Re: Fail-safe for keyless entry

["Robert Green" <robert_green1963@xxxxxxxxx>]

"Ian Shef" <invalid@xxxxxxxxxxxxx> wrote in message
news:Xns9D5F7CDCCC507vaj4088ianshef@xxxxxxxxxxxxxxxxxx
> "Robert Green" <robert_green1963@xxxxxxxxx> wrote in
> news:hqi2ge$6a7$1@xxxxxxxxxxxxxxxxx:
>
> > I've got the typical minivan keyless entry keychain fob but it's got a
> > serious design problem.
> <snip>
> > The real estate inside the fob is very cramped.  It runs off a CR2032
> > coin cell battery, and the circuit traces are simple enough so that I
> > could scrape off the 3VDC+ line and bridge it with a small switch.  That
> > way, I could just put in a tiny pushbutton that had to be depressed as
> > well as the desired command button each time a command was sent.  That
> > approach makes it a two-handed operation, whereas a slide type ON/OFF
> > switch could still be worked with one hand.
>
> Before you scrape circuit traces or install switches, please try a simple
> experiment:
>
> Remove the battery for a few hours (as if you had turned off the switch),
> then reinstall the battery and see what it takes to unlock your minivan.
>
> Many keyless entry fobs use a trickle of battery current to maintain state
> information (such as where they are in a rolling code).  The system can
> reacquire the state, but it may take multiple presses.  Furthermore, since
> the state after power-off/power-on is likely to be the same each time (for
> the same fob under similar conditions), you may be defeating the purpose
of
> the rolling code.
>
> I may be wrong, but the experiment is a lot easier than all of the wasted
> work if I am right.

Even if there's only a "remote" possibility of the unit losing its
programming (I couldn't resist the pun!) it's a good thing to check.  Very
good contribution, Ian, and it's a good reminder to always get someone to
review project plans!

Before reading your post, my latest thought was to cut out some thin plastic
the same size as the battery, place a contact on it, and slide it between
the battery and the case so that the battery can no longer touch the contact
on the board and then lead the contact that I attached to the plastic
insulator, run it to a tiny switch (maybe a small membrane switch attached
to the back of the fob) and then solder a return wire to the battery contact
on the board that would now be prevented from making contact with the
battery except through the new pathway.  Only one point on the fob circuit
board needs to be touched doing it this way, and no traces need to be cut.

Now I'll just unload the battery from one of the keyfobs and see what
happens.

Your post got me thinking that there has to be some way of keeping the
battery alive but still require a second switch to complete the circuit.
Perhaps there's a common line from the switch to the ground that could be
interrupted so that it needed to be closed when the desired button was
pressed.  Thanks for reminding me it's a good idea to try to come up with a
solution that doesn't affect any trickle current going into the IC.

I guess it's time to take pictures of both sides of the board and come up
with a rudimentary circuit diagram.

Thanks for the heads up.

> Good luck!

I'll need it.  Experience suggests I may, as Tony predicted, end up with a
dead fob but you know the old saying - experience is gained proportional to
the amount of equipment ruined.

--
Bobby G.