00% bgcolor=ccccff> « back to results for ""
Below is a cache of http://ax84.com/media/ax84_m35.pdf. It's a snapshot of the page taken as our search engine crawled the Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
DUAL RECTIFIER Rev B
1
THE AX84 AMP (REV 10)- THEORY OF OPERATION
Let's take a walk through the inner workings of this amp. Along the way, I will be including some math and electronic theory. To fully
understand how a guitar amp works, even a simple one like the AX84, you must know how to read schematic diagrams, and know what
things like resistors and capacitors are. You also need some knowledge of electronic theory, but not much. If you don't know what
resistors and capacitors are, or don't know what Ohm's law is, I suggest a quick trip to your favorite search engine or library and find
out.
HOW THE HECK DO TUBES WORK ANYWAY?
So you run to the bookstore and buy a book on basic electronics. Hmmm, no tube chapter. Even some 'good' tube amp books don't do
a good job of explaining how these work. There are many different tube types out there, but guitar amps use three types of tubes:
triodes, pentodes, and rectifiers.
RECTIFIER TUBES
Let's start with the rectifier, because it is the simplest type of tube. Inside the glass bottle, there are a few metal parts: filament, cathode,
and plate. The filament is sometimes called the heater, because that is exactly what it does - it heats the cathode. Like the filament in a
light bulb, the tube filament is a thin length of wire that gets quite hot when electricity flows through it.
When the filament heats the cathode, electrons 'boil' off the cathode and flow toward the plate.
Electrons flow from negative to positive, so the cathode is negatively charged and the plate has
a positive charge. This flow is a sort of one way valve. Remember that electron flow is the
opposite of 'conventional' current.
(The ancients did not know which way the electrons really flowed. They knew that there were
invisible bits of charge that flowed through wire in a circuit, but had no way of determining the
direction the bits traveled. They took their best guess. Bzzzzt! Wrong! By the time the not-so-
ancients realized the mistake, it was too late. To this day, when we talk of electric current, we
really mean 'the negative flow of electron current'.)
So, current flows from the positive charged plate to the negative charged cathode, just like
current flows through a solid-state diode, from its anode to cathode. But the rectifier tube used
in the AX84 has two plates. Just think of it as two diodes with the cathodes connected. In our
circuit, it is used as a full-wave rectifier.
TRIODE TUBES
The grid in a triode is a mesh of thin wire or wires positioned between the plate and cathode - very close to the cathode. If you were to
connect the grid to the cathode, the tube would behave somewhat like a rectifier. Most of the electrons flow right past the grid on their
way to the plate. Please do not try this. Most triodes are not designed to be operated this way.
When charge on the grid is made negative with respect to the cathode, the electron flow starts to
get 'pinched off'. The electrons are repelled by the negative charge on the grid. Now, make the grid
voltage negative enough, and the electron flow will stop.
This DC voltage applied to the grid is called bias. The voltage that cuts plate current to zero is
called cut-off bias. Increase the bias voltage and more plate current flows, up to a limit. The point
where plate current no longer increases, regardless of increases in bias voltage, is called the
saturation point. Between cut-off and saturation, a triode behaves as a linear device.
How do we make sure the tube operates in this linear range? By setting the grid bias about halfway
between cut-off and saturation.
PLATE 1
PLATE 2
FILAMENT
CATHODE
DUAL RECTIFIER
PLATE
GRID
CATHODE
TRIODE Rev B
2
BIASING V2B
Take a look at Preamp Stage 1 in the AX84 schematic. When
no signal is present at the input, the grid of V2B is connected
to ground through resistors R9 and R12. So the DC voltage at
V2B's grid is 0V. The data sheet of a 12AX7 has a graph
showing plate current as a function of grid voltage for different
plate voltages. Looking at the schematic, we see the plate
voltage on V2B is 159V. Back to the 12AX7 data sheet - at a
plate voltage of 150V, plate current is 0mA when grid voltage is
about -2.2V. Saturation occurs when the grid voltage is 0V. So
we want a bias point halfway between these two points: -1.1V.
Uh oh! Didn't we just determine the grid voltage at V2B is 0V?
How do we make the grid voltage negative? Don't have to! We
raise V2B's cathode voltage to 1.1V. Look at Preamp Stage 1
in the AX84 schematic. The voltage across R4 is:
Using Ohms law, the current through R4 is:
So, 0.74mA flows into the plate of V2B. This current has to go somewhere. It just does not disappear inside the tube. A tiny fraction of
the plate current exits through the grid, but it is such a miniscule amount that we can disregard it. For practical purposes, all of the plate
current flows out through the cathode. We use this to our advantage to set the bias on V2B. V2B's cathode current flows through R13.
Now, using Ohms law again, we can determine V2B's cathode voltage:
Remember that the bias is the grid voltage, referenced to the cathode voltage. Now, we can easily calculate the bias of V2B:
Good. That is the bias voltage we
wanted. So you see that having a
resistor between the cathode and
ground is used to set bias - the
technique is called self-bias or
cathode-bias.
Wasn't it convenient that we knew
what the plate voltage was! If you
were designing your own preamp
stage from scratch, you would have
to calculate the plate voltage.
There are a few ways to do this.
Perhaps the easiest way would be
to model the circuit using PSPICE
(models for popular tubes are
available on Duncan Munro's
website).
How about doing it the old way,
which would mean looking at the
tube data books. An example of an
Ep-Ip chart for the 12AX7 is shown
here.
Once built, it is very easy to
measure actual voltages and current, then fine tune a design.
74V


