A minimalistic chip amplifier

by Mick Feuerbacher, July 2006. Updated August 2006, August 2006.

This project is a minimalistic LM3886 based power amplifier. It has only the minimum number of parts in the signal path:
Three resistors - nothing else. These three resistors are required for the functionality of the amp so further simplification
in this sense is impossible. Referring to the schematic below, the three resistors are Rg, which sets the input impedance
of the amp, and Rf and Ri, which set the gain. In particular in this design the use of an input coupling cap is avoided. This
greatly improves clarity and transparency, in particular if the preceding stage has output coupling capacitors. These would
otherwise be unnecessarily cascaded with the input capacitors, which is not good for the sound. If you want to build this
amp, be however careful if you are not sure if the preceding stage in your system has output coupling caps. If it has not,
DC may enter the amp, which will be amplified and sent to your speakers! Also, be careful if you have difficult to drive
speakers. The amp has no Zobel network or any other stabilizing components.

Of course finishing the amp requires additional parts, but these are not in the signal path. The LM3886 has a muting
function which requires the addition of a resistor Rm to V-. Also power supply bypassing is required, which here was done
with a 2200 uF electrolytic parallelled with a 100 nF film capacitor.

The amp is built point-to-point, i.e. no PCB or matrix board was used. The geometry was chosen according to the
following criteria. i) I wanted to have it extremely compact and ii) I minimized the area of any current loops, which lead
to unwanted incuded voltages in the circuit.

In the following you find an illustrated description of the build process of my amp. It is rather detailed and step-by-step,
which may be helpful for beginners. You are invited to use it as a guide for building your own amp - if you do so, I would
be grateful for your feedback..

Fig. 1: This is the schematic of the amp. The

chip is a National Semiconductors LM3886.The
following additional parts are used:

Part Value Rating, type

Rg 22k 0.35 W metalfilm
Rf 22k 1 W carbon
Ri 680R 0.35W metalfilm
Rm 10k 0.35 W metalfilm
2200 uF 63 V electrolytic;
Cs
|| 100 nF 100 V MKS

Please note that Rf is overrated. This is just my

personal choice. The minimum power rating for
all resistors is 1/4 W.

The gain is chosen fairly high, it amounts to

about 33, which makes it possible to use this
amp without a pre. For use with a pre, lower
the gain to about 20 by changing Rf or Ri.

Optionally, you can add a resistor in series to

the input. This resistor should have something
like 220R.

Fig. 2: Before you start building, select the

parts you will use. Use a multimeter to match
the parts to be used in the left and right
channel.

In the back you see the Cs electrolytics, in the

middle (from left) the pot (I use an Alps RK09
for testing), the resistors and the bypass caps.
On the picture the optional input resistors are
also shown. In the front are the wires I use for
connection of the amp to the supply terminal.

Fig. 3: Make a careful layout of the position of

the parts and the connectors. I always build the
amp on a test jig first (I use an aluminum book
end), which makes soldering and modifying, if
necessary, much easier and the finished amp
will be moved to the final case as a whole. On
the test jig, the positions of all connectors
should be exactly as they later will be in the
final case.

As you see, I build my amp identically for both

channels, i.e. it is not mirror-symmetric as in
most other designs. This way, wiring can be
optimized and be done equally efficient for both
channels. Note that channel matching is not
only important for the parts used, but also
involves wire lengths, surroundings for the
parts and leads, etc.

Fig. 4: Rf is soldered between pin 3 (out) and 9

(in-). The left connector is not cut as it wil be
used to connect the pin to the positive output
jack.

Fig. 5: Rm is also soldered directly on the pins

(8 to 4). I added a connection from pin 4 (V-)
to pin 11 (nc), so that I could make the V-
connection there and get a more symmetric
layout.

Fig. 6: The signal ground star will be placed on

the RCA connector. Therefore Ri and the
ground connection are directly soldered on pins
9 and 7, respectively, and the leads on the
right are left uncut.

Fig. 7: Here Rg and the optional input resistor

is soldered on pin 10.

Fig. 8: The chip with all connections from the

front...
Fig. 9: ... and attached to the jig. You can see
that I have attached the parts such that they
nicely connect to the corresponding connectors.

Note added: Add a jumper connecting pins 1

and 5. See update.

Fig. 10: Now solder the bypass caps directly on

the pins of the Cs electrolytics.

