| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 2168.1 | It's in .2 | RUTILE::MACFADYEN | it's only a dream | Mon Jan 20 1992 04:25 | 12 |
| Hi Graham,
Thanks for putting your note in. So that everyone else can read what's
being discussed, I've put the full text of the article into the next reply.
DECwindows notes users, be aware that it's a long note, ~ 260 lines.
As for what sparked this journalism bit, I dunno. Just an unstoppable
desire to shoot my mouth off, how about that. Besides, you get pocket
money too.
Rod
|
| 2168.2 | That article in full | RUTILE::MACFADYEN | it's only a dream | Mon Jan 20 1992 04:27 | 260 |
| Automatic Gear-shifting for Bikes-a Proposal
I've been interested in bikes for ages, but I've been
interested in cameras for longer than that. In the late
seventies I got my first serious camera, a Voigtlander
made in 1961. It was totally mechanical, beautifully made,
and had all the features you needed: adjustable shutter,
aperture and focus. I learned a lot from it. By the early
eighties it was obvious that a revolution was taking place
in camera design. Electronics were automating manual
features and allowing brand new ones. Things have moved
on since then and the 1991 equivalent of my old Voigtlander
has autofocus, autoflash, autozoom, auto-everything. You
can rely on it to take sharp, well-exposed photos.
In 1961, had you been able to glimpse the future, you'd
hardly have been able to recognise a 1991 camera as a
camera.
But you'd have recognised a 1991 bike as a bike.
OK, that's a little unfair. Wheels have to be round, and
the upright cycling position is hard to beat, so the basic
structure of a bike is less open to change than that of a
small hand-held device like a camera. What I am saying is
that mechanical devices can be radically changed by the
introduction of electronics. Even bikes. In a small way
it's happening already as cycle computers become a standard
accessory. So what's next?
Cadence is the Key
I think what's next is the production of a good automatic
gear-shift system for bikes. It would help beginners who
don't know when or why they should shift gear, or indeed
any cyclist who finds gear selection a chore. They would
all welcome a reliable automatic system.
There have already been mechanical systems, but they have
not been successful, either surprising the rider with
unwanted shifts or being over-complex. This only shows how
tricky gear-shifting is! So before designing a new system,
we should analyze the process of gear selection. How do we
know when to change gear?
I've presented this analysis in the diagram. It shows how
speed, cadence and gear selection change as a competent
cyclist approaches, climbs and descends a hill. The key
point is that the one thing which remains constant is the
cyclist's cadence: cyclists change gear when they feel
their cadence becoming uncomfortably high or low.
Cadence is the rate at which you spin your legs. The fact
is that our legs turn the cranks most efficiently over
just a narrow range of rates, around 90 rpm for a good
cyclist. But you must cycle at a wide range of road speeds,
as dictated by fitness, inclination and circumstance: 10
mph panting up a hill, 20 mph on a good flat road, 30 mph
speeding down the other side of the hill. The idea of using
gears is to keep your legs spinning at the same rate while
travelling at any of these road speeds.
So my proposed automatic gear-change system depends on
cadence. It would monitor your cadence and when it detected
a significant change, trigger the gear-shift necessary to
maintain your cadence around a preset constant value. I'm
not going to propose a detailed mechanical design (that
would only emerge after development and testing), instead
I'd like to speculate on the control program needed.
Necessary Technology
There are three devices necessary to realise the system:
o A device that measures cadence and sends a signal to
trigger gear-shifts when required.
o An electrically-operated front shifter.
o An electrically-operated rear shifter.
As I see it, the system would build on modern and familiar
technology.
The first device is no more than a slightly modified cycle-
computer. Cadence is already a standard feature and the
only modification required is a trivial amount of computing
ability and an output to trigger gear-shifts.
The second device exists. The Browning electrically-powered
front shifter came out in 1988 and is now made by Suntour.
In its current form it shifts over a triple chainring and
is aimed at the demanding top end of the mountain-bike
market. It is reported to be very reliable under load.
The third, an electrically-operated rear shifter, is the
only new device required. It could be very similar to
current rear derailleurs, requiring only the addition of
a small actuator mechanism inside the upper parallelogram
portion of the derailleur to move the derailleur in and
out. The development of indexed shifting over the past
few years has led to equipment which is just waiting to be
automated like this, and I'm surprised it hasn't already
been done.
In summary, the system would use a cycle-computer to
measure cadence and transmit a signal when necessary to
the electro-mechanical front and rear shifters. You the
cyclist would notice the gears changing of their own accord
as you speeded up or slowed down, the cadence staying
firmly fixed. You could concentrate more on the world
around you and less on the bike.
