[dougtyler]

If I could work around the problem using macros, I wouldn't have asked the question.
I need to find someone who has dealt with this problem before.
I'll be getting in touch with the Fanuc rep. Just wanted to throw a line out in the forums to see if I could find some answers in the mean time.

[dougtyler]

If I could work around the problem using macros, I wouldn't have asked the question.
I need to find someone who has dealt with this problem before.
I'll be getting in touch with the Fanuc rep. Just wanted to throw a line out in the forums to see if I could find some answers in the mean time.

[sinha_nsit]

Use increment system C, instead of B.

[dougtyler]

How would I do that?

[dougtyler]

Amazingly enough, theres a parameter setting that changes the accuracy to exactly what I need.

After talking to the Fanuc rep. he discovered parameter #11630

Bit #0 (FRD) of this parameter changes the resolution of the R word in the G68 code.

It changes it from .001 to .00001

Thanks Eric (fanuc rep)

Thanks to the ones on the forum for your help.