DR #55 - Restoring the Bass on an Adcom GFA 5500 Power Amplifier

Duality Repair
132 followers
Mar 12, 2025
This Adcom GFA 5500 is suffering from a common problem: weak bass. This video documents how I restore the bass, as well as the theory behind why it was lost.
this is the Adcom GFA 5500 power
amplifier and this unit is suffering
from what seems to be a relatively
common issue with these models and
that's weak Bas indeed I connected this
to my test speakers and it does sound
really Hollow especially at the lower
end so let's see what it takes to fix
it the first thing to do is to verify
that we have an issue now I've already
done this myself with the test speakers
but you're definitely not going to be
able to hear a weak base issue through
my microphone so I thought I'd show you
there are at least a few ways to do this
first I'm injecting a 1 khz square wave
into the amplifier this is the input
signal in a healthy amplifier the output
will look just like the input maybe with
a little bit of distortion but here's
what ours looks
like much much different than the input
this sloping downward on the top half
and sloping upward on the bottom half
are classic signs of poor low frequency
response we can also use sine waves so
now my signal generator is sweeping from
20 HZ to 20 khz or the entire audio
range theoretically the amplitude of
these waves should remain constant
throughout the entire sweep but
obviously they're not at the low end
we're sitting at about one or two
divisions on the scope and at the high
end we're at about six divisions so
there's some very obvious attenuation or
signal reduction happening here at the
low
end that's enough Jib Jab for now let's
see what it looks like inside
all right let's take a look around
before we get started in the front we
have this giant toal Transformer labeled
Adcom on the bottom in the back here we
have four very large filter capacitors
these will smooth out the high voltage
rails on the top we have eight smaller
filter capacitors to smooth out the
lower voltage rails along with some
additional components on the board here
and then on either side is the amplifier
board so this is the left channels board
this is the right now I'm fairly
confident I know exactly what's causing
this issue and so I'm not going to
actually do any live troubleshooting
right now and I'm going to focus on just
one channel since both channels are
exhibiting the same problem so for now
I'm going to remove this entire
amplifier module with the heat sink and
we can take a closer look
[Music]
okay here's the right channels amplifier
board and it's in really good shape now
I've studied the schematic quite a bit
and I'm fairly confident that there's
just one component responsible for this
issue and that's this component here
this is C3 it's a 47 microfarad at 50vt
electrolytic
capacitor now I'm going to remove the
board from the Heat syn and we'll pull
and test this capacitor
here is our suspect capacitor
C3 and here is a brand new highquality
audio grade capacitor of the same exact
values 47 microfarad at 50 volts you can
see we're measuring
45.8
microfarads so no problems there for
reference the Tolerance on these types
of capacitors are usually plus or minus
20% so on the low end that's going to be
in the high
30s this is obviously well within spec
as expected let's measure C3
our C3 is measuring just3 narat so not
even close to the spec thousands of
times less capacitance than the brand
new one this is definitely a problem
let's install the new capacitor test it
and if it works we'll explain what's
going on
here all right we're all back back
together again or at least good enough
for testing let's run our Square wave
test again see if we've made any
Improvement after the capacitor change I
have the same 1 khz square wave at the
input that's this signal here let's take
a look at the
output it's much better it's very very
close to the
input so let's run the sine wave sweep
test
now once again we're sweeping from 20 HZ
to 20 khz the entire audio range only
this time the amplitude at the low end
is exactly matching the amplitude at the
high end we have our base
back so great we've resolved the base
issue but why was C3 causing that issue
this is the schematic for the amplifier
board and the schematic is the same for
both channels here's our failed C3 up
here and this is part of the input
section it's actually part of the
feedback Network which I've highlighted
in Orange so we have our output signal
being fed back through this 49.9k Ohm
resistor it's then going through our C3
here which is in parallel to another
capacitor C4 goes through a 10K Ohm
resistor before going to ground now it's
a little bit busy over here so I've
redrawn this so it's easier to
see here's our feedback coming from the
right I rounded the 49.9 kohm resistor
to
50K here are the two capacitors in
parallel C3 and the other capacitor is
just 0.22 microfarad and then we have
the 10K Ohm resistor to ground now
capacitors block DC of course but they
do pass AC so for the signals we're
concerned with audio signals we can just
think of these two capacitors as a
resistor in series with this
resistor now I've replaced the two
capacitors in parallel with a resistor
and under normal circumstances when C3
is equal to 47 microf far this resistor
is equal to 170 ohms at the low end at
20 Hertz and less than 1 ohm at the high
end at 20 khz now 170 ohms might sound
like a lot but really you have to
compare that to this resistor 170 ohms
to 10K is almost nothing so this is
virtually a dead short at 20 HZ and it's
certainly a dead short at 20 khz so
really this resistor goes away when the
capacitor is at 47 microfarads of
capacitance and we have just a simple
voltage divider we go through 50k ohm
and then 10K ohm to ground that's under
normal circumstances so let's see what
happens when C3 is much much lower than
47 microfarads when we measured C3 it
was nowhere near the 47 microfarads we
expected it was unbelievably low at just
0.3 nanofarad at this capacitance at the
high end or 20 khz this resistor is
still just 36 ohms so we can still
consider this a short at the high
frequencies at the low end however at 20
htz this resistor turned into 36,000
ohms so we can no longer ignore this
resistor if we add it in series to this
resistor we get one resistor with a
value of 46,000 ohms what that means
practically is the feedback signal
that's sent at the low end to the input
is about three times as large as what's
fed back at the high end and so the more
feedback you provide at the input the
lower your signal at the output so
hopefully that should make sense as to
why you get a poor base response because
we have such a low value for
C3 it's outside the scope of this video
but I'll let you know what else I plan
on doing with this unit certainly I'm
going to clean all of the dust out I
only replaced one of the three
electrolytics on the right amplifier
board I'm going to replace the other two
as well there is a overtemp switch
mounted to the heat sink right here when
I had this apart I noticed that the
thermal compound had started to dry up
so I will clean and replace that and
then I'll do the exact same thing to the
left amplifier
board I will measure all four bulk
electrolytic capacitors here for ES R
and capacitance I'd be very surprised if
there were problems with these they are
very high quality and they usually last
quite a long time but if there are any
problems or signs of problems we'll look
at replacing those I'll also measure the
ESR and capacitance of these eight
smaller filter capacitors on top I would
not be as surprised if there were
problems with these when it's all back
together I will set up the bias and DC
offset for both channels as per the
service manual after that when
everything's done I think this unit will
run really really well and for a very
long time so that's it for this video
thanks for watching