Theoretically Speaking
The diagram shows how gears are used to maintain a constant
cadence over varying gradients, just as my proposed system
would do for automatically. The diagram also hints at some
key features of the computing necessary to decide when to
trigger shifts...
o Cadence must be allowed to vary, without triggering gear
shift, within a "dead zone" around the nominated ideal
cadence (shown here as plus or minus 5%). If the system
tried to hold an exact cadence it would be triggering
gear shifts all the time; very undesirable. The width
of the dead zone depends on the application. On racing
bikes with close ratio blocks you can hold a cadence
closely; on bikes with wider blocks you must tolerate a
wider variation between gears. There's a balance to be
struck between excessive gear-shifting on the one hand,
and on the other allowing cadence to vary over too wide
a range.
o In percentage terms, the dead zone should be a little
wider than the difference between adjacent rear gears.
This means that the change in cadence caused by a gear
shift will not be so great as to move the cadence out of
the dead zone and so trigger another shift.
o Front gear shifts may well cause a change in cadence
large enough to trigger a compensating rear shift
(the diagram shows one occurrence of this). However,
this matches cycling experience: if you make a front
shift, you must often make a simultaneous rear shift to
maintain a comfortable cadence.
While we're on the subject of front shifts, how would the
system minimise their frequency? Cyclists usually prefer to
do most of their shifting on the rear gears. Let's imagine
the strategy for a gear system with two front chainrings
and seven rear gears.
On the small chainring, the system would use only rear
shifts while gears 1 to 5 were adequate. When a higher gear
than small-5 was required, the system would trigger a front
shift to the large chainring. Once on the large chainring,
the system would again use only rear shifts while gears
3 to 7 were adequate. When a lower gear than large-3 was
required, the system would trigger a front shift back to
the small chainring.
This strategy would keep front shifts down to the minimum
necessary and, again, I think it matches the way cyclists
actually use gears.
Practical Points
Seeing as we're not robots, some aspects of real cycling
don't fit the theory I've outlined above. How would the
automatic system cope with contrary human demands?
First, the preferred cadence should be set by the cyclist.
Not everyone is a racer who can spin his legs into a blur.
With a little experimentation you could find the cadence
that suited you.
The permitted variation (dead zone) should also be user-
settable, so that the system could cope with different
ranges of gears. Perhaps an intelligent system could sense
the variation in cadence caused by a rear shift and adjust
the dead zone accordingly. A more technically complex
approach would be a system that knew precisely what gear
ratios it had. This would require either a system built
with unchangeable ratios, which would be inflexible, or the
cyclist to enter the gear ratios onto the computer. Given
the trouble I have setting my watch, that might not be such
a good idea either.
What would happen when you freewheeled? The system could
be programmed not to make shifts while the cadence was at
or close to zero. Alternatively, there could be a button
to put a temporary freeze on shifting. This would be useful
for out-of-the-saddle climbing.
The idea of freezing shifts could be taken further. You
could have a fully manual mode, with push-buttons on
the handlebars to initiate shifts. That would be very
flash, and I for one would appreciate gear controls on
the handlebars of drop-handlebarred bikes. The existing
mechanical versions of this fail to convince.
And what would happen if you had to go slower than the
slowest speed obtainable with the lowest gear at your
fixed cadence, or faster than the highest speed obtainable
with the highest gear at your fixed cadence? Simple, the
system would simply leave you in the lowest or highest
gear. There's nothing else it could do.
Will It Fly?
The system I've outlined here is based on familiar
and effective technology. The difference is that the
introduction of electronics enables it to achieve something
that purely mechanical systems have failed to do. I'm
willing to bet that an automatic system embodying similar
ideas to those sketched here will be launched by some
manufacturer within the next year or two.
Initially, it's liable to be expensive and have teething
troubles, like any new product. It may never be a low-
end system. It will introduce new worries to cycling. You
might initially find it strange for your cadence to be held
remorselessly constant, and for shifts to happen of their
own accord without warning. You could be grounded by a flat
battery!
After technology enthusiasts, the first real customers
are likely to be triathletes, who have shown themselves
to be more receptive to new ideas than more traditional
racers. They will appreciate a system that removes one more
distracting element from the serious business of working
at the limit. When the price comes down, the system will
spread into the leisure market, where newcomers to cycling
will love it. Traditionalists who stay with mechanical
systems will then be able to feel a satisfying glow of
superiority. The RTTC will ban it.
Whatever form a successful automatic gear system takes,
it's worth looking at two precedents. The change having
the biggest impact on the cycling world in recent years has
been the introduction of mountain-bikes. This new product
brought people into cycle shops who'd never been there
before and took unprecedentedly large amounts of money off
them. Secondly, to return to my analogy with the camera
world, modern-day cameras sell in quantities undreamt of in
the Sixties. Under the urge of constant change, the market
has exploded.
