Frame Design: Then and Now

The above picture is me riding in the British National 12 Hour TT Championship in 1953; dating the picture is the spare tubular tire worn around my shoulders as the pros did in that era.

My bike frame size back then was 22 inch; (56cm.) compare that to the 51cm. I rode at the end of my racing career in the early 1980s, and still ride today. That is a whapping 5cm or 2 inches smaller.

If I rode a 56cm. frame today it would way too big for me, and yet looking at the above picture my bike looks fine and not too large at all. So what happened; did I shrink over the years? I was certainly a lot slimmer back in my youth, but my legs are pretty much the same length as they are now.

A clue is in the amount of seat post showing; about 2.5 inches (6.3cm.) in the above picture; I have 4.5 inches (11.4cm.) on my current bike. This accounts for the 2 inch (50cm.) difference in frame size.

The reason my 1953 bike does not appear too big for me is because it wasn’t; it was a totally different design than today’s frame. My bottom bracket height was only 9.25 inches; (23.5cm.) today’s bike has a BB height of 10.625 inches (27cm.)

I would also point out that all racing cyclists in the 1950s, including the pros, pedaled much lower gears, sat more upright, and rode with their saddles set lower by today’s standard.

See picture of Fausto Coppi on right.

I can remember that my bottom bracket was low enough that I could lower my heel and actually touch the ground.

Our cycling shoes had real leather soles, and had steel tips on the heels to prevent wear when walking.

While out training after dark, and coasting down hill; we would sometimes lower our heel so the steel tip made contact with the road, sending out a shower of sparks. A pretty spectacular visual effect, especially if several riders did it together.

When you lower the bottom bracket on a frame you also lower everything above it, the top tube and the height of the saddle from the ground. You do not necessarily lower the saddle in relation to the pedals. That will be whatever the rider sets it at.

However, the handlebars are not lowered by as much. The reason being that the size of the front wheel and therefore the length of the front fork are constant no matter what frame size. Above the front fork there is a head tube and head bearings. 

It would be impossible to build a 51cm. frame with a nine and a quarter inch BB height, because with a level top tube there would be no room for a head tube; which is why I rode a much larger frame back then. Or not so much a larger frame, but one with a lower BB and a longer seat tube.

Over the years the bottom bracket height on racing frames has increased; not because striking a pedal on the ground was an issue. (I never found it to be a problem.) But rather one of making the BB higher makes the chainstays and down tube shorter, and therefore makes a stiffer frame.

Also probably the main factor driving frame design is the change in riding positions of today’s racing cyclists, over those of their predecessors in the 1950s. I have already mentioned the 50s riders sat more upright because the handlebars were higher in relation to the saddle.

Today’s racing bicycle has a large saddle to handlebar height difference; which is how most racing frames sold today are designed. However, the majority of the frames are bought by non raciing leisure riders; using them purely for exercise and pleasure riding. Many of them like myself are older, and are not flexible enough to get down in those horizontal, low tuck racing positions.

Today’s frame design with its sloping top tube does not restrict a frame designer/builder like the level top tube did. No matter what the BB height and seat tube length, the head tube can be any length. So anyone having a frame built for leisure riding by an independent builder might consider lowering the BB height.

I did this when drawing up the specs for my New Fuso that Russ Denny is building for me. This is probably the last frame I will ever need. I designed it with an 8.5cm. drop; which is a 10 inch. (25.3cm.) BB height.  (Drawing below.)

This does two things; by sitting closer to the ground it will be easier to put my foot down when stopping. But more importantly, a lower bottom bracket means the saddle is lower in relation to both the ground and the handlebars.

Not the other way round of having the seat high to begin with, then raising the handlebars to achieve the desired position.

I have come to realize, the racing position of the 1950s is probably the ideal leisure riding position for today. I will have my frame built lug-less; (Welded.) this means there is no restriction on angles, and because the modern design has a sloping top tube it is no longer necessary that I go to a larger frame.

                         

Recognition

Someone asked me the other day, “Who built the bike from the 1950s with the double bend in the front fork blades?”

