Anybody seen a motoman robot follow a cad/cam contouring type of movement sequence?
Close
Login to Your Account
Results 1 to 13 of 13
  1. #1
    Join Date
    Oct 2006
    Location
    Mediapolis, Iowa
    Posts
    1,713
    Post Thanks / Like
    Likes (Given)
    364
    Likes (Received)
    387

    Default Anybody seen a motoman robot follow a cad/cam contouring type of movement sequence?

    Curious about an application where a robot such as a motoman kr series could be used to contour to some pretty loose tolerance for an end result of a “sculpted” piece?

    For example if you fastened a router at the end of the robot arm and used say 3 different router bits, could it sculpt a faceted end piece (picture a dog size piece of foam)?

    Anybody ever seen a video of something similar? Links?

    Program code would be from some kind of drip feed laptop like we often see on limited memory based cnc milling machines?

  2. #2
    Join Date
    Jul 2014
    Location
    Ontario, Canada
    Posts
    1,026
    Post Thanks / Like
    Likes (Given)
    731
    Likes (Received)
    662

    Default

    I’ve seen this firsthand but not using Motoman. It was a Kuka, path programmed was an intricate die polishing job. Buff wheel mounted to effector.

    Here’s a Motoman application.
    https://m.youtube.com/watch?v=OFQSAFZF91A

  3. Likes powerglider liked this post
  4. #3
    Join Date
    Oct 2006
    Location
    Petaluma CA 94952
    Posts
    486
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    75

    Default

    Ive got a UP50 with XRC control. There is a gibbscam Motosim plug in available from motoman that lets you run the arm as a multiaxis cutting machine. At least thats what the literature indicates. It acts like a post processor and creates motoman compatible programs directly from gibbs. I got a quote of 1000 for the plugin directly from motoman about 2 years ago. You have to have Motosim or Motosim EG (offline programming software) to use it (I think). Not sure if its required or if I got a better quote as I already owned a seat of Motosim. XRC control is the minimum that will work. I may get it to try out later in the year. WIll post results if I do

  5. Likes powerglider liked this post
  6. #4
    Join Date
    Jul 2006
    Location
    Hillsboro, New Hampshire
    Posts
    7,173
    Post Thanks / Like
    Likes (Given)
    1675
    Likes (Received)
    4931

    Default

    Here's some links to work being done by the Fraunhofer Institute:

    Robots as Tool Machines (RoboCasting) - Fraunhofer IFF

    Robot on demand: Mobile machining of aircraft components with high precision - Fraunhofer IFAM

    New kinematics for milling – customized, high-precision manufacturing

    I've seen other examples, but there's still some work to be done before articulated arm robots are really stiff and accurate enough to do precise work.

  7. #5
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    7,591
    Post Thanks / Like
    Likes (Given)
    377
    Likes (Received)
    6340

    Default

    I've seen this done since the 70-80s.
    Don't know if I've seen a Motoman but have seen many.
    Once upon a time you programmed all robots in G-code.

    The problem is defection out on the end. Robots are not "stiff" like machine tools.
    Foam is a big plus over trying to cut steel.
    The thing with robots is the fact that everything rotates and then you get errors due to distance away from the pivot point. (witness Faro arms vs CMM machines)

    For sure what you state can be done If the tolerance fits the robot.

  8. Likes powerglider liked this post
  9. #6
    Join Date
    Feb 2005
    Location
    Akron, OH
    Posts
    1,824
    Post Thanks / Like
    Likes (Given)
    277
    Likes (Received)
    1367

    Default

    Most 6 axis robots have good repeatability (order of 20-50 microns) but absolutely garbage accuracy (order of 2000 microns) which makes doing it... difficult.

    This sort of thing will get you there. Robot calibration - RoboDK

    But if you're just trying to carve a dog sized piece of foam, why not just use a CNC router?

