Re: How do STI brifters sense the gear position?

Jasper Janssen <jasper@xxxxxxxxxxxx> wrote:

> On 6 Oct 2005 14:39:53 -0700, "spin156" <spin156@xxxxxxxxxxx>
> wrote:
>
>>Some day I will get a defunct set of brifters and reverse engineer
>>exactly how this works (but I'll bet it's just as I described).

I'm surprised nobody here has ever done that.

> If it's actually a positive indication, probably, yeah. It might
> just be a system that sends "gear lever up pushed" and "gear lever
> down pushed" signals instead, which'd just work with two switches
> and three wires.

But then it would get out of sync with the lever if anything went
wrong -- or if the head were disconnected. In my case, it's usually in
sync. If the display doesn't change when I shift, it eventually will
change, and it will be right, regardless of how many gears I shift.

> Although if it was a set of 9-10 resistors, it'd need only 2 wires
> (the Vsig/Gnd wiring can be inside the head), so actually I think
> my hypothesis above is more likely.

There are four contacts between the head and the mounting bracket, but
one of them doesn't do anything. Disabling any of the other three (by
covering it with electrical tape, for example) causes all functions to
stop working -- the speed sensor, the gear indication, and the two
buttons.

I measured resistances between various pairs of contacts. All were at
least several megohms, except one pair involving the unused contact was
2.8 kilohms in one direction, "infinite" in the other. None of the
resistance values changed much when I changed gears, but I was using a
not-very-good meter with not-very-good test leads, so that's a bit
questionable.

On the computer head, three of the four contacts are at +2.8 VDC
relative to the fourth contact, which has an o-ring around it.

Maybe one of these days I'll get energetic enough to hook up a scope.

--
Ray Heindl
(remove the Xs to reply)
.