Shimano are about to offer an electronic
upgrade to their top of the range dura-ace component set.
This system will be 67 grams lighter than the current mechanical
system, though a bit heavier than the system they plan to deliver
next year. It runs on a 7.6 volt lithium-ion battery.
For a cycle commuter like me cable actuated gears are a total
PITA. They dry out, they stick and they fray. You get them
adjusted just right then it rains and the oil washes away and you
are back to square one. Murphy's law means that they
invariable fail on a monday and leave me replacing the things in
the middle of the night after work. Yes, I have a lot of hate for
gear cables and several times I have considered throwing them out
and putting in my own stepper motors.
zogger, thanks for the wired link in your Toyota Winglet story.
They linked to the dura-ace one and I decided to submit an
article.
I never seemed to have problems with my old seven speed bikes.
Every shift was very precise. Now I have a nine speed system.
After shifting two clicks left on the cluster the next click to
the right invariably does nothing. The cable for the front
derailleur fails inside the shift lever which gets ripped up when
I replace it.
Maybe we could eventually have a semiautomatic system. Something
tied to pedal cadence. If so, mine will be set to 120.
You are a very decent pedaler. That's hauling @$$ets. With
that said, not trying to change the subject but to offer an
alternative to that, back when I built my prototype mountain bike
and the design apparently got "borrowed" by a big bike
factory the next little town over, the tards didn't realize
that was only the freekin easy and stupid half of the design, the
cool part, which I was carefully and sneakily having built
by different guys at different shops so they wouldn't bingo
to what I was doing I just stopped developer work on, was
*hydraulic drive*. Unlimited ranges and gears, totally enclosed
system. You could even make it with regenerative braking and
coasting for a fast powerful boost for uphill and from dead stop
situations, and use the internal frame structure for a lot of the
reservoir you would need. Plus make it two wheel drive fairly
easily. Like you, I think cables and chains are sorta 18th
century old tech, because they just stretch and get dirty and get
sucky. I guess some guys really like electronics, and this new
shimano stuff is dang nice, I used a lot of dura ace way back,
never could see any advantage of note with campy for the price,
but I like simpler tech and more robust, electronics are always
the major weak points on moving machinery out in the dusty and
rainy real world, little bitty do dads get a bad connection and
it ruins your whole day.
I took my current mountain bike to the Mallee
shortly after I bought it and totally destroyed the driveline. I
eventually got rid of the cluster and chain because it got
covered with sand and basically eroded the moving parts away.
So anything which puts the moving parts inside away from the dirt
is a good thing. I wonder how much efficency you can get out of
hydraulics?
BTW I submitted a story about the Falcon 1 launch but it is out
of date now because the launcher worked perfectly. You could
probably drop that one if it hasn't been posted yet.
What was the efficiency of your hydraulic drive zogger? Did it
end up in any product?
I'm quite interested in alternative drive systems for bikes.
I haven't built anything yet, but have done lots of back of
envelope sketches and calculations. With my background being
electrical engineering I would dearly like to experiment with an
electric transmission on my bicycle. CSIRO have built a motor with 98%
efficiency. They have even made limited technical details
available. Perhaps it is possible to build a scaled down version
of this motor and put two back to back, one acting as a
generator. Pedals go on the generator and the motor goes on the
wheel. Voila! Electric/electronic transmission with an efficiency
of 96%, comparable to a chain drive. I'm not not sure if
CSIRO's numbers include losses in the electronics and how
their "best case" numbers relate to reality.
Once a bike has electric drive, why not make it intelligent?
Regenerative braking is an obvious addition but why stop there?
Why not add control systems to make the regen. automatic,
decoupling the rider from the terrain? The rider pedals at a
constant (optimal) cadence, generating constant power, equal to
the average power required for the trip. The transmission matches
the average power to the instantaneous power requirements.
How about we put the electric transmission on a unicycle? We add
an angle sensor to the unicycle and the transmission is now
responsible for doing all the fiddly pedaling to keep the
unicycle upright. Not that much different to Trevor Blackwell's
effort, except this one would still be human powered. Next
step is to add a weight to the unicycle, rotating in a horizontal
plane. The weight is driven by a stepper motor, under computer
control. Now the computer can steer the unicycle and we have a
unicycle that is stable in both front/back and left/right
directions. Blackwell's is only stable front/back and
requires the same level of skill as a normal unicycle to stop it
from tipping left/right. Would this one be ridable by an
unskilled person?
