RandolphCarter

Nice!

Is that a Curtis Odom front hub in your mockup? (drool)

iab

Yes. I had to model something. I was thinking a Son generator hub but since I want to use a Li-ion pack, that is not an option.

I would also consider a high flange FB hub for the front.

Catnap 

regarding the lug fabrication, you can save a ton of money and time by making them yourself. buy stainless tubing that has a diameter slightly larger than main triangle tubing. then mitre the ends to match the correct main triangle angles, and either fillet braze or TIG the tubing together. then you overlay your lug design on the blank and cut it out using a Dremel, drill press, and finish with hand files.

this dude's video here explains the process clearly: https://www.youtube.com/watch?v=9eAtHXjTbMM

following this method will get you build-ready lugs with around 12 to 16 hours of work, and maybe $20-$50 in materials costs. if you want that three-dimensional aspect to the lug surfaces, you just could down a layer of brass on top of the steel and then sculpt from there.

suggesting this because going down the 3-D printed or lost-wax route is going to eat up months of time and hundreds, if not thousands, of dollars.

Centaurious

Chain stays are too long. You only need room for a fender between the rear tire and the seat tube, any thing more makes the bike slower.

iab

I thought about that. And in a previous life, I probably would have done it. When you are young, you have more time and less money, but when you are old, the opposite is true (at least in my case).

Also, and I'm not sure it matters, there is no way a human could match the precision of 3D printing.

iab

Huh? Slower?

What kind of chainstay magic makes a bike faster?

Rocket-Sauce 

Just posting to subscribe. Enjoying the process and looking forward to following this.

Centaurious

I tried to argue for longer chain stays in the Framebuilders forum and the consensus was that two identical bikes, with longer chain stays on one, with identical twins for riders putting out the same power, the bike with the shorter chain stays is faster. They will admit that longer chain stays give a softer ride and more heel clearance for panniers. I guess Something automagicaly makes the bike slower.

iab

Did they provide any actual data or just say "I sez so."?

Did they quantify faster?

Did they signify a type of ride when it is "faster"?

Did they show any example of identical twins as riders putting out the exact (I added that bit) same power?


btw, this is a commuting bike. Why exactly would "faster" be better? I can also show you an example of a bike with the exact, and I mean exact, same chainstay length ridden by a young Gino Bartali. That young Gino Bartali on the too long chainstays would drop everyone, and I mean everyone, in the Framebuilders forum on their short chainstays.

rhm

@iab, have you ridden a bike with the Gates belt drive? I have that on my latest folding bike, along with an Alfine 8 hub (manual shift, though). I'm just not thrilled with the Gates belt drive. I'd be cautious building a bike around it.

iab

@rhm what don't you like about Gates?

elcraft

iab,
Where on your design is the removable section of either the seat or chain stay to allow for the install/removal of the Gates belt?

iab

The rendering is poor quality and it doesn't show the part line. The ds dropout is split, here is an exploded view.

Gregario28 by iabisdb, on Flickr

bikemig 

Maybe they know something you don't? I'm not being argumentative; just sayin' this might be a place where a few compromises might lead to a better product since builders--good builders--have a knowledge base that I suspect few of us do.

iab

I wouldn't disagree at all.

My point is that I want a collaboration, I don't want to just toss it over the wall and I get what I get. I have shown my drawings to a couple of builders. I won't name names, but one of them just wants my money and no input from me. That is unacceptable. Another one pointed out a structural flaw in an early lug design. That is acceptable.

easyupbug 

Now retired I do custom woodworking and found that when I started burning my name on the piece the pieces changed. Now I sweat details that I am positive no one will ever appreciate or even be aware of, just because my name will be there.

iab

I buy into that psychology. But if I had my druthers, I would prefer the guy who's going to sweat the details whether the maker's mark is there or not.

rhm

Well, I wonder about that myself, and I'm not sure I will give you a satisfactory answer. I think we all know the feeling, that a bike isn't as much fun to ride as its predecessor, and it's hard to pinpoint its shortcoming(s). But here is my sense of it:

First of all, the new bike has the Shimano hub whereas the old ones had Sturmey Archer. I like the Sturmey Archer better. The gears are more evenly spaced. So I start out with a prejudice.

Next, obviously, the new bike is geared too low. The designer had to pick a combination of chain ring, cog, and belt that fit the chain stay length, and evidently they don't make them in a sufficient variety of sizes. I'm sure he picked the best one, but it's too low. I spin out at 25 or 26 mph coming down off the Queensborough Bridge where on the old bike I could go 32, and now I'm getting passed by guys on fixies whom I just passed on the way up. Argh!