I won't go as far as saying that my system will do that for
the bike world, but it might give it a little shove. And
anyone sitting on a bike appreciates that.
�1992 Rod MacFadyen
|
| 2168.3 | | LJOHUB::CRITZ | | Mon Jan 20 1992 11:56 | 15 |
| All this is very interesting. We've had auto trannies in
cars for a long time. Within the last couple of years,
some of the top Formula I racers have started using
something similar, although they control when the shifts
are made.
I remember watching Nigel Mansell a while ago from an in-car
camera and wondering why his right hand never left the
wheel. Interesting to hear the engine revs change and never
see the hands leave the wheel.
I thing .2 has a great idea. Kinda reminds me of the differences
between my old Minolta and my (much newer) Pentax.
Scott
|
| 2168.4 | where's the benefit? | WUMBCK::FOX | | Mon Jan 20 1992 13:57 | 13 |
| With the rule of thumb (that's I've heard) being "shift before
you have to", I'm not sure an auto transmission would be accepted.
Often I maintain or shift in anticipation of upcoming terrain. An
auto seems to void that option.
What about sprints? Would the tranny suddenly do something
contrary to what the racer is trying to accomplish?
With ergo shifters, the time wasted is less and less a factor.
An auto transmission is not going to save more time, and will
take control away from the rider. I don't think we'll see them
as a serious choice for most applications - most likely not
at all, imo.
John
|
| 2168.5 | | CSCOA1::HOOD_R | | Mon Jan 20 1992 14:24 | 19 |
|
I believe that it is a possibility for a couple of reasons. For one,
the idea would sound really appealing to people who just can't ever
get the hang of shifting. We don't know many of these people because
they usually give up biking because they just can't get the hang of or
the reason for shifting. Yes.... such people do exist. I saw somebody
the other day that paid a couple of hundred bucks for a bike that
does some kind of mechanical-automatic shifting in the hub. A top
notch, well thought out electronic shifting mechanism would be very
appealing to this type of person. The other reason that I think that
such items are a good possibility is that it is one more way for
bicycle manufacturers to hook you on their products. Once you have
their computer that measures cadence, you will have to buy their
electronic transmissions. You will be financially tied to their line
of electronic bike products and their upgrades. There is no evidence
to suggest that they will agree to any kind of standard.
doug
|
| 2168.6 | | DANGER::JBELL | Zeno was almost here | Mon Jan 20 1992 14:35 | 11 |
| I don't mean to discourage you, but....
How are you going to get it to shift up when you stand to pedal?
How are you going to keep it from shifting down when you coast?
How does it know to downshift at a light?
Haven't automatic shifters been introduced dozens of times already?
I recall reading about them about every three years or so since
the late seventies.
-Jeff Bell
|
| 2168.7 | its ok to learn and work hard ,still. | WLDWST::SANTOS_E | | Mon Jan 20 1992 15:50 | 15 |
| Hi , Its a good idea but thats not cycling anymore to me.
I think its easier to learn how to shift than learn how to find
a bad component or a bug in the cpu or a bad servo , and imagine
if you crashed and the mini nuclear power generator housing splits
in two and kills you and a 10 mile radius around the site ? Oh boy
that could hurt. the same bicycle frame design has not been changed
for a 100 years well at least 80 cause its simple and its good.
I just dont agree in hanging more gizmos than needed and paying more
and more for parts etc.
Perhaps someone would come up with a motor that acts as the bottom
bracket spindle , so we can mimick pedaling on our own power and the
supre record model has turbo for sprints. I am sorry but your dream
is my nightmare.
Ed S.
|
| 2168.8 | | CSCOA1::HOOD_R | | Mon Jan 20 1992 17:05 | 19 |
|
> How are you going to get it to shift up when you stand to pedal?
> How are you going to keep it from shifting down when you coast?
> How does it know when to downshift at a light?
Pressure sensors in the pedals and seat. Establish some minimum
push/pull load that defines when a cyclist is really cranking and when
he is just coasting. Put a sensor in the seat to determine when he
stands. Personally, I pretty much hate the idea of anything electronic
that could prohibit me from cycling. It sort of ruins it in my mind.
I do believe that the means exist to take a subtle queues from
the cyclist and make a good guess as to which gear he/she should be in.
This guess may not be good enough for racing, but might appeal to the
masses who really can't do any better.
doug
|
| 2168.9 | | RUTILE::MACFADYEN | wreathed in joy | Tue Jan 21 1992 03:04 | 18 |
| Well I never. Who'd have thought Digital was home to all these technophobes?
