I am looking to build a new front triangle for my bike. I don’t have a proper workshop so my thought is that I could order 3d-print/CNC parts and somehow glue these together with aluminum pipes. I would do a full design in fusion360 and somehow order parts by exporting from there.
Basically I wonder how viable a project like this is? Is it super expensive, will it break easily, is it hard to get the tolerances right? Any general advice? Should I go another direction?
Here is a picture of the current bike which I want a new front triangle for:
Lotus sports cars use (or used to use) a bonded aluminium chassis (Elise, Exige, VX220, Tesla Roadster et al). It’s certainly capable of being strong enough. For a home builder it’s probably quite simple, once you’ve figured out the key parts: specifically, the right bonding agent, material prep, tolerances, and fixturing.
I worked at Lotus for a while, not bonding chassis myself but I did bonding of other parts (GRP, plastics) and worked alongside guys working on carbon fibre chassis for Aston Martins. It’s a pretty simple process really - we had custom fixtures to hold everything just so, you clean and prime the surfaces, squirt on the glue (usually a 2-part epoxy but we had several different types for different jobs), clamp it in place and wait the allotted time. I was young and didn’t know much at the time, but it’s a very exacting process; we had to squeeze the glue on in just the right way so there was the correct amount, even going so far as having a little jig to cut the nozzle tips exactly right (to be fair this was mostly down to cost and meeting ISO cert, not because otherwise the joint would have failed; in your case I’d just overdo and clean off the excess).
You’ll need to find the proper surface prep and bonding agent, and design parts with enough surface area and correct bonding gap. I doubt a direct tube-to-tube set-up will work, but a lugged joint should work fine.
You should look through this build HERE for some inspiration. Eva built a killer gravel bike with 3D printed portions and carbon tubes bonded in-between them. Not the exact same materials, but the process is the same.
They did quite a bit of research into this project, and its all quite fascinating.
That thread was very informative! Even had prices and everything. Definitely looks doable.
When saying 3d printing in that thread, is it actually CNCd? Or is it actually possible to 3d print metal?
I was thinking about buying a plastic 3d printer to print my joints at home for fitting trial and error, before I send everything to be printed for real, increasing the likelihood of getting things right the first time. Makes sense or not?
Many details to figure out. In that thread he/she used some kind of oven to treat the glue. I dont have access to such things. Will need to learn about glues
I wonder if ordering 3d printed joints with sockets for bearings will be precise enough? I guess it worked out in that thread with the headset and BB
If you haven’t done any frame building before, it is difficult to read between the lines with the work that others do. I haven’t done a lot, and I have very few tools. I have to rely on other frame builders for certain tools. It’s likely that any 3D printed metal part will require some treatment to work well with bearings, threads, etc. THIS THREAD has a lot of information on 3D printing metal parts. Most frame builders will use services that print in Stainless Steel or Titanium, though there are other materials, these are just the easiest for smaller outfits like frame builders to use for bike frames.
I learned now that 3d printing of metal is actually a thing. I wonder if the tolerances are better with actual CNC? My plan is to use aluminum since its familiar and I like it, if there is no strong reason to use something else
Welcome to 2024 framebuilding! 3D printing lug tolerances will be worse than CNC. However, they would be good enough to build a one-off frame. Printed lugs will also be cheaper than CNC, unless you have access to a CNC machine.
Gluing a bike together is not a bad idea at all. In fact, I think you could argue that gluing a bike together is stronger than welding a bike. Welding creates stress risers and distorts the frame.
I am very curious about 3D printing aluminum alloys. I know some are specifically designed for that application. I just never had the opportunity to experiment with it. We know titanium works well, but expect to pay $3/g (a titanium lugset would probably cost $1000).
For more inspiration, check out Kova Bike’s thread. They made a killer CNC machined lugged bonded frame:
Do you think 3d printed pieces (titanium or aluminum) would have good enough tolerances for bearings and headset? Or what kind of post processing would I need to do?
Do you have any comment on the idea to have a plastic 3D printer for prototyping? I like the idea of being able to build a cheaper replica of the frame before spending that $1000
That Kova bike thread is insane, very impressive. Very cool to have found this forum
I would recommend consuming as much content as possible on Atherton Bikes (previously Robot Bike Co). They have been making 3D printed titanium lug / carbon tube bonded MTB frames for many years and just came out with a CNC alu lug / alu tube bonded MTB.
Not 3D printed or Alu but a builder working in bonded tube to tube is Framework Bicycles.
You can find some info on their site, Instagram and also some articles but the 3hr interview they did on the SUABB Podcast is the one I can recommend. Dives deep into tolerances and process stuff, including some great info on glueing up a frame. With a socket/lug style build you’ll especially need to consider fixturing since you kinda need to push the whole frame together at once which is totally different to welding. The bonus seems to be there is a small amount of self fixturing that will occur.
Framebuilders will ream and face headtubes after welding. 3D printing should be no different. I think the crux of the problem is the bottom bracket junction. It is a big, expensive print, and I don’t know how well the threads behave.
Definately print it in plastic first! Printing a lugset takes about a month, so you can waste a lot of time if you overlook something and have to re-print.
I 3D print things to test fit and visualize all the time:
I see, I can probably find some old local machine shop to that About the bottom bracket I think I can either get away without threads or thread it afterwards
On aluminium printing, I’m just dipping my toe into this too…recently had these small parts made for a racing wheelchair contract. 3.25 × 6.72 × 2.24 cm
Hollow parts, internal webs not lattice, with walls of the main body at 1.5mm, and down to 0.5mm on the feathered edge. Quoted tolerance was +/-0.1mm (100 microns). None of my model needs anything tighter than that. Repeatability of dimensions seems very good, with little variance at all.
Pricing for these was $8.10 USD each @75 units (May 2024), but numbers made only a small difference in cost. They weigh 14g each, so effectively $0.60 per gram.
Alloy was AlSi10Mg. Aluminium_AlSi10Mg_-Data-sheet.pdf (50.7 KB).
They were post-print glass bead blasted to give a good smoother finish. Finish is still slightly rough, so good for bonding or painting, but might need some more work to be commercially presentable bare.
4 drain holes to let the powder out and the central hole thread was modelled for the print and then tapped out M4. This face was bonded onto a deep section CF rim, so left bare, the rest was acrylic metal primer and painted.
Possible, but I’m certainly not ready to give it a try yet.
Quoting the supplier’s data sheet, then the properties are reasonable compared to 6061-T6, but let down a little by the lower Yield. There doesn’t appear to me much XY directional difference in properties, which is good.
Well behind compared with 7075-T6 billet which is frequently used for CNC machined cantilevered parts on bikes, though.
Metron use the Scalmalloy material which is a scandium doped alloy, for their printed stem and presumably also Ganna’s hour record bike.
IIRC, Trek was doing lugged and bonded aluminum bike frames in the early 90s as a lower cost option to their carbon frames using the same construction method. My buddy rode one for years without any issues. I think his younger brother is still riding it.
As mentioned before, this allows the home builder to work around the issues welded aluminum frames have, like the need for heat treatment on most alloys. I think it’s a very cool idea.
About the bottom bracket I think I can either get away without threads or thread it afterwards
