Posts

Yesterday I shared the very first batch of images and the project outline online, and honestly, the warm feedback from the community felt great. Along with that came a mountain of critique—some of it completely fair, some unintentionally hilarious, and some clearly driven by mild panic. And yes, posting a short riding clip without a helmet was… not my finest moment. So I went through every comment and suggestion, and with a bit of help from GPT I pulled everything together into the summary you’ll find below this post. ...

Developing a bicycle frame from scratch is one thing — getting it officially certified for real-world use is another. Anyone who’s ever looked into ISO certification for bicycles knows how complex (and expensive) that process is. The exact standard depends on the category of bike — for example, ISO 4210 applies to safety and performance requirements for road and mountain bikes — and obtaining approval involves rigorous fatigue, impact, and assembly tests. ...

With the first frame under my belt, I can finally redesign the lugs with confidence. The head lug turned out best of the bunch, so I’m using that approach as the template for the rest. That means: surface modeling all the way. If this pans out, it’s also the route I’ll take for the online configurator. I still see a future where anyone can easily configure their own lug, set the dimensions, and download it—no hurdles, no gatekeeping. ...

It’s been a while since I last wrote an update. I kind of lost track of the devlog for a bit, but the good news is that I now have plenty to talk about. Last week was mostly about research and printing. I redesigned the head lug using surface modeling, which not only calculates faster but is also far smoother to work with. I’ll probably look into adding this method to the WebGL configurator at some point, but that’s something for a later chapter, which, for now, feels a bit far away. This week, I printed all the parts and everything is now ready for assembly. I also spent quite a bit of time on social media. Ever since creating openframe.cc, I slipped into a bit of a doomscrolling phase. Still, I don’t really consider it wasted time. On the contrary — I came across a ton of inspiring content, from branding and marketing ideas to video editing tricks, bike projects, and 3D printing innovations. There are so many creative people out there; it’s truly impressive. With all the parts printed, the wood cut to length, everything gathered in the workshop, the right epoxy on hand, and the downtube already laser-branded, there’s really no excuse left not to assemble the bike. Before doing so, I brought my small scale to the workshop and found that the first frame version weighs less than 2.5 kg. That was a pleasant surprise. I also included the 51 grams of the FSA adapter, since the frame can’t really be used without it. Another nice discovery is that the total of all 3D-printed parts doesn’t even exceed a single spool of filament. Altogether, it’s about 686 grams, and if I add roughly 10% extra for skirts, brim, and support, which I try to avoid, the total comes to 754.6 grams. That means the weight is quite reasonable. I didn’t expect it to be much heavier, but it’s not ultralight either. ...

It’s a great day—because we’re really getting close to a finished product, and I couldn’t be more excited about it. The frame feels surprisingly solid, even though it’s just PLA at this stage. To check the fit, I quickly assembled everything, and so far, it looks right. That’s a good sign. What’s Next on the Design Side Back brake holder → Still needs to be designed. Seat post → Needs to be solid and sturdy. 3D scanning experiments → Trying out scanning for some tricky parts. Bamboo bars → Still experimenting with slightly smaller diameters. Luckily, the resin is very forgiving, so using a little extra material is no problem. Upcoming Plans High-resolution lug set → Printing a full set in PA12CF. Design isn’t final yet, since I’ll be making a lot of changes over the next week. Laser engraving → Testing on the bamboo tubes to see what kind of detail and look we can achieve. Head lug orientation → Current setup isn’t optimal, so adjustments are on the table. Color experiments → Thinking about giving one lug a different color. I’ll also be trying out more colors soon, since these will tie into the website configurator and the overall branding style. Every iteration gets us closer to the vision—and this week feels like a real step forward. ...

The end result should be perfect. Some problem can be solved by investing in better or new equipment. Others by doing more research and more testing. Today it’s mainly the latter. I noticed some quality flaws in the print in the first couple of layers. I am not sure what’s the cause of this yet. But I am eager to find out today. First I am going to do a speed calibration. Each material is unique, and PA12CF isn’t really the most easy material. I read a lot online about the printing speed of PA12CF, and that it should be slow… But how slow are we talking about. Do we need 30mm/s - can we go with 60, 120, 300 … this speed is very important. It’s the difference between waiting two days and 12 hours for a complete lug set, or just 10 - 15 hours. That’s the difference between 2-3 lugsets a week to 10-14 sets… (wishfull thinking here). ...

The first prints are here! Feels amazing. I’m really excited, but I also immediately learned a lesson: don’t go overboard with support material. Most of it wasn’t even needed, and taking it off was a pain. Let’s just say it’s not my favorite workout. Also, there’s now an official email address: hello@openframe.cc. Feel free to reach out if you have questions. With a proper address I can finally start setting up socials too, to share photos, videos, and timelapses (like the one below). ...

Alright, I was definitely getting ahead of myself again. Time to take a step back and keep things a little simpler – at least for now. I’ve got all the materials needed to build a bike… except the actual knowledge to build one. Which is kind of a key ingredient. Trying to set it up with a rear cassette was blocking my progress. Turns out it’s way more complicated than just drilling a hole in the dropout and bolting it on. Extra wiring, less space, questions about strength – suddenly a whole list of issues I wasn’t ready to tackle this early. ...

Loading 3D model… 📏 Measure ↺ ⛶ Please enable JavaScript to view the 3D model.

Today I pushed forward with the CAD design. Every step makes it more obvious why every single part needs to be modeled with exact dimensions. In the screenshot below you can spot the main issue: the back tire clashes with the frame. On top of that, the front gear has barely a millimeter of clearance from the frame – way too tight. Definitely need more tolerance there. Tomorrow I’ll continue with the wheels, the front fork, and the saddle. Stay tuned! ...