Reminds me of STEP-NC/AP238. http://www.steptools.com/products/stmachine/benefits.html http://www.steptools.com/library/stepnc/faq/faq_04.html http://www.isd.mel.nist.gov/projects/stepnc/omac-orlando-2005/pilot-scenario.pdf
Mark --- Kirk Wallace <[EMAIL PROTECTED]> wrote: > On Sat, 2007-10-13 at 01:07 -0600, [EMAIL PROTECTED] > wrote: > > Hi Ken > > > > Today manufacturing ? machining moves into specialize targeted > market. > > There are big size manufacturing company that can produce part > that > > cost more than that middle size brand new cnc mill! > > > > Some NC program if print out in letter size 10, will be 200 > pages plus > > long with use canned cycles! > > Because program is so big -200 pages it is impossible manually > alter > > the code. To change something need use Catia and to check if > > everything ok need use another software. > > > > It is super big risk to let machinist to change any parameters > in > > controller. I am sure that it is unique situation and that will > bring > > unique solution. > > CAD /CAM become less expensive ? $15 000 for all 5 axis milling > and > > parts become more complex and expensive. So PC that runs > CAD/CAM > > becomes extension of machine controller. > > From this perspective codes that I mentioned can be good > foundation > > in generation of universal G codes. I am interesting in > specific and > > targeted criticism of my proposal > > > > Thank you, > > Aram > > I am trying to understand the gist of what you saying, so I will > try to > restate it. > > Problem: > Most CNC produced parts are so complex that the g-code that > produces > them, becomes too complex for manual editing or validating. > > Solution one: > Use additional software to do the editing and validation. > > Problem with solution one: > Must manually use additional software. > > Solution two: > Merge CAD/CAM with CNC controller. > > Problem with solution two: > CAD/CAM g-code won't match g-code for various CNC machines, hence > the > need for universal g-codes. > > Solution to the problem with Solution two: > Create universal g-codes. > > My thoughts on: > > The Problem: > I agree that g-code files are becoming complex to the point of > being > unmanageable and will get worse. > > Solution one: > This seems reasonable. Just as C came about due to the increase > in > assembly code complexity, and C++ from C, and so on. > > Problem with solution one: > Here is were it gets interesting and I have little CNC experience > in > this. I think you are indicating that CNC operators are having > problems > with manually using g-code and would also have problems with > manually > using a g-code validator/editor or using the software that > created the > g-code to begin with. In the shop I briefly worked in, that is > what the > machinists did. They would get a work order with the part's file > name, > bring up the part in CAD/CAM and review the drawing and any > revision > notes. Then they would use the original code as-is, or invoke the > CAM > procedure and send the g-code to their machine. The operators > that could > do both CAD/CAM and machining got to do the more interesting > parts. > Occasionally, small parts of the g-code would have to be tweaked. > The > biggest problem usually was in finding were in the code to look, > which > may be were the real problem lays(sp? gr?). I found, if the > part's > original g-code was well commented, it was much easier to locate > the > relevant code. Maybe, finding a way to automate the commenting of > code > would be of value. I guess canned code helps here to. There is a > divergence here in that, you suggest a simplified g-code, which > would > need more commenting to decipher, and I am thinking more complex > canned > code might be better because it tends to be more self-indicating. > > Solution two: > When I first used EMC I used tkEMC and became attached to its > simple > interface. I then started using AXIS and found the integrated > backplot > window to be valuable. It provides some of the functionality of a > CAD > program in validating code, especially by successively clicking > through > the code. From your conversation. it suggests that it would be > nice to > be able to edit or place features and way-points on the backplot > and > have an EMC plug-in create the appropriate machine commands in > either > g-code or NML. > > Problem with solution two: > I don't think you will ever see universal g-code beyond G0, G1 > and a > handful more. There is no incentive, for the powers that could > influence > such a thing, to pursue it. > > It would be interesting to figure out who has how much power in > the > industry and how it could be influenced. EMC on the surface, > seems like > a bunch of guys (and gals?) on a mailing list playing with > machines and > computers, but I wonder, how many, if any, high power lurkers are > out > there gleaning nuggets of insight from this activity. I remember > when > the idea of a personal computer was a ridiculous idea. > > Solution to the problem with Solution two: > See - "My thoughts on: / Problem with solution two:" above. > > I think that since there is no one major player in the CNC > industry, > that for the foreseeable future, standards will remain fragmented > and > redundant. Evolutionary solutions will come about through > converters, > filters and plug-ins. > > I do beleive it would be valuable to merge at least some if not > most > CAD/CAM functionality into the CNC machine controller. Maybe > g-code > could go away and the common denominator between machines from > different > manufacturers would be one, or a collection of, drawing formats. > > --- Caution: Side rant ahead --- > > I think the issues you are dealing with are political rather than > technical. > > If you use PC operating systems as an example, it would be nice > for a > computer user to need to learn either one operating system or > have all > operating systems be very similar in how they are used. When the > PC was > new, there were nearly as many operating systems as there were > manufacturers. As a result of power shifts over time we have > shifted > towards a more universal (Microsoft) operating system. Because of > the > incompatibilities between operating systems, most people have had > to > make a choice and use only one platform, PC, Mac, Linux, etc. And > the > pressures on individuals to be compatible with other individuals > has > driven the market to a single operating system. Thank goodness, a > small > minority of Mac and Linux users (geeks) have figured how to be at > least > partially compatible with Microsoft. > > My thinking is that, in the beginning, the PC operating system > consumer > and operating system producers were pretty level in power. > Microsoft was > able to provide a product that satisfied the majority of the > individual > consumers, so it gained the majority of power among > manufacturers. It > then used that power to destroy its competitors, which gave it > power > over consumers. Fortunately, there is a feedback loop, in that > Microsoft > has lost interest in keeping its customers happy, so many people > are > looking for alternatives. More individuals with a common > interest, I > hope, will gain more power over a group of individuals commanded > by a > central authority (Bill Gates). I think that individuals that can > provide things like the operating system compatibility can, help > facilitate a shift in power, but unfortunately, they can't drive > it. > > -- > Kirk Wallace > (Hardinge HNC lathe, California, USA > http://www.wallacecompany.com/machine_shop/ ) > > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a > browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