Belt tension is tricky. If I make it too tight, it feels really inefficient, but if I make it too loose it skips when I pedal hard, even in the lowest gears. There should be a happy medium, where it feels efficient and doesn't skip, but it seems I have to pick a compromise position where it doesn't skip much and turns fairly easily.

And finally there's alignment. The belt drive is unforgiving of poor alignment. Get the rear wheel askew by the tiniest amount, and the belt becomes noisy (and even less efficient). I don't know how many times I diddled with the dropout adjusters in search of the sweet spot, and it's never quite as sweet as I wish.

Please don't take this as a rejection of the system. All in all, I'm in favor of it, at least in principle. I understand the compromises the designer made, and I applaud his ongoing attempt to make the perfect bike. But I would be reluctant to build my dream bike around this system. Hence my question: have you tried it?

iab

No. I haven't tried it. But yes, I will before any I buy anything. Thank you for detailed review, it will help in the final decision.

Have you switched your bike to a chain drive?

rhm

Incidentally, here's part of what Mike Burrows has said on the subject:

"Modern toothed belts are much more durable and efficient than they once were. Materials such as urethane and Kevlar are used. Millions upon millions of car camshafts are driven by toothed belts, where chains were once the standard solution. But car engines have plenty of power to spare and the slight loss of efficiency through using belt drive is much less significant than it is to a cyclist." (Mike Burrows, Bicycle Design, p. 136). He has more to say on the subject, most of which you have already anticipated. If anyone following this thread hasn't read Burrows' little book, I recommend it.

rhm

No, and I don't expect I will. All in all I'm very happy with the bike. I only mention my misgivings because you seem to be on a quest for perfection, which I admire, but... well, the word "quixotic" comes to mind.


I should probably mention that no matter how hard I try, my pants always get ruined one way or another. If I could make the belt drive bike my main commuter bike, perhaps I could spare my clothes a bit. But I only ride that bike in the summer, when I'm wearing shorts anyway.

Darth Lefty 

What a beautiful design and a great project. I'm envious.

I'm curious to know what you learned about the strength of the SLS steel vs. investment casting. I'd have thought the printed part would be not nearly as strong, even if it's completely solid, just because it has a grain direction.

I am also curious about your split slot design. Most of the ones I've seen so far have the dropout and the split as separate features. With the split in the middle of the slot like that, the faceplate carries the entire load across the split, both from the frame and from the axle, and the thinnest cross section that takes the brunt is by the bolt hole and countersink.

Here's the bike I recently won

iab

As with everything, it depends. If you look at a straight 316, the printed part have about 85% the strength of a non-printed part. But I also can get other printed stainless that will have a higher strength than 316. Also, with printed parts, horizontal strength will generally be higher than vertical strength. So you do your build with keeping the stress points of the lug in mind.

For the dropouts, those will be machined. While searching the interwebz, I saw the split in the opening and not. I personally like the look of what I currently have over the others I have seen. btw, mine is a copy of something I saw, it is not original. The screw plate supporting the split is 3.5mm thick. I am confident at this time it will take the stress. That said, I have not had a detailed discussion about the design with a framebuilder. This is a perfect case where I would take their recommendation over what I have seen on the internet. No specification has been locked at this point.

Catnap 

this is bad science. shorter chainstays will change the handling characteristics of a bike but they will not make it "faster". only the rider makes a bike faster or slower. and for a bike that's being used as a commuter, the few extra grams that one or two centimeters of chainstay adds would be unnoticeable to the rider.

on a general note...

@iab if you have lots of time and money to spend on this project, i would suggest building a proof of concept. you're trying to do a couple eccentric things on one bike (your own belt dropout design, unique lugs, integrated lock, etc.) that almost guarantees that you'll either be less than satisfied with the final product or drive away builders by refusing to compromise. better to have a prototype built with your special dropout first, see how that works for you, as well as the relationship with the builder. then take what's learned there and evolve it to the next level. just a suggestion!

repechage

The rear dropout will be fine if it gets trapped by two plates, not just tied together on one side.
I don't see big danger.
I do see the point about throwing it all together in one prototype, but from a fabrication perspective, elements of the concept might benefit from say fabricating a sub assembly prior.
The general geometry is I think very close to many Italian city bikes of decades past.
Executing the fork crown in plate material is what I would think be best but might be a trick to get it done in one attempt. The laser sinter process there is more than I would chew.