The end result is a pole with a seat and pedals. You sit on the
seat, in the same manner as you might sit on a bar stool. Put
your feet on the pedals, turn them at a constant rate and off you
go. Easier than riding on a bike, since you can't fall off
and you are cycling on perpetual (virtual) flat terrain! If the
automatic control systems can be made to have a faster response
time than a human then why not make the wheel as small as
possible, keeping in mind that is has to be big enough to go over
rough terrain? That way the whole thing could be folded up and
carried like an umbrella.
Continuing the thinking aloud... Let's replace the wheel with
a ball. The ball can be driven in two axes: front/back or
left/right, a bit like an oversized mouse ball in reverse, with
the ball being driven by the "mouse". Maybe the ball
can actually be the rotor of an induction motor, a two
dimensional version of Tesla's
Egg of Columbus?
I'm sure there are lost of holes in this (eg. final idea
would require extremely strong fields to get decent torque), but
it would be interesting to see how much of it could be built. I
reckon it is probably all realistic apart from the final
paragraph, the biggest unknow being whether the efficiency could
approach that of a chain/sprocket transmission.
The rider pedals at a constant (optimal) cadence, generating
constant power, equal to the average power required for the
trip.
It would be handy to not have to stop working at red lights. My
pulse rate tends to drop if I have to wait more than a couple of
minutes at a time and then it is hard to get going again. Thats
why runners tend to jog on the spot when waiting to cross the
road.
i'd just like to see more CVT on bikes, but there are
just a few (unecessarily) expensive systems out there for the
high end rich snooty types, I just hate the derailleur
thing. Cause of all my troubles....
Bicycling is my main means of transportation and has been for 35
years. Anyone who really rides regularly knows that the
constant adjustment needs and vulnerability of derailleurs
(I've used them all) is one of the main barriers to broader
adoption of the most efficient and healthiest way to travel.
For about a year I have been riding a bike with the new
'Nuvinci' CVT, am quite happy with it and think this is
the first 'alternative' transmission system that has a
chance of persisting in the market (although the current weight
will limit it in the beginning to serious commuters).
http://www.bikecommuters.com/2007/09/29/nuvinci-hub-review/
Electric shift systems of the past have been design failures, but
there's a compelling reason to make it happen, and it has
everything to do with the complaints already put forward.
In a traditional system, you set the low and high limits, then
get the tension just right so when you throw the lever, it moves
the cage over X millimeters and cross your fingers that it's
gone just far enough, no more or less. The electric system
knows where the top is, and moves the derailleur cage until the
chain engages the next requested cog. When you're nine
shifts down from the top (on a ten speed cassette), you've
obviously hit bottom.
This is simple embedded programming (that has taken years to
develop, I'm quite aware), and could probably evolve into an
autoshift program that the market could actually accept (if CVTs
don't catch on).
On an individual level, its an interesting project/product.
On a social scale, in view of the role of the bike in what is
coming towards us, its irrelevant. We had many years ago
transmission systems which had evolved to be fit for all weather
commuting.
The first thing is a fully enclosed chain. This means that
it doesn't get grit in it, lubrication lasts, reliability is
greatly increased. The second thing, of course, if you have
a chain enclosure, you can no longer have derailleurs. So
you have hub gears. This limits the number of speeds, but
it doesn't much matter in ride to work scenarios for the
masses, because 99% of the people will be riding on terrain
without extreme hills.
The comments on this thread about shifting issues and wear of
sprockets are thought provoking in this connection. One can
see why click shifting occurred - its because of the effort to
make derailleurs usable by a mass public that will not listen to
the bike and know when their shift has gone a bit too far or not
quite far enough, and adjust till its silent. But its the
result of making a technology which is optimised for racing into
a mass market transport technology. I can see why they went
for click shifting, but its halfway to the real solution, and if
you are used to derailleurs, listening to the bike, remembering
where you are in the gears, making the fine adjustment, click
shift is infuriating.
For mass commuting, hubs are the optimal solution, given the
requirements, not all of which are about cycling feel and
pedalling cadence and speed. Its like automatic
transmission for driving in traffic. Its just so much
simpler when you are getting from A to B with as little thought
as possible about how the vehicle is working.
