17 April, 2009

Bending Forks and Designing Boats


Before VO I started and owned a company where one of my jobs was designing small boats. The photo is of an open water sliding seat rowing boat I designed. Like all boat builders and designers I developed a keen eye for curves. The various curves that make up a hull, the sheer, the stem, the buttocks lines, the diagonals, and a lot more, define both the performance and the appearance of a boat. I spent days tweaking lines in hull design software, on paper drawings, and in the shop to ensuring that each one was as fair and pretty as could be. Sadly there are few curves on bikes and my skills are now little practiced. So I devote considerable thought to the one curve there is, the bend in the fork. I'm rather proud of the nice bends on VO semi-custom frames, though the frame builders are ultimately responsible for them.

Traditionally the French builders had particularly nice fork bends. The blades were thin and the bend was low and smooth. A good example is the fork on my Motobecane Le Champion frame as shown below:


A really pretty fork bend is one of the bigger problems when having frames made in Taiwan. Heck, many custom builders in the USA and Japan don't get it right. It's pretty tough to explain the importance of this subtle detail to a fork factory that makes thousands of forks a month that their customers are perfectly happy with. That's right, forks are usually made in a separate specialty factory, not by the frame builder. One look at the fork delivered with our sample Rando frame illustrates the problem. This is not a bend: it's a bloody dogleg.


Still it's been possible to get a reasonably nice bend as on our city bike. It's not perfect, but certainly better:


We are going to continue working with the frame and fork factory, but I'm, frankly, not sure if they can do much better than the above. I hope to make it back to Taiwan this summer and visit a small fork manufacturer I met with last month who seem interested in this issue. I'll come bearing sample French forks and a check for the required tooling. If all goes well we'll have perfect forks, but not until next year.

On an unrelated topic, we have the rare and elusive canti rack mounting bolts again.

47 comments:

Anonymous said...

Chris, can you elaborate on the technical issues here? To the extent of my understanding, this is unfortunately a classic form vs. function dilemma. A lower bend looks better, but a higher bend is easier to create and is a much stronger design. It's a matter of the tubing wall thickness and diameter. A tighter, lower bend requires a smaller diameter tube, which has to have thicker walls for the same strength. Is that correct, or is there more going on?

Velo Orange said...

Anon, What you say is true, but in this case it's largely a matter of setting up the proper tooling, the right bending mandrel. And they are simply not accustomed to this style of bike.

D.L. said...

While I can appreciate the bend in old French forks, I've always thought that it was the smoothness of the bend — and not how low the bend starts — that counts. On some of the older French bikes I've seen, the low, sharp bend is quite jarring (they sometimes remind me of praying mantises). High rake/low trail forks, by my reasoning, need to start bending higher up.

Gunnar Berg said...

I believe that the bends which are really pleasing to my eye have the bend going all the way to the end of the fork. Normal tube bending processes leave a short straight section at the end. This requires the manufacturer to make them a little long and then cut off the end in a secondary cut. Factories don't like the word "secondary". More than I know. And your eye may vary.

Gunnar Berg said...

And D.L.'s eye varies from my own.

Studs Turqel III said...

the only curves I concern myself with are the ones I see lying next to me when I wake up in the morning. Curves and bike frames just don't seem to go together in my opinion.

Le Liquier Fissure said...

Curves and Bikes ?
http://www.instructables.com/files/deriv/FUZ/28XN/L7REP27RCOG/FUZ28XNL7REP27RCOG.MEDIUM.jpg

curves on bikes are like parsley sprigs on a plate.

Le Liquier Fissure said...

http://www.instructables.com/files/deriv/FUZ/28XN/L7REP27RCOG/FUZ28XNL7REP27RCOG.MEDIUM.jpg

Dad said...

Chris, I don't know if this entry stems from my post on the last thread, but at any rate it's good to air this issue out. I don't care quite so much for their bikes, but I notice that most of the Rivendells have proper French-style bends. Aren't those frames sourced from Taiwan for the most part?

