Adcom GFP-555 User Manual

solid state

Back

Adcom GFP-555 Preamp Upgrade

In the best tradition of DIY audio, this author’s modifications significantly

upgrade a proven preamp design.

modified my Adcom GFP-555 in
a manner similar to Gary Galo’s
GFP-565 project in his series of

I

articles in audioXpress
do a “purist” upgrade like Gary—I kept
the tone controls, which now supply
a new set of output jacks, and all the
available inputs but with an upgrade to
gold jacks. I changed the tone control
op amp from the original NJM2041 to
the OPA2604. The LAB and NORM
outputs are now both DC-coupled. I
also removed the headphone amplifier,
which was a mediocre design with an
NJM4556 op amp, so I decided to save
the available watts for the new AD744/
AD810 output stage, which draws a
lot of current. The GFP-555 has even
higher gain (22.5dB) than the 565, so
the volume control wiper is well down
at normal listening levels.

The lower 14dB gain of Gar y’s
line stage was a welcome change. The
AD744/AD811 line stage was a bit
more difficult to install because the -555

1-5

. I did not

By Charles Hansen

does not have the dual-composite op
amps that the -565 has. Victor Cam­pos recalled that the -555 was probably
designed by Nelson Pass before Victor
arrived at Adcom.

While I did not use an external power

FIGURE 1: Block diagram comparing original with modification.

supply, I did upgrade the power supply,
changing to a 36V CT 2A secondary,
with a copper flux band and steel end
plates. I fit in the larger transformer by
removing the line voltage selector switch
and its extra wiring.

AE-adcom-01

FIGURE 2: Power supply schematic.

44

audioXpress

9/06 www.audioXpre ss . co m

AE-adcom-01

I kept two of the existing plastic dual­dual phono jack modules, which make
a desirable mechanical connection be­tween the PC board and the rear panel.
These are the Tuner In/Tape 1 In pair
and the Signal Processor In/Out pair.
The former two will be active, while the
Processor pair will be strictly mechanical
and bypassed with jumpers.

Parts cost for this upgrade was about
$115, based on the parts list in Table 1.
I happened to have the GFP-555, but
I recommend using the –565 if you are
considering this type of upgrade. Follow
Gary’s articles for the purist version, but
for a “less-than-purist” modification that
retains the original I/O flexibility of the
Adcom design, this article will provide
guidelines.

I strongly suggest you obtain the ser­vice manual, for the model of preamp
you are modifying (readily available as
reprints on eBay). I mapped all the PC

board jumpers onto the schematic so I
could replace them with discrete com
ponents where necessary. I must thank
Gary Galo for his comments on my pro­posed modifications; he was very helpful
and provided some interesting insight
into the Adcom preamp designs of that
era.

MODIFICATIONS

Prior to making any modifications, I
verified all the DC voltages shown on
the schematic (it had sat unused for a
few years), checked that all the controls
and input/outputs were functional, gave
the Adcom a listening session, and made
a few measurements.

Figure 1 shows block diagrams for

the original GFP-555 and for my modi­fications. In both cases the CD, Tuner,
Tape 1, Tape 2, and Video/Aux inputs
are processed through the Listening
input selector switch, with the phono

preamp section providing one of the in

-

puts. There was also a separate Record­ing selector switch that coupled to the
Tape 1 and Tape 2 tape loop jacks.

Audio signals proceed from the input
switch to the Signal Processor I/O jacks,
which are furnished with removable
metal links. The next step is the Balance
control, the Mono/Stereo switch, and
the Volume control, which is equipped
with a Contour switch to the tapped vol­ume control that could select the loud­ness contour function for low listening
levels. Finally, the selected audio signal
was sent on to the active line stage.

The output of the line stage can then
take a number of series paths. When the
tone controls and filters are bypassed, the
line stage is direct-coupled to the LAB
output and capacitively-coupled to the
NORM output. The headphone amp
input receives this same signal. Three
additional switches in the line output

-

PHOTO 1: GFP-555 original power supply rear.

PHOTO 2: GFP-555 original power supply front.

PHOTO 3: GFP-555 mod power supply rear.

PHOTO 4: GFP-555 mod power supply front with new
transformer.

audioXpress

September 2006 45

signal path allow selection of the tone
controls—a Lo filter and/or a Hi filter.

The original phono preamp and line
stage ICs had a house number, “5E,”
which Victor thought was an OP37.
This would make sense, because the
OP37 is one of the few op amps whose
maximum supply rail voltage was ±22V,
and the Adcom schematic shows the
5E operating on ±21.5V rails. On the
other hand, Gary observed that with the
phono gain switch set to the lower gain
position, the circuit does not satisfy the
gain-of-5 stability requirement of the
OP37, so maybe the “5E” was an OP27.

The modified block diagram at the
bottom of Fig. 1 is essentially the same
up to the output of the line stage. At this
point the amplified signal takes two dis­tinct paths. The first is direct-coupled to
both the LAB and NORM output jacks.
The second path uses all the switched
tone control and filter circuitry of the
original unit, but goes to another set of
output jacks that I added called Tone
Out.

You can also see that I changed all
the ICs to reflect improvements made
over the intervening years. The phono
preamp went from the 5E to an AD845.
I replaced the line stage 5E with the
aforementioned AD744/AD811 com­posite amplifier. The NJM2041 tone
control IC became an OPA2604. Finally,
I removed the 4556 headphone ampli­fier and all its passive parts, although I
left the jack in place on the front panel.
I tried unsuccessfully to fit my Head­room headphone module board6 in the

space between the transformer and the
front panel.

POWER SUPPLY

Gary used a separate power supply chas­sis for his GFP-565 modification, which
has the distinct advantage of lower
phono stage hum and noise. However, I
wanted to retain a one-box preamp. The
original power supply regulators were
composed of discrete transistors Q901
and Q902 that used ±32.5V raw DC
rails to derive the ±21.5V DC op amp
rails. Another set of transistor regulators,
Q905 and Q906, further stepped this
down to ±15.6V DC for the headphone
op amp, and provided 31.2V DC for the
output delay relay circuit.

Photos 1 and 2 show the front and

rear views of the original supply. In the
foreground of Photo 1 I have already

FIGURE 3: Phono preamp schematic.

removed the components for the head­phone amplifier.

My modified power supply is shown
in Fig. 2. After mapping all the jump­ers onto the schematic, I replaced all the
power supply jumpers with larger wire
gauges. Next, I removed all the power
supply components from the PC board
and I replaced the Jamicon bulk filter
caps with Nichicon KZ and Panason­ic HFQ types. I used LT1085CT and
LT1033CT linear regulator ICs for the
new ±16.2V DC op amp power supply
rails.

Because the delay relay has a 24V
coil, there was no need for 31.2V DC.
The original design used a 274Ω series
resistor to drop the coil voltage to 24V.
I removed the resistor and dedicated
Q905 and Q906 to a reduced ±10V DC,
which saved some power dissipation in

AE-adcom-03

FIGURE 4: Line stage schematic.

46

audioXpress

9/06 www. au di oX pr es s .c om

AE-adcom-04