expensive road bikes fussy and unreliable?

Discussion in 'Cycling Equipment' started by Wle, Oct 27, 2003.

Thread Status:
Not open for further replies.
  1. x

    x Guest

    RE/
    >Is it really, when you consider all the parts of a deraileur drive train (hub, freehub, cassette,
    >rear mech, front mech, extra rings, extra chain, extra changer)? If so, by how much?

    On my bike, almost *exactly* two pounds (like 1.97 or something....)

    That's bottom-line: bike before vs bike after. The der system it replaced was SRAM 9.0.
    --
    PeteCresswell
     


  2. Matt O'Toole

    Matt O'Toole Guest

    "Simon Brooke" <[email protected]> wrote in message
    news:[email protected]...

    > I've been thinking and doodling about ideas which are based on the Nicolai Nucleon frames, which
    > have the Rohloff not at the rear hub but at the suspension pivot. Lots of advantages, starting
    > with lower unsprung weight and less vulnerability to getting bashed on rocks and so on. If you do
    > that, and you manage to make the suspension pivot perfectly coaxial with the output sprocket[1],
    > then you can have a box section swing arm which can double as a chain case (you could even run the
    > chain in an oil bath, but certainly you could protect it from dirt). If you did that, then it
    > isn't much more difficult to make the box section arm sufficiently stiff that you don't need an
    > arm on the other side... so now you've got a monoblade rear fork, and you don't have to remove the
    > rear wheel to change the tyre.
    >
    > Finally, you permanently assemble the stub axle, splined inner hub and disk brake to the swing
    > arm, and lace the wheel onto a splined outer hub. The splined outer hub slips onto the splined
    > inner hub and is retained by a washer and split pin, or some other fairly simple mechanism. Hey
    > presto, a wheel which is even quicker and easier to change than a conventional bicycle wheel.
    > Don't have to release or adjust the brakes. Don't have to mess with the chain or the gearing. Pull
    > out one pin, pull off the wheel. Push the wheel (or another wheel) back on, drop in the pin,
    > spread the ends of the pin, ride.
    >
    > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't
    > have to be epicyclic - in fact, it's easier to design and construct if it isn't. A much simple,
    > more conventional, motorcycle type gearbox layout will do. So the real question in my mind
    > boils down to, how light is it possible to build a gearbox which will survive and transmit the
    > 200 odd Nm of torque that a cyclist puts out. If Rohloff can do an epicyclic at 1.7Kg that's a
    > benchmark to aim at.

    I've thought of similar designs, and I'm sure many others have too. Frankly, I'm amazed that no one
    has tried it yet, especially with composites.

    I don't think the unsprung weight is an issue. It's not like a car or motorcycle, with big masses
    reciprocating at potentially high velocity/frequency, due to high vechicle speeds. There isn't that
    much kinetic energy to keep under control. Downhill racing bikes have big, heavy tires and go very
    fast, and it doesn't seem to be a problem there.

    But I like the idea of a completely internal drivetrain, with the chaincase a structural part of the
    frame. And the stub axle thing too...

    I haven't seen the Nicoli frames you're referring to. But wouldn't they have more drivetrain
    friction, from the extra chain?

    Also, though it might seem correct at first glance, suspension pivots usually should not be
    concentric to the drive sprockets. Draw a free body diagram of the forces at the rear axle (most of
    us get it wrong the first few times) and you'll see what I mean. That's why there are so many
    monkey-motion suspension designs.

    Other than these few nits, it's a great idea.

    One more thing to throw into the mix -- have you seen that 2 speed epicyclic crank? I forget who
    makes it, maybe Rohloff too.

    Also, with composites, a whole conventional drivetrain could probably fit inside the "chainstay." It
    would definately look unconventional, but it would be doable.

    Matt O.
     
  3. Jay Beattie

    Jay Beattie Guest

    "David Damerell" <[email protected]> wrote in message
    news:cNl*[email protected]...
    > wle <[email protected]> wrote:
    > >i;ve seen it in here, "commuting on a racing bike is hard duty". but, if you are not really
    > >pounding the pedals, it seems to me that the only hard thing is mileage.
    >
    > And luggage; you're usually carrying quite a bit of junk. Other people have discussed the road
    > surface issues, but as well as work equipment
    and
    > clothes I carry tools and suchlike because there's no way to call for assistance if I have a
    > breakdown.
    >
    > >but if you start spending over a certain amount, maybe 1000 or so, the focus starts being on
    > >weight reduction, exclusively. it seems reliability is not the main thing any more, over a
    > >certain price. so, under 22 lbs, last at least 25,000 miles,
    >
    > "under 22 lbs" and "reliability" aren't necessarily working towards
    the
    > same goal, y'know. For instance, I got fed up of breaking cheap
    aluminium
    > luggage racks, and got a steel one. It's heavier, but it hasn't broken (yet :). Got fed up of
    > battery lights flaking out? Then you want a
    dynamo
    > hub - but that's heavier. Fed up of drivetrain problems? Then you want
    a
    > Rohloff - but that's heavier. Don't want your laptop bag getting wet
    (a
    > heavy object in and of itself)? Then you want Ortlieb luggage - but,
    guess
    > what, it's heavier...

    I've been commuting for over 30 years on every conceivable kind of bike -- even on a fussy racing
    bike when the weather is good. There is no need for a special commuting bike. Any bike that will
    accept fenders is good enough (not even a requirement if you live in the Southwest). If you really
    want to carry a lot of stuff in panniers -- then get a rack.

    My first-line commuting bike is a 1986 Cannondale Black Lightning former racing bike with futzed-on
    fenders, flashers, headlight, 23mm slicks (because 25s don't fit). Sometimes I put a rack on it (I
    drilled and tapped the drop outs). This is a great bike that is durable, but not so heavy that I
    dread going home over the hills. Sometimes I ride my touring bike, which does allow me to use
    panniers and gives me bigger tires for those really wet days, but it is a slug through the hills.

