3D printing a set of hubs (or what ever I think of)

Posted on
Page
of 4
Prev
/ 4
Last Next
  • i was also thinking about putting a race on the carbon axle so that it fits snug wiht the bearings.

    Do other sealed-bearing hubs have an axle like the carbon axle that you've shown above? I had a Hope hub a while back that definitely didn't, since you could knock the cups out by putting a drift in through the opposite axle hole.

    If you have the carbon axle as shown then at least one cup will have to come out of it every time you want to change the bearings (assuming you'll put a flange on the inside of the hub body to retain the bearings). When you fit the bearings you'll then be trying to interference fit the cups into the carbon axle while pushing them into the bearings. You could get round this by fitting a Hirth coupling to the centre of the carbon axle, but I don't know whether that would work in terms of axial loadings on the bearings.

  • One more thing: from the lack of a spigot on the end of the hubs it looks like you're planning to make this a bolt on hub with an internal threading on the cups (rather than QR). My Paul hubs are like this and use stainless M8 socket cap allen bolts. Because dropouts are 9 or 10mm in diameter you need something like the Phil Wood Chrome Dome to sleeve the bolt and increase its diamter up to 9 or 10mm where it goes through the dropout. The annoying problem that I have is that the sleeve part of the Chrome Dome equivalents for my rear hub aren't long enough to go through a chain tug and into the dropout, which leaves me with +/-1mm of slop when locating the hub in the rear dropout. If you can make something like the Chrome Dome with a longer sleeve it would solve this and I would buy them from you!

  • It's a sintering process this, isn't it? How does it affect the strength of the final material? Let's say ya boggo standard pressed mild steel for a flange is something like 500MPa in tension, what would this stuff come out like?

  • If you have the carbon axle as shown then at least one cup will have to come out of it every time you want to change the bearings

    yup, pretty much, but the new hope hubs use the same system.

    It's a sintering process this, isn't it? How does it affect the strength of the final material? Let's say ya boggo standard pressed mild steel for a flange is something like 500MPa in tension, what would this stuff come out like?

    this is a really good question. in all honesty, i have no idea. i was chatting to our printing expert the other day, and he claimed the in plane material properties of sintered alloy, could actually surpass that of non-sintered alternatives. He explained why, but there were alot of words i didnt recognize.

    really i should use reliable sources when choosing material properties to model with, but im going wiht this site for speed

    https://www.stratasysdirect.com/blog/material-nuances-metal-3d-printing/

    the simulations are on, heres the mesh:

    theres going to be two flavors of simulation, first will be tensioning the spokes to 200kg, just to see how it reacts. the second will be taught spokes, and then a single spoke will be pulled until failure. this is quite unusual because its not realistic, however it is basically the worst case scenario, which is what im interested in.

    these computations are pretty heavy so im running it through a small cluster, however, it still looks like itll at least take 10 hours.

    when the results do come in, i can have a look at areas that are highly stressed, and those that arent. then material will be stripped away accordingly. unfortunately, this thing is currently HEAVY. the stainless steel parts alone are 180g, and its a front hub!

  • OK, so stress analysis is done. time to see what we have.

    Firstly, with the visualisation, i have set the maximum visualised load to 200MPa, anything above that will appear as grey. The yield strength of this steel is likely to be in the 600MPa region, by aiming at 200MPa, we have effectively set a factor of safety of 2. That is, I have to be wrong by a factor of 3 for it to fail (easily done when youre guessing quite a lot of stuff :P).

    Heres a flange with a radial spoke pattern, loaded purely by a spoke tension of 150kg:

    What we are looking is the von Mises stress (MPa) applied to the hub, red is bad, blue/green is good (or is it? well come to this later).

    this is all fine, no high stress except around the spoke holes. The spoke hole stress is probably an analytical error assocaited with the manner to which im applying the load.

    Now, lets imagine im on the bike. I weigh 70 kg, but in the near future i may put a load of weight on, so im assuming i weigh 100kg. the worst loading manner possible, is 100kg applied through a two spokes, one on each flange (this is not realistic, its a massive load in comparison to what you would typically expect, but over estimating is completely fine, its just under estimating).

