I'm considering a Rohloff Speedhub for my next bike, but I like drop bars. The factory Rohloff shifter is a grip-shifter that mounts on 22.2mm diameter straight bars, rather than 23.8mm drop bars.
Because the indexing for the hub is all in the hub itself, rather than in the shifter, I think I could design and print a pretty nice grip shifter for drop bars, or possibly some sort of fat lever shifter, even. The most reasonable drop bar mounted solution that I'm aware of is Comotion's $250 shifter, which looks very nice, and seems to be milled out of nice metal parts, but frankly, I'd rather design and print something, so that a not-$250-solution will exist.
I don't think the actual CAD bit will be all that difficult; I have a fair bit of OpenSCAD experience, and the parts aren't all that complicated. I think. But, I'm wondering if anyone here has any tips about designing long-wearing all-weather printable bike parts.
I'm also wondering if anyone around here owns a Speedhub and would be interested in beta testing. I don't have that bit yet, and while the spec sheet from Rohloff is quite detailed, there's no substitute for trying it out.
Any update on this?
I haven't played with the design in a little while. I did buy a bike with a Rohloff hub (it's quite nice), and while the grip shifter on Jones bars isn't bad, I'd still like to get this to work well on drop bars.
In the back of my head, I'm thinking about changing the assembly and handle, so that it's more of a lever-shifter than a grip shifter. I might even abandon having it mounted axially on the bars, and just do a clamp, like you would a friction shifter. I also might play with making the grip smaller diameter than the pulley, which would make the physical hand movement necessary to shift smaller.
Posted an Update. U-shaped cable grooves, better (at least more printable) clamping mechanism, better cable guides and cable guide sockets.
Also, posted a build-plate STL, since Thingiverse's SCAD previews don't seem to be working (neither is the customizer).
The grub screw to hold it in place is pretty concerning, no matter what bars you're running - short of really cheap, thick aluminum bars. If a conventional clamp isn't viable, at least have something under the grub screw so it's not digging into handlebar material directly and the pressure is distributed.
Also wondering if the cutouts for the shift cable barrels are deep enough to keep them from jumping out. Didn't realize it was two separate cables, thought it was one that looped around. Guess that would mean you have to properly adjust the tension on two cables to get shifting right?
I lifted the grub screw idea from a how-to video on installing old-style grip shifters. Looks like that's what that particular model used. But then, that was on straight bars, and possible steel straight bars. I'm certainly open to alternatives for securing the thing; I don't much like the idea of grinding a grub screw into my bars.
I think ideally, the shift cable end stops wouldn't have enough headroom to leave their places (they'd bump into the outer housing), but you're right, that is a concern. I'll have to be careful about the geometry there, and I haven't been so far, because I don't have accurate dimensions for the shift cable ends. I should probably buy a set of Rohloff cables and design to those.
The two-cable thing is how Rohloff hubs operate. Uniquely, the indexing and detents are all inside the hub, and the second cable removes the need for a spring. So if you cut the shift cables, the hub won't shift; it'll stay in the gear it was in (you can even shift it while stopped, by pulling on the broken cables). Everything outside the hub is just dumb pulleys and cable guides, which is what makes this remotely feasible.
Too much slack would mean play in the shifter, which would be annoying, but there's a barrel adjuster near the hub for that. So you can kind of think of this as a friction shifter, except without the friction and the spring, and it feels like an indexed shifter, because there are detents in the hub itself. The other odd thing about Rohloff shifting is the amount of cable travel: 7.4mm per shift, which is rather more than a derailleur uses. That means the pulley diameter at the shifter needs to be large, to fit that travel (times 13, for 14 gear positions) into less than a full circle. Which is part of why the factory shifter is a grip shifter.
I could probably get creative with mounting and handle design, and make it a lever-style friction shifter (which will feel indexed, because of the hub), it would just have to be rather large. Instead of the 25mm or so outer diameter of a normal friction shifter, it'd be closer to 60mm. Probably worth a try at some point, and it would provide more mounting options. Maybe a pair of M4 captive nuts in the stator core, to receive bolts through a clamp. Except then the tension on the clamp is pulling the stator's layers apart.
Excellent explanation, thanks! Knew some of it but not all together. Haven't had the opportunity to work with one yet.
I'd say at this point, definitely get some precise measurements from the Rohloff cables - looks like they're comparable to normal shift cables but I guess that might not be the case - and just increase the ridge that they sit on a little bit. I think cable tension should help keep them seated.
Regarding the handlebar clamp, I would think a pinch clamp like you find in brake levers or shift levers might be a better approach, but figuring out how to fit that inside the existing design might be tricky. I agree that copying an old Grip Shifter isn't a bad idea. At the very least, have a piece under the grub screw that takes the abuse but still clamps the bars.