    Neither bike has had unusual drive train problems or comfort problems -- or any problems that
    require an exotic solution (e.g. internal gearing, special wheels or tires, sling saddles, cross
    frame, etc., etc.). My usual problems involve lubrication because I only do maintenance on my
    commuting bike about twice a year, even though half that time is spent riding in the rain. But even
    that is not a problem because I own an entire case of 30 weight, and I am not afraid to use it. I
    used to ride a fixed gear, but I got tired of flailing my legs on the down hills . .
    . . but that is another thread. -- Jay Beattie.
     
  4. Jay Beattie <[email protected]> wrote:
    : My first-line commuting bike is a 1986 Cannondale Black Lightning former racing bike with
    : futzed-on fenders, flashers, headlight, 23mm slicks (because 25s don't fit).

    damn.

    my 1986 cannondale black lightning road bike ..

    http://www.visi.com/~reuteler/bicycle_cannondale.html

    sniffle, was stolen august 10 of 1996 from in front of a bar in northeast minneapolis. i commuted
    through several minneapolis winters (scrubbed off the al oxide every spring!) on that bike. a great
    commuter bike to be sure. the only good thing about losing that bike was that my next bike was my
    first fixed gear. i get pretty nostalgic about that bike, tho. owned it for 7 years.
    --
    david reuteler [email protected]
     
  5. Carl Fogel

    Carl Fogel Guest

    Simon Brooke <[email protected]> wrote in message
    news:<[email protected]>...
    > "Matt O'Toole" <[email protected]> writes:
    >
    > > "Antti Salonen" <[email protected]> wrote in message
    > > news:[email protected]...
    > >
    > > > Simon Brooke <[email protected]> wrote:
    > > >
    > > > > Is it really, when you consider all the parts of a deraileur drive train (hub, freehub,
    > > > > cassette, rear mech, front mech, extra rings, extra chain, extra changer)? If so, by how
    > > > > much?
    > > >
    > > > I'm geeky enough to like these kind of questions, so I did the math:
    > > >
    > > > The hub itself is 1700 g (cross-country version). I'm not sure if that includes the bolts. In
    > > > addition you need the crankset with one chainring (Deore XT, about 760 grams), and some chain
    > > > (about 250 grams). Depending on the frame you might need a chain tensioner, but I'm assuming
    > > > that you don't. Then there's the twist shifter which probably weighs about a hundred grams
    > > > plus cables. The Rohloff hub uses two cables just like a conventional setup, so I'm ignoring
    > > > cables and casings in both cases. Total: ~2800 g
    > > >
    > > > Then a conventional Deore XT setup: A rear hub (370 g), cranks (860 g), cassette (264 g), rear
    > > > derailer (232 g), front derailer (165 g), chain (300 g) and shifters (Deore LX, 240 g). Total:
    > > > 2430 g
    > > >
    > > > The Rohloff setup would be more expensive, so for the same budget you could afford some XTR
    > > > parts and the weight difference would be greater, but it's still not going to be significant.
    > > > More relevant is the fact that you can go cheaper than XT without any significant penalty in
    > > > performance, in which case the Rohloff setup would easily be about $500 more expensive.
    > >
    > > So in other words, it's about a pound (for us Americans).
    > >
    > > One thing to consider is the lack of a quick release with the Rohloff, and that
    > > removing/remounting the wheel is a bit harder.
    >
    > Yehbut...
    >
    > I've been thinking and doodling about ideas which are based on the Nicolai Nucleon frames, which
    > have the Rohloff not at the rear hub but at the suspension pivot. Lots of advantages, starting
    > with lower unsprung weight and less vulnerability to getting bashed on rocks and so on. If you do
    > that, and you manage to make the suspension pivot perfectly coaxial with the output sprocket[1],
    > then you can have a box section swing arm which can double as a chain case (you could even run the
    > chain in an oil bath, but certainly you could protect it from dirt). If you did that, then it
    > isn't much more difficult to make the box section arm sufficiently stiff that you don't need an
    > arm on the other side... so now you've got a monoblade rear fork, and you don't have to remove the
    > rear wheel to change the tyre.
    >
    > Finally, you permanently assemble the stub axle, splined inner hub and disk brake to the swing
    > arm, and lace the wheel onto a splined outer hub. The splined outer hub slips onto the splined
    > inner hub and is retained by a washer and split pin, or some other fairly simple mechanism. Hey
    > presto, a wheel which is even quicker and easier to change than a conventional bicycle wheel.
    > Don't have to release or adjust the brakes. Don't have to mess with the chain or the gearing. Pull
    > out one pin, pull off the wheel. Push the wheel (or another wheel) back on, drop in the pin,
    > spread the ends of the pin, ride.
    >
    > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't
    > have to be epicyclic - in fact, it's easier to design and construct if it isn't. A much simple,
    > more conventional, motorcycle type gearbox layout will do. So the real question in my mind
    > boils down to, how light is it possible to build a gearbox which will survive and transmit the
    > 200 odd Nm of torque that a cyclist puts out. If Rohloff can do an epicyclic at 1.7Kg that's a
    > benchmark to aim at.
    >
    > [1] of course if you aren't interested in suspension this becomes even easier

    Dear Simon,

    At least your dream stops short of a shaft drive. I think that a bit of searching in the archives
    will turn up some convincing posts from Jobst Brandt about how the low-speed, narrow-rpm-range,
    high-torque human-powered bicycle is pretty much stuck with the chain and derailleur system, which
    as he points out is used by nothing else in the mechanical world.

    Still, I'd love to see a drawing of your scheme, even if pesky practical matters prevent it from
    becoming reality. But I'd be dubious about any enclosed chain--in the 1960's, the off-road
    motorcycle world finally admitted that the damned things were a great idea until the housing smashed
    on a rock and you had to untangle the mess and fix the chain in the field.

