Shop OBEX P1 Docs P2 Docs Learn Events
What do you use to do mechanical prototyping for your robotics? — Parallax Forums

What do you use to do mechanical prototyping for your robotics?

Too_Many_ToolsToo_Many_Tools Posts: 765
edited 2014-09-07 21:11 in Robotics
In the past, I have used components from sets of Capsela, Legos, Tinkertoys, Erector and Meccano to prototype for small robotic projects.

The different variety of mechanical components allow for a large build set to draw from.

I also have a large collection of raw metal, plastics and wood that I can draw from to make custom components..which of course take longer to construct than just reaching for an ready made component.

So..what works for you for mechanical prototyping?

Comments

  • ercoerco Posts: 20,257
    edited 2013-11-05 07:21
    I prefer to scratch build, but I know I'm a dying breed. Most people like premade or very simple kits. Ask your local hobby shop how many balsa & tissue airplanes they sell vs ARF foamies. Kind of sad.

    Lego and Vex are popular with the school crowd, where build speed and uniformity is important. And 3D printing is taught in schools now, that's a game changer.
  • mindrobotsmindrobots Posts: 6,506
    edited 2013-11-05 08:13
    erco wrote: »
    I prefer to scratch build, but I know I'm a dying breed. Most people like premade or very simple kits. Ask your local hobby shop how many balsa & tissue airplanes they sell vs ARF foamies. Kind of sad.

    Lego and Vex are popular with the school crowd, where build speed and uniformity is important. And 3D printing is taught in schools now, that's a game changer.

    I bet if you could find Lego and Vex parts made from nomex, high temperature tungsten and high impact carbon fiber, you'd be using more of those! :smile:

    Depends what I'm doing at the time. Pre-made robots (S2), kits, baltic birch plywood, foam board, plastruct and metal tubing from the hobby shop, long screws and rods....I haven't figured out how to incorporate stained glass into a robot yet but given enough time.......
  • NWCCTVNWCCTV Posts: 3,629
    edited 2013-11-05 14:08
    I use whatever I have lying around that I feel would work for what I am trying to accomplish. Metal, foam, wood, plastic, etc. Once I have something that works I try to stick with plastic or aluminum, depending on the application.
  • WBA ConsultingWBA Consulting Posts: 2,934
    edited 2013-11-05 14:32
    I am with Erco. I don't really prototype much in my projects. I either build from scratch to a completed item or start my "prototype" with what will be the finished item. In either case, it's usually due to the amount of time I have to get it done.
  • Duane DegnDuane Degn Posts: 10,588
    edited 2013-11-05 14:45
    I used foam board a few times to make sure I have the shape of a structure correct before I commit to build it with some other material.

    My recent Vex tread robot has one tread module made from expanded PVC but before I built it, I first made one from foam board. The foam board module will be replaced with an expanded PVC module.
  • Martin_HMartin_H Posts: 4,051
    edited 2013-11-05 16:38
    My last few robots (4x4, drawbot, and tiny arm) have been plywood. I'm working on a delta robot that is a mix of metal, plywood, and printed plastic parts. In the pipeline is a scratch build from EPS.
  • Too_Many_ToolsToo_Many_Tools Posts: 765
    edited 2013-11-06 15:12
    erco wrote: »
    I prefer to scratch build, but I know I'm a dying breed. Most people like premade or very simple kits. Ask your local hobby shop how many balsa & tissue airplanes they sell vs ARF foamies. Kind of sad.

    Lego and Vex are popular with the school crowd, where build speed and uniformity is important. And 3D printing is taught in schools now, that's a game changer.

    Scratch building works well but takes time.

    I find a combination of custom and premade parts work best for my prototyping.

    And sometimes the prototype components end up in the finished robot..okay..who am I kidding..a REAL hobbyist robot is never finished. ;<)
  • Too_Many_ToolsToo_Many_Tools Posts: 765
    edited 2013-11-06 15:14
    I find that hot glue and duct tape work best for attaching parts..with velcro working best for heavier components.

    Oops..forgot about Super Glue...how could I forget that!
  • Too_Many_ToolsToo_Many_Tools Posts: 765
    edited 2013-11-06 15:17
    I am curious...what "stuff" do you keep on hand for building...screws, bolts, etc.?
  • ajwardajward Posts: 1,130
    edited 2013-11-06 18:32
    Mostly I scratch build my odd 'bots and a lot of designs use very simple shapes... squares, rectangles and circles. If I have a complex shape or or parts that have to interlock, I'll plot a full size drawing in AutoCAD and trace onto the material.

    =Usually= everything is bolted together, but I keep an assortment of adhesives on hand for things that are going to be permanently connected. I also keep Velcro strips and patches on hand.