As to weight: I doubt that even 10% of professionals in
blind tests could tell one case from the other, when the
difference in weight of the e-version from the standard version
is 70gm. Weigh it in your hand, its under 3 ounces.
In commuting use, no-one will notice. In fact, in commuting
use, probably steel should replace alloy for lots of parts.
Stronger, wears better, if it gets bent, its easier to
straighten, and the weight is immaterial. There's also
a lot to be said for steel rims and hub brakes, in an era of mass
commuting.
There are several issues here, I'll deal with them
separately.
Chain and sprockets is the most efficient method of power
transmission out there, and when you only have a couple of
hundred watts of human power on tap, efficiency is king.
Chain wear is down to lack of lubrication and maintenance,
period, I can get 25,000 miles EASY out of a relatively weedy
5/8" chain on a 1,000+ cc motorcycle, but the closest I have
ever seen a pushbike owner come to "routine
maintenance" is when they take the whole bike back to
Richards for a service maybe once a year.
As someone else said, enclose the chain and it will
effectively last a lifetime, enclosed chains make derailleur
tough to do but as they say, 3 or 4 speeds are enough for any
road bike.
Of the three pushbikes in this family one is a 12 speed and
two are 18 speed, most of these speeds are literally never used.
On motorcycles cables worked for years with no problems,
because unlike the crap fitted to bikes the cable was entirely
enclosed from end to end.
On motorcycles the device that the cable plugged into was
designed with some care, both to protect the cable, maintain
minimum radius, keep egress of dirt away, and so on and so forth,
so a periodic oiling was all it ever needed... please note we did
this with 1960's era lubricants, oil in other words.
On motorcycles cables were NEVER used in the same way, eg
they were NEVER used to position something like a gear selector,
if you are going to use a cable for a gear selector you must
always use two, when japanese motorcycles did away with the slide
carb they introduced twin cable throttles, one cable to open the
butterflies and the other to close them, and the tension between
the two was balanced.
SO basically all the "problems" associated with bikes,
which are evidenced by two out of the three bikes in this
household, eg the two that are not maintained properly and which
have not been modified properly (that bike is used daily all
weathers, and has never once failed in operation in 4 years of
daily use, and it was bought from Richards second hand for 60
quid) are down to one thing, piss poor design and piss poor
quality.
The two crap bikes have sprockets cut from what appears to be
recycled aluminium, the roller chains themselves are generic and
piss poor quality, ditto the cables and ditto the cheap and nasty
plastic cable controllers / gear selectors, likewise the
derailleurs themselves are cheap and nasty, plus as mentioned
above there are several basic fundamental engineering design
flaws, put it all together with users who lack clue #1 about
maintenance and you get what you get, a pile of crap and people
casting about for alternatives.
You don't NEED alternatives.
A good quality derailleur running good quality sprockets and
chains (and bearings) and a good two wire selector mechanism will
work as smoothly and as trouble free as a dream, and with minor
regular maintenance you'll wear out several tyres and sets of
brake pads before you replace anything in the drive train.
Pop into richards (url below) and he can sell you all the fancy
crap, but he also builds stuff and services stuff for customers,
using all these "obsolete" technologies, and I'll
tell you it works as smooth as silk (lot better than the girls
bike for instance) and is a mechanical joy to behold, effortless,
damn near silent, and highly efficient.
Electronic gear shift for bicycles
Shimano are about to offer an electronic upgrade to their top of the range dura-ace component set. This system will be 67 grams lighter than the current mechanical system, though a bit heavier than the system they plan to deliver next year. It runs on a 7.6 volt lithium-ion battery.
For a cycle commuter like me cable actuated gears are a total PITA. They dry out, they stick and they fray. You get them adjusted just right then it rains and the oil washes away and you are back to square one. Murphy's law means that they invariable fail on a monday and leave me replacing the things in the middle of the night after work. Yes, I have a lot of hate for gear cables and several times I have considered throwing them out and putting in my own stepper motors.
zogger, thanks for the wired link in your Toyota Winglet story. They linked to the dura-ace one and I decided to submit an article.
I never seemed to have problems with my old seven speed bikes. Every shift was very precise. Now I have a nine speed system. After shifting two clicks left on the cluster the next click to the right invariably does nothing. The cable for the front derailleur fails inside the shift lever which gets ripped up when I replace it.
Maybe we could eventually have a semiautomatic system. Something tied to pedal cadence. If so, mine will be set to 120.