    Carl Fogel
     
  6. "wle" <[email protected]> wrote in message news:[email protected]...
    > i;ve seen it in here, "commuting on a racing bike is hard duty".
    >

    I think the problem is not so much excessive wear and tear or maintenance issues commuting on a
    'racing' bike, nor the riding position (after all, most races are longer than most commutes), nor
    pot holes (most road races are run on the road!). It is simply that dressing up something so
    exquisitively lovely as a lightweight racing bike in all the accoutrements required for commuting
    (rack & panniers for things like office clothes & documents, wet weather gear, pump & puncture
    repair kit, multi tool etc; fenders, lights, lock mount and heavy duty lock (or two - I usually
    carry two locks) seems like an insult to beauty. There is something so wonderfully delightful at
    lifting a lightweight bike over the doorstep and appreciating its lightness. You just don't _want_
    to load it up with the boring necessities of commuting!

    Rich
     
  7. Simon Brooke

    Simon Brooke Guest

    [email protected] (Carl Fogel) writes:

    > Simon Brooke <[email protected]> wrote in message
    > news:<[email protected]>...
    > > I've been thinking and doodling about ideas which are based on the Nicolai Nucleon frames, which
    > > have the Rohloff not at the rear hub but at the suspension pivot. Lots of advantages, starting
    > > with lower unsprung weight and less vulnerability to getting bashed on rocks and so on. If you
    > > do that, and you manage to make the suspension pivot perfectly coaxial with the output
    > > sprocket[1], then you can have a box section swing arm which can double as a chain case (you
    > > could even run the chain in an oil bath, but certainly you could protect it from dirt). If you
    > > did that, then it isn't much more difficult to make the box section arm sufficiently stiff that
    > > you don't need an arm on the other side... so now you've got a monoblade rear fork, and you
    > > don't have to remove the rear wheel to change the tyre.
    > >
    > > Finally, you permanently assemble the stub axle, splined inner hub and disk brake to the swing
    > > arm, and lace the wheel onto a splined outer hub. The splined outer hub slips onto the splined
    > > inner hub and is retained by a washer and split pin, or some other fairly simple mechanism. Hey
    > > presto, a wheel which is even quicker and easier to change than a conventional bicycle wheel.
    > > Don't have to release or adjust the brakes. Don't have to mess with the chain or the gearing.
    > > Pull out one pin, pull off the wheel. Push the wheel (or another wheel) back on, drop in the
    > > pin, spread the ends of the pin, ride.
    > >
    > > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't have
    > > to be epicyclic - in fact, it's easier to design and construct if it isn't. A much simple, more
    > > conventional, motorcycle type gearbox layout will do. So the real question in my mind boils down
    > > to, how light is it possible to build a gearbox which will survive and transmit the 200 odd Nm
    > > of torque that a cyclist puts out. If Rohloff can do an epicyclic at 1.7Kg that's a benchmark to
    > > aim at.
    >
    > At least your dream stops short of a shaft drive. I think that a bit of searching in the archives
    > will turn up some convincing posts from Jobst Brandt about how the low-speed, narrow-rpm-range,
    > high-torque human-powered bicycle is pretty much stuck with the chain and derailleur system, which
    > as he points out is used by nothing else in the mechanical world.

    Yes, but as we've seen in other discussions, Jobst writes with great authority and confidence about
    matters on which he is expert, and also writes with great authority and confidence about matters on
    which he is blissfully ignorant, so his opinion is not of great utility. Empirically, epicyclic hub
    gears work, whatever Jobst may think. They are arguably not as efficient as deraileurs, and under
    road conditions are unlikely to displace deraileurs. But off road cycling is significantly different
    and the deraileur is working in significantly sub-optimal conditions.

    Protecting the transmission of an off-road bike is going to provide increased reliability and
    resilience. It may not cost much in terms of efficiency, since the deraileur is not working at its
    peak of efficiency. Even if there are efficiency costs, these may be outweighed by the reliability
    and resilience benefits, even under racing conditions. And even if it did not turn out to be a
    competitive solution for racing, high reliability is still a great benefit for many non-competitive
    cyclists, both off-road and on - as is demonstrated by the success of the Rohloff and other
    epicyclic hub systems, despite their cost.

    In my opinion these issues can only be resolved by (i) some very careful maths and engineering work,
    and (ii) empirical testing. The question boils down to: can you create a gearbox in an enclosure not
    more than 100mm wide (you can use the width conventionally taken up by chainrings) by 180mm long by
    about 200mm tall, which has a sufficient range of gears to appeal to the cycling marketplace, which
    can reliably transmit the torque a cyclist can generate, and which has an efficiency competitive
    with a good epicyclic.

    The downside of this scheme is that while an epicyclic can be retrofitted into a conventional
    bycycle, the gearbox I am describing cannot.

    > Still, I'd love to see a drawing of your scheme, even if pesky practical matters prevent it from
    > becoming reality.

    I'll post a drawing somewhere this evening.

    --
    [email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/

    For office use only. Please do not write or type below this line.
     
  8. Simon Brooke

    Simon Brooke Guest

    "Matt O'Toole" <[email protected]> writes:

    > "Simon Brooke" <[email protected]> wrote in message
    > news:[email protected]...
    >
    [snip: gearbox fantasy]
    >
    > I've thought of similar designs, and I'm sure many others have too. Frankly, I'm amazed that no
    > one has tried it yet, especially with composites.
    >
    > I don't think the unsprung weight is an issue. It's not like a car or motorcycle, with big masses
    > reciprocating at potentially high velocity/frequency, due to high vechicle speeds. There isn't
    > that much kinetic energy to keep under control. Downhill racing bikes have big, heavy tires and go
    > very fast, and it doesn't seem to be a problem there.

    Yes; personally I'm not very interested in downhill bikes. Something which is too heavy to ride to
    the top seems to me boring.

