[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Help With Mockingboard Amplification

On 9/4/2010 2:28 AM, Polymorph wrote:

To: Michael J. Mahon
Michael J. Mahon wrote:


The original Mockingboard was also pretty noisy, so don't think that
will cure the problem!

It's quite possible that your card has always had a fair amount of
background digital noise, and that the 386's are just amplifying it.
(Did you have the treble control turned down on your powered speakers?)

To get reasonably quiet audio out of an Apple II will require good
filtering on all supplies--including the +5v which powers the source
of the audio signal.

Extra bypassing may help, but the ultimate solution is to double
regulate down from the +12v to get quiet +5v for the 8910's.

You may also want to increase the value of the capacitors shunting
the input to the 386's. That will roll off the high frequencies,
attenuating the most annoying noise and making the sound richer
and less shrill (a big issue with squarewave sound generation).


Thanks for the help.


You're welcome--I hope you get it working to your satisfaction!




OK, so I bit the bullet today and un-did all of the amplification
circuit and put things back to how they were before (as described on my
web site).

I found the noise from the LM386's unacceptable. I tested the LM386's in
isolation (with no input signal) with only power and ground wired up as
per the Mockingboard V1 (i.e. with the 12V line attached to a 220uF
filter capacitor) and the noise was the same as noted previously.

I notice that I have LM386N-1 chips as opposed to the Mockingboard V1's
LM386N-3 chips, but I don't think this would make any difference from
reading the specs (slightly different voltage tolerance and output
IIRC). So I have no idea how they've got the Mockingboard V1 to not
produce the noise I was getting. The only thing I can think of is that I
noticed that the power and ground rails on the Mockingboard V1 are
thicker. Could using thicker wire for power and ground have reduced the
noise I was getting? I'm not going to try it, but would be interested to
hear thoughts on this.


Thicker isn't so much the issue, but a ground plane might help.

If noise is entering through the power bus, then neither wire nor
a ground plane will make any difference.

If the noise is originating in the 386's, then *nothing* will help,
except fixing the 386's.  ;-)  It is possible that the 386's are
oscillating at a very high (inaudible) frequency, and that is the
source of the noise.  Decoupling close to the 386 and care to reduce
coupling between the input and output are important.  It is also
useful to use a high-frequency "snubbing" ciruit across the output
(which is otherwise dominated by the LC circuit of the speaker coil).

Many 386 circuits show a 10 ohm resistor and a 0.05uF capacitor,
connected in series, shunting the output to ground.  I don't recall
if your circuit has this or not, but it's worth a try.

It's worth keeping in mind that the 386 has a high gain, so good
analog circuit layout techniques are in order to prevent oscillation.

For diagnostic purposes, try building up a 386 amplifier on a separate
piece of perfboard, powered by a known quiet source--like a battery.
Then see how it sounds.  If the noise is still present, then you have
a bad 386 (or very high standards ;-).

I would expect this 386 amp to be sufficiently quiet, and then you
can hook it up to your board and see if the noise is on the input.
If it isn't, then cleaning up the Apple +12v is the only thing
that needs to be done.  If the noise *is* on the input, then the
problem isn't the 386 stage at all.


Without the LM386's in place, I can crank the volume/treble/base up to
maximum on my powered speakers and I can only then just faintly hear
noise when no music is playing.


That's good.  That bodes well for getting a 386 stage to work, too.


Based on my findings, anyone looking to make a Mockingboard, my
recommendation is to forget the LM386's and look to another chip, or do
what I did and just forget amplification altogether and run it with
amplified speakers.


From a practical point of view, running with amplified speakers makes
a lot of sense--they're about the only game in town these days.  ;-)

The 386's provide for headphone use, though, and that may be enough
reason to include them.

-michael

NadaNet 3.1 for Apple II parallel computing!
Home page:  http://home.comcast.net/~mjmahon/

"The wastebasket is our most important design
tool--and it's seriously underused."