People in a *computer* company want to ignore electronics and stay
mechanical...
Anyway, back to the point. The criticisms that have been raised in the
preceding replies mostly relate to a racer's worries. But honestly, racers
are a small segment of the market. There are a *lot* of cyclists who don't
sprint for corners, or frequently exert themselves out of the saddle.
Perhaps they might appreciate an automatic system. More than that, there
are a *huge* number of people who don't cycle at all. Maybe that's because
they're put off the by the same technicalities that appeal to you lot?
If they could buy a bike that could handle all that gear stuff by itself
they might be pretty happy. Maybe it could bring a second crop of newcomers
into bike shops, in the way that ATBs have brought people in who had no
previous interest in cycling. That would be good for all of us.
Rod
|
| 2168.10 | A LITTLE BLASPHEMY | WMOIS::GIROUARD_C | | Tue Jan 21 1992 06:27 | 13 |
| I agree with John (to a point). They will never be a reasonable
inclusion to the racing game (unless they can do something with
telepathy).
The other point about the market size is well taken. Recreational,
touring, and the lazy folks who do about 5 miles a month would
probably enjoy the elimination of the distraction...
Me, everyone probably has me pegged for a techno-weenie already, but
I can't see myself ever entertaining something as blasphemous as
relying on a set-up like that...
Chip
|
| 2168.11 | loss of control? | SHALOT::ELLIS | John Lee Ellis - assembly required | Tue Jan 21 1992 07:54 | 26 |
|
The "technophobic" reactions here go beyond practical limitations
of the device. I've pointed out some of these limitations and others
in handling not only racing tactics but competent-cyclist strategies
in dealing with hills, curves, etc. Rod has solutions for some, none
at the moment for others. Telepathy seems like the easiest fix. :-)
But, as I say, there's something more underneath the "technophobic"
I think. My suspicion is that a big reason we enjoy cycling is
mastering the equipment, and being in touch with it. We relegate
(perhaps unfairly) Rod's idea to neophytes who otherwise wouldn't
cycle, because maybe we fear it would rob us of one of the pleasures
of the sport. (Is Indexed Shifting the same sort of thing? Different
order of magnitude hence more acceptable? An aid to the cyclist's
control, rather than taking control away from the cyclist?)
At the risk of being a "technogrouch" my gut reaction is the same:
"Keep me in control; give me something to do; don't make me into just
a pedalling engine." (BTW, autofocus SLR's brought a similar reaction
and debate, which continues to this day.)
Having said that, I like Rod's thinking, and believe it could help
a sizable number of people use bikes for some of the best reasons:
day-to-day transport, laid-back touring and "promenades," etc.
-john
|
| 2168.12 | ATB's can be had for >200, can this? | WUMBCK::FOX | | Tue Jan 21 1992 10:12 | 12 |
| I don't think you'll see this as an invitation to the non-cycling
masses. If this device is going to shift on its own, it'll have to
work very well under load. You don't find drivetrains that shift
well under load for entry-level prices - never mind ones that
have an auto-tranny as well! No non-cyclist is going to get off
the couch for a bike costing more that $500. And to bring the cost
of this down, it has to be around for a few years. To do that, it
has to be popular among the current crop of cyclists. Anyone who
rides now isn't about to drop out cuz they hate shifting. It won't
work.
John
|
| 2168.13 | I "fixed" it | NEMAIL::DELORIEA | I've got better things to do. | Tue Jan 21 1992 10:29 | 21 |
| I found the answer.
Have everyone ride a fixed gear.
No more worring about shifting and all that bother.
Actually, I remember taking a vacation down to the Cape
and the Inn supplied bikes to ride around Chatham(sp).
They were single speed beach bikes. The big white wall
tires, bull horn handle bar and coaster brake, brought
back the fun and joy of a simple ride.
Sure some hills you were better off walking up rather
than trying to ride this 30+lbs beast up but thats all
part of the experience.
This is were an auto-shifter would really fit into
cycling. Rental shops would like to see a bullet proof
auto-shifting system.
T
|
| 2168.14 | My attorney owns one and a bmw too | WLDWST::SANTOS_E | | Tue Jan 21 1992 11:08 | 18 |
| the market that I thought of is not for the non riders Heck my
wife won't ride because she's scared of fallin and breaking her
nails. And look at our kids they shift 2 out of 10 too high or
too low . Oh how about the recreational cyclist who only rides on
top gears ( it mimicks walking ) One guy asked me for gears that
will let the bike go real fast while he pedals less rpm . and not
too much load.
