Fully Integrated Design

Car drivers are lucky: They can buy sports cars that are fully equipped to be driven in the rain and even at night. And their fenders rarely rattle loose, their lights don’t fall off, and most car owners think little of driving 1200 km without having to tighten bolts or do other maintenance. The secret to this reliability and performance is a fully integrated design.

It wasn’t always like that. In the early days of the automobile, car makers provided just the chassis. Below is a brand-new Bugatti sports car ready for delivery.

Of course, the owner didn’t drive it like that, but took it to a body builder, where a body and all the other accessories were added. The owner got to choose the body style, the fender shape, lights and many other details. The resulting cars could be breathtakingly beautiful and elegant.

For the time, they offered great performance, too. Equipped with a 3.3 liter compressor engine, this Bugatti Type 57 could go 180 km/h (112 mph), and few cars could keep up with it on the open road during the late 1930s.

Fast forward 12 years, and suddenly you had a new breed of sports cars with smaller engines and much fewer horsepower that were not just able to keep up with the Bugatti, but outperform it with ease on a curving road.

What had happened? The Lancia above no longer used a separate chassis, but featured unibody construction. The body was the load-bearing structure. The fenders were integral to the design, and so were the lights. Not only was this structure lighter and more rigid, but it also greatly improved the reliability of the car. Gone were the days when fenders rattled loose, lights vibrated until their bulbs broke, and bodies cracked because they flexed independently of the chassis. Everything was conceived as a unit, everything worked together, and the result was a car with modern performance.

In the bike world, we are still stuck in the 1930s. If you buy a “real-world” performance bike at most bike stores, you get the equivalent of a rolling chassis, whether it’s a cyclocross, a touring or a “randonneur” bike. Here is an example:

The manufacturer expects you to add the various parts you need to make the suitable for real-world riding. The result usually looks something like this:

This bike has all the drawbacks of a 1930s cars and then some: The many clamps and adjustable sliders add a lot of weight. They are likely to come loose. You are bound to get rattles and resonances from the fenders. Lighting wires are zip-tied to the outside of the frame, where they are vulnerable. And the resulting bike doesn’t even look as elegant as a 1930s Bugatti. No wonder many cyclists prefer to ride just the rolling chassis, rather than deal with all the clamped-on “accessories.”

Now imagine a bike where all the parts you need have been integrated into the original design. For a loaded touring bike, it would look something like this bike from our book The Golden Age of Handbuilt Bicycles (as always, click on the photo for a more detailed view):

The racks are custom-made for the frame. They fit exactly, with no adjustments necessary or even possible. That makes them lighter and above all stiffer, which improves the handling of the bike. It also means that they are unlikely to need tightening – ever.  The fenders attach directly to the frame, without spacers or tabs, providing a more solid attachment that is unlikely to flex,  resonate or crack. The lights are integrated into the rack (front) and fender (rear).

In the photo at the top of the post, you see a detail of such a fully integrated bike. Note how the fender attaches directly to the frame, with a reinforcement at the spot where the stresses are greatest. See the remote control lever for the generator, so you don’t need to stop if you want to turn on the lights. The fender has a small bulge to make room for the generator’s wheel. The lighting wires run inside the fenders and frame. This is not only prettier, but also reduces the risk of the wires getting snagged and broken.

The result is a lighter, stronger, more reliable and more elegant bike. It also offers better performance. It’s the bicycle equivalent of a modern car. Once you have experienced a fully integrated bicycle, it is hard to go back to the equivalent of a 1930s car.

About Jan Heine, Editor, Bicycle Quarterly

I love cycling and bicycles, especially those that take us off the beaten path. I edit Bicycle Quarterly magazine, and occasionally write for other publications. One of our companies, Bicycle Quarterly Press publishes cycling books, while Compass Bicycles Ltd. makes and distributes high-quality bicycle components for real-world riders.
This entry was posted in Our Bikes, Testing and Tech. Bookmark the permalink.

25 Responses to Fully Integrated Design

  1. That’s a beautiful bike and a great example for integration done well.

    I’m somewhat ambivalent about internal wire routing. Aesthetically invisible cables are certainly a plus and the added protection is nice. On the other hand, this comes at the cost of serviceability. If you do have to replace a cable, the in-frame routing can be a big pain. As you probably know, a good proportion of the German street legal “trekking” and “city” bikes had and continue to have internal cable routing, but in my experience they all inevitably fail at some point and then are a pain to fix, especially on the road. In addition, the holes necessary for the cable can be a problem for frame strength (see here, for example).