That was Bates an East London builder; it was called a Diadrant fork. (Left.)  It was introduced in the mid 1930s and remained popular into the 1950s.

The next question was. “What was the advantage?” None that I can see, although I’m sure Bates claimed there was. Designs like this were done for recognition.

“Cycling” was the main publication for the sport in the UK and a picture of a top rider in this weekly magazine on a certain bike was very good for business.

When you saw a picture of a bike with this distinctive front fork, or even when you saw one on the road, you instantly knew it was a Bates.

Hetchins, another London builder had their famous “Curly” stays (Picture right.) for the same reason; it was instantly recognizable.

An interesting story I first heard back in the 1950s.

It concerned an English rider competing in a road race in France, on his Curly Stay Hetchins.

He crashed and was rendered unconscious for a few minutes; when he came around, he found some local French farmworkers trying to straighten his bike.

I doubt this incident actually happened, the story became one of those urban legends and everyone claimed to know someone who it had actually happened to.

I do know this bike was a source of amusement for the French cycling establishment. I remember in the 1950s seeing a picture of a Curly Stay Hetchins in l’Equipe a famous French sports paper that always covered the Tour de France. I didn’t get the exact translation of the caption under the photo, but it mentioned something about “Queen Anne Legs.”

Another trend of that same era was the short wheelbase frame, or rather a short rear end with short chainstays. The idea was to make a stiff and more responsive ride, but if you overshorten the chainstays the rear wheel touches the seat tube.

To overcome this brought about some very interesting frame designs. Probably the most famous is the Bains “Flying Gate.” (Below.) Actually its official name at first was the ”Whirlwind,” but was nicknamed the Flying Gate by cyclists; the name stuck and later became the new official name.

First built in the 1930s, Bains ended production in 1953, but in the late 1970s the design was resurrected by Trevor Jarvis a Burton on Trent builder. The frame can still be ordered today from T.J. Cycles.

Another design was the Saxon Twin Tube. (Picture below.) That deraillier by the way is called an Osgear.

Jack Taylor Cycles achieved the same ends with a curved seat tube. (Picture left.)

One of my favorites was the Paris “Galibier,” an interesting cantilever design with a large main strut in the center of the frame and small diameter twin tubes at the top. (Picture below.)

The Galibier frame construction method is known as Bilaminated or Bilaminates. Steel sleeves cut in fancy shapes are brazed over the tubes, and the actual joint is then made by a fillet of brass (Fillet brazing.) It has the finished appearance of a lug but it is not really a lug. (Pictures below.)

Each of these frame designs were distinctive and all instantly recognizable; I'm not sure if every builder patented their design, but I'm sure each claimed a definite advantage over all others.

The shortened chainstays were popular in the 1930s and 1940s when most British club riders and time-trialists used a single fixed wheel. If derailleur’s were used all that was available was a single chainwheel and three or four sprockets on the rear.

By the 1950s gearing had advanced to double chainrings and five speed freewheels; this type of frame without the conventional seat tube made the fitting of a front derailleur difficult or impossible. The trend died a natural death.

Today these bikes are collectable and make interesting conversation pieces. They came from an era when there was much competition amongst the many framebuilders; each was clamoring for their own little piece of the market.  

One way to stand out in a crowd was to build something different and distinctive. Bike riders have always looked for an edge, and a different design could be claimed as beneficial to the rider. But recognition was the main objective; somthing that would be instantly identified as a particular brand.

I previously wrote about Paris Cycles here

I touched briefly on the various framebuilders here; you can read more and view pictures of these and other classics on ClassicLightweights.co.uk

                        

Fractals and Framebuilding

I watched a program on Tuesday evening on my local Public TV Station. It was called “Hunting the Hidden Dimension,” and was on the Nova series of programs.

The subject of the broadcast was “Fractal Geometry.” (First aired in October 2008.)

Fractal is a word coined by mathematician Benoit Mandelbrot (Left.) who published his findings as recent as the 1970s.

Up until that point, throughout history geometry was all about straight lines, triangles, pyramids, circles and cylinders, etc.

