How To: Assemble TryCatchThrow@‘s replacement TMBR axle


I’m @TryCatchThrow and I’ve been getting into the world of 3D printing and modeling. A project I’ve been working on for the last day or two, inspired by @GnarlyCharlie and @Pun1sh3R, is to create a suitable replacement for a TMBR axle, as they are no longer being produced.

Here is a picture of the component we intend to replace:

I was writing up a Google Doc to print out and include as directions for how to assemble what I’ve got here, but I figured I would drop it on the forums so it’s more freely available.

Here is the kit as it currently stands:

The kit contains:
2x axle adapters
2x caps
1x gap shim

You will need to provide your own 8mm dowel axle cut to 18mm length. You will also need some sort of glue (I use JB Weld 2 part epoxy), sandpaper or a file, and a clamp.


  1. Prepare axle. Cut to roughly 18mm, sand if you choose to do so. I personally cut the axle to 18.1mm and sand down to the right length.
  2. Test for fit. If you press the adapters on the ends of the axle, put the shim over the gap, you should not see any space in between (pic below means the axle is still long.

  1. At this point, file or sand down the axle a little at a time until you get a good tight fit when you press together with your fingers, and the shim is difficult to pull out.

  1. Glue. A little goes a long way, but I do try to get a small amount on both the inner walls and the bottom of the axle holes.

  2. Clamp. Give it ample time to dry, as this is going to be the thing holding your yoyo together. I let the JB Weld two part epoxy cure for 18ish hours.

  1. Remove shim. Axle should look like this:

  1. Install. When installing an axle in a TMBR, I like to install the axle first into one half just a little loose of where I want it, then install the cap, then tighten from the axle, then repeat on the other side. You may use the original wooden caps if you would prefer; mine are provided for convenience.

I’m currently sending out some samples to a few folks to get some feedback. If all goes well, I may wind up making a run of these fully assembled to sell. Stay tuned!


Heck yeah! Cant wait to test it out and compare!

1 Like

Awesome thread

I made some similar a couple years ago. I found the flat head screwdriver slot to be a little weak and easily damaged (just like the wood ones). If you are having an issue with durability maybe try a Torx pattern instead. Good job @TryCatchThrow


This looks so awesome! Great work, Keith!!!


Thanks! I’ve been thinking about improvements like changing the slot to a Philips head or even bumping out a hex head on top of the cap. Torx is a great idea; I think it would be less likely to strip the plastic.

For the first revision I decided to keep it to the original form (which I agree were a bit weak if you used a smaller than slot sized bit). I have been mostly hand tightening TMBRs with either an original or replacement axle and haven’t been having any issues yet. The only time I’ve had an issue was the first time I tried to open my first TMBR :grin:. I could also easily print a key for this one…


Now that you have the axle design figured out and printed, have you ever considered 3D printing a mock TMBR? Or using the axle system for your own designs?


I have not gotten that far. I’m still mostly designing fixed axles with glued in wooden axles. I do feel that a metal axle capped with nuts and a wooden axle sleeve is probably a more reliable system and would give better opportunities to experiment with response. That said I don’t have a source for axle sleeves, so I guess maybe?


Ohh, like a no-jive system? That would be very cool as well.

1 Like

This is amazing! I’ve been reluctant to throw or even open some of my tmbr throws because I don’t have any spare axles in case they break.

I’d be very interested in picking up a couple of these if you ever sell any.


Yeah this is awesome :ok_hand:, I also appreciate the use of a wooden axle as that will be a big part of play feel. It would also be cool for a bearing option as well.

1 Like

I do not immediately know how a bearing would work but I guess I could think about it a little…

1 Like

This was a cool conversation and project to watch.

Total time from inception to testing was what 48 hours?


About that. I think there’s still more testing and iteration to do yet, but that might be the perfectionist in me speaking.