    The other issue with integrated designs is customization. As long as the bike’s specification work for you that’s great, but if you have different needs or preferences fully integrated solutions will create difficulties.

    For a more modern example of an integrated touring design (which the BQ readership will probably hate :-) is the Riese und Müller Delite.

    • I agree that internal wires done poorly can be a reliability problem. However, done well, they should not require maintenance. I have ridden a number of Rene Herse and Alex Singer bikes that were 50+ years old, and the lights worked perfectly, without anybody touching them. In my car, the wiring hasn’t needed repair in the almost 20 years it’s been on the road, either.

      Poor integration is worse than none – whether it’s cars or bikes. Unfortunately, the German “trekking” bikes I have seen fall into this category. In most cases, the wires are only partially internal, and exit in the worst spots, where they are likely to get snagged. Integration done well is a wonderful thing.

      You are right about customization: The Lancia in the blog post came ready-made. If you wanted a different body (or just different fender shapes), you were out of luck. Well, you’d just sell the car you had, and bought another that better met your needs. I find that with bicycles, it’s often easier to start from scratch than trying to rig up a bike for something that it wasn’t intended to do.

  2. Brian Campbell says:

    So when will Compass be offering a fully integrated bicycle for sale?

  3. Richard James says:

    How about publishing a constructeur framebuilding guide or series of articles? At least to me, the fittings (placement, purpose, etc.) and design considerations (decisions made before mitering the first tube, e.g. combining narrow cranks and wide tires, effective positioning of bridges, fork heights, etc.) are the distinguishing characteristics of such a bicycle. Neither of these areas is magic or requires much knowledge of the rider or their riding style.

    The Kogswell blueprints floating around the internet are particularly useful in this regard, but given BQ’s thorough treatment I think this could be a great subject.

    • We have published a good number of articles on subjects like how to integrate a front rack into a bike to how to mount fenders… You can find a list at here. We used to publish the specs on bridge spacing for our test bikes, but we found that the exact details depend on so many variables that I now prefer builders to figure it out themselves. Once they have built a few fully integrated bikes, they’ll know exactly what to do.

  4. Erik says:

    I understand the point you are trying to make, but I think the example is only the first step: this is not yet integrated design, this is still the old stuff made better: several parts put together but in a brighter way than normal. Just take a look at the Greenmachine by Flevobike (http://www.flevobike.nl/en/products/greenmachine) which, I believe, is a perfect example of modern integrated design: the hollow frame protects the drivetrain from dirt and dust. The several components are integrated in such a way that the whole bike appears to be one component. I haven’t ridden one (and I can’t afford one) but it certainly looks like one step in the good direction.

    • AC says:

      Wow, that’s a pretty amazing design. I’d love to see more outside-the-box thinking like this in “conventional” bicycles as well.

  5. jimmy says:

    Jan, how do you feel about racks directly brazed to frames? It seems like they would remove even more unnecessary hardware from the total integrated design. Bottle cages are another candidate for direct brazing to the frame. Depending on frame and fender material, they could be integrated together, too. We already have integrated headsets, where the frame itself provides a surface for the cartridge bearing, though I’m not sure if many custom integrated rando builders are using them.

    • Some constructeurs brazed their racks to the frames in the 1930s – see this Reyhand, for example. The constructeurs stopped doing this. I suspect this was because racks can get damaged, and it’s nice if you can replace one without having to repaint or rechrome the frame. A well-made custom rack doesn’t come loose, and the four bolts weigh very little, so there is no real advantage to brazing the rack to the frame. The same applies to bottle cages – they eventually break, and then it’s nice to just bolt on a new one. Bottle cages are a nice example of integrated design: They used to be clamped to the tubes, but today, almost every bike has dedicated braze-ons for the bottle cages. The spacing of the bolts is standardized, so you can easily fit any cage to any bike. Integrated headsets are interesting, but I am more enthusiastic about integrated bottom brackets. The new BB30 standard is pretty much the same as the old René Herse bottom brackets, with bearings pressed straight into a special BB shell.