My old beater Fuji '70s touring bike has the most beautiful, round fork bend. Is it a dying art?

I mention it because this detail seems vital to distinguish the frames from the zillions of other substitutes out there. A fundamental branding issue.

Cheers,

David

rod hungh said...

now this bike has some nice curves : http://www.flickr.com/photos/17479504@N07/2676146640/

Hank G. said...

A lower tighter bend will absorb shock better which is probably why the French builders used it considering the condition of post war roads.

The problem with getting a tight low bend today is that all the benders typically in use have way to large a radius. The builder would have to custom make special tooling. I don't know that any tool supplier sells benders with small enough radius for a classic French bend.

patates frites said...

You want curves? Just do some Hetchins-style chainstays. It don't get much curvier than that!

Uncle Ankle said...

Here's branding for ya

Dad said...

Anonyme, you raise an interesting point but I don't think I agree with it. I seriously doubt that a higher bend is a "much stronger design." I say that because the point of greatest stress on a fork is right above the crown. Think about it as a cantilever with the hub end being the free end. Very little stress occurs out near the free end. The stress occurs where the cantilever starts. Empirically this makes sense too: every single fork breakage I've experienced, seen or heard of in racing is right above the fork crown -- including George Hincapie's infamous crash at P-R.

So, ease of manufacture I agree with. Strength and resiliency, I'm seriously dubious.

David

Tom said...

I was just telling my wife the other day about seeing at Belle Haven Marina a boat like the one pictured. It's the most beautiful rowboat I've seen.
--Tom

Anonymous said...

Surely you could find a metal worker in your neck of the woods with the suitable skill to bend fork rakes? Even if she just bent the blades in blank blades, then boxed them up and shipped them to the production facilities, you would get what you want. From the picturs I have seen, the operation involves little more than a maple block the right bend, a clamp for the tip, and the lever bit to get the radius. IS this the sort of thing that is best 'hand crafted?' I bet there are enough people who value such lines to warrant this sort of business.
M Burdge

tadd said...

I do agree that the fork bend is lacking in the prototype. I believe the Kogswell forks are made in Taiwan, and it is avail. with several types of forks with different bend. It does not seem like it would be a technical issue.
If the bike does not have that nice french fork bend, it would look rather generic to me, even though it may ride well.

DT bosses makes it much more versatile. I would add it, and as someone wrote earlier, some will be using this as a budget randoneusse.

tadd

Anonymous said...

I agree that having the bend continue to the drop-out is the key to a graceful line.
Personally I prefer a higher bend though.
Gillotts were famous for their graceful forks
check out these!

http://www.flickr.com/photos/makfreak/sets/

Simon

Joel said...

Chris: I wish you would start selling that boat!

David_NJ: I believe all Riv forks are made by Mark Nobilette. At least he makes the forks for the US and Japanese made Rivs.

Mark also makes the beautiful forks for the new Rene Herse bikes.

Perhaps Mark would make forks for VO as well if we all ask nice.

Joel said...

Simon: Those Gillotts are very attractive bikes. Thank you for the link.

jim g said...

Chris, your fork shop needs a bender like this or this. I'm wondering if maybe they're using a bending mandrel w/o a lever-mounted roller, like this one. It seems that a bender with the right-diameter mandrel and a lever-mounted roller are key for the French-style, low fork rakes...?

Anonymous said...

I'm a word person, not a technical person, so I don't know what's stronger or anything like that...

But I do appreciate a pretty fork, and that's one that has a nice, smooth, lowish curve that goes all the way down to the drop-out.

The sad thing today is that few people seem to care about how a bicycle looks anymore, and if that's so, no manufacturer will care about taking the extra trouble to make things pretty.

There's more to life than pure technicity. What is anything in life without an equal measure of art?

marcus17 said...

Is that the annapolis wherry, i've thought about buying the build kit for a while now, beautiful boat.

robatsu said...

