First frame build-along: review my thinking on carbon frame

Hey guys,
I’m a software engineering guy from Belgrade, and I love cycling. I decided do make a carbon frame, and I want to do this as a hobby project. My goal is to make a functional, and safe frame. It does not have to be pretty, I am not optimizing for the cost, I am optimizing for a fun and learning experience.

I was actually inspired by @Sideways_Viola, @Lester and @Pol. This forum seems like a great place to do a build-along, and get my thoughts peer reviewed.

I chose carbon because I started learning about composite materials just for fun and I’m really enjoy the topic, so it seems like building a carbon frame is a natural cross section with my passion for cycling. And it honestly looks like carbon fiber has a lower barrier to enter than welding! That is the only constraint, I am willing to change anything else based on your guys feedback!

The bike’s mission: I commute to work every day, and on weekends I like do to like 50km or 100km loops, away from the city. It’s usually all paved, or cracked pavement like 95% of the time. But in the summers I like to shred the hills of green Šumadija where you can’t avoid segments of doubletrack (with huge grain size). So I want to do an All-Road geometry, with big tire clearance.

Current design:

  • I usually switch between 700x32 tires, and 700x47, so naturally I want to accommodate that bigger tire clearance with my new bike.
  • Rear axle - thru axle 12x142, as I already have a wheelset
  • BB86 - I want to go with pressfit, in order to minimize the need for carbon to metal bonding, as every carbon-to-metal bond is a possible failure point. I want to go with bb86 specifically, because it’s kinda wide, which I believe gives me more options when designing the chainstays.
  • Headtube - ZS44/28.6 | EC44/40 - So a straight tube 44mm, again press in cups because it seems safer. Also it’s a straight tube as it seems easiest to produce, and I have a fork ready as well.
  • Integrated seatpost - I wanna go with an integrated seatpost because I guess it’s one less worry, getting the inner diameter of the tube right. I am aware that seat mast topper offering is slim, but I guess something can be hacked up here. I actually skimmed through some forum posts saying this is possible.

I managed to source some cheap tubes from m-carbo.

Current plan: Initially I wanted to go down the lugged frame approach, where I would somehow create carbon lugs and then bond the tubes to the lugs, however based on some feedback I decided to switch focus to mitered tube-to-tube design, with roll wrapped lugs. So the general steps are creating a jig, mitering the tubes, creating bb and headtube cylinder, bonding tubes and those parts, then wrapping. Because of tire clearance I will have to do a bit more complex chainstays, for this I am considering doing a foam core, or 3d printing with abs, and then wrapping carbon around, and then dissolving the core in acetone.

Before I start building the frame I want to practice the wet layup on something simpler, so I decided to try to make a carbon mudguard for my 47 tires.

Current progress:

  • I designed most of the bike in onshape. I still need to design the dropouts, and I actually need to think more about how I will do the entire rear triangle.
  • I bought the jig from Simple Bicycle Framebuilding Fixture by ideas2cycles
  • I’m waiting on some aluminium extrusion which I will need for the jig.
  • I’m waiting for carbon tubing from m-carbo to arrive.
  • I sourced some epoxy, mold release.
  • I bought some carbon, and ancillaries from easycomposites.
  • I have to pick up some 3d prints of a mold I designed for the mudguards.
  • I started contacting some cnc shops to see who can help me make precise metal mandrel cylinders for BB shell and Headtube.

Basically I am still buying stuff and planning ahead, trying to decide as much as I can upfront.

I just wanted to kick off this intro post today, I will followup with more thoughts and questions, as I like thinking about all the details of this project, and for every step of the way so far I have so many things I would like to double check.

4 Likes

Welcome! Great CAD work so far. I agree that carbon has a lower barrier than metal bikes.

If you have some time, check out Ride July. Ben documents his work very well. He uses mostly tube-to-tube mitered with overwrap construction to build his bikes. He’s also a professor so he gives very clear instructions and explanations:

4 Likes

I was thinking the same thing, having seen this presentation at the Philly Bike Expo. It’s really interesting.

Ford

2 Likes

Thank you for you reply @Daniel_Y! I actually learned a lot from this video, it filled out some knowledge gaps for me.

Many things happened in last two weeks. I will do a bigger write up when I have some more time. Let’s go through most important items:

I revised the bike design a bit. Dropped BB to 80mil, elongated the chain stays.
I designed my first version of the dropouts. I’m generally happy with how they look, except I was struggling a bit with filleting/face blending. I designed the dropouts for the UDH standard, 174mm axle length, for 142 OLD, and for 160 mil rotors. I am amazed with documentation for UDH, even though it was a pain to design symmetrical dropouts. Is there even an alternative to UDH for my use case?

I consider this v1.0 of the bike design, and I will work from here. Here is the onshape link if anyone is interested!

I laminated my first part!!

It’s a segment of a bike fender. I learned a lot. I am happy with the results, but for the bike I will need to do better. I think I will have to do vacuum bagging of the wrapped joints in order to get the compression I need.

Currently I am thinking of doing clamshell mould and forged carbon for dropouts, chainstay will be most complicated, with 3d printed inner core, vacuum bagged outer part. For the BB and HT I am considering doing prepreg roll wrapped around a steel mandrel.

4 Likes

All seems good :slight_smile: I have considerable experience making carbon lugs and its not really worth in most cases, tube to tube is a way more sane approach.
For the chainstays I would consider making a 3D printed outer mold and then eg. making a bladder using an inner tube (here is an example i found) - dont recommend doing a soluble core, i dont know anyone who had a great time with it.