    > But I like the idea of a completely internal drivetrain, with the chaincase a structural part of
    > the frame. And the stub axle thing too...
    >
    > I haven't seen the Nicoli frames you're referring to. But wouldn't they have more drivetrain
    > friction, from the extra chain?

    The Nicolai Nucleon models have one chain running from a conventional chainwheel up to the
    epicyclic, and a second chain running from the epicyclic to the rear hub. You'd think that this
    would have a friction penalty. They also have tension devices on the rear chain because the gearbox
    is not perfectly at the pivot.

    It was the complexity and lossiness of this that made me think of doing away with the chainwheel
    altogether and make the bottom bracket shaft the input shaft of a gearbox.

    > Also, though it might seem correct at first glance, suspension pivots usually should not be
    > concentric to the drive sprockets. Draw a free body diagram of the forces at the rear axle (most
    > of us get it wrong the first few times) and you'll see what I mean. That's why there are so many
    > monkey-motion suspension designs.

    This is one of those engineering tradeoff things again. If you can make the distance between the
    gearbox output sprocket and the rear hub fixed, you can do away with an active chain tension device
    and its attendant friction - but (if you do it by making the suspension coaxial with the output
    shaft) you will get bob, which you will have to try to control with suspension damping. If you put
    the pivot above the output sprocket you can reduce the bob, but you will need an active chain
    tension device.

    --
    [email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/

    There are no messages. The above is just a random stream of bytes. Any opinion or meaning
    you find in it is your own creation.
     
  9. Jay Beattie <[email protected]> wrote:
    >If you really want to carry a lot of stuff in panniers -- then get a rack.

    But then your machine has to be suitable for a rack, unless you really fancy drilling holes in it
    (which I realise you were willing to do.)
    --
    David Damerell <[email protected]> Kill the tomato!
     
  10. "David Damerell" <[email protected]> wrote in message
    news:dD*[email protected]...
    > Jay Beattie <[email protected]> wrote:
    > >If you really want to carry a lot of stuff in panniers -- then get a rack.
    >
    > But then your machine has to be suitable for a rack, unless you really fancy drilling holes in it
    > (which I realise you were willing to do.)
    > --

    Or simply attach with P-clips....

    Rich
     
  11. Jay Beattie

    Jay Beattie Guest

    "Richard Goodman" <[email protected]> wrote in message
    news:[email protected]...
    >
    > "wle" <[email protected]> wrote in message news:[email protected]...
    > > i;ve seen it in here, "commuting on a racing bike is hard duty".
    > >
    >
    > I think the problem is not so much excessive wear and tear or
    maintenance
    > issues commuting on a 'racing' bike, nor the riding position (after
    all,
    > most races are longer than most commutes), nor pot holes (most road
    races
    > are run on the road!). It is simply that dressing up something so exquisitively lovely as a
    > lightweight racing bike in all the
    accoutrements
    > required for commuting (rack & panniers for things like office clothes
    &
    > documents, wet weather gear, pump & puncture repair kit, multi tool
    etc;
    > fenders, lights, lock mount and heavy duty lock (or two - I usually
    carry
    > two locks) seems like an insult to beauty. There is something so wonderfully delightful at lifting
    > a lightweight bike over the doorstep
    and
    > appreciating its lightness. You just don't _want_ to load it up with
    the
    > boring necessities of commuting!

    Which really speaks to how many bikes you want to own. Shifting equipment on and off a racing bike
    is no fun, and I would not want to use my racing bike as my full-time commuter. I like having a
    dedicated commuter and have an old racing bike that I can use for the task. If I were starting from
    scratch, I would buy an old steel frame/fork off of Ebay -- something from the second tier of '70s
    racing bikes. These things have fender clearance, but are still pretty fast. I might also get a
    sport tourer from the same period with fender eyelets but no caniti bosses. I would get some of the
    Shimano long reach dual pivots. This is because I use STI, and cantis are not the best match-up with
    STI levers. If I had an essentially flat commute (which I do not), I would set it up as a fixed
    gear. -- Jay Beattie.
     
  12. Simon Brooke

    Simon Brooke Guest

    Simon Brooke <[email protected]> writes:

    > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't
    > have to be epicyclic - in fact, it's easier to design and construct if it isn't. A much simple,
    > more conventional, motorcycle type gearbox layout will do. So the real question in my mind
    > boils down to, how light is it possible to build a gearbox which will survive and transmit the
    > 200 odd Nm of torque that a cyclist puts out. If Rohloff can do an epicyclic at 1.7Kg that's a
    > benchmark to aim at.

    ...and, of course, when an idea is this obvious someone else has had
    it... it's those ruthlessly efficient leather shorts wearing people again - vorsprung durch technic,
    and all that.

    <URL: http://www.g-boxx.org/_2-english/_2-index-engl.html >

    --
    [email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/

    'Victories are not solutions.' ;; John Hume, Northern Irish politician, on Radio Scotland
    1/2/95 ;; Nobel Peace Prize laureate 1998; few have deserved it so much
     
  13. Matt O'Toole

    Matt O'Toole Guest

    "Simon Brooke" <[email protected]> wrote in message
    news:[email protected]...

    > Yes, but as we've seen in other discussions, Jobst writes with great authority and confidence
    > about matters on which he is expert, and also writes with great authority and confidence about
    > matters on which he is blissfully ignorant, so his opinion is not of great utility.

    I have not found this to be the case. Jobst seems very much a no-BS kind of guy, with the
    perspective of a long career in mechanical engineering and knowledge of history of same that most
    of us lack.

    > Empirically, epicyclic hub gears work, whatever Jobst may think.

    I've never seen Jobst argue against them, except to point out certain problems -- like how old
    Sturmey Archers fail under high torque loads. And he even explains why.