THE MARKET is for Le mond and others during their time trial
events so they can tuck and hammer down without having to change their
riding positions. They also use the leading edge of the techonlogy,
sponsors have deep deep pockets, don't have to get off the saddle
unless starting or end sprint. Their aero/composit frames can hide the
parts , the rear disk can be solar panels for power source . And they
will do (buy?) anything to help them win.
Yeah what a good note
Ed S.
|
| 2168.15 | | RANGER::WASSER | John A. Wasser | Tue Jan 21 1992 17:33 | 24 |
|
The only technology needed to make this workable is a rear
cluster you can shift UNDER FULL LOAD. Unfortunately that
is not a simple variation of the existing derailleur. Pushing
a roller-chain sideways to get it to mesh/unmesh is something
you don't want to do under full power.
Someone brought up the argument "how can it 'shift before
you have to' like I was taught?". The answer is that it
doesn't have to. You shift before you have to because you
have to unload the drivetrain to shift... going up a steep
hill and unloading the drivetrain are incompatible (you
loose speed rapidly!). If you could downshift WHILE you
were pumping full-steam up a hill you wouldn't have to
downshift BEFORE you start up the hill.
Someone else brought up the argument "how will it know I
am standing?". The answer is: it doesn't need to. If
you stand up to put more power to the pedals your cadence
will increase (more power = faster). Get your cadence up
high enough and the machine will upshift for you. Slack
off and it will downshift for you.
Sounds like a great idea... Now we just need the technology.
|
| 2168.16 | Compare bikes with cars (if you know what I mean!) | GALVIA::STEPHENS | Green Eggs and Ham | Wed Jan 22 1992 03:25 | 19 |
| One should probably compare the concept of automatic gears with cars. Automatic
gears are only popular with those drivers who just use their veichle as a means
of getting from a to b. Serious drivers wouldn't touch them with a barge pole.
And you see very few automatics in Europe, where the roads mean you need to
change gear a lot.
By that logic, only those cyclists who use bikes as a means of transport,
without having much interest in the bike as such, will be interested in automatic
gears.
And as for pros, the technology will need to advance to the stage where
automatic gearing is accurate enough to improve on their own reactions, before
they will be used. Ever see Ayrton Senna using an auto gearbox? ('though I
think the Williams team do?)
Patrick
|
| 2168.17 | | MOVIES::WIDDOWSON | Rod, VMSE-ED013. 824-3391 | Wed Jan 22 1992 04:07 | 15 |
| A flight of pure fancy:
Perhaps what is needed is continuously variable transmissions. This
brings to mind nightmarish Heath-Robinson type setups with belts and
the like (and lots of lost power) but this needn't be the case. For
instance it could be using a hydraulic pump/motor pair. Some of these
have the capability to convert a fixed amount of energy into enormous
amounts of force over a short distance or a small force over a long
distance. With enough (hydro-dynamic) intelligence you could `program'
this to support spinning or stomping and this could be varied during the
ride. There is none of the leg-braking sudden and unexpected changing
and you are always in a `perfect' gear.
Mind you, This would not be light, even if you made it efficient
enough....
|
| 2168.18 | IT'S NECESSARY (FOR ME) | WMOIS::GIROUARD_C | | Wed Jan 22 1992 06:35 | 23 |
| I know this isn't a "to shift or not to shift" topic, but I have to
to disagree with John's statement that "you don't have to shift before"
when approaching a hill. I agree that in recreational riding and
touring it isn't too important. My experience has been, while racing
anyway, it is. I think there's a pshychological aspect (for me) to
it as well.
One thing you don't want to do is be shifting under load (particularly if
someone is attacking) and lose a shift/gear. You can lose a lot then
spend a lot trying to make it up and get back in the game. There is
much more distance and fuel at stake on a hill then there is on the
flats/downhills.
Then there's the "avoiding the interruption of continuous flow" aspect
to this. You're in a nice comfortable cadence. Then the hill
approaches. What I don't want to do is get into something (or stay
in the gear I've been in) too small or too big. I'd much rather "spin
into the hill" then shift when I get there or after (when my cadence
begins to drop).
My $.02
Chip
|
| 2168.19 | hill and training strategies | SHALOT::ELLIS | John Lee Ellis - assembly required | Wed Jan 22 1992 07:46 | 18 |
|
My experience about hills (alas, as a non-racer) is much like Chip's.
In addition, I look at the hill as I go into it and judge whether
I can "power" over it (avoiding a gear change) or not. There are
advantages if you can power over it. Others may have said this before.
I said "look at the hill" on purpose. Sounds like we'd need to add
radar (in addition to telepathy?) to take care of this.
People also choose different gears going into hills depending on
whether they're training for power, speed, whatever, or are just
tired. I may take the same hill in a high gear one day, to work
on strength development; spinning another, to work on speed and
cadence. Training means at times *avoiding* what is most efficient.
