[HH] dead IR remotes

[Tom Metro]

I had an IR remote that was behaving like the battery was dying (slow
response; short range), so I opened up the battery compartment and found
one of the batteries was leaking.

No big deal. I tossed out the batteries. Took a slightly dampened scrub
brush and cleaned out the battery compartment. Dried it out with paper
towel, and then finished off with some canned air.

I inserted new batteries, and then the remote was completely non-functional.

I happened to have another identical remote, so I tried it. It worked,
but had been sitting around a long time, so I checked its battery
compartment. Again it had a leaking battery.

I repeated the same process as above. This time the remote functioned.

I came back a few hours later, and now the 2nd remote was no longer
functioning. The 1st remote still didn't work.

So what happened to kill both remotes?

The obvious guess is that mixing battery electrolyte with some water
results in a corrosive mixture that oxidizes the battery contacts. To
test that out, I crack open the shell of one of the remotes, find that
the contacts do look a bit dull and oxidized, so I clean them. I also
inspect the interior, and find that there is no stray moisture and
nothing else looks out of place.

I partially reassemble enough to test it, and still no functionality. I
check with a volt meter at the PCB where the battery terminals attach,
as well as right at the microcontroller, and read a good 3.2 VDC in both
places. Correct polarity.

Two days later, both remotes are still not functioning, so it can't be
blamed on residual stray moisture.

(I found yet a 3rd remote, which fortunately didn't have leaking
batteries, which is working, but is an older design that has fewer
buttons, and thus I'd still like to repair one of the newer style remotes.)

At this point I'm out of ideas.

If you're curious, you can see a picture of the type of remotes in
question and the device they control here:
http://mvpmc.wikispaces.com/Mediamvp

My IR testing has been to observe the IR activity light on the device
being controlled. I have no reason to question this, though I ought to
take a look at the output of these remotes through a video camera to see
if they are perhaps producing a signal that the receiver is completely
rejecting.

Like most modern IR remotes, it's a really simple design: a 33 uF
electrolytic cap filtering power, a 24-pin fine pitch DIP surface mount
micro (no part numbers I recognize), crystal, few resistors, a few
transistors (appear to be used to multiplex the keyboard), an IR LED,
and a driver transistor.

It's a single sided board with the only through-hole components being
the cap, crystal, LED, and the battery contacts. The cap and crystal are
close enough to the battery compartment that they could have been
exposed to some moisture and compressed air, but I can't see that
killing them. I know cap isn't shorted, and if it failed open, the
remote would still work, but would have reduced range due to starving
for current...or it might have a truncated pulse train if the current
spike causes a "brown out" for the micro. Viewing the output on a video
camera might shed some light on that.

Short of probing things with a scope (checking that the oscillator is
running, etc.), I'm not sure what else to do.

(On a side note, I noticed the silicone keyboard has a sticky residue
all over the surface that contacts the PCB (and on the PCB), which I've
found when disassembling pretty much any IR remote that's more than 3 or
4 years old. No, soda or other liquids haven't been spilled into it. It
seems like the silicone naturally degrades in this manner. And yet it's
always on the surface touching the PCB, as if something in the PCB
coating causes it. I've tried researching this phenomenon online in the
past and found nothing useful...other than products for repairing the
carbon pads on silicon membranes. I'd sure like to understand why/how.)

 -Tom