    > They are arguably not as efficient as deraileurs, and under road conditions are unlikely to
    > displace deraileurs. But off road cycling is significantly different and the deraileur is working
    > in significantly sub-optimal conditions.

    This is true.

    The Rohloff's efficiency vs. standard Shimano 3x9 drivetrains has been published on the web.
    Unfortunately, I've lost the link. I remember being surprised at the results -- overall, the Rohloff
    is pretty close. The derailer drivetrain is worse than I thought, and the Rohloff better. Of course,
    a derailer drivetrain suffers greatly with a dirty chain, enough to be felt by the rider. So under
    these conditions there's definately potential for the Rohloff.

    But all of this is old news. You're preaching to the choir.

    > Protecting the transmission of an off-road bike is going to provide increased reliability and
    > resilience. It may not cost much in terms of efficiency, since the deraileur is not working at its
    > peak of efficiency. Even if there are efficiency costs, these may be outweighed by the reliability
    > and resilience benefits, even under racing conditions.

    Yup. I'm sure a Rohloff could be an advantage in a muddy MTB race.

    > And even if it did not turn out to be a competitive solution for racing, high reliability is still
    > a great benefit for many non-competitive cyclists, both off-road and on - as is demonstrated by
    > the success of the Rohloff and other epicyclic hub systems, despite their cost.

    No one would argue against this. Epicyclic hubs have always been great for utility cyclists, and
    still are. They seem to be getting popular again -- note that Shimano just came out with a new 8
    speed Nexus. And yes, the main hurdle to their becoming more popular is cost. The kind of cyclist
    who would benefit most from a Nexus hub can buy a whole new bike at Wal-Mart for less than what just
    the hub costs. And a Rohloff costs as much as a mid range MTB.

    > In my opinion these issues can only be resolved by (i) some very careful maths and engineering
    > work, and (ii) empirical testing. The question boils down to: can you create a gearbox in an
    > enclosure not more than 100mm wide (you can use the width conventionally taken up by chainrings)
    > by 180mm long by about 200mm tall, which has a sufficient range of gears to appeal to the cycling
    > marketplace, which can reliably transmit the torque a cyclist can generate, and which has an
    > efficiency competitive with a good epicyclic.

    I think so, at least for a niche market.

    Matt O.
     
  14. Matt O'Toole

    Matt O'Toole Guest

    "Simon Brooke" <[email protected]> wrote in message
    news:[email protected]...

    > "Matt O'Toole" <[email protected]> writes:

    > > "Simon Brooke" <[email protected]> wrote in message
    > > news:[email protected]...

    > > I don't think the unsprung weight is an issue. It's not like a car or motorcycle, with big
    > > masses reciprocating at potentially high velocity/frequency, due to high vechicle speeds. There
    > > isn't that much kinetic energy to keep under control. Downhill racing bikes have big, heavy
    > > tires and go very fast, and it doesn't seem to be a problem there.

    > Yes; personally I'm not very interested in downhill bikes. Something which is too heavy to ride to
    > the top seems to me boring.

    But you did get the point about unsprung weight?

    > The Nicolai Nucleon models have one chain running from a conventional chainwheel up to the
    > epicyclic, and a second chain running from the epicyclic to the rear hub. You'd think that this
    > would have a friction penalty. They also have tension devices on the rear chain because the
    > gearbox is not perfectly at the pivot.

    Lots of friction!

    > It was the complexity and lossiness of this that made me think of doing away with the chainwheel
    > altogether and make the bottom bracket shaft the input shaft of a gearbox.

    Well, as I mentioned, there's already a 2 speed epicyclic crank. If you're enamoured of this idea I
    suppose you could develop a 14 speed version. But if you're putting the gears in the crank instead
    of the hub because of unsprung weight, I think you're barking up the wrong tree. It's not
    significant.

    > > Also, though it might seem correct at first glance, suspension pivots usually should not be
    > > concentric to the drive sprockets. Draw a free body diagram of the forces at the rear axle (most
    > > of us get it wrong the first few times) and you'll see what I mean. That's why there are so many
    > > monkey-motion suspension designs.

    > This is one of those engineering tradeoff things again. If you can make the distance between the
    > gearbox output sprocket and the rear hub fixed, you can do away with an active chain tension
    > device and its attendant friction - but (if you do it by making the suspension coaxial with the
    > output shaft) you will get bob, which you will have to try to control with suspension damping. If
    > you put the pivot above the output sprocket you can reduce the bob, but you will need an active
    > chain tension device.

    You'll get bob either way. If you're going to live with bob like that, you might as well use a
    URT design.

    Matt O.
     