(Guess that means training is at odds with a system designed to
make pedalling most efficient?)
-john
|
| 2168.20 | Sounds like my SIS (Sometimes It Shifts) | NOVA::FISHER | Rdb/VMS Dinosaur | Wed Jan 22 1992 08:55 | 5 |
| Didn't Eddie B have some advice that went something like: "When you're
climbing and you think you're going to die, upshift! Noone else will
be able to react in time."
ed
|
| 2168.21 | | RUTILE::MACFADYEN | cast into despair | Wed Jan 22 1992 09:36 | 23 |
| I suspect that we're all sloppier about gear-changing than we admit or
even know about. I'm prepared to bet that my system could offer performance
improvements. It would hold an exact cadence for you and you wouldn't need
to break concentration to shift. I'm sure it would be good for time-
trialling. The plain fact is we don't have the experience to know what it
would be like to ride such a system, and even if we had slagged it, we might
find to our surprise that we liked it. Photographers have been through this
process many times: through-the-lens metering was going to destroy real
photography; automatic exposure was going to destroy real photography;
autofocus was going to destroy real photography. Somehow these changes didn't
do that, and almost all the doubters got converted. For those who didn't
there is still plenty of enjoyable and *pure* equipment.
What's the big deal about knowing how to change gear anyway? Is there some
sort of closed shop in operation for cyclists who know how to change gear?
I certainly seem to have touched some vein of disgust. I can sympathise with
it, but I'm not impressed.
I wish I could put a demo system together myself, but I'm not a skilled
electro-mechanical engineer with time, money and good workshop facilities.
Rod
|
| 2168.22 | autoshifters | WLDWST::SANTOS_E | | Wed Jan 22 1992 10:53 | 22 |
| I read an article last night in the feb. issue of bicycle guide .
It was an interview with Shinpei Okajima the r&d Director for
Shimano. The core of the article is about shimano's advancement in
the technology and how US the traditionalists did not accept change.
Any ways now they have SIS perfected and 80% of the market . Wouldn't
DEC like to have that market share.
His comment was we are at the top of the mechanical evolutionary
cycle ( what ever that means ) . He then said that electronic systems
are on the way . Using the SLR camera aproach he showed how it evolved
from mechanical to electronic. Acording to Okahima they have been
working on this subject and can come up with a system in a few weeks
though it will not be reliable . The biggest problem to him was the
power required to make the shift , batteries are heavy. He do not
think that we will be there soon . But who knows
On the other offshoot discussing when to shift .Its best to keep your
speed the same and not shift until you start to bog down . Notice in
a pack if the person in front of you shifts too you tend to climb on
top of him ? cause he lost 4' of travel during the shift.
Do you shift your car from 4th down to 3rd at the a. bottom b. when it
slows down ? c. neither, d. all of the above?
theanks Ed S.
|
| 2168.23 | | WUMBCK::FOX | | Wed Jan 22 1992 11:06 | 7 |
| I think John Ellis said it best. The same cadence doesn't always
apply, and the prefered cadence may change throughout the course
of a ride. Shifting based on cadence alone won't solve that.
Now if there was a neural input to this sucker, and it would know
when you *thought* about changing gears, then I'd buy one. :-)
John
|
| 2168.24 | Semi-automatic? | OROGEN::BODGE | Andy Bodge | Wed Jan 22 1992 11:13 | 11 |
| If you already have the electronic shift system, it would be pretty
easy to put a spring-loaded switch on the bars to force an upshift or
downshift. The switch would also have to tell the computer, "Trust me
and don't pay attention to my cadence for a while." The system could
be multi-mode: pure automatic (no override switch); semi-automatic;
electronically-assistesd manual (the switch controls shifting, not the
computer). The parallels to modern cameras are clear.
Me, I enjoy shifting the gears, same as I do in a car.
Andy
|
| 2168.25 | Show me the working model | UKCSSE::ROBINSON | Twitching the night away... | Thu Jan 23 1992 04:12 | 15 |
| Re .21:-
>I certainly seem to have touched some vein of disgust. I can sympathise with
>it, but I'm not impressed.
I don't think anyone's disgusted Rod. What I see from some of the
replies to this interesting note, is a doubt that your system could be
made to work intelligently enough to be of use (at least to the style
of cycling most of this conference use).
It's also a little unfair to draw a parallel between autofocus, etc. in
photography. They are tried and tested technologies which we
know work; whereas your ideas (Shimano's plans notwithstanding) are
still only a concept.