159V)

-
(233V


R4)

across

(voltage
=
=
0.74mA


OHMS
100K
74V


R4)

through

(current
=
=
1.10V


OHMS)
(1.5K

*

(0.74mA)


V2B)
of

cathode

at

(voltage =
1.10V
-


voltage)

(cathode

-

voltage)

(grid


voltage)

bias

grid

(V2B
=
=
5
4
3
2
1
0 0
50
100
150
200
250
300
350
PLATE VOLTS
PL
ATE CURRENT (MA
)
0
-2.5
-1
-0.5
-2
-4
-3.5
-3
-1.5
GRID VOLT
S
12AX7/ECC83
233V
C4
0.01UF
R21
470K
VR3
1MEG LOG
159V
1.10V
J2
R12
1M
R9
68K
C8
0.1UF
R13
1.5K
V2B
12AX7
R4
100K Rev B
3
AMPLIFYING THE INPUT SIGNAL
We carefully calculated the bias point to keep the tube operating in the linear part of the curve. In plain English, this means the tube
circuit is designed so the input signal is amplified with no distortion. Now, a tube is not a perfect linear device, so there is a slight
amount of distortion. We won't worry about this here, it would be too complicated to add this non-linearity into the calculations below.
The three tube stages in the AX84 amplify the input signal. V2B amplifies the input, V2A amplifies this some more, and V1 boosts this
even more - enough to drive the output transformer/speaker load. The gain of each stage can be determined mathematically. To do
this, we need to understand tube parameters.
In the linear region, the ratio of change in plate current to the change in grid voltage remains constant. Well, not perfectly, but close
enough for rock-and-roll. This ratio is called the transconductance (or mutual conductance) of the tube, and is usually specified in a
tube data sheet. The formula for transconductance is:
Transconductance (Gm) = (change in plate current)/(change in grid voltage)
Transconductance is measure in units called MHO. It is the opposite of resistance, thus the name (OHM spelled backwards). A tube
datasheet usually specifies the transconductance at a couple different plate voltages. For instance, a 12AX7 data sheet I have
indicates:
Plate Voltage
100
250
VOLTS
Transconductance (Gm)
1250
1600
uMHOS
We have to estimate what the transconductance will be. V2B has a plate voltage of about 150V, so Gm should be about 1400 uMHOS.
Another important parameter found in a tube datasheet is plate resistance. Think of a tube as a variable resistor. When the grid voltage
is made more positive, more current flows through the tube. This means the effective resistance between the plate and cathode
decreases. Make the grid voltage more negative, and this resistance increases. The formula for plate resistance is:
Like any other resistance, plate resistance is measured in ohms. My 12AX7 data sheet specifies plate resistance at two different plate
voltages:
Plate Voltage
100
250
VOLTS
Plate Resistance
80K
62.5K
OHMS
We have to estimate the plate resistance