  15. Carl Fogel

    Carl Fogel Guest

    Simon Brooke <[email protected]> wrote in message
    news:<[email protected]>...
    > [email protected] (Carl Fogel) writes:
    >
    > > Simon Brooke <[email protected]> wrote in message
    > > news:<[email protected]>...
    > > > I've been thinking and doodling about ideas which are based on the Nicolai Nucleon frames,
    > > > which have the Rohloff not at the rear hub but at the suspension pivot. Lots of advantages,
    > > > starting with lower unsprung weight and less vulnerability to getting bashed on rocks and so
    > > > on. If you do that, and you manage to make the suspension pivot perfectly coaxial with the
    > > > output sprocket[1], then you can have a box section swing arm which can double as a chain case
    > > > (you could even run the chain in an oil bath, but certainly you could protect it from dirt).
    > > > If you did that, then it isn't much more difficult to make the box section arm sufficiently
    > > > stiff that you don't need an arm on the other side... so now you've got a monoblade rear fork,
    > > > and you don't have to remove the rear wheel to change the tyre.
    > > >
    > > > Finally, you permanently assemble the stub axle, splined inner hub and disk brake to the swing
    > > > arm, and lace the wheel onto a splined outer hub. The splined outer hub slips onto the splined
    > > > inner hub and is retained by a washer and split pin, or some other fairly simple mechanism.
    > > > Hey presto, a wheel which is even quicker and easier to change than a conventional bicycle
    > > > wheel. Don't have to release or adjust the brakes. Don't have to mess with the chain or the
    > > > gearing. Pull out one pin, pull off the wheel. Push the wheel (or another wheel) back on, drop
    > > > in the pin, spread the ends of the pin, ride.
    > > >
    > > > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't
    > > > have to be epicyclic - in fact, it's easier to design and construct if it isn't. A much
    > > > simple, more conventional, motorcycle type gearbox layout will do. So the real question in my
    > > > mind boils down to, how light is it possible to build a gearbox which will survive and
    > > > transmit the 200 odd Nm of torque that a cyclist puts out. If Rohloff can do an epicyclic at
    > > > 1.7Kg that's a benchmark to aim at.
    > >
    > > At least your dream stops short of a shaft drive. I think that a bit of searching in the
    > > archives will turn up some convincing posts from Jobst Brandt about how the low-speed,
    > > narrow-rpm-range, high-torque human-powered bicycle is pretty much stuck with the chain and
    > > derailleur system, which as he points out is used by nothing else in the mechanical world.
    >
    > Yes, but as we've seen in other discussions, Jobst writes with great authority and confidence
    > about matters on which he is expert, and also writes with great authority and confidence about
    > matters on which he is blissfully ignorant, so his opinion is not of great utility. Empirically,
    > epicyclic hub gears work, whatever Jobst may think. They are arguably not as efficient as
    > deraileurs, and under road conditions are unlikely to displace deraileurs. But off road cycling is
    > significantly different and the deraileur is working in significantly sub-optimal conditions.
    >
    > Protecting the transmission of an off-road bike is going to provide increased reliability and
    > resilience. It may not cost much in terms of efficiency, since the deraileur is not working at its
    > peak of efficiency. Even if there are efficiency costs, these may be outweighed by the reliability
    > and resilience benefits, even under racing conditions. And even if it did not turn out to be a
    > competitive solution for racing, high reliability is still a great benefit for many
    > non-competitive cyclists, both off-road and on - as is demonstrated by the success of the Rohloff
    > and other epicyclic hub systems, despite their cost.
    >
    > In my opinion these issues can only be resolved by (i) some very careful maths and engineering
    > work, and (ii) empirical testing. The question boils down to: can you create a gearbox in an
    > enclosure not more than 100mm wide (you can use the width conventionally taken up by chainrings)
    > by 180mm long by about 200mm tall, which has a sufficient range of gears to appeal to the cycling
    > marketplace, which can reliably transmit the torque a cyclist can generate, and which has an
    > efficiency competitive with a good epicyclic.
    >
    > The downside of this scheme is that while an epicyclic can be retrofitted into a conventional
    > bycycle, the gearbox I am describing cannot.
    >
    > > Still, I'd love to see a drawing of your scheme, even if pesky practical matters prevent it from
    > > becoming reality.
    >
    > I'll post a drawing somewhere this evening.

    Dear Simon,

    I suppose that I'm just mesmerized by the evil rule of thumb from my motorcycling days that a good
    chain loses about 3% of the engine's power, while each gear-to-gear interface loses about 5%.
    Perhaps I'm mistaken on the power losses involved.

    Putting a chain between two enclosed gear-to-gear systems may be a needlessly elaborate way to tame
    the fury of my heroic thighs. A gentle slope is usually more than enough.

    A search for "jobst shaft drive gear" on my Google archives of the newsgroup turns up a fair number
    of what seem like well-reasoned comments on various gear, shaft, v-belt, and other schemes intended
    to improve on the derailleur. Jobst predicted that they would come to naught at least as far back as
    1995, and so far his prediction seems sound.

    (His theories about puncture vine, on the other hand, are blasphemous and I intend to found a church
    eventually and rent pews to refute his heresies--real soon now.)

    That said, am I right in thinking that your scheme involves a double-arm on one side, with the chain
    running back inside one arm and returning through the other?

