Module 6 2 PA Classes

Berkeley
The ABC’s of PA’s
Prof. Ali M. Niknejad
U.C. Berkeley
Copyright
c 2014 by Ali M. Niknejad
Niknejad Advanced IC’s for Comm

Class A Review

Class A Amplifiers
Class A amplifiers have a collector current waveform with

100% duty cycle. In other words, the bias current IQ > io ,
where io is the current swing.
The advantage of Class A amplifiers is high linearity, clean
sinusoidal waveforms (little harmonic filtering needed), and
simple design (not relying too much on transistor
non-linearity).
The disadvantage of Class A is low efficiency and maximum
power consumption at zero output power!
Dynamic Class A (modulate bias current) is an attractive
solution for some applications. Envelope following Class A
combined with dynamic Class A is even better.

CE/CS Class A
VCC
RL
vo
vin
The CE amplifier has the advantage of higher power gain

(there is voltage gain and current gain).
The collector effciency is given by
1
PL vo io 1
ηc = = 2 = v ×i
Pdc VCC IQ 2
The efficiency is maximized if we can set the voltage swing
and current swing independently.
CE/CS Voltage Swing
VCC − VCE!sat VCC
VQ,high
VQ,mid VQ,mid
VQ,low
VCE!sat VCE,sat
We see that to avoid clipping, we should bias the transistor at

the midpoint between VCC and VCE ,sat (or Vdsat ).
Thus
VCC
vo ≤
2

CE/CS Current Swing
IQ
0A
The current in the transistor cannot go negative. Therefore,
the maximum current is set by the bias current
io ≤ IQ
The efficiency is therefore limited to 25%

1 1
η =v ×i = ×
2 2
Optimum Load
It’s important to note that to achieve these optimum

efficiencies, the value of the load resistance is constrained by
the current and voltage swing

vo v VCC
Ropt = =
io i IQ
Since the load resistance is usually fixed (e.g. antenna

impedance), a matching network is required to present the
optimum load to the amplifier.

Inductor Loaded CE/CS
VCC
RF C
C∞
vo
vin RL
If we AC couple a load RL to the amplifier, then the Q point

of the amplifier collector voltage is set at VCC through the
choke inductor.
The maximum swing is now nearly twice as large. Notice that
the collector voltage can swing above the supply rail.

Swinging both ways...
VCC
−
VCC
VCC
+
−
+
vo
+
vin VCC RL
−
Recall that the voltage polarity across the inductor is given by

dI /dt, which can go negative. Thus the collector voltage is
equal to the supply minus or plus the absolute voltage across
the inductor.

Improved Class A Efficiency
Since the swing is almost double, the efficiency now

approaches 50%
vo ≤ VCC
1 io vo 1
η= ≤
2 VCC IQ 2
In practice, due to losses in the components and back-off from
maximum swing to minimize distortion, the actual efciency
can be much lower.
Package parasitics (see later slides) also limit the voltage
swing.

Complete Class A Output Stage
VCC
Lres
Cm
vo
vin Lm R0
Cpar
In practice the collector inductor can double as a resonant

element to tune out the collector parasitics.
The coupling capacitor Cc can be replaced by a matching
capacitor Cm .

Class A Power Capacity
A good metric when comparing PA’s is to measure the output

power versus the peak voltage and current stress imposed on
the device
These peak stresses are set by breakdown considerations. For
Class A operation, we can apply
Pout
PC =
Vmax Imax
Vdrain,max = 2VDD
Idrain,max = IQ + IL = 2IQ
1
Pout = VDD IQ
2
1 1 1 1
PC = · · = = 0.125
2 2 2 8

Emitter Degeneration
VCC
RF C
Emitter inductance has a C∞
detrimental effect on PA vo
efficiency since it reduces the
swing. The voltage across vin RL
the emitter is given by +
VE = jωLE io VLE
−
For a current swing of 1A at 1 GHz, a typical parasitic

inductance of 1 nH will “eat” up 6.28V of swing! We need to
reduce LE or to use a much higher VCC .
In practice both approaches are taken. We choose a
technology with the highest breakdown voltage and the
package with the lowest LE .
Class B PA’s

Class B
+
vin vbias VCC
−
The above circuit utilizes two transistors. Each device only

delivers a half sinusoid pulse and the full sinusoid is recovered
by phase inversion through the transformer.
The base and collector bias voltages come from the
transformer center tap. The base (or gate) is biased at the
edge of conduction (threshold).

Class B Waveforms
ic (t) vc (t)
VCC
p(t)
0V
Since the voltage at the load is ideally a perfect sinusoid, the
voltage on the collectors is likewise sinusoidal.
The power dissipated by each transistor is thus the product of
a sine and a half sine as shown above.

Class B Efficiency
The average current drawn by each transistor is given by

Z T Z T /2 Z π
1 Ip Ip Ip
IQ = ic (t)dt = sin ωtdt = sin θdθ =
T 0 T 0 2π 0 π
Where Ip is the peak voltage drawn from the supply.

