I couldn’t find any actual examples of a framebuilder using a setup like this, but they seem quite easy to make and could be hugely timesaving if you could completely process your tubes this way.
Nice playlist. I’m always amazed at what comes out of people’s garages. I caught myself staring at a 15min long CNC cutter video haha.
I think on the Cobra Poddy, Corey at Kruch Customs mentioned he laser cut his tubes in batches on his production bike? Laser = plasma? Maybe someone can chime in on this.
I am always surprised by how time-consuming the cleaning of the tube is. You might lose the time gained by cleaning the slag off the tube. That being said, a CNC, bike-specific tube notcher would be pretty rad. Or even just a CNC bike tube cleaner haha.
Does anyone know how they mass-produce frames in Asia? Do they have any CNC tube-cutting machines?
That sounds super cool. It’s not the same medium, but in the realm of alternative tube mitering I saw grizzly makes an abrasive drum tube mitering machine and was wondering if anyone had any experience or thoughts on those?
The homebuilt machines really show how simple it is.
Apparently, the challenge getting a clean plasma cut on thin wall tubing is going to be having a machine that can cut fast enough. Which sounds like a good problem to have.
Laser would be superior, cleaner, more detailed, but unfortunately I don’t think framebuilding alone could pay for a laser tube cutting machine… I’m guessing it’s going to be well over 100k for one, but i really don’t know.
As far as cleaning, I think it’s possible to get a ready to weld joint out of plasma, but I could be wrong.
This video of the cnc lathe Co-Motion has really got me scratching my head on how I could improve tube mitering in my future shop… It’s been one of my least favorite parts of framebuilding and where i’ve made the most mistakes, so handing the main tubes to a machine sounds ideal to me.
I have looked into abrasive tube notching a bit too! Seems like it could be a great system, but I’m not a fan of the noise, dust and wasted belts they would produce.
I have seen a few framebuilders using abrasive notching though, and it would produce a nice clean miter.
Given that a cheap mill/lathe and some $8 hole saws can cut a miter in ~30 seconds that requires minimal cleanup (and is quiet and not particularly dangerous for eyeballs/lungs/etc) I’m not sure it would make sense for anyone not trying to make, say, thousands of bikes at a time.
Those abrasive systems are awful IMO, unless you just don’t have space for anything in your shop. They make a ton of noise, get the tubes super hot (not good in some cases) and make a crap-ton of gnarly dust.
But yes, I am thinking about this for production runs in particular. I am a 3d printer and have CNC experience, and this is basically the same thing but with a plasma torch tool holder.
I don’t know how any other builders feel, but I always felt my miters could be better, regardless of the setup I was using. I’ve used a knee mill and horizontal mill and some other less savory setups as well.
It would be a huge time savings to do all three downtube miters and cut the water bottle boss holes in one pass. It’s been a while, but I’d say a downtube would take me 30 minutes on the mill at the very least, with a lot more room for error (if you are me).
I look forward to seeing how many tubes I ruin with a CNC plasma cutter. I’m sure there will be a few.
If that’s accurate, just optimizing your tooling/process with your existing equipment might be a better use of time. If doing a batch of downtubes (something I rarely ever did) I’m pretty sure I didn’t have more than 5 minutes into each one, including brazing in the water bottle bosses (you can do that while the mill is cutting the next downtube).
Taylor
I used to have an abrasive notching set up. It was fun, but loud, dirty, and maybe most annoying of all… Not super consistent. The belt tries to bend the tube, so you have all the same issues that you do with a mill. But usually not sad good a way of clamping the tubes.
Also the two issue for most people is rear triangle mitering. Front triangles are pretty simple.
Hahn Rossman
Hey good to see you here Hahn!
I remember that belt mitering machine… you hadn’t tried it yet when I was there. I don’t think belt is a great option either, i’d like to use as little abrasives as possible because of the dust.
I don’t know if you remember but you were actually using my Sputnik fixtures for your rear miters for a while…
Do you have any thoughts on cnc plasma tube mitering?
I would love to make mitering more automated when I set up a new shop, which is why I am considering CNC.
Yeah 30 minutes not including brazing on bottle bosses. My last setup was on a Bridgeport clone with a Kurt vise and Paragon tube blocks. Maybe my guess is way off… I have a poor sense of time. It took a while.
At any rate, I am looking forward to optimizing my business, process and tooling as much as feasible. Automated tube processing is one of my personal grails, so I’m excited to explore it further. I’m ok if it’s fruitless.
I saw on the another thread that you are using a Anvil mitering fixture? That’s a much faster setup than I ever got my hands on.
Yes I have all the first gen Anvil stuff. It would be relatively easy to make something like that fixture yourself and just mount it in a mill vise with a digital angle finder glued on.
Alternatively just 2 tube blocks per tube and a mill vise is not at all slow. Should be very easy to get a downtube done in 10 min or so.
I fully agree with Walt here, spend some time refining your process so you do the same thing every time. It gets fast……fast with repetition. Alternately, I never build only one frame at a time, mitering a downtube is about 3-5 minutes start to finish, each additional is about a minute even if the miter angle changes provided each tube is the same diameter. It becomes 2 minutes if I have to go grab a different clamp block.
As is true with any skilled trade or craft, the only way to get better is practice.
The old Carl Strong rule of thumb is that if you can’t make a complete hardtail/road/gravel type frame (do all the fabriation from raw tubes to ready for paint) in one 8 hour work day, you should not be trying to sell bikes.
For really artistic or weird stuff that obviously doesn’t apply, but regardless, you can’t be spending 30 minutes mitering one tube. It might be worth having some outside person come watch what you’re doing/count steps/critique (even if they’re not a builder) to help figure out what’s taking so long.
I’d like the purpose of this thread to be exploring the idea of applying cnc plasma cutting to mitering frame tubes.
I will also have a mill setup for mitering - this is just an experimental idea for now
I have about a year before I can even think about setting up a shop and figured that I might be able to design a small benchtop plasma mitering system in that time (currently living in a 520sqft condo). Maybe modifying another open source design?
Im pretty sure i can build this for around $600, I already own the plasma cutter. I’d give the plans away. It also sounds fun to me.
*To clear up some possible confusion:
It’s been about 8 years since I did a miter, 30 minutes may be a dramatic overestimation.
That said, I had a inefficient and ill equipped framebuilding operation. I never really had the funds to get it off the ground and started when i was in my early 20’s.
I knew back then that I wanted to change the process, and now I have a broader skillset to apply.*
Please drop a note here if you run across anything you think would be helpful to this project or have ideas or insights.
They use an (off the shelf) laser system to etch graphics onto aluminum tubes. I wonder if that roller wheel system could work well with your plasma cutter:
Hey Taylor!
I got great results with the belt grinder, but it was fussy, loud and dirty.
Then I got a horizontal mill and populated it with sputnik jigs. Super easy to use, easy to verify, repeatable etc.
I actually made the CNC machine of your dreams for a different project. It uses a rotary axis and a linear axis to allows the machine to handle pieces of infinite length by shuffling the stock along. Like cnc saws do on their infeed tables.
The one we made was only for marking a series bends on rod stock, but it would be trivial to put plasma/laser on the axis that currently has a fine point sharpie.
But really the horizontal with dedicated fixtures is fast, accurate, and cheap in the long run.
Hahn