  6. I think you pose an interesting question about “integration.” What makes something count as integrated? Being brazed on? Lack of all mounting hardware but screws? Dedicated and standardized braze-ons? Being part of the frame?
    What I would be interested in: integrated lighting. With the current state of LED technology it should be possible to integrate at least rear lights into the seat stays, eliminating the problem of theft and damage. It would be great to also do it in the front, but I don’t really see how it could work there, because of the size of the reflector. But if someone could make it work it would be awesome. I think the Pereira bike recently reviewed in BQ showed some interesting ideas about integration, with the rear light and the integrated lock.

    • To me, integrated means that the components are considered as part of the initial design, rather than being added later as a retrofit. It also means that they fit together in the best possible way. This means that sliders, clamps, spacers and other “crutches” are out of the question. It doesn’t mean that you have to braze or glue everything together as a one-piece unit. On my car, I can remove the headlights with three screws if they break, but that makes them no less integrated into the overall design.

  7. superfreak says:

    u use a bike with a bottle gen as an example of integrated design? i always thot botte gens were so mickey mouse but mebbe cause i only used cheap ones. bottle gens seem like such an after thought. no car would ever run the altenator off the tire.
    thx superfreak

    • I agree, generator hubs are much better than bottle generators. The French randonneurs talked about generator hubs in the 1940s, hoping that they soon would be available. Alas, it took 50 years to get them…

      • Erik says:

        I’d like to add that Raleigh started commercial production of their Dynohubs in 1936 until the 1980’s. The dynohub had several problems: weight and limited power being the most obvious. However they were very reliable and offered a great alternative to the bottle generators. It is unclear to me why no other producer copied and improved the Raleigh design.

      • I believe that the rare earth magnets needed to make a small and powerful alternator (which is what a generator hub really is) were very, very expensive until recent decades. Even cars only switched from dynamos to alternators during the 1960s, and they aren’t as constrained by concerns about weight and size. A switchable dynamo hub has a number of problems, most importantly sealing the mechanism and wear on the clutch.

        I also believe that a big issue was a lack of market. When the French cyclotouring magazines were predicting in the late 1940s that soon there would be powerful generator hubs, nobody considered that the market for high-end bicycles would collapse just a few years later, which effectively ended most development of non-racing parts.

  8. The non-chromed Singer above features tubing that follows the line a mixte frame would have. Do you find that this helps with loads noticeable more than a touring bike without them? How does this affect the bike ability to ‘plane’? Most frame builders I talk to suggest that over-sized tubing is more desirable for larger frames and greater loads–how does this compare? Lots of questions, sorry…

    • Camping bikes often featured the extra triangulation with two long stays running from the head tube to the rear. Tandems used similar configurations.

      I haven’t tested bikes with and without the triangulation, back-to-back. The Singer is very stiff and handles very well no matter the load. Better than my Mercian that had similar frame tubing without the extra stays, but the Mercian was most handicapped by the flexible Blackburn racks. So it’s hard to say whether the stays are needed or not.

      The Singer is not very nice to ride unloaded, as it’s harsh and unresponsive. However, with a load of 35 lb, it comes to life. I climbed Old Blewett Pass with it one day, and it was “planing” wonderfully. I haven’t had a bike from OS tubing work that well, but again, that doesn’t mean it cannot be done. Many questions, few answers. The solution may be to build a frame without triangulation, test it, then add the triangulation, test it again, and perhaps cut the extra stays out again to re-test. Then we might know. Since loaded touring and camping bikes are favorites of mine, we may well do this some day.

      • I have noticed that Rivendell’s “mountain bike,” the Bombadil, is very similar (at first glance, at least) to the Singer camper you have pictured above. The larger sizes have the extra stiffener from the headtube to rear dropout. It also is sized for 650b. I would guess that the main frame tubes are stiffer, though. I wonder if it would be a good camper for rough forest roads, which as we all know are the very best place to go bicycle camping!

      • The latest Bombadil uses a single tube in the main triangle, but otherwise is similar. It’s an interesting design that may have originated with French women’s bikes from Reyhand, later copied by others.

        The twin-tube mixte, without a top tube, tends to be very flexible, because the tubes’ diameter is small. So Reyhand put a bigger single tube in the main triangle, and then braced that with two extra seat stays. After the war, most constructeurs copied this style.

        On a bike with a top tube, it’s not clear whether adding more tubes in the same plane does much good. The front triangle remains two-dimensional. I have the same concerns about modern tandems… The answer probably lies in making a finite element analysis model and comparing the frames’ stiffness in various directions.

