From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Waterproof bike computers?
Date: 14 Dec 1998 18:31:28 GMT

John Abatte writes:

> Now that's really a puzzler.  I have a Cateye Mity2 I bought well
> over a year ago on my MTB which acts flaky as all get when it gets
> wet.  The one time I intentionally wanted to ride in the rain I'd
> gotten about a mile from home when I noticed the speedo reading
> jumping all over the place. At one poiint it read about 50 mph, and
> while stopped at a traffic light, it read 14 mph.

The problem here is not water getting into anything but rather that
the sensing is done by such a high impedance circuit that water and a
drop of sweat as electrolyte will short the input pins.  This makes the
instrument think it is seeing reed switch closures when there are none.

Using a reed switch is the easiest and most accurate way to sense
wheel revolutions but it has the drawback that parking the wheel with
the magnet near the sense switch can drain the battery.  Reed switch
units send a sense current down the wire to see whether the magnet is
in front of the reed switch that closes in the presence of a magnetic
field.  To prevent battery failure, or even excess drain in operation,
a high impedance RC circuit is used.

In contrast the Avocet 35, 45, 50 use an inductive system with which
the pickup generates a voltage high enough not to be masked by a
little water.  To make up for that, their units are not uniformly
sufficiently waterproof to prevent water intrusion in to the
instrument although many of them do not leak.  So there we are, suck
between the usual rock and hard place.

Jobst Brandt      <jbrandt@hpl.hp.com>