This geometry applied to everything man-made, buildings, bridges and other engineering projects, and of course bicycle frames fit right into this category, being made up of straight lines and triangles. Up until Mandelbrot’s findings, everything outside of that which is man-made could not be explained by geometry and mathematics.

Mandelbrot changed all that when he discovered that shapes and forms in nature. For example cloud formations, trees, mountains, river flow and even blood vessels in our bodies, were a series of repeating irregular shapes, which could be recreated and proven mathematically.

Benoit Mandelbrot was at first scoffed at by other experts, but now with improvements in computing his theories are being proved mathematically. Now taken seriously, fractals is being studied and used in the medical and the environmental fields.

One of the places we also see fractal geometry in practice is in special effects for movies. Not only are images of nature being created, like landscapes, plants, trees, and even human and animal forms, but fire and explosions can be created and animated on a computer. 

This subject has always fascinated me, because on the one hand you have the bicycle which is traditional man-made geometry, and the task of the frame builder is to match that to the human body, so the two become one, and the bicycle becomes an extension of its rider.

The human body, like all forms in nature, appears to be chaotic and infinite in its makeup. Yet it was possible for me to build a series of production frames, the Fuso for example, in a range of sizes that would fit just about anybody.

From as far back as the late 1960s I found I could fit someone to a frame “Intuitively.”  I did not let this be widely known for fear of being labeled a crack-pot. I was basing my estimation of frame size primarily on a person’s height.

Long after I left the bike business, and therefore the effect of the “Crackpot” label had diminished, I wrote an article here in February 2006, stating that frame size could be estimated around a person’s height.

I came to this conclusion, not so much by what I could do in sizing a person, but more by what I could not do. A person who is six feet tall would normally fit on a frame around 58 or 59 centimeters (Measured center to top.)

However, it is quite a common occurrence to find a person six feet tall (183 cm.) with a 30 inch (76 cm.) inside leg measurement. You cannot put a person like that on a 51 centimeter frame as his inseam would suggest then build a long top tube to accommodate his long body.

I would simply drop the frame size down to a 56 or 57 centimeter because of the short legs, and leave the top tube as standard for that size frame. (55cm. or 55.5 respectively.) This same frame would also suit a person 5’10” tall, (178 cm.) with an inseam around 33 inches.(84 cm.) The difference being the taller guy with short legs would have his saddle lower and possibly use a longer stem.

I knew this was so, but never knew why; it all became clear to me on watching the Nova episode. Towards the end of the program a group of environmentalists were studying rainforests. They cut down a large tree, then measured and documented the dimensions of all its branches, overall height etc.

They then found that a seemingly random pattern of trees of all sizes growing throughout the rest of the rainforest followed the same pattern as the branches of the one tree they had documented; both in the position of their branches, and their position in the forest relative to other trees.

Watching this, it occurred to me that if you took a large group of humans all the same height, you could fit them all to the same size bicycle frame. (Within a centimeter or so.) This is why this theory works; although on the surface it appears that my group all the same height are each different in every other way, they are no different than the trees in the rainforest. They all follow the rule of fractals that can be plotted mathematically.

Just a little something for you to think about, to agree or shoot down. If you haven’t seen the hour long program, you can watch it here. It is fascinating stuff.

                          

Building small frames

Frame design, in many ways, was much simpler when I was building back in the 1980s.

Top tubes were always level; it was not acceptable, to me or my customers, to build a frame with a sloping top tube.

Once a rider had established his correct frame size, he set the handlebar stem about 4 or 5 centimeters above the head bearings.

Then he set his saddle to the correct height and he was for the most part, good to go.

The handlebar stem could be adjusted a centimeter up or down as the rider wished.

The other point that made everything simpler was the fact that a person could go buy a frame of any make, in the same size, and the seat to handlebar height ratio would be the same.

If the top tube length varied slightly it could be corrected with a longer or shorter stem. Handlebar drop never even entered into the equation because it was automatic once you had the frame size right. The level top tube was in fact a point of reference.

Many years before I started riding, the wheel size for a racing bicycle was set at 27 inches diameter, or 700c as it is known today. This means that there is a fixed distance from the ground to the bottom of the head tube. This is always the same for any size frame with 700c wheels.