A lot of the higher, wider radius, non-French style bends, almost always end up straightening out for a ways before the dropout. This is one thing that makes the French ones so elegant, the continuous curve into the dropout.

Here is an example of a wider radius curve that continues to the drop, with the radius of the curve increasing to where is is practically straight at the end. On the stop side, the curve also fades in, so there is none of the dogleg effect. It is pretty subtle. The frame is a 1972 Fuji Finest.

I think that this sort of curve could be an alternate for those who don't like the low/tight French style.

FWIW, I like the tighter French style as well.

Anonymous said...

a production steel fork with a nice, elegant crown, eyelets, and traditional blades? Does anyone remember how many decades we had to wait for that?

Ok, so the bend doesn't look French, probably because it's not, but let's try to keep all this in perspective.
mw

james said...

A lot of people, myself included, would assume that the tight and decreasing radius bend helps with bumb absorption, but I have to wonder how significant this is. I've never done an experiment with two otherwise identical bikes, differing only in rake radius (and not amount of rake).

I'd also point out that this isn't a uniquely French thing, it appeared on all over western europe and become pretty common in the 40s at the same time that low trail became popular, again not just in France. You guys rely on too few sources for historical information.

Andrew F said...

Even Toei took years to sort this out. The design of the mandrel that Hoshino built was something that he spent quite some time perfecting in search of the perfect bend...

Big & Tall said...

Since we are talking about forks, is there any chance VO might one day sell aftermarket forks? I would really like to find a well made steel 700c fork with a 300mm threadless steer tube in 1" diameter. If it featured a nice looking crown, the kind of bend being discussed here, and eyelets on the dropouts, I might have to buy more than one. (Trail is unimportant to me.} It amazes me that such a thing seems very hard to find.

Anonymous said...

For want of the perfect bend a ride in the sunshine was lost.

Gene Hunt said...

http://www.youtube.com/watch?v=EqHlfGk2Kcg
I need a nancy french bender.

Joel said...

"You guys rely on too few sources for historical"

Pretty natural when you think of the history of European imports to the United States.

Bike companies from France, England and Italy have long pursued the U.S. market. From the rest of Europe it was pretty much catch as catch can.

English and Italian bikes - at least those that sold in big numbers in the U.S. - do the curve different than what Chris is looking for here. Had the Swiss, German, Spanish, Polish, etc. companies figured out a way to get their bikes to the U.S. in numbers, I am sure we would be calling it the Continental curve.

Drink Beer said...

See you all at the annual st john's v USNA croquet match !!!!! Croquet is french, oui ?

Anonymous said...

There is a very intersting thread on the Framebuilders list at bikelist.org on this issue the past couple of days. Part of the problem may be a change in the way the fork blades are drawn. They may not be constant thickness, but rather thicker at the narrow part making bending that much harder. At one time Reynolds sold fork blades that were pre raked at the factory. I suspect the same may have been true of the Vitus and other French tubing manufacturers. Perhaps Pacenti or JPWeigle or someone like that could be persuaded to do some bending en masse.

By the bye, the boat angle is Wherry Interesssting;}

Anonymous said...

Live without beauty! How do you resist despair?

Anonymous said...

Live without beauty! How do you resist despair?

Anonymous said...

You can obtain the visual quality of the tight french curve all the drop out as long as the curve of the bending mandrel is right and you practice. No high tech tooling is needed. As far as ride quality the radius of the curve is only one factor. Wall thickness, diameter, and tubing shape are also as important in determining ride quality. This is the beauty of fork design...there are so many variables.

doc said...

Ahh! Chesapeake Light Craft. Back in the late '70's and through the '80's I was a competitive sculler. Still have a Staempfli racked behind the shed. I always thought about the idea of building a touring shell, and I think I even picked up some literature on those boats. Never enough time though...

doc said...

btw: Gary Piantidosi is now taking his oar building/sliding rigger making skills and constructing Pine Car Derby tracks. Quite the pleasant surprise when I saw one of his famous shipping boxes at my son's PCD event.

