Open Sourced Carbon Fiber Rod Ends

Modellers and makers who have been around the block for a few decades generally have their preferred materials. Balsa wood, sheet metal, brass tube… these were all staples of the hobbyist workshop. Composites are very much the new kid on the block and are starting to gain more of a foothold in the hobby marketplace. [Anthony] has been experimenting in this area, and has created some useful attachments for carbon fiber tubing.

The fittings are designed to be lasercut from aluminium or 3D printed. The rod ends are a simple two-piece design that slots together, before insertion into the carbon fiber rod. [Anthony] shows off a series of rods being used as linkages with a stepper motor, before performing pull-out tests on the links. Installed with cyanoacrylate glue, the link holds up to a pull load in excess of 180 lbs. The strength is impressive, and [Anthony] also talks about how to install the appropriate bearings to use the links for motion projects.

Overall, these links will likely prove useful to anyone using carbon fiber rods in a build, and helpfully, the required files are all available on GitHub. The source material is now cheap and readily available online, and is strong and resilient when used properly. We’ve seen carbon fiber popping up in a lot more projects recently, too. Video after the break.


18 thoughts on “Open Sourced Carbon Fiber Rod Ends

    1. Anodized aluminum won’t be a problem as it is already an aluminum oxide. and in this case, I won’t be concern with gap in the protective layer…. as it is glued at the end… so no rubbing that can break it!! .

    2. Yes but using aluminum and composites together is so common. Your linked article even mentions ways to reduce the corrosive effect. The company I work for uses the fiberglass layer between the aluminum and composite part method, and the corrosion is negligible for the life of the product.

    3. I am more worried about using a square profile inside of a round tube, it seems like he is really asking for stress fractures… One could argue that you can compensate with heavier wall tube, but at that point why not just make the whole thing out of aluminum–carbon fiber is only twice as stiff as aluminum and aluminum is much more resilient against fracture. Plus can you safely drill holes it in rather than relying on glue….
      Speaking of, CA is just about the worst possible choice of glue for a system like this, it is extremely brittle and not itself mechanically robust (it bonds to CF and Al well, but the cured glue itself is weak and it easily breaks leaving glue on both sides of the joint) so if the system sees any significant shock loading I expect the rod ends will start popping off…

      Using a loaded epoxy (ex, JB weld or Hysol 1C) would solve the joint cracking problem, and if a suitable fillet were made at the rod ends would help mitigate the stress fracture risk. I have never tried to bond CF with either of those adhesives but with proper surface treatment I think it should hold better than CA does.

      1. jrfl – I’m using 5052 aluminum which is much softer than 6061. That sharp profile on the aluminum immediately rounds over when pressed the tube. I couldn’t remove the rod end that was inserted with CA glue. I couldn’t pull or twist it out with the equipment I have. I’m sure JB weld would work better but for 99% of hobby projects, it will not matter much what type of glue you use. If I were hanging babies from these or something else critical, I would drill a small hole all the way through the rod after the glue has cured and insert a spring pin.

    4. There is such a minimal amount contact between the aluminum and the carbon fiber that this will not be an issue unless you are using it in salt water. Even then, only the sharp corners are making contact. Once glued in place, the mating piece is essentially air and water tight. The article even suggests using epoxy to mitigate the corrosion. You can always make these out of acrylic, other plastics or even carbon fiber.

  1. I stopped the video when he demolished his CF tube with pliers in an attempt to “grab” it.
    Before that I frowned my brow on the first pull test.
    A decent connecton to 6x8mm CF tube should be able to withstand a pull test of probably some double digit kN. Way more than his small scale is able to measure.
    I’d be impressed if he made a connection that was strong enough to pull the tube in 2 instead of being pulled out of the tube by hand.

    1. That scale can measure up to 300kg/600lbs. I was not expecting the connections to be that strong so my setup was clearly not prepared. I would like to build a hanging set up for testing things like this in the future. I think you’re missing the point of my video. Most people do not need connections strong enough to pull CF tubes in 2. I just designed an easy way to make customizable rod ends that are plenty strong for hobby projects/robotics.

  2. If you’re on a budget and can handle a taper in your design… thrift stores have carbon fiber and even titanium reinforced clubs where those salvaged materials can be hacked into other items for a dollar or two at some Goodwill Stores I’ve found. Same goes for ski poles sometimes made from aluminum.

  3. I see the issue here is “glue” the Alumunium and the CF, how about 3d waving cf or creating the mold replacing the Alumunium ?
    it become 100% CF ?
    or to complex and expensive to do ?

    I don’t see much CF 3D Waving for hobbiest around.

  4. 3 problems with your design:
    1) bonding surface must be maximized – an aluminium tube insert would do.
    2) bonding sufferance finish does impact the bond strength
    3) adhesive thickness does severely impact the bond strength (

    Should you address the 3 problems, you would be looking into 10-fold increase in bond strength. In the end the aluminium bolt slot should break, not the CFRP tube or the bond.

  5. What I noticed about the test: when you pull *on the rod* with a vise-grip, the rod will be deformed (flattened) tremendously. Thus the epoxy between aluminum and carbon fiber might have been loosened.
    Why did you not pull with the vise-grip on the aluminum insert at the *other* end and use the same glue on both ends?
    One rod with epoxy and one with super glue, like you did with the non-glued version.
    Then you can also use a safer ‘pulling mechanism’, like a screw or a car jack, where pieces of metal cannot flynin your face.

    Another weird thing: you have a newton meter which goes to 3000N, but the supplied hook is so weak it bends halfway! ????
    Perhaps use a sturdy bolt & nut next time.

    Thanks for the designs and video!

  6. So two things, instead of machining square to fit in the round tube, use a hole saw to make a round end, cut material off on mill, then machine z slots to fill with carbon fiber and aluminum compatible glue p200 is a two part adhesive. This prevents corrosion and uses everything as it’s intended.the reason for the z slots is for the glue to have more surface area to bond like trying to pull a key out of a lock when it’s turned in the lock.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.