    Carl Fogel
     
  16. David Kerber

    David Kerber Guest

    In article <[email protected]>, [email protected] says...
    > Simon Brooke <[email protected]> wrote in message
    > news:<[email protected]>...
    > > [email protected] (Carl Fogel) writes:
    > >
    > > > Simon Brooke <[email protected]> wrote in message
    > > > news:<[email protected]>...
    > > > > I've been thinking and doodling about ideas which are based on the Nicolai Nucleon frames,
    > > > > which have the Rohloff not at the rear hub but at the suspension pivot. Lots of advantages,
    > > > > starting with lower unsprung weight and less vulnerability to getting bashed on rocks and so
    > > > > on. If you do that, and you manage to make the suspension pivot perfectly coaxial with the
    > > > > output sprocket[1], then you can have a box section swing arm which can double as a chain
    > > > > case (you could even run the chain in an oil bath, but certainly you could protect it from
    > > > > dirt). If you did that, then it isn't much more difficult to make the box section arm
    > > > > sufficiently stiff that you don't need an arm on the other side... so now you've got a
    > > > > monoblade rear fork, and you don't have to remove the rear wheel to change the tyre.
    > > > >
    > > > > Finally, you permanently assemble the stub axle, splined inner hub and disk brake to the
    > > > > swing arm, and lace the wheel onto a splined outer hub. The splined outer hub slips onto the
    > > > > splined inner hub and is retained by a washer and split pin, or some other fairly simple
    > > > > mechanism. Hey presto, a wheel which is even quicker and easier to change than a
    > > > > conventional bicycle wheel. Don't have to release or adjust the brakes. Don't have to mess
    > > > > with the chain or the gearing. Pull out one pin, pull off the wheel. Push the wheel (or
    > > > > another wheel) back on, drop in the pin, spread the ends of the pin, ride.
    > > > >
    > > > > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't
    > > > > have to be epicyclic - in fact, it's easier to design and construct if it isn't. A much
    > > > > simple, more conventional, motorcycle type gearbox layout will do. So the real question in
    > > > > my mind boils down to, how light is it possible to build a gearbox which will survive and
    > > > > transmit the 200 odd Nm of torque that a cyclist puts out. If Rohloff can do an epicyclic at
    > > > > 1.7Kg that's a benchmark to aim at.
    > > >
    > > > At least your dream stops short of a shaft drive. I think that a bit of searching in the
    > > > archives will turn up some convincing posts from Jobst Brandt about how the low-speed,
    > > > narrow-rpm-range, high-torque human-powered bicycle is pretty much stuck with the chain and
    > > > derailleur system, which as he points out is used by nothing else in the mechanical world.
    > >
    > > Yes, but as we've seen in other discussions, Jobst writes with great authority and confidence
    > > about matters on which he is expert, and also writes with great authority and confidence about
    > > matters on which he is blissfully ignorant, so his opinion is not of great utility. Empirically,
    > > epicyclic hub gears work, whatever Jobst may think. They are arguably not as efficient as
    > > deraileurs, and under road conditions are unlikely to displace deraileurs. But off road cycling
    > > is significantly different and the deraileur is working in significantly sub-optimal conditions.
    > >
    > > Protecting the transmission of an off-road bike is going to provide increased reliability and
    > > resilience. It may not cost much in terms of efficiency, since the deraileur is not working at
    > > its peak of efficiency. Even if there are efficiency costs, these may be outweighed by the
    > > reliability and resilience benefits, even under racing conditions. And even if it did not turn
    > > out to be a competitive solution for racing, high reliability is still a great benefit for many
    > > non-competitive cyclists, both off-road and on - as is demonstrated by the success of the
    > > Rohloff and other epicyclic hub systems, despite their cost.
    > >
    > > In my opinion these issues can only be resolved by (i) some very careful maths and engineering
    > > work, and (ii) empirical testing. The question boils down to: can you create a gearbox in an
    > > enclosure not more than 100mm wide (you can use the width conventionally taken up by chainrings)
    > > by 180mm long by about 200mm tall, which has a sufficient range of gears to appeal to the
    > > cycling marketplace, which can reliably transmit the torque a cyclist can generate, and which
    > > has an efficiency competitive with a good epicyclic.
    > >
    > > The downside of this scheme is that while an epicyclic can be retrofitted into a conventional
    > > bycycle, the gearbox I am describing cannot.
    > >
    > > > Still, I'd love to see a drawing of your scheme, even if pesky practical matters prevent it
    > > > from becoming reality.
    > >
    > > I'll post a drawing somewhere this evening.
    >
    > Dear Simon,
    >
    > I suppose that I'm just mesmerized by the evil rule of thumb from my motorcycling days that a good
    > chain loses about 3% of the engine's power, while each gear-to-gear interface loses about 5%.
    > Perhaps I'm mistaken on the power losses involved.

    Keep in mind that a motorcycle chain (or a bike's if it's a single speed) is perfectly in line. I
    believe it's the bends it makes when you are on chain ring/sprocket combinations which don't allow
    the chain to be straight, which cost you the most energy losses.

    A chain and sprocket drive is extremely efficient method of transferring power when everything is
    correctly aligned.

    --
    Dave Kerber Fight spam: remove the ns_ from the return address before replying!

    REAL programmers write self-modifying code.
     
  17. Carl Fogel

    Carl Fogel Guest

    Simon Brooke <[email protected]> wrote in message
    news:<[email protected]>...
    > Simon Brooke <[email protected]> writes:
    >
    > > The final step in this fantasy is that if the gearbox is at the suspension pivot it doesn't have
    > > to be epicyclic - in fact, it's easier to design and construct if it isn't. A much simple, more
    > > conventional, motorcycle type gearbox layout will do. So the real question in my mind boils down
    > > to, how light is it possible to build a gearbox which will survive and transmit the 200 odd Nm
    > > of torque that a cyclist puts out. If Rohloff can do an epicyclic at 1.7Kg that's a benchmark to
    > > aim at.
    >
    > ...and, of course, when an idea is this obvious someone else has had
    > it... it's those ruthlessly efficient leather shorts wearing people again - vorsprung durch
    > technic, and all that.
    >
    > <URL: http://www.g-boxx.org/_2-english/_2-index-engl.html >

    Dear Simon,

    Fascinating gadgets, lots of nice pictures. I fear that the actual engineering might not be all that
    we'd wish, but I have friends who'd try to wheedle a ride on these contraptions.

    It reminds me a little of a long-lost Honda fluid-transmission motorcycle. We were all impressed
    that it worked, but never saw any production models.

    Carl Fogel
     
  18. Simon Brooke

    Simon Brooke Guest

    "Matt O'Toole" <[email protected]> writes:

    > "Simon Brooke" <[email protected]> wrote in message
    > news:[email protected]...
    >
    > > "Matt O'Toole" <[email protected]> writes:
    >
    > > > "Simon Brooke" <[email protected]> wrote in message
    > > > news:[email protected]...
    >
    > > > I don't think the unsprung weight is an issue. It's not like a car or motorcycle, with big
    > > > masses reciprocating at potentially high velocity/frequency, due to high vechicle speeds.
    > > > There isn't that much kinetic energy to keep under control. Downhill racing bikes have big,
    > > > heavy tires and go very fast, and it doesn't seem to be a problem there.
    >
    > > Yes; personally I'm not very interested in downhill bikes. Something which is too heavy to ride
    > > to the top seems to me boring.
    >
    > But you did get the point about unsprung weight?

    I got it, but I didn't entirely agree with it. Yes, the unsprung weight may not matter as much as it
    does in (e.g.) motor cycle design, but it still increases the inertia or the unsuspended part of the
    vehicle, preventing the suspension from reacting as rapidly as it could. So I agree it may not be
    very important, but less is still better.