Charlie said...

Here's an modern production bike that has a very nice bend (IMHO), from Olmo.

Rivendell has a description of what makes a good fork bend that agrees with most of us--radius continues to the end.

Velo Orange said...

Don't forget that what works for a custom frame builder making 20 frames per year is often far too time consuming in a production environment. We order 300 frames and expect the factory to have them ready in 3 months. That time includes having the tubing drawn to order and painting and packing.

Dad said...

Respectfully, I'm finding it a little hard to believe that these are harder to make in a production environment than on a bespoke basis. I mean if the designers at Toyota decide that they want a body panel curved a little differently, they call the plant, fiddle with the tooling, and start stamping the newer versions out.

Isn't it just a matter of equipping them with the right mandrel? The press just goes whomp and bends it to that shape. End of story. I don't fancy that forks are hand-bent in a production environment, but even if they are, same deal.

In my business we just submitted a detailed design for a rather complex electronic gadget to a bunch of Far East manufacturers, and in 24 hours we had several quotes emphasizing that we could have it any way we wanted it. Not a big dollar item; about US$70 each to manufacture to spec, in quantity. They all seem to understand full well that there's a lot of competition for business amongst manufacturers.

All I'm saying is, let's get these forks made up the way we all want 'em! For sure, nothing is as easy as it should be in this world, and I readily concede that, but still.

keithwwalker said...

Timely blog post, relative to the new issue of Bicycle Quarterly's latest issue:
Spring 2009 IssueArticle entitled:
Builders Speak: Making a Mandrel to bend fork blades, The challenge of getting a tight bend at the bottom of the blade.

Send that article to your suppliers!

Tom said...

The 'in quantity' part of your statement is the one that matters the most. 300 forks is not a large amount- it represents maybe 2 days for 3 workers in Taichung. Some factories do well with small runs of under 500 units. Others won't touch those projects because they don't make a nickel off it. If velo orange was selling 10,000 bikes with french style forks, it would be a different conversation. As it is now, we just can't snap our fingers, provide drawings, and wish them to be here in 3 months. i would not want to work with a supplier who could promise something like that anyway. those $70 widgets whipped up in 3 months from anonymous factories via alibaba give you no IP protection or quality assurance once it's on the boat. And forget about covering warranties when something does go wrong.

Hank G. said...

To David NJ:

Try and get a factory that produces thousands of forks a year (and works on narrow profit margins) to create new tooling and change it's production methods to satisfy your order of 300 forks without having to pay a stiff premium for the privilege. Low pricing comes from being able to leverage the numbers of what factories are geared to produce. You want something special? no problem at all... for a price.

Dad said...

Totally fair Hank. That being said, at least if the upcharge isn't ridiculous, it's worth it! For me it would be the difference between buying the bike and not. Without a correct fork bend, it's an also-ran; with one it's something special. Although I sure would like to see some centerpull brake bosses on there ...

Anonymous said...

"It's a matter of the tubing wall thickness and diameter. A tighter, lower bend requires a smaller diameter tube, which has to have thicker walls for the same strength. Is that correct, or is there more going on?"

I'm not sure any of that is true. Frk tubing tends to be thick in the narrow bits anyway, since that is what happens to the wider stock when it is tapered, the wall thickness picks up, so most fork blades are like that.

The lower bend can be made in any ocnceivable material, really, just a mater of the tooling.

I don't think lower curves look better, unlike boats, water pressure and drag isn't going to sort this one out. It is mostly just a looks thing.

Anonymous said...

The forks in all the 3 photos don't look good.

The first is the least worst, it has the bend lower down, which is good but then it's straight above the dropout, making it look like it's kinked.

The second and third are just ugly, typical of any modern forks.

It's quite simple; the bend should be a continous curve starting low down the fork blades, continuing all the way to the dropout without straightening out.