    "Stuff"... beyond that mentioned above:
    Sheets of various kinds of plastic from Tap Plastics scrap bin
    Aluminum stock... flat strips, angles, channels, sheets, spacers
    Brass stock... flat strips, angles, telescoping tubing, threaded rod
    Steel stock... plain rod (for axles)
    Plastic stock... flat strips, angles, channels, plain rod, spacers, leftovers from Tamiya kits
    Wood stock... various sized dowels
    Hardware... various Parallax aluminum L/Universal brackets, 2-56 thru 1/4-20 screws, nuts and washers (mostly 3mm), various self tapping/wood screws
  • mick444mick444 Posts: 1
    edited 2014-08-31 13:12
    For years I used cardboard and duck tape. Eventually I wanted things to look more professional so I tried out all the various building kits like legos, k'nex, and erector set and after some finagling I could get my prototypes to look somewhat presentable. Most importantly, I would spray paint the final assemblies so they would have a consistent color, like a krylon plastic spray paint from home depot.

    At work, we typically use T-Slotted aluminum since we have the budget for it and we have a small shop with all the required cutting and deburring tools.

    As a mechanical engineer with 10+ years of design experience I decided to create a simple building kit that had a consistent professional look like T-Slotted aluminum but didn't require extra tools and wouldn't break the bank. I'd love to get some feedback from the prototypers here on if they think it would make prototyping easier. www.gridlinksinc.com
  • LeonLeon Posts: 7,620
    edited 2014-09-01 06:07
    This is a nice, but rather expensive system:

    http://www.vslot-europe.com/index.php
  • prof_brainoprof_braino Posts: 4,313
    edited 2014-09-01 06:59
    I started using pink foam insulation from the home center. A 4x8 foot sheet costs a couple bucks, and can 0.5 to 2 inches thick. Cuts with a utility knife or saw. Drywall screws can be screwed in by hand. Slots can be cut using threads from alag bolt. Doesn't burn well unless you are really determined. If a section fails, I laminate on craft sticks (popsicle sticks) with expoxy cross the crack until it becomes wooden dibond.

    I also use drinking staws. If one straw is not rigid enough, I slit a second straw, andslip this over the first. More layers are added until sufficent strength is attained. seven layers is just under seven times stronger than 1 layer (who'da thunk?). I 3D print a socket end, and touch it with a hot soldering iron to spot weld it to the end.

    Not plywood, but may approach erco-cheap.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2014-09-01 10:31
    mick444 wrote: »
    As a mechanical engineer with 10+ years of design experience I decided to create a simple building kit that had a consistent professional look like T-Slotted aluminum but didn't require extra tools and wouldn't break the bank. I'd love to get some feedback from the prototypers here on if they think it would make prototyping easier. www.gridlinksinc.com

    Interesting product, not interesting web site. I stayed about seven seconds. The concept seems a good one, but what's more important are pictures (and/or videos) showing REAL designs, not just 3D renders of things that maybe could work.
  • photomankcphotomankc Posts: 943
    edited 2014-09-02 08:18
    GordonMcComb wrote: »
    Interesting product, not interesting web site. I stayed about seven seconds. The concept seems a good one, but what's more important are pictures (and/or videos) showing REAL designs, not just 3D renders of things that maybe could work.

    Agreed. I see what they are going for but that site needs some clear examples of real-world chassis and mechanisms created from it.


    Also in those systems the hardware weight of the screws can become excessive.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2014-09-02 09:18
    If it's a shortage of funds to pump out enough parts to make the physical prototypes, this is what KickStarter is all about. Create enough pieces to build something for the video, and hold out the hat for more $.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-09-04 11:50
    I am seriously considering buy at least a pair of Dynamixel servo motors for my next project.
    At bit expensive, but the are very sophisticated and rugged.

    http://blog.trossenrobotics.com/tag/dynamixel/
  • Duane DegnDuane Degn Posts: 10,588
    edited 2014-09-04 12:57
    Loopy Byteloose wrote: »
    I am seriously considering buy at least a pair of Dynamixel servo motors for my next project.
    At bit expensive, but the are very sophisticated and rugged.

    http://blog.trossenrobotics.com/tag/dynamixel/

    Dynamixels are really nice. My one complaint is they use a pot for position feedback like other servos. These pots will wear out with a lot of use. If you use your Dynamixel in something that you want to move a lot of the time, you'll likely wear out the pot before too long.

    I keep trying to find a better way to sense position than using a pot. My experiment adding a magnetic encoder to a servo didn't work very well buy adding an optical encoder seems promising. The problem with the optical encoder is the lack of precision. I'm only getting 378 transitions per rotation which is a long way from the 10-bit precision of the Dynamixels.
  • Too_Many_ToolsToo_Many_Tools Posts: 765
    edited 2014-09-07 21:11
    prof_braino wrote: »
    I started using pink foam insulation from the home center. A 4x8 foot sheet costs a couple bucks, and can 0.5 to 2 inches thick. Cuts with a utility knife or saw. Drywall screws can be screwed in by hand. Slots can be cut using threads from alag bolt. Doesn't burn well unless you are really determined. If a section fails, I laminate on craft sticks (popsicle sticks) with expoxy cross the crack until it becomes wooden dibond.

    I also use drinking staws. If one straw is not rigid enough, I slit a second straw, andslip this over the first. More layers are added until sufficent strength is attained. seven layers is just under seven times stronger than 1 layer (who'da thunk?). I 3D print a socket end, and touch it with a hot soldering iron to spot weld it to the end.

    Not plywood, but may approach erco-cheap.

    Good ideas....need to add it to my component mix.

    Thanks for posting it
Sign In or Register to comment.