The peak current drawn from the supply is just the load
current swing reflected to the collector, Ip = io × n.

1 Ip vo
η=
2 2IQ VCC
Note that the total DC current draw is twice IQ since both

devices draw current from the supply.

Class B Efficiency (cont)
Since the collector voltage swing can be as large as VCC

(similar to an inductively loaded Class A), the efficiency is
bounded by
1 Ip 1 Ip
η≤ =
2 2IQ 4 IQ
π
η ≤ ≈ 78%
4
This is a big improvement over the peak efficiency of Class A.
Note that the average current naturally scales with output
power, and so efficiency drops more gracefully as we back-off
from peak power.

Efficiency versus Back-Off
η(V )
ssB 50%
cla
s A
clas
VCC
The efficiency drops linearly as we back-off from the peak

output voltage
where vc is the collector voltage swing, which is just n times
smaller than the load voltage, vc = vo /n.

Tuned Class B
VCC
RF C
C∞
vo
vin RL
vbias
A tuned Class B amplifier works with a single devices by

sending half sinusoid current pulses to the load. The device is
biased at the edge of conduction.
The load voltage is sinusoidal because a high Q RLC tank
shunts harmonics to ground.

Class B Tank
In a single transistor version, the “minus” pulse is in fact

delivered by the RLC tank. The Q factor of the tank needs to
be large enough to do this. This is analogous to pushing
someone on a swing. You only need to push in one direction,
and the reactive energy stored will swing the person back in
the reverse direction.
The average current drawn from the supply is the same as
before, IQ = Ip /π. The harmonic current delivered to the load
is given by Fourier analysis of the half pulse
Z π Z π
2 1 1 − cos 2θ
Iω1 = Ip sin θ sin θdθ = Ip dθ
2π 0 π 0 2
1π Ip
= Ip =
π2 2

Class B Waveforms
ic (t) vc (t)
VCC
0V
We see that the transistor is cut-off when the collector voltage
swings above VCC . Thus, the power dissipated during this
first half cycle is zero.
During the second cycle, the peak current occurs when the
collector voltage reaches zero.

Class B Efficiency (again)
The efficiency is therefore the same

1 Iω1 vc 1π π
η= ≤ =
2 IQ VCC 22 4
The DC power drawn from the supply is proportional to the

output voltage
VCC Ip
Pdc = IQ VCC =
π
vc nvo
Ip = =
Ropt Ropt
The power loss in the transistor is given by
Z π
1
pt (t) = Ip sin θ(VCC − vc sin θ)dθ
2π 0

Transistor Power Loss
Integrating the above expression

π
VCC Ip vc
pt (t) = (− cos θ − 2π
2π 0 Ip

1 vc Ip
= 2VCC Ip − π
2π 2
Ip vc Ip
= VCC −
π 4
vc Iω1
= IQ · VCC −
2
= Pdc − PL

Class B Power Capacity
Recall the definition for PC

Pout
PC =
Vmax Imax
Vmax = 2VCC
Imax = Iout
Note that the output power is related to the fundamental
coefficient of a half sine wave, or 1/2 the peak
1 Iout
Pout = VCC
2 2
1 Iout
2 VCC 2
PC = = 0.125
2VCC Iout
Same as Class A
Dynamic PA
Supply DC-DC converter
Envelope
Cm
vo
vin Lm R0
Dynamic Bias
Envelope tracking supply and dynamic class-A

Efficiency always close to peak efficiency of amplifier (say
30%) regardless of PAR
Need a very fast DC-DC converter

Class C PA’s

Conduction Angle
Often amplifiers are characterized by their conduction angle,

or the amount of time the collector current flows during a
cycle.
Class A amplifiers have 360◦ conduction angle, since the DC
current is always flowing through the device.
Class B amplifiers, though, have 180◦ conduction angle, since
they conduct half sinusoidal pulses.
In practice most Class B amplifiers are implemented as Class
AB amplifiers, as a trickle current is allowed to flow through
the main device to avoid cutting off the device during the
amplifier operation.

Reducing the Conduction Angle
ic (t) vc (t) ic (t)

vc (t)
The most optimal waveform is shown above, where a current

pulse is delivered to the load during the collector voltage
minimum (ideally zero)
As the pulse is made sharper and sharper, the efficiency
improves. To deliver the same power, though, the pulse must
be taller and taller as it’s made more narrow. In fact, in the
limit the current spike approaches a delta function.

Class C
VCC
RF C
C∞
vo
vin RL
Vbias < Vthresh
Class C amplifiers are a wide family of amplifiers with

conduction angle less than 180◦ . One way to achieve this is to
bias a transistor below threshold and allow the input voltage
to turn on the device for a small fraction of the cycle.