Fig. 11: The outer contacts of the Cs caps are

connected to a thick copper wire. The copper
wire is isolated where it passes the other
connectors of the caps, and skinned in the
middle, where the power ground will be placed.

Fig. 12: Now connect the caps assembly to pins

1 and 11 of the chip. Make sure that the
polarity is correct: The positive connector of the
left cap and the negative connector of the right
cap are soldered to pin 1 and 11 respectively.

Between the caps you can see the solder tag of

the speaker return, which has to be connected
to the power ground star.

Fig. 13: At this point I decided to use a straigth

solder tag oriented upwards instead of the 90°
bent one shown in the previous figure and in
figure 3. This allows me do connect it directly to
the power ground and to save another piece of
cable length.

Fig. 14: V+ is connected to pin 1.

Fig. 15: The power ground star.

Fig. 16: V- is connected to pin 11. Amp

completed.
Fig. 17: As you see, the complete amp is very
compact. It takes only a few cubic centimeters
space. On the lower left you see the supply
terminal.

Fig. 18: I use a snubberized supply to power

this amp. See a description here.

Fig. 19: The grounding scheme. The mains

earth is connected to the PSU case directly. The
two ground connections (+ and - grounds) are
lead to the amp separately. The cable used for
the PSU-amp connection is a shielded four-wire
cable. The shield is connected to both the amp
and the PSU cases.

The ground connections coming from the PSU

are split (left and right channel) in the amp and
both are connected to the power star ground of
the respective channel. The power stars, as
shown below, are connected to the signal stars
by a thin wire. One power star is connected to
the case.

At each power star the PSU grounds, the

speaker return and the connection to the signal
star are connected. All other grounds are
connected to the signal star.

Note that the separation of signal and power

star ground is another deviation from the
original Gaincard layout.

Fig. 20: The signal ground is on the minus pole

of the RCA connector. The power ground star is
seen in Fig. 15. The two star grounds are
connected by a thin and relatively long copper
wire in order to lift the signal ground by a small
amount.
This grounding scheme is done independently
for both channels.

Fig. 21: The second channel is now also

finished. The green wire you see on the right is
the connection of the power ground star to the
chassis.

Fig. 22: On the left you see the small pot for
testing. Note that I have changed the
connection of the optional input caps with
respect to Figs. 7 and 8: a longer connector
end is attached to the pin, such that I can
solder the input wire before or after the
resistor itself in order to try the configurations
with and without this resistor.

Also note that it is easy to change the amp

from integrated to power amp (i.e. with and
without a pot). Connect the pot (I to the RCA
connector, O to the input resistor lead (see
above) and G to the signal star ground for an
integrated amp. Or connect the input resistor
lead directly to the RCA connector for a power
amp.

Before you connect the amp to your system check the DC offset: Connect the speaker outputs to a load (I use a 4R7 10
resistor, power up the amp, and measure the DC voltage across the resistor. In my amp I found about 30 mV for both
channels, which is fine. Then connect it to some crap speakers and run it for a while. Only if no problems occur, connect
it to your precious speakers.

I should mention that for the first channel I built I used a chip which was damaged. The DC offset was very high (0.3 V)
and the sound was distorted in the bass. So I built the amp again using a new chip and it worked fine.

Update August 2006: Pin 1- pin 5

A user at diyforum.com (SpittinLLama)

convinced me, with some well founded
arguments, that a jumper should be soldered
between pins 1 and pin 5 (both +V), as
otherwise the full current in the chip is lead
over a narrow internal structure.

As I was about to recase my amp anyway, I

followed his advice and made a jumper. There
was no change in the sonics, but maybe I am
more on the safe side concerning the operation
of the amp.

The jumper is seen on the lower right side of

the amp. I isolated the jumper in order to
avoid unwanted contact with pin 2 or 4.

Update August 2006: Recasing

I chose a very simple industrial case as a new

home for my amp. It is an Aluminum diecast
case (afaik Deltron builds these) of about 18 x
8 x 5 cm.

This is the raw case...

... and here it is after drilling the holes and
polishing.

The amp can be removed as a whole from the

test jig.

Here it is shown just lying in the new case...

... and after making all connections.

This is the finished amp. The knob was taken

from an old piece of electronic equipment.

And from the back. Note the non-mirror

symmetric arrangement of the connectors.