It looks like you need insets for the housing to slide into the cable guides. I think 5mm is what you need to fit most cable housing ferrules. Either that or build in a place you can thread in a barrel adjuster
Definitely. Insets in the cable guides for the cable ferrules. Thanks!
Here's the Thing: https://www.thingiverse.com/thing:2719485
That's a never-been-printed first cut. Probably doesn't work yet, but I'd love feedback on the design.
Maybe take some screen shots or export some STL files for those of us who aren't running OpenSCAD?
Two things so far on the design, both related to printing feasibility.
The cable retainer and guides will be very difficult to print being a totally flat overhang. Instead of a square shape trough for the cable to lay it, it should probably be u-shaped so the overhang isn't as severe.
The other weakness I see so far is the clamping mechanism. I see the slits allowing the clamp to pull tight around the bar, but if the part is printed where that is horizontal, it will have almost no resistance to shearing off due to layer alignment. The act of clamping it would almost be enough to make it break the rest of the way around if it was printed in PLA or ABS. Maybe something with two bolts that pull separate halves together?
Regardless, the best part about 3D printing is it's super cheap to try out an idea! Best of luck.
Thanks! Those are both good points. U-shaped pulley troughs won't be too hard.
The clamp. That's the thing I'm most uncertain about with this design. And you're absolutely right about the shear strength issue. I'd like to try to keep the weather out of the enclosed area (eventually; that seems like a version 2 feature, with o-ring seals and such), so I'm reluctant to do a two piece, two bolt design with a split down the middle for the whole non-moving part.
Perhaps I'll try a captive nut and grub screw. It seems like MTB grip shifters used to use that with success.
The old Gripshifters would probably be a good place to start on how to do this using plastics. Maybe even to the point of using their old spring and detente mechanism.
As far as plastics go, ASA and ABS for any exterior parts would help reduce any UV breakdown effects. I'd recommend using a Nylon based filament for any parts where they move against each other. I like ABS for how easy it is to finesse via filing or sanding after the part is printed. PETG always seems so gummy to mess with post printing and I wonder if that could happen if you have two PETG parts moving against each other.
Shifters need to be able to pull the same precise amount of cable each time and with as little friction as possible. Good luck and keep us posted with this project.
One of the great things about the Rohloff is that the detent and indexing mechanism is inside the hub, rather than in the shifter. The mechanism has two cables that wind around a pulley on both ends. Turning the shifter pulls one cable (loosening the other) which turns a rotary selector on the hub itself. So the shifter itself is super simple, and won't need springs or anything.
I am a bit concerned about friction, and nylon would probably be ideal, for both durability and low friction. I haven't printed with nylon, but I might try it for this.
I've worked on bikes quite a bit throughout the years, but never had the pleasure to work on one equipped a Rohloff. I forgot they run the push-pull control cables which would allow the shifter to solely operate via turning friction and have no need for indexing. That's one less thing you will have to worry about.
I've printed some bike parts ( and motorcycle parts too) I've used good quality ABS and really high density infill and walls. On the motorbike I tried the carbon fiber PLA with 70% infill ,indestructible, but this material doesn't love friction or high temeprature, so my tip is to use ABS. I've never printed PET but could be an alternative to ABS, similar property
Sounds like a great project! I don't have any experience with the hub, but have heard good things about it. The Comotion option does look nice, but yikes, yeah, that's pricey.
I think PETG is probably the best from what I've read, but I have some PLA printed parts that have worked pretty well - although to be fair most of them I swap out long before they could be considered "long wearing". If I really wanted durability, I'd definitely go with the PETG or ABS, or the ASA mentioned by fantasmag0ria although I've got no experience with that.
I'll be really interested to see how this project progresses for you!
My bike has a lot of printed parts, they all are in PETG. The bike is outside in every weather so I can confirm PETG is good for the job. The fenders will break after about a year, but they are printed in vase-mode, a single outline perimeter so they are very flexible.
I also have a Rohloff on my bike, I'd be happy to beta test your designs. I am also considering building my own electrical shifter for the Rohloff. Since Rohloff released theirs, but you can't buy it, it will only be bundled and compatible with bosch.
Maybe we could even collaborate a bit here. What do you use for design? Hopefully Fusion360.
I'm planning to use OpenSCAD. The parts to to Comotion one look simple enough, and I like the idea of a parametric scalable shifter.
I'll be sure to post here when I have draft of the design. I have a little time off coming up, so I'll sketch it out then.
And thank you to everyone for the material pointers. I'll give PETG a try; I have a spool of that.
For material try ASA (ABS type material design for outdoor use) or PETG. On that thought I have a few parts on my bike printed with cheap PLA and it works just fine in summer heat and winter colds. But those are not 'safety' concerned parts.