Need G-Code for three axis chamfer path. - Page 2
Close
Login to Your Account
Page 2 of 2 FirstFirst 12
Results 21 to 25 of 25
  1. #21
    Join Date
    Apr 2018
    Country
    UNITED KINGDOM
    Posts
    2,877
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1341

    Default

    I was thinking a couple of different ways if you didn't have any cam ... one would be to calculate .005" steps in excel for each plane, then combine them. That wouldn't be too hard, just a little tedious. Could probably automate that if you were good with excel.

    Or you could use helical, but split it up into a bunch of arcs with a z that you calculate for each small arc. Probably get away with a lot less steps. There'd be a fair amount of typing and room for error tho, with the i's and k's for each arc.

    Another would be to program straight line moves in xz and add a rotary command in a, but have to figure your xz points in a similar way ... not sure which would be easier ? None of them fun but once done, heck, no more deburring

    Even doing this the hard way, you could be done in a couple hours or less. Probably worth it.

    Photo looks like it's in a lathe, btw, which would blow up all the axis designations ... 'zat really a mill, Kyle ? Looks strange. This'd probably be a good part for c axis lathe with live tooling ?

  2. #22
    Join Date
    Mar 2013
    Country
    UNITED STATES
    State/Province
    Oregon
    Posts
    1,913
    Post Thanks / Like
    Likes (Given)
    766
    Likes (Received)
    2268

    Default

    Quote Originally Posted by EmanuelGoldstein View Post
    Done. Simple as apple pie, just as easy as writing g-code.
    But you aren't done yet! Now go run those APT statements on his Haas...

    Now you're in the world of dealing with rickety CNC utilities that haven't been updated since 1998, or downloading uncompiled code from SourceForge. Why not? Because nobody who has invested in a 5 axis machine tool hasn't also spent 5% more for some decent 5 axis software in 20 years.

    Look, APT still has a place in the modern world - why do you think Siemens still offers it in NX? But it is absolutely not a tool that you're going to pick up and start using in 10 minutes. You're on a batan death march of figuring out how to get a bunch of outdated software utilities to work and spit out G-code for your machine.

    Or you could take about that same amount of time and get a small-biz copy of Fusion 360 for free and have a complete tool that could do this, and 90% of all the machine tool programming you could ever desire.

    The APT is a niche tool and is a jewel of ingenuity and elegance, but for most everyone who isn't doing some totally wackadoo jazz? It has a very poor ROI.

  3. #23
    Join Date
    Apr 2013
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    213
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    104

    Default

    Quote Originally Posted by Kyle Smith View Post
    I have been playing around with it, but it is 2.5 axis.
    I use InventorHSM now, but also had solidworks HSMworks years ago, and I think the Trace toolpath is included in the free 2.5 axis package. I use it all the time for 3d debur with a ball mill.

  4. #24
    Join Date
    Apr 2018
    Country
    UNITED KINGDOM
    Posts
    2,877
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1341

    Default

    Quote Originally Posted by gkoenig View Post
    But you aren't done yet! Now go run those APT statements on his Haas...
    I get it that you cannot stand to see any other method used, but oh well ... I have my own but as you pointed out, there's an interpreter on Sourceforge.

    Not as you claimed, only source. There is a Windows binary. I haven't tried it but probably will as a test ... see if it runs in Parallels or Softwindows, just for fun.

    Now you're in the world of dealing with rickety CNC utilities that haven't been updated since 1998
    I hear this often. "Rickety" ... tell me something - does 2+2 still equal 4 ? That's all that software does, add and subtract. As far as I know, Fortran code from 1968 still gives exactly the same answers to the same questions as it did in 1968. Has drilling holes and driving around contours changed ?

    Kinda funny on that subject, on the front page of this site are the same articles that Modern Machine Shop ran in 1980. The exact same stuff. Sure, the feedrate is higher but it's the same two flute endmills shearing off chips of aluminum. And the same questions on this forum as people were asking in 1980. Or dumber ones, in some cases.

    I don't see software wearing out. In fact, a lot of older software is better.

    or downloading uncompiled code from SourceForge.
    Has binaries. Also, compiling your own code is good. You can adjust it to fit your own needs. But that's a separate subject ...

    Why not? Because nobody who has invested in a 5 axis machine tool hasn't also spent 5% more for some decent 5 axis software in 20 years.
    Better not tell NCCS Or Tim Markoski, rest in peace.

    Is it the situation that the average person running five axes are stupider now than they were twenty years ago ? All they can do is point and click ? hmm.

    Look, APT still has a place in the modern world - why do you think Siemens still offers it in NX? But it is absolutely not a tool that you're going to pick up and start using in 10 minutes. You're on a batan death march of figuring out how to get a bunch of outdated software utilities to work and spit out G-code for your machine.
    I think I mentioned that most people will not go this way. But it is not anything like what you claim. It's as easy or even easier than learning g-code. Plus software does not get outdated. It is what it is, not like a cantaloupe that rots on the shelf behind your back. Write the program, run it through the interpreter, then postprocess. Exactly the same as now. Biggest difference is you have to type instead of click on icons. Just as easy to learn as Mastercam.

    btw, your NX will post a program I've written in APT ... just run the CL file through the post and double bass, "modern" software does the job In fact, they probably are the same posts because NX was (UG + I-DEAS) was UG was UniAPT, 90% odds that they haven't changed that part since 1980 because there is no need to. So much for rickety

    Come to think of it, I had a full tower PDP-8 with UniAPT tapes, so I am an NX user also, version 1

    Or you could take about that same amount of time and get a small-biz copy of Fusion 360 for free and have a complete tool that could do this, and 90% of all the machine tool programming you could ever desire.
    You have not had the same experience as me. Any American software company will eventually fuck you. Apparently all our reliable stuff went overseas, which I guess is good, I trust the Germans more than the worthless pricks leading US software companies.

    Let us note that copper-roofed Autodesk is located in San Rafael, California, on the wrong side of the railroad tracks. That's in the US still. They have a history of fucking people over, e.g. Delcam in our world and several other in various other fields. They are worthless cocksucking bastards.

    If that is whom you are recommending, thanks but no thanks. If I want a ten inch dick up my ass I'll go to San Quentin and get it free.

    The APT is a niche tool and is a jewel of ingenuity and elegance, but for most everyone who isn't doing some totally wackadoo jazz? It has a very poor ROI.
    Hmm. Would have done this part in ten minutes and cost nothing. I can't agree with your assessment. Certainly not for the masses but for the few people with an independant streak, it can work very well indeed. If you are working from prints instead of models, it's a no-brainer.

    I agree with you that it is not for most people. But your reasoning of why does not hold water. It's a viable tool for a few. More than viable, 5-axis capable from the very beginning, with a lot of features that make good sense to machining-type people. And as demonstrated right here, would be a very good answer for many current jobs. I could program this toolpath in ten minutes or less, at no cost for software and no Carol Bartz fingernails in my balls.

    You don't have to use it, it's okay but there are other people in the world ...
    Last edited by EmanuelGoldstein; 02-16-2020 at 06:24 AM.

  5. #25
    Join Date
    Apr 2008
    Location
    Helmer, Indiana, USA
    Posts
    1,208
    Post Thanks / Like
    Likes (Given)
    294
    Likes (Received)
    486

    Default

    Quote Originally Posted by Wsurfer View Post
    I use InventorHSM now, but also had solidworks HSMworks years ago, and I think the Trace toolpath is included in the free 2.5 axis package. I use it all the time for 3d debur with a ball mill.
    Thanks, I'll check it out.


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •