Post by James McClaveWow , lots of good info there. The resistor idea sounds kind of tedious.
Give it a go, you have absolutely nothing to lose,
if the chip is indeed stuffed.
Whenever you get random counts, numbers missing etc, this is always
the cause. The memory unit cant find the "77" or "27" numbers, or
detects the number sequence problem, and keeps searching. As this
device isnt particularly intelligent, it has no "time out" etc and
just turns forever.
Its from the wipers making bad electrical contact with the disc, they
can connect and disconnect as the disc turns (its probably warped very
slightly or something)
I would not advise doing anything at all to the disc as it is gold
plated, The resistors fix it by providing more current through the
wipers (note this is now standard practice in ANY design involving an
IC and a mechanical switch) making sure they will conduct and send the
right signals to the IC
Post by James McClaveNot crazy about the MP3 idea either. Yes, the numbers when scanning did
not read in sequence either. Too bad, from what I know this memory unit
was only used in this model and maybe the R-82? Jim
Was used in everything ROWE made in solid state prior to R-84
I personally have seen this system in RI-1, R81, R82, R83
I think that the 1975,6 models were available as either solid state
(with memory unit) or electro-mechanical.
I have no idea how many were sold in each type or how popular they
were.
There are few US made solid state ROWE here in Australia prior to
R-81.
Last week I copped a service call from a private owner to service a
"ROWE electronic" that was in a woodgrain cabinet with a lift up lid.
I thought originally it was a kentwood (1979 with central control
computer) and that it would be a simple repair, but to my despair it
had the old board set, and a totally dead memory unit, scanned
indefinitely and showed nothing on the display - (except when you
punched in a selection).
With the cost of an exchange unit, and 2 way international postage
approaching the best part of $400 AU he still seemed keen to get it
fixed for some reason, and didn't want to go with the MP3 option.
Had one R-83 machine come in this week for repair. After I fixed other
faults in the unit, it turned out that it had a "sort of" working
memory unit, however it repeatedly plays "293" over and over
indefinitely after playing the selections that you DO want (until you
hit the "clear memory" button.)
While we had the chance (with a working memory unit in captivity), we
took it apart, monitored the serial interface, drew up timing diagrams
for the serial interface, and the sequence of operation of the IC and
feel more confident now about it being feasible to make a substitute,
since there seems to be enough demand (and dead machines piling up,
taking up room in my workshop). Once I get a few jobs out of the way,
might start on a prototype