"Upgradable" Remote, Loses Upgrade data!
Loren Coe
RE: "Upgradable" Remote, Loses Upgrade data!
i am looking for suggestions as to the cause and possible cure for
a serious defect that i have discovered in my Universal Electronics
Model URC-6050 OneForAll(8) ir remote control (if you are not
electronically inclined, skip the technical stuff, but please read
the last paragraph).
the company is aware of this problem but refuses to rectify it,
offering only to "re-program". i am not posting to flame UEIC,
(although, in this case, they richly deserve it) but to solicit opinions
and suggestions about what can be done to fix it. the capabilities of
this particular design, AFAIK, does not exist in any current product
from any manufacturer (and i purchased _three_ of them).
the details:
40pin dip, Motorola SC410971P
8pin dip, ATMEL605, 24C08, *PC18 B (the * is actually a "bullet")
2N3904 (volt reg?), power supply is 4xAA alkaline batteries
(nominal 6.0 volts)
two 2N4403 (led circuits)
several gold tol. resistors, diodes, caps (including one 100uf, 10v
(electrolytic)
the symptom:
starting from new (after factory upgrade), units perform flawlessly
until batteries are spent (about 11weeks, batteries actually have about
50% capacity left). after replacement of batteries, unit works for
some small number of key strokes (devices added in the upgrade still
respond) then shutdowns completely. nothing will reset it except for
battery removal. afterward, the added devices are no longer present
(un-recognized, just like before the upgrade).
after dying in the identical sequence, all three remotes were sent
back in for "reprogramming" (no charge except for shipping to UEIC),
received back and all worked fine, for _TWO_WEEKS_! after which
followed the _same_scenario_ , with one minor variation: the fresh
alkalines still measure 1.51 volts each, and instead of going "flakey",
(with "low" batteries) the units just shutdown.
the question(s):
who can give any details on the Motorola and Atmel(?) chips. is there a
single pin that i can isolate to prevent any changes to the on board
memory? does this memory require some kind of "keep-alive" voltage?
how is that accomplished? the unit comes back from the factory sans
batteries.
there is a three-pin jack in the battery compartment that is used by the
factory to install the "upgrade". is there something that i can do here
to insure that the changes are not lost?
is this design just plain broken? or is there something wrong with a
batch of the Mot. or Atmel chips? is there a "common" design error,
based on bad data from the chip manufacturers (like these chips need a
different voltage)? is the ATmel chip part of the circuit that allows
the memory change? it seems clear from the few traces that i can see that
the EEPROM(?) is on the MOT. chip, and that the ATMEL chip has nothing to
do with keyboard decoding.
any suggestions comments, or, what the hell, blind speculation, would
be welcome. i am more interested in fixing these things than i am in
bad-mouthing Universal Electronics for ignoring this issue, but i would
welcome any anecdotal evidence from folks who have had this problem andor
any responses that you have had from UEIC (which is why i cross posted).
thanks,
loren
Michael House
The Atmel 24C08 is an 8K EEPROM. This is where some of the memory
is held. Without a schemo, I wouldn't know if the memory it holds is
the data that you enter, or if it's the operating system that
Universal Electronics downloads into it.
In my experience, these chips are notorious for giving memory or
data problems.
Michael