So my question is, what stupid mistakes am I making right now? Design isn’t finished yet, but I would love feedback on the idea, as well as holes poked in the design.
What are you trying to solve by doing that? You’ve increased total weight, CoG height, friction, complexity, cost, more points of potential failure from fatigue for no perceivable benfit.
Nice CAD! In the class I taught, there was one rule: There are no dumb ideas.
Nick and I have had countless ideas that we thought were dumb, but we tried them out anyways. Most end up staying with dumb ideas, but the few that succeeded ended up being really innovative tech and workflows.
From my research, I found a few precedents:
Zerode G2:
This one has an Alfine hub mounted in the center of the frame
My advice for approaching this project is not to try to do it in one shot. I am pretty fluent in CAD, but a full 12s drivetrain from scratch would be overwhelming for me. It would be very hard to model the chain paths or where my calves or knees may rub.
I think a singlespeed version could be a very good proof of concept and a cool bike to ride around.
Even a “real” company like ceramic speed built a singlespeed prototype before they figured out how to make it shift:
Solving the derailleur being exposed to damage + the wind, instead centered in frame and could be potentially be covered with a shroud. As far as complexity goes, rear wheel removal is less complex without derailleur in the mix, so maybe on balance less complex. Drive train is now mostly mounted to the frame rather than the wheel so less likely to need futzing with.
Solving the rear hub needing hub shell flanges of asymmetric heights, different spoke lengths, etc. Makes a stronger hub with lighter weight.
Solving poor chainline as chainring + cassette need to be offset to avoid wheel/tire interactions. Chainring can be centered on cassette in this design. Reduced efficiency due to friction may be offset by increased efficiency here.
IDK about the impact to the bike based on raising the COG, but when moving the cassette off the wheel and rear of bike, the “Center”-of-gravity moves more to the center of the bike so it might be negated?
Total weight addition is essentially 4 short tubes, a new Mid Hub, and a second chain. Those tubes are going to be short so total weight addition doesn’t seem to be massive. Additionally this was designed to be welded, but future iterations would be carbon beam structures, maybe a design similar to Cervelo P5x beam design. This design is straight tubes that I can weld, whereas I have no experience molding carbon.
@DEVLINCC No perceivable benefit is maybe a bit harsh? Not evident to you, but there are plenty of benefits when you take some time to think about it.
I’ve seen the Trinity bike, but their location limits cassette size pretty severely. They have a newer version that is more contained below the frame which is also compatible with gearboxes, but it has some big limitations.
That Zerode is a great idea! Using what’s available on the market, although those alfines are not efficient. I’m trying to avoid internal if possible.
Honda bike seems to be pretty much the same design. Guess I need to look into patents a bit more
The idea for a single speed prototype and tester bike makes sense. I will mess up the first frame regardless of what I’m doing.
@terraformbicycles No, the idea is to use existing 12 speed 1x drivetrains, so no new componentry there.
The issue I see is the width of a 12 speed drivetrain.
I just threw a tape measure on my 12s axs cassette + derailleur and we’re looking at 130mm / 5” from the top of the cassette to the outside of the derailleur when in the 10t.
A lot of riders (myself included) have a hard time not polishing top tubes with their knees, so having any sort of extra bulk that high up would be a hard sell.
I really like the Honda and Trinity. They have the drivetrain more ‘out of the way’ and those DH drivetrains are easily 50mm / 2” narrower with the added bonus of wide DH q-factors.
You said to poke holes in the design. So while not specific that was what I was doing. I comment with an eye to be helpful not critical.
To your specific points (again just spitt balling so don’t take it to heart)
Solving the derailleur being exposed to damage + the wind, instead centered in frame and could be potentially be covered with a shroud. As far as complexity goes, rear wheel removal is less complex without derailleur in the mix, so maybe on balance less complex. Drive train is now mostly mounted to the frame rather than the wheel so less likely to need futzing with.
I see this come up all the time so maybe it is a real concern, personally I’ve munched maybe 3 or 4 derailleurs from trail strikes or debris in 30 years of mtb riding. My anecdotal evidence would suggest it isn’t a real concern but other people may ride differently or on different terrain that can cause these issues. I don’t know. I see more complexity in that you need to release the chain tension to get the wheel out. I don’t struggle with removing my rear wheel with a derailleur so for me that isn’t an issue I would need to solve. More bearings, more chain to keep lubed, more material in the frame to make it work. Chain passing by the thigh which is wider thatn my lower leg. A deilleur poking out right where my knee comes up. I can guarantee my knee would smash that. Higher top tube so stand over clearance is reduced.
Solving the rear hub needing hub shell flanges of asymmetric heights, different spoke lengths, etc. Makes a stronger hub with lighter weight.
definitely merit in that and don’t have any argument against that at all.
Solving poor chainline as chainring + cassette need to be offset to avoid wheel/tire interactions. Chainring can be centered on cassette in this design. Reduced efficiency due to friction may be offset by increased efficiency here.
Looking at a normal chainline the chainring is lined up with roughly the middle of the cassette already so you aren’t changed anything here other than lengthening the center distant, so the angle is a touch less at the extremes of the cassette. There is definitely gains to be made there but for me they would not be enough to chase after.