  9. TS WU says:

    This is a wonderfully provocative article- thank you Jan for your care in thinking about this. I suppose one could argue (as someone seemed to in an earlier post that I no longer see here…??) that the MTB is the other stark choice offered to American cyclists, as if to say our choices are either a Formula 1 race car or a Humvee.

    What kind of mindset do bike makers have to have in order to start building in other features for a more complete product? One issue is powerplant. Unlike autos, the bicycle powerplant isn’t improving as everything else is (um, drugs notwithstanding). So there will always be a stark tradeoff between additional product features that add weight (fenders, lighting, racks, etc.) and total weight. Much less so in the auto world as car engines became more powerful.

    Are there other examples of important product evolution in the face of a hard, unchanging parameter? Laptops are an example: everyone wants a laptop to be light and long lasting in battery operation. All product design is restrained by these two parameters, and, like the rest of the computer industry, we’ve seen these improve in their power and performance while getting lighter and using less power.

    This would mean that if we want more complete bicycles- fully integrated as you say-, given that the human powerplant isn’t changing much, they’ll need to be made of yet lighter and lighter materials.

    Yet there must be a role played by the infrastructural context of the products as well. Ironically, it was early cyclist that led the movement for better pavement, making life much easier for cars. I currently use a B-cycle at work, and while it’s heavy and clunky and never a bike I’d want to hoist on top of my roof rack, b/c it is well integrated to its pickup/ dropoff stations, it works really well for what it’s meant to do. I’m not exactly sure how the general bicycle infrastructure would change to make fully integrated design more likely, but there may be some role played by it.

    • The performance of a bike with well-integrated fenders and lights is the same as that of a bike without. A set of fenders weighs about 2/3 of a full water bottle. Lights add even less weight. Few cyclists notice the difference in performance whether their bottles are full or half-empty. In our testing, we consistently have found that a 3% increase in weight (of the bike/rider/luggage total) doesn’t make a significant difference in performance. Other factors, such as the flex characteristics of the frame, make a much, much bigger difference.

      I can see two reasons the bike industry pushes bikes without fenders and lights:
      1. They are easy to make. Spec’ing, making, assembling and shipping a racing bike is easy. You don’t need to have clearances that are spot-on; you don’t need to worry about the whole integration; and the bike easily fits into a UPS-sized box.
      2. They are easy to sell. The public equates “no fenders” with “fast.”

      The car industry has been down the same path with truck-based SUVs. Easy to make, easy to sell. American car makers were champions of that approach, but it turned into something of a dead-end. The number one vehicle destroyed under the “Cash for Clunkers” program, was the mid-1990s Ford Explorer, then the most popular truck-based SUV. 83% of the vehicles traded in were trucks. (See the story here.) The resale value of racing bikes also is much lower than that of bikes that can be used for “real-world” riding.

      As far as infrastructure goes, infrastructure has adopted to cars, but cars also have adopted to the infrastructure. It would be laughable if today’s cars couldn’t go across streetcar tracks or ride through a small pothole without endangering the driver. There are cars that cannot cross speedbumps (older Lamborghinis and such), but they aren’t very common, and if you rip off your Lamborghini’s front airdam, you won’t get much sympathy. (Newer Lamborghinis have a lift system that allow you to raise the car for speed bumps.) The racing bikes many people ride are poorly adapted to the infrastructure, and there have been many injuries as a result.

  10. Henry G says:

    For the recreational weekend warrior the ability to drive the equivalent of an F1 race car on public roads actually sounds really attractive. However this sort of bike should occupy the same niche as a Lamborghini or a stripped down Porsche. Even these integrated and adapted for the public roads performance cars are not considered practical transport for the majority of drivers.

    In the USA bicycles are not considered a serious mode of transportation so practical considerations carry little weight. As we see more cyclists using bikes as an alternative to cars you will see more practical and better adapted to tramsport bike designs.

    • The idea of the Formula 1 racecar is appealing, but in practice, you wouldn’t enjoy driving it on a real road like this one. Formula 1 race cars are designed for smooth racecourses, and don’t have enough suspension travel for real roads.

      Racing bikes similarly are designed for very smooth roads. For real roads, the equivalent of your Lamborghini or stripped-down Porsche is a better choice. And those cars have lights, fenders and even can carry a few things in their trunks. They are the automobile equivalent of randonneur bikes. (The station wagon is the equivalent of an urban bike.)

Comments are closed.