    > > The Nicolai Nucleon models have one chain running from a conventional chainwheel up to the
    > > epicyclic, and a second chain running from the epicyclic to the rear hub. You'd think that this
    > > would have a friction penalty. They also have tension devices on the rear chain because the
    > > gearbox is not perfectly at the pivot.
    >
    > Lots of friction!
    >
    > > It was the complexity and lossiness of this that made me think of doing away with the chainwheel
    > > altogether and make the bottom bracket shaft the input shaft of a gearbox.
    >
    > Well, as I mentioned, there's already a 2 speed epicyclic crank. If you're enamoured of this idea
    > I suppose you could develop a 14 speed version. But if you're putting the gears in the crank
    > instead of the hub because of unsprung weight, I think you're barking up the wrong tree. It's not
    > significant.

    The key issue is not unsprung weight, it's keeping twigs, stones, grit etc out of the tranmission.
    But if you're going to do that then you might a well reduce unsprung weight at the same time, since
    it's actually easier to put the gearbox at the frame end of the chainstay (where it doesn't _have_
    to be epicyclic) rather than at the hub end (where it does). Epicyclic gearboxes, while elegant, are
    rather complex bits of technology. This has the useful side-benefit of reducing unsprung weight.

    > > > Also, though it might seem correct at first glance, suspension pivots usually should not be
    > > > concentric to the drive sprockets. Draw a free body diagram of the forces at the rear axle
    > > > (most of us get it wrong the first few times) and you'll see what I mean. That's why there are
    > > > so many monkey-motion suspension designs.
    >
    > > This is one of those engineering tradeoff things again. If you can make the distance between the
    > > gearbox output sprocket and the rear hub fixed, you can do away with an active chain tension
    > > device and its attendant friction - but (if you do it by making the suspension coaxial with the
    > > output shaft) you will get bob, which you will have to try to control with suspension damping.
    > > If you put the pivot above the output sprocket you can reduce the bob, but you will need an
    > > active chain tension device.
    >
    > You'll get bob either way. If you're going to live with bob like that, you might as well use a
    > URT design.

    Well, I've ridden plenty of cantilever type single pivot bikes with the pivot a chainwheel radius
    above the the bottom bracket, and my experience is that good ones don't bob much except when you're
    out of the saddle. I haven't ridden one with the pivot concentric to the bottom bracket, but I'm
    prepared to believe it would bob badly, and the same applies to a URT (although of course that is
    very simple to engineer).

    --
    [email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/ ;; "If I were a Microsoft
    Public Relations person, I would probably ;; be sobbing on a desk right now" -- Rob Miller,
    editor, /.
     
  19. Simon Brooke

    Simon Brooke Guest

    [email protected] (Carl Fogel) writes:

    > Simon Brooke <[email protected]> wrote in message
    > news:<[email protected]>...
    > > In my opinion these issues can only be resolved by (i) some very careful maths and engineering
    > > work, and (ii) empirical testing. The question boils down to: can you create a gearbox in an
    > > enclosure not more than 100mm wide (you can use the width conventionally taken up by chainrings)
    > > by 180mm long by about 200mm tall, which has a sufficient range of gears to appeal to the
    > > cycling marketplace, which can reliably transmit the torque a cyclist can generate, and which
    > > has an efficiency competitive with a good epicyclic.
    > >
    > > The downside of this scheme is that while an epicyclic can be retrofitted into a conventional
    > > bycycle, the gearbox I am describing cannot.
    > >
    > I suppose that I'm just mesmerized by the evil rule of thumb from my motorcycling days that a good
    > chain loses about 3% of the engine's power, while each gear-to-gear interface loses about 5%.
    > Perhaps I'm mistaken on the power losses involved.

    Well, that's the rule of thumb I learned too; so the minimum loss in the gearbox is about 8%, which
    is not good. But this does not mean it's a non-starter, just that the engineering needs to be good.
    Engineering is always about trade-offs.

    > That said, am I right in thinking that your scheme involves a double-arm on one side, with the
    > chain running back inside one arm and returning through the other?

    No, out and back in one arm (which would be tall but not wide). Interestingly Nicolai have a design
    where the chain runs out and back inside two arms as you suggest.

    --
    [email protected] (Simon Brooke) http://www.jasmine.org.uk/~simon/ ;; "If I were a Microsoft
    Public Relations person, I would probably ;; be sobbing on a desk right now" -- Rob Miller,
    editor, /.
     
  20. On Wed, 29 Oct 2003 06:35:03 -0500, Simon Brooke wrote:

    >> that a bit of searching in the archives will turn up some convincing posts from Jobst Brandt
    >> about how the low-speed, narrow-rpm-range, high-torque human-powered bicycle is pretty much stuck
    >> with the chain and derailleur system, which as he points out is used by nothing else in the
    >> mechanical world.
    >
    > Yes, but as we've seen in other discussions, Jobst writes with great authority and confidence
    > about matters on which he is expert, and also writes with great authority and confidence about
    > matters on which he is blissfully ignorant, so his opinion is not of great utility. Empirically,
    > epicyclic hub gears work, whatever Jobst may think. They are arguably not as efficient as
    > deraileurs, and under road conditions are unlikely to displace deraileurs. But off road cycling is
    > significantly different and the deraileur is working in significantly sub-optimal conditions.

    And lest we forget, the original post was about a commuting bike. And for many commutes, epicyclic
    hubs have advantages:

    - shift when stopped, at a light or stop sign, a big issue for commuters
    - weather protection, again more likely to be an issue for an every-day commuter than a fair weather
    recreational rider
    - less maintenance (certainly true for the old reliable SA, which merely needed oil once in a
    while) required
    - may be easier to shift in the dark and in the rain (1 shifter only, simply run up and down the
    gears, no worries about front shifting)
     
Loading...
Thread Status:
Not open for further replies.
Loading...