home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky rec.bicycles.misc:1761 rec.bicycles.tech:2848
- Newsgroups: rec.bicycles.misc,rec.bicycles.tech,aus.bicycles
- Path: sparky!uunet!gatech!destroyer!cs.ubc.ca!uw-beaver!pardo
- From: pardo@cs.washington.edu (David Keppel)
- Subject: Re: Automatic transmissions for bikes
- Message-ID: <1992Dec29.002126.20206@beaver.cs.washington.edu>
- Sender: news@beaver.cs.washington.edu (USENET News System)
- Organization: Computer Science & Engineering, U. of Washington, Seattle
- References: <1992Dec27.224239.2958@sifon.cc.mcgill.ca>
- Date: Tue, 29 Dec 92 00:21:26 GMT
- Lines: 59
-
- marc@gauss.math.mcgill.ca (Marc Sokolowski) writes:
- >[Most responses have been pessimistic, but still I wonder: an
- > automatic transmission works for cars, why not bikes?]
-
- The basic issue is that the demands of a transmission on a
- high-performance bicycle are much harder to deal with than
- the demands on a high-performance car.
-
- - Cars generally get by with 4 to 6 speeds. Bikes often
- have more to choose from so there is more work involved in
- deciding which gear to be in. There are more gears because
- human engines are much sensitive about gearing, which makes
- it all the more important to be in the right gear.
-
- - In cars, both the engine and transmission are under
- machine control. The bike transmission must either signal
- the rider to relax, or it must shift under load. Though
- modern derailleurs are far better than those of a decade
- ago, bike transmissions still screw up sometimes when
- shifting under load. Upshifts on the front are especially
- hard (except for the Browning).
-
- - Car shifting can rely on a few key inputs: engine and
- road speeds, some estimation of engine load, and desired
- speed. Bicycles are far more affected by wind: riding in
- to a given wind load requires different gearing than an
- equivalent hill load (I don't understand why, but people
- I've asked agree). Bicycles are also more affected by
- vehicle loading.
-
- - Car shifters are designed to work with a given engine
- and transmission and can be tuned and tested then do not
- need to be retuned. A bike transmission needs to work
- with a variety of riders, which has implications not only
- for the power output but also the total vehicle loading.
- Worse, the engine power characteristics of any given rider
- change rapidly as the ride progresses depending on heat,
- exhaustion, rider position, etc.
-
- A truly great automatic transmission would probably need to
- monitor the rider's physiology in order to decide when to
- shift. While approximations might be successful, all fully
- automatic transmissions to date have been far too crude to
- be used for high-performance cycling.
-
- A transmission could use hints from the rider to decide when
- to shift. But because the rider would need to update those
- hints frequently, the rider would be as busy as if s/he were
- doing the shifting manually.
-
- For low-performance markets where the loss of efficiency is
- less important, a simpler transmission (e.g., only a rear
- derailleur) that has a modest range (11-34) and shifts well
- under load is probably a first step in the right direction,
- instead of trying now for an automatic transmission.
-
- Is this a FAQ yet?
-
- ;-D on ( Pessimism? ``Differently optimistic.'' ) Pardo
-