Chris
|
| 2168.26 | autofocus - give it time... and money! | SHALOT::ELLIS | John Lee Ellis - assembly required | Thu Jan 23 1992 06:00 | 22 |
|
Actually, autofocus may be a good example. Ten years ago, autofocus
devices were heavy, slow, and cumbersome. Autofocus didn't leave the
laboratory for years, and the common man could hardly envision how it
could become fast enough, smart enough, and compact enough to be used
on a hand-held SLR. Yet, as we see, most SLR's today are autofocus,
and the market for them is huge.
Behind this success is obviously some huge companies who could devote
years of R&D before seeing a return - Nikon, Minolta, Canon, and others.
Of course, I'm sure there are lots of other concepts that got lots of
investment and in the end just never worked out.
-john
PS: Autoexposure is an example of an ever-more-sophisticated campaign
to account for a variety of conditions - we went from simple center-
weighted to lower-frame biassed (for landscapes) and so on to a choice
of "programs" (individual portrait, group portrait, landscape, blimps
with large birds in the frame, etc.). The biggest challenge for the
photographer is to figure out what the icons mean. :-)
|
| 2168.27 | money needs demand | PAKORA::GGOODMAN | Number 1 in a field of 1 | Sat Jan 25 1992 20:00 | 35 |
|
Rod,
I'd like to apologise for setting you up against this barrage of abuse.
Just a few things that these notes have made me think of...
I think maybe we've all been too concerned with what this system will
do to our own particular branch of cycling. I think that it's a fair point
that the average punter who takes his bike out with the kids every couple of
months might find an advantage with the system. He's enjoying the country air
and not the actual cycling. We enjoy (?) both. By all means, make the cycling
easier for him.
But (you knew that I was going to say but, didn't you?), I think that
you may have over-estimated the size of that group. Although the large
numbers of non-cyclists would join that group, the reasons that they don't
are purely down to fiddly gears. Most non-cyclists that I know, are put off
by the money involved; this system is going to make that even worse. Not only
is the initial outlay going to be higher, but a lot of Mr.Average's get their
punctures repaired by cycle repair shops. These shops will make a killing
with something that will always be sensitive like an automatic gear system.
Also, I think my initial statement about front changes still stays. The
system could only ever work with one chainwheel at the front.
So, maybe I was too abrupt at first. However, although I am willing to
recognise a market out there, I don't believe that it's big enough to justify
this sort of investment.
Still, you've made me think about publishing some of my thoughts. In
the meantime, we'll just need to rename this conference "Graham's Soapbox'.
Graham.
|
| 2168.28 | Critamatic (TM) Shifting | CIMNET::MJOHNSON | Matt Johnson | Thu Jan 30 1992 13:28 | 17 |
| How about a different tack on this problem, more suited to racers?
I could imagine a system that you could switch on to "record" your
shifts around a criterium course, which could then be tuned to
auto-shift to the right gear at every point. (Odometer and altimeter
could readings could be compared to minimize drift.) Then there could
be a "differential" shifter that would shift from "2 higher than tuned
setting" to "2 lower than tuned." Hook this up to a heart rate monitor
and a seat sensor so some heuristics can be added. Provide manual
override. Automatically raise the gearing in the last lap.
Use hyperglide to allow shifting under load.
Charge at least $3000 a unit.
MATT
|
| 2168.29 | | SHALOT::ELLIS | John Lee Ellis - assembly required | Thu Jan 30 1992 14:01 | 9 |
|
Yes! Yes! Yes!
Matt has addressed the two big problems (technocomplexity and price)
by finding the audience to whom those are not problems but attractions!
Well, Rod? Maybe a market after all.
-john
|
| 2168.30 | Automatic Reply | CIMNET::COSTELLO | | Thu Jan 30 1992 14:53 | 50 |
| I kind of think that the "automatic shifting" (AS) part and the
"electronic shifting" (ES) part of this debate should be separated for
purposes of discussion. All the (AS) stuff applies to the
cyclocomputer/microprocessor/semi-automatic/cyborg-heart-sensor stuff
that draws ire and criticism based on who would want it and how would
it accomodate different riding styles etc. There has been alot of
creative discussion on AS that shows it probably could be done, but it
basically applies to helping someone decide *when* to shift - you could
put a little processor on the bike now that could give you this
information. And certainly, to have the bike act on this information
without rider intervention would require ES. But what of ES itself, as
a replacement for normal mechanical shifting?
The ES part is the one that interests me, a racer, the most. The news
that an electric front deraileur has been developed for mountain biking
is new, but not surprising, because weight is much less a factor in
serious mountain bikes than in serious road bikes. (I've seen some new
front mountain shocks that with longer travel could be used on a
motorcycle). What STI seeks to accomplish is put shifting at the
fingertips of the roadracer, for out of the seat shifting. Try shifting
up in a full out of the seat sprint and you'll identify the problem
this is seeking to solve - out of the seat climbing included.
