David Farber wrote:
"Gareth Magennis" wrote in message
...
"N_Cook" wrote in message ...
Gareth Magennis wrote in message
...
"David Farber" wrote in message
...
This Bose CD-3000 tabletop player was brought to my attention
because intermittently there were popping/static noise was coming
out of the speakers. It was quite loud. When I powered it up via
the 12 volts input, the problem didn't seem so noticeable. I took
it to the shop, opened it up, and noticed there were two car
stereo-type speaker amplifier chips powering the speakers. Using
these two datasheets,
http://www.datasheetcatalog.org/data...onics/1512.pdf and
http://pdf1.alldatasheet.com/datashe...LECTRONICS/TDA
7375.html I
was able to figure out that the TDA7375 powers the left and right
speaker and the TDA7396 powers the subwoofer. I checked the prices
of replacing these parts and each one was less than $10. Since the
pc board where they resided was difficult to access and make tests
while powered on, I decided to replace the stereo amp chip. It
didn't help. I figured then it had to
be the other chip. I replaced the other chip with no improvement.
Then I
figured out a way to remove all the pc boards and cables and power
it up
outside the unit where I could make some meaningful tests. First of
all,
there was some small dc offset going to the subwoofer. I soldered
a 330 ohm 0.5 watt resistor across the subwoofer output terminal
so it would have some load. It was about 340mv. Not great but
certainly passable. Then after about 15 seconds, the dc shot up to
12 volts and my bench meter showed a spike in current. This
voltage swing delay occurs every time I power it on. If I just
disconnect the subwoofer, the left and right speaker outputs are
fine. A schematic would be of immense help here or perhaps someone
has had some experience with this unit. Any and all suggestions
are welcome. Thanks for your reply.
--
David Farber
Los Osos, CA
Could this be yet another "conductive glue" problem?
(See the "TDA7295" thread here, first post 18/3/13, last post
confirming this problem 29/03/13)
Gareth.
Either that or a failing minor supply to an op-amp farther back, eg a
small
cap on one rail, is my guess. Either way requires tracing back from
the input/s to the output device
If this unit normally waits 15 secs or so before coming off Standby,
it would support the conductive glue theory, as this is perhaps when
the Standby voltage gets into an IC pin via the glue, and causes the
output to DC.
A brand new TDA device should not have 300mV of DC anyway, this must
be a clue, there is a leak somewhere.
Gareth.
I was leaning toward a conductive glue problem because there are a
few wires that are fed from the amp board to the main board that are
protected by a flimsy sleeve that is covered with what looks like a
foam material. It was sticky to the touch. I cleaned it off from the
parts of the pc board that it was touching but it made no difference.
Then I replaced the two input caps going to pins 1 and 2 of the TDA.
The originals were .47uF film type. I replaced them with .1uF ceramic
type just to see if it made any difference. It did. Noise and popping
stopped. Voltage swings stopped. And I even tested it for ten whole
minutes. I'll give it a full burn in tomorrow but usually it wouldn't
hold steady even for one minute, especially powered by AC. I think
the AC was more of a headache because it put a higher supply voltage
(15VDC) than my bench supply which I only powered up to 12VDC. Maybe
not. Anyway, I'll replace it with the original valued capacitors if
this problem stays away for good. By the way, the reference numbers
of the caps are C62 and C64.
Thanks for all your suggestions.
I powered it back on this afternoon. It lasted for about 30 seconds before
it acted up again. Back to the drawing board for me. I'm going to see if I
can figure out a way to just power up the TDA chip by bypassing the cpu
power up signal with everything else disconnected and then see what the amp
output does. It would be nice to know if the problem is on the amp board or
coming from somewhere else. Would I need to put some dc on the two signal
input caps to bias this correctly?
http://www.datasheetcatalog.org/data...onics/1512.pdf
--
David Farber
Los Osos, CA