  10. Likes powerglider liked this post
  11. #7
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    7,591
    Post Thanks / Like
    Likes (Given)
    377
    Likes (Received)
    6340

    Default

    Quote Originally Posted by Comatose View Post

    But if you're just trying to carve a dog sized piece of foam, why not just use a CNC router?
    How cheap or common are 5 and plus axis cnc routers this size?
    A robot can get around the back side a router never will and into spots even a 5 axis machine can not.
    The "spindle in a robot hand" has been the elusive dream of so many.
    It seems like it should be so doable and easy. Then...the devils in the details.

    With a mapped robot it's not such a crazy idea in foam since deflection won't be a deal killer.
    I think the software pre-processor would have to be very custom and not for the light hearted.
    Bob

  12. #8
    Join Date
    Feb 2005
    Location
    Akron, OH
    Posts
    1,824
    Post Thanks / Like
    Likes (Given)
    277
    Likes (Received)
    1367

    Default

    A... lot more common and cheaper than what's being discussed here?

    I doubt you're going to get meaningfully better performance than a shopbot 5 axis (read: real-damn-bad-but-not-useless) that costs $45k new ShopBot 5 Axis Tools

    If someone can get equivalent performance out of a 6 axis robot (all new parts, and I'll even allow them a rental on the laser tracker) for less money I'd hire them on the spot.

  13. #9
    Join Date
    Apr 2018
    Country
    UNITED KINGDOM
    Posts
    1,930
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    892

    Default

    Quote Originally Posted by Comatose View Post
    I doubt you're going to get meaningfully better performance than a shopbot 5 axis (read: real-damn-bad-but-not-useless) that costs $45k new ShopBot 5 Axis Tools
    We sold a Forest-Line 5-axis gantry that shoulda been better. A lot better And it wasn't that expensive ...

    If someone can get equivalent performance out of a 6 axis robot (all new parts, and I'll even allow them a rental on the laser tracker) for less money I'd hire them on the spot.
    Whatever happened to the spiderman things ? Almost infinite axes on three sets of ballscrews ? I thought those were the coolest, can't the highly-advanced software keep up with the geometry ? I forget the name but Giddings & Lewis built one, along with one or two other places.

  14. #10
    Join Date
    Feb 2014
    Location
    Sunny South West Florida, USA
    Posts
    2,679
    Post Thanks / Like
    Likes (Given)
    10284
    Likes (Received)
    3107

    Default

    Quote Originally Posted by EmanuelGoldstein View Post
    We sold a Forest-Line 5-axis gantry that shoulda been better. A lot better And it wasn't that expensive ...


    Whatever happened to the spiderman things ? Almost infinite axes on three sets of ballscrews ? I thought those were the coolest, can't the highly-advanced software keep up with the geometry ? I forget the name but Giddings & Lewis built one, along with one or two other places.

    You talking about the Hexapod Mills?

    Yeah, they're neat to look at.

    YouTube

  15. #11
    Join Date
    Apr 2018
    Country
    UNITED KINGDOM
    Posts
    1,930
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    892

    Default

    Quote Originally Posted by TeachMePlease View Post
    You talking about the Hexapod Mills?
    Yes. Seems like portions of my memory are going south

    They are neat. I wonder why they have never caught on ?

  16. #12
    Join Date
    Jul 2006
    Location
    Hillsboro, New Hampshire
    Posts
    7,173
    Post Thanks / Like
    Likes (Given)
    1675
    Likes (Received)
    4931

    Default

    Quote Originally Posted by EmanuelGoldstein View Post

    They are neat. I wonder why they have never caught on ?
    I think the basic design sees a lot of use in pick-and-place robots (light duty), as well as in motion platforms (low precision requirements). But for machine tool use it may be that the relative limited positional envelope coupled with programming issues makes it less attractive.

  17. #13
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Tennessee
    Posts
    445
    Post Thanks / Like
    Likes (Given)
    139
    Likes (Received)
    116

    Default

    Panasonic or fanuc had one at fabtech a few years ago pushing a form drill. It looked like a giant snake with overkill amount of axis so it could drill/tap a hole on the other side of a car door, where it went thru the window hole.


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
  •