On a level top tube frame, it doesn’t matter that the builder changes the bottom bracket height. If he raises it, and the seat tube length remains the same, then he also raises the top tube and the head tube becomes longer. This is because the bottom of the head tube remains in a fixed position.

The rider’s saddle height is measured from the pedals (Or BB center.) to the top of the saddle. So although the rider is sitting higher because of the high bottom bracket, because the head tube has become longer by an equal amount, the seat to handlebar height difference always remains the same for any given size.

Each frame size will have its own saddle to handlebar height difference, which increases as frames become larger, decreases for the smaller sizes. On a modern sloping top tube frame, raising or lowering the bottom braket height will not necesarily alter the head tube length.

With today’s design, the bottom of the head tube is still in a fixed position, but the top of the head tube can be anywhere; it is not governed by a level top tube as it used to be. There seems to be no standard point of reference between the different manufacturers.

Where today’s design has an advantage it is in building very small frames. The bike pictured top left, is my personal bike; it is a 51 centimeter. (Center to top.) You can see in the picture that if the position of the bottom head lug is fixed, a framebuilder can only lower the top tube another 2 cm. and the lugs merge. To all practical purposes a 49 cm. is the smallest level top tube frame he can build.

The only way to go smaller is to shorten the seat tube by raising the bottom bracket. This really goes against the requirements of the rider, because the last thing a person with short legs needs is to be higher from the ground.

I was asked just this week how would I go about designing frames for women. If the woman was 5’ 4” or taller it was no problem; I would just build according to the customers measurements as I would for a man.

If the woman was less than 5’ 4” then it was not so much a case of building a frame to fit, but one of how small can I build this frame? There is not only a limit to how short can I make the seat tube, but there is a limit the how short one can make a top tube.

The whole problem is fitting two large wheels into a frame that has reached the limit for those size wheels; it restricts what you can do. Smaller wheels are available but rims and tires are limited to a much narrower choice than for the standard 700c.

Frames are now sized like tee-shirts; extra-small, small, medium, and large. My advice to female under 5’ 4” would be to buy a frame in the smallest size possible. There will be no problem with the seat tube length, and there is now a far wider range of handlebar stem lengths and angles than were ever available in my day.

With this wider range of handlebar stems, it should be possible for most riders to dial in a near perfect position. The only problem as I see it is that another dimension has been added to the equation, and that is handlebar drop.

In the old days one only had to concern themselves with saddle height and reach, drop took care of itself with the correct size frame.

I always maintained that a rider’s arms (On the drops.) should be in direct opposition to the legs. Choose a combination of drop and reach that will achieve this. 

Back in 2007 I wrote an article which included a chart that gave a drop measurement and was based on my old fit philosophy from the days when top tubes were level. Some people have found it useful.

Your comments and input as always are appreciated.

                         

Why Disc Brakes?

I fail to see the point of a disc brake on a road bike. A conventional caliper brake is a disc brake, with the rim being the disc.

The rim is the largest possible diameter and therefore has the most leverage when the braking force is applied.

With a disc at the hub the braking forces are transmitted through the spokes. What is more the driving stress is in one direction, followed by the braking stress in the opposite direction. Stressing spokes in this manner could possibly lead to breakage or coming loose.

With a normal caliper brake acting on the rim there is little or no stress on the spokes, (Apart from the normal weight bearing stresses.) the stress is directly from the rim to the tire.

A disc brake means you can’t have radial spokes. Not only must there be crossed tangential spokes on the non drive side of the rear wheel, you must also have tangential spokes on the front wheel.

Tangential spokes are longer, therefore more weight; remember also this is spinning weight which is more important. A radial spoke wheel has marginally less wind resistance.

The disc and its caliper has to be heavier than the conventional brake, and more expensive to produce.

I can see an argument for a disc brake on a mountain bike or a cyclo-cross bike, because it would possibly collect less mud than a brake that operates near the tire.

But on a road bike; I just don’t get it. Why make something more complicated than it need be.

What is your view?

Update 11/17/10

An interesting piece by a commuter from Minnesota who rides through the winter in the ice and snow

More about the bike and its brake here