IDK about the impact to the bike based on raising the COG, but when moving the cassette off the wheel and rear of bike, the “Center”-of-gravity moves more to the center of the bike so it might be negated?
anything moved will affect the CofG. On a road bike there is certainly less need more mass centrelisation. On a mtb you could argue there is benefit but these days the high spec drivetrains are pretty light plus you have a big fork hanging off the front so hard tails a pretty well balanced. Going higher means teh pendulum of teh weight above the tyre contact patch will make the bike feel heavier in the side to side motion. It may prove inconsequential but also might be too much.
Total weight addition is essentially 4 short tubes, a new Mid Hub, and a second chain. Those tubes are going to be short so total weight addition doesn’t seem to be massive. Additionally this was designed to be welded, but future iterations would be carbon beam structures, maybe a design similar to Cervelo P5x beam design. This design is straight tubes that I can weld, whereas I have no experience molding carbon.
What ever material you use will have to be beefed up because you are making inherently less of a structure. So instead of using geometry to handle load paths you are using material thickness. Usually you end up with a heavier structure by doing that. Plus you have all the extra components adding to the total.
@DEVLINCC No perceivable benefit is maybe a bit harsh? Not evident to you, but there are plenty of benefits when you take some time to think about it. [/quote]
Don’t mean to be harsh but sometimes it feels shit when people poke holes in your design when you’ve obviously put a chunk of thought and time into it. I’ve been there way too many times myself and even with my four bar horst platform people are trying to poke holes in it despite it being a proven suspension design thet ahs won many wold level events. (that ignores the geometry of the bike obviously)
10-12 years ago I designed and modelled pretty much what Trinity is doing now. I’ve kept quiet because I’ll cop flack for even mentioning it. I messed around with that design for a couple years in CAD poking holes in it and trying to make it conceptually as perfect as I could. In the end I shelved it because to me the extra weight and complexity didn’t give me any perceivable benefit over what was available at the time and drive trains are infinitely better now.
With the tech avaiable now and the fact they built the platform they may prove me to be completely wrong and the same with your concept. Im not saying don’t build, because you may do your prototype and it works exactly how you want it to and you can come back and give me the bird . Just being the critical preactical sounding board to bounce ideas off so you can have a chance of making it work. With a concept like this you need to be super critical because you are stepping way outside the box and there is a lot of work in it to make it have a chance of working.
@DEVLINCC From the other side of a computer screen its easy to construe your first comment in a different tone. But you’re likely right in that I’m also sensitive to critical feedback as this is not an area I feel competent in.
Appreciate the genuine feedback, there seem to be quite a few concerns of width that have thus far been minor concerns to me.
———-
Regarding @wzrd, i currently have placed the cassette at SRAM’s min chainstay length of 395mm for a road 1x12. I can drop it much closer to the bottom bracket similar to the Trinity and Honda Drive systems, but shifting performance will likely suffer as it was not designed for such a short distance. I figured my first prototype would be to manufacturer spec, and then we experiment with getting closer to the BB. Would y’all recommend just making this change in the first place?
I measured an 12 speed non-axs drivetrain, but may have though about pieces too independently and not at their extremes.
Regarding @DEVLINCC about never damaging rear derailleurs, there are always going to be people that ride their bikes with precision and don’t damage rear derailleurs or bend hangers. But I remember plenty of people that damage these while learning to ride, know plenty of triathletes that don’t do bike maintenance, and remember being a collegiate rider and shoving 35 bikes in a trailer to travel 6 hours to a race.
Weight considerations make sense. One application would be removing the seat tube to bottom bracket connection is going to change any kind of saddle-steer input and necessitate oversizing associated tubes get a similar output. I will definitely run into issues with the width of these tubes if I move the cassette closer to the bottom bracket.
That 395 mm minimum chainstay is no joke. I built a 350 mm chainstay 20” wheeler for my kid and had to remove one cog from her eight speed cassette because the chain angle from the rear would shift the chain off the front chainring.
After installing a narrow wide ring on the front and shimming the cassette a little I made it back to all eight gears, but that twelve speed cassette is wider still. Chains just aren’t that bendy in that direction.
So that particular constraint is not very negotiable, but unfortunately shortening that distance would help solve many of less desirable things this design introduces— like the width location and CoG.
Last week the wind blew over my road bike while I was drying clothes on top. It ended up bending my derailleur hanger, which I found out on the climb. The gears were skipping so badly I had to go back home
I just want to remind everyone that this is the education forum, so it is a safe space to explore new ideas and ask questions. All ideas and constructive feedback are welcome.
I think the design solves some problems with traditional bikes. Maybe it won’t improve them all at once, but every idea has to start from somewhere. A lot of times the spark of innovation does not come from the completed product, but from some small sub-assembly or workflow. That is why its important to try crazy, new, controversial ideas.
Even though this design is formed as a road bike, this sort of architecture could be very valuable in recumbents, e-bikes, mountain bikes, etc…
Ideas:
To get back on track with constructive feedback, from the frame design perspective, I really like how it looks, taking queues from the new Sirrus.
Using my eyeball FEA, I think the tube diameters you have are adequate for a road bike. And as per @WZRD’s comment, I think the cassette cluster/ single-speed (if you go that route) could be moved lower. You can use a trick from the mountain bike world, and utilize the space in front of the crankset. It is not nearly as pretty though.
I run into this problem all the time: trying to make everything perfect before I even start. I end up never doing what I set out to do because it’s never perfect. Making this design light, efficient, aero, gear range, compact, etc… is possible, but really hard to do. If you break it into small steps and accept that some things won’t be perfect the first pass (like a singlespeed), then you can have a platform to test your ideas.
, but really what this project needs is a Sawzall and a MIG welder, not intricate trinkets.