It is most likely a machine setting. It the machines are identical, then it may be possible to switch between G90 and G91. If they are not identical, then who knows?
What machines and controllers?
Hi,
Switching from one machine to another this problem popped up and I havn't been able to solve it.
Running G-code fine on one machine yet showing errors on the other.
My g02 programmed as X Y I J with a G90 works on one machine. Reading all the functions as absolute.
On the second machine the I and J will stay incremental no matter what I try.
The simpler the change is to rewrite the same codes would be best. If this is possible ofcourse. Might it be
that I miss crucial information on however the systems work? Could it be a machine setting?
Thanks in advance,
Similar Threads:
It is most likely a machine setting. It the machines are identical, then it may be possible to switch between G90 and G91. If they are not identical, then who knows?
What machines and controllers?
Jim Dawson
Sandy, Oregon, USA
It is very common nowadays that IJK are incremental by definition. Not only sinumerik, but Fanuc and many other languages works like that.
hy, if you wish, i can create a simple app, that should fix your code
deliver code as is, and also as it should be, so to spot the differencies, and write the code for it
we are merely at the start of " Internet of Things / Industrial Revolution 4.0 " era : a mix of AI, plastics, human estrangement, powerful non-state actors ...
Thank you all for helping.
For now i'm rewriting the code manually. An app would be awesome, yet I am not to sure what takes more time. Also there are some inconsistency's in the code regarding rounding off values. Thus I'm not too sure an app will work for that, yet I do not know anything about apps, how they work etc. What would such an app cost me?
hello look, don't worry, i am here to help; if i can solve your task, then we talk
example's about how it could work :An app would be awesome, yet I am not to sure what takes more time
... you simply drag your file over an icon on desktop, and then an app will fix it automatically
... or you put in inside a folder, that is always scanned for content, thus you file is detected automatically
... or you open a progam, drag your file in it, run a few checks ?!?
however, it can be done simple, but, if you wish to go that way, then simply provide what is, and how it should be
truncation problems, and such, when it comes to radius, is not a new problem; some cnc machines are built in such a way, to somehow detect and compensate such anomaliesAlso there are some inconsistency's in the code regarding rounding off values
i don't say that i can fix those, but at least i can try; depending on your actual error level, it may be possible to aproximate it for example, at the end , you receive a report, about what has been fixed, and how much shift was needed
Last edited by deadlykitten; 10-13-2021 at 02:20 PM.
we are merely at the start of " Internet of Things / Industrial Revolution 4.0 " era : a mix of AI, plastics, human estrangement, powerful non-state actors ...
hy in case you are still arround, below is a code generated by whatever cam software, that i process futher, detect&display deviations in the last column; later, those can be minimized, or even eliminated
it happens automatically, in background, thus it works, and you may forget about it / kindly
Code:G03 X 207878 Y 327098 I - 30444 J 9632 e0.052 X 207989 Y 339764 I -194548 J 8038 e0.028 X 207385 Y 352557 I -287669 J - 7171 e0.002 X 206234 Y 362484 I - 99825 J - 6544 e0.012 X 197269 Y 386603 I - 69579 J -12137 e0.074 X 187866 Y 399547 I - 80569 J -48641 e0.006 X 179012 Y 407785 I - 43880 J -38285 e0.125 X 164127 Y 416200 I - 40003 J -53387 e0.08 X 155614 Y 419110 I - 29250 J -71661 e0.027
we are merely at the start of " Internet of Things / Industrial Revolution 4.0 " era : a mix of AI, plastics, human estrangement, powerful non-state actors ...
You did not tell us which type of sinumerik your are using.
If it is not a very old model you can try the following:
If for example you have J17 in your NC-program replace it by J=AC(17) (and likewise for I and K).
I, J, K parameters are incremental by default. With the AC syntax (AC means Absolute Coordinates) the value given will be interpreted as an absolute value.
Parameter is set wrong in the 2nd machine so find out what ones related to G02 and G03 and away you go
If I understand correctly the two machines do have different controls i.e. only the new machine has got a sinumerik control.
Though it will be rather impossible to compare parameter sets of both machines.
I don't know of any parameter in sinumerik controls which allows to interpret I, J K for circles as absolute values per default.