    With this loading, the hub now looks like this:

    No prizes for guessing which spoke is loaded! the spoke width appears to be about spot on (complete fluke there...). We can also see an additional stress concentration on the outter most edge of the hub next to the loaded spoke.

    Now, heres the fun part. lets look at what isnt under much load, and cut the material away. Looking at it, not much is going on around the body of the hub, or the material that connects adjacent spokes on the outer most edge. So i removed material from these two areas, and i will resubmit the geometry. The updated geometry is right here:

    Just so we are all on the same page, this hub is still predominantly made of steel. its going to be heavy. i would be pretty happy if it ended up sub 200g :)

  • might want to do a loading analysis for cornering which will add greater out of plane loading to the spoke flange? Furthermore have you accounted for surface finish?

    You should really do a convergence plot of multiple tests to make sure the model is working and tending to an ultimate stress value. If you exploit the symmetry of the hub the simulation time will be reduced as you only have to run a fraction of the nodes.

    Great thread though!

    Maybe you could make an awesome personalised seat clamp too

  • This is so cool. I played around with FEA tools with fusion 360, as it not only shows the stress areas but the plastic deflections on the item. Might be worth a look.

  • seat post clamp with double bottle opener on the back! this is a game changer.

  • I want one

  • that is genius. I think that could make you a few quid

  • Yes please.
    Oh no. It would only work if mounted sideways as there's not enough space to move a bottle between clamp an rear tyre.

  • that is genius. I think that could make you a few quid

    this might go on kickstarter. how much do you think people would pay for it?

    Oh no. It would only work if mounted sideways as there's not enough space to move a bottle between clamp an rear tyre.

    with this design, the bottle is horizontal, then you pull it downwards towards the tyre. for me thats no issue because my frames weird. if i took it to production i would probably make a few tweaks to make sure it universally works :)

    edit: when its made ill get a video :)

  • when bottles are horizontal the beer falls out.

  • It'd never be open and horizontal though. You'd get a very small amount spilled.

  • Dibs a free one for making you think of bottle opener in your other thread

  • Iteration 2 of the bottle opener. better opener positioning, it means you can functionally use both at the same time. plus this will fit all frames, not just my one :)

  • Stupid question, but will you be able to do up the bolts still?
    Also - could you have "LFGSS" cut out/ embossed on the front of it?
    Love the idea though!

  • Stupid question, but will you be able to do up the bolts still?

    thats actually a pretty good question. its something we should be able to do in theory,but no one at our work has actually done it....

    strictly speaking, theres no reason why we couldnt use the printer itself to cut the thread. i thought that was a bit ambitious though, so im going to try tapping it by hand instead.

    Also - could you have "LFGSS" cut out/ embossed on the front of it?

    yes we can. the printer can probably achieve a minimum line width of 0.1-0.15mm, whch would allow for very intricate lettering. the model i sent to the print centre actually says "Refuel" on it.

  • would buy one

  • Brilliant stuff, would also buy one

  • Any chance this might overstress the top cap/stem assembly?

  • Subbed, look forward top finished results

  • That's gonna snag shorts.

    How about exploiting the gap so the bottle approaches from the side of the bike? Fewer tyre clearance issues (or toptube issues for this with the bolt at the front)

  • That's gonna snag shorts.

    im gunna prototype, then if this is a problem ill alter the design. although im not that acrobatic on my bike.

    How about exploiting the gap so the bottle approaches from the side of the bike? Fewer tyre clearance issues (or toptube issues for this with the bolt at the front)

    theres actually a company that tried to kickstart this design:

    http://www.swarmeffect.com/designs/seatclamp

    It was actually successful, but they arent producing them anymore :(.

    its actually quite a hard design challenge because a bottle opener is quite large relative to a seat post clamp. the bottle cant be sideways either. plus, both CNC and SLS must be capable of manufacturing the thing. my mark 2 design cant actually be CNC'd. Which is a massive fail on my behalf because it easily could have been, and missing this also makes it much more likely to fail.

  • Post a reply
    • Bold
    • Italics
    • Link
    • Image
    • List
    • Quote
    • code
    • Preview
About

3D printing a set of hubs (or what ever I think of)

Posted by Avatar for Isotropic @Isotropic

Actions