However, the shift in this case is almost always a rear-deraileur
operation. I believe that someone mentioned that it would be possible,
if normal pedal feathering during shifting were observed, to use an
electric actuator to achieve the shift. This would trade the weight of
an STI system for the weight of the electric actuator and the power
supply (battery?), and put an electric control somewhere on the
bike handlebar. That this could be accomplished more efficiently,
durably and with less weight using competing STI, bar-end, or bar-top
shifting becomes the focus of the debate. Something tells me that STI
is going to be evolving for a while, shaving weight and improving
accuracy. I won't buy it in its present form for $500, but you will
see me in line when the price is $250, the weight is cut by 30%, and it
*never* misses a shift. I don't think an ES will be able to compete
with that.
However, I will take the AS, when it has a heads up display that gives
shift status, heart rate, power output, cadence, altitude,
speed, drink fluid reminder, miles to go, body temperature,
caloric output, blood analyser (glucose level), air temperature,
and the weather channel on Oakley Razorblade Sunglasses. Cool, huh?
Maybe even a directional probe that can tell me those things about
the guy I'm chasing down? And a running percentage probability
that I'll catch him, or win the race, using strategy A or strategy B?
|
| 2168.31 | How much would Greg pay for this in March? | CIMNET::MJOHNSON | Matt Johnson | Fri Jan 31 1992 09:04 | 4 |
| Following Barney's thread tangentially.... I bet you could make money
selling a device that would INTERCEPT the readings of other
competitors' heart-rate monitors, joule-counters, and cadence meters,
and tell you which rider you should draft.
|
| 2168.32 | hmmm | NOVA::FISHER | Rdb/VMS Dinosaur | Fri Jan 31 1992 09:51 | 8 |
| Hey, and then even more money if you could intercept and alter the
signal so that the competitor would think he was killing himself
and back off a tad.
Of course, most of US would just figure the danged computer's
gone nuts and ignore it... :-)
ed
|
| 2168.33 | My technology is better than yours! | AD::CRANE | I'd rather be on my bicycle! | Fri Jan 31 1992 09:54 | 7 |
|
Boy! You guys are beginning to take all the fun out of the sport.
;-)
John C.
|
| 2168.34 | Electric shifting is here! Auto can't be far behind... | JURA::PELAZ::MACFADYEN | All I want is all I want | Mon Aug 03 1992 07:09 | 37 |
| From 'Cycling Weekly', 1/8/92. I'm going to sue for royalties!
" Gears Go Electric
-----------------
Major innovation of this year's Tour was electric transmission,
developed by Mavic as an answer to Shimano's STI, and tested by the
ONCE team in time-trials.
The radical gear-changing system is powered by a small battery
concealed in the handlebars, powering tiny toggle switches. The bike
Stephen Hodge rode had one behind the brake lever and another on top
of the triathlon handlebars. Gear changing was possible from both
positions. [A photo shows a tiny switch on the bars. The bike also
appears to have conventional gear-levers.]
The derailleur has also been modified, with an electro-magnet moving
the indexed changer across the sprockets.
An ONCE mechanic reckoned the system gave a considerable weight saving
compared to Shimano's STI - which top riders rarely used in the
mountains this year due to its bulkiness. The main drawback as yet is
the absence of effective waterproofing, which means the changers
cannot be used in the rain.
Hodge was extremely impressed with the system in the time trial [Tours
tt]. 'Changing is unbelievably fast and smooth,' he said, and this was
borne out by a quick static test before he faced the timekeeper. The
toggle switch is easy to move, giving very quick changing with
virtually no finger movement necessary. Hold the switch down and the
chain glides across all the sprockets with virtually no noise. The
electric age may be just around the corner. "
Rod
|
| 2168.35 | imagine the havoc possible. | NOVA::FISHER | Rdb/VMS Dinosaur | Mon Aug 03 1992 08:00 | 8 |
| All users in a race must register their frequencies so there's no
incidental cross talk.
:-)
Or. No shifting near demolition sites. "Turn off all two way radios."
ed
|
| 2168.36 | Probably not radio... | RANGER::WASSER | John A. Wasser | Mon Aug 03 1992 11:16 | 10 |
| > All users in a race must register their frequencies so there's no
> incidental cross talk.
Do you realy think they use a radio connection instead of wires?
That would mean two sets of batteries, a transmitter and a receiver
instead of some fine wires.
When they said "static test" I believe they meant "non-moving test"
instead of "electromagnetic interference test".
|