Class C Linearity
Q
010
011 001
100 000 I 2 + Q 2 = A2
I
101 111
110
The Class C amplifier is very non-linear, and it is only

appropriate for applications where the modulation is constant
envelope. For instance, FM uses a constant amplitude carrier
and only modulates the frequency to convey information.
Likewise, any digital modulation scheme with a constellation
on a circle is constant envelope

Polar Modulation
A(t) C∞ B(t) cos (ωt + φ(t))
A0 cos (ωt + φ(t)) RL
Vbias < Vthresh
While the amplifier is a non-linear function of the input

amplitude, the Class C amplifier can be made to act fairly
linearly to the collector voltage.
By driving the amplifier into “saturation” in each cycle, e.g.
with a large enough swing to rail the supplies, then the output
power is related to the voltage supply. Collector modulation
then uses the power supply to introduce amplitude
modulation into the carrier.

Class C Approximate Analysis
IDD
2y IDQ
Assume current pulses are sine wave tips, conduction angle is

2y
−1 IDQ
IDD cos y = IDQ y = cos
IDD
−IDQ + IDD sin θ IDD cos θ ≥ IDQ
iD (θ) =
0 otherwise
Class C Average Current
To make the integral easy to calculate, change variables to

the range of y to 0.
IDC from supply:
Z 2π
1 y
Z
1
IDC = iD (θ)dθ = (IDD cos θ0 − IDQ )dθ0
2π 0 π 0
1
IDC = (IDD sin y − IDQ y )
π
IDD
IDC = (sin y − y cos y )
π

Class C Analysis (cont)
Output voltage is: VOM = Fund{−iD (θ)RL }
Since iD is an odd function, the Fourier series will only yield
sine terms
−2RL 2π 2RL y
Z Z
VOM = iD (θ) sin θdθ = (IDD cos θ−IDQ ) cos θdθ
2π 0 π 0
2RL IDD y IDD sin 2y
VOM = ( + − IDQ sin y )
π 2 4
IDD RL
VOM = (2y − sin 2y )
2π
Note: IDQ sin y = IDD cos y sin y = IDD
2 sin 2y
DC power from prev page:
PDC = IDC VCC = IDDπVCC (sin y cos y )
Assuming that max swing ∼ VCC :
IDD RL
VOM ≈ VCC = (2y − sin 2y )
2π
Class C Efficiency
2
2VCC sin y − y cos y
PDC ≈
RL 2y − sin 2y
2
VCC
Power to load (assuming VCC swing): PL = 2RL
Ideal Efficiency:
PL 1 2y − sin 2y
η= =
PDC 4 sin y − y cos y
lim η(y ) = 100% (ideal class C)
y →0
η(π) = 50% (tuned-circuit Class A)
η(π/2) = 78.5% (single-ended Class B)
For small conduction angle current pulses approach
delta-function.
Many practical issues make Class C difficult to design.
Power Capacity of Class C
0.4
Vdrain,max = 2VCC
0.3
Power Capability
Class B Class A ID,max = IDD − IDQ

0.2
= IDD (1 − cos y )
0.1
Po
PC =
2VCC IDD (1 − cos y )
0.0
0 50 100 150 200 250 300 350
1 2y − sin 2y
Conduction Angle =
4π 1 − cos y
The power capacity of Class C is low at low conduction
angles. Peak is Class AB.

Class D and F

Class F Circuit
3ω0
ω0 RL
The above circuit isolates the fundamental resonant load from

the drain at the third harmonic, allowing the drain voltage to
contain enough third harmonic to create a more square like
waveform.
It can be shown that just adding a third harmonic boost the
efficiency from 78% (Class B) to 88%.

Class F
ic (t) vc (t)
Since it’s difficult to create extremely narrow pulses at high

frequency, we can take a different approach and attempt to
square up the drain voltage.

Quarter Wave Class F
RF C
ℓ = λ0 /4
Z0
ω0 RL
In this circuit a quarter wave transformer converts the low

impedance at the load (due to the capacitance) at harmonics
of the fundamental to a high impedance for all odd harmonics.
Even harmonics are unaltered as they see a λ/2 line.
In theory then we can create a perfect square wave at the
drain of the transistor and thus achieve 100% efficiency.

Ideal Class D Switching Amplifier
VDD
S2
RL
S1
If give up linearity, then we can create some intrinsically

efficient amplifiers using switches. An ideal switch does not
dissipate any power since either the voltage or current is zero.
By varying the switching rate, we can impart phase/frequency
modulation onto the load.
A real switch has on-resistance and parasitic off capacitance
and conductance.
MOS Class D Inverter
VDD
RL
Switching amplifiers are realized with transistors operating as

switches. MOS transistors make particularly good switches.
The input voltage is large enough to quickly move the
operating point from cut-off to triode region.

Class D Waveforms
vd (t) id (t)
VDD
0V
The MOS drain voltages switches from VSS to VDD at the

rate of the input signal.
A series LCR filter only allows the first harmonic of voltage to
flow into the load. Since current only flows through a device
when it’s fully on (ideally VDS = 0), little power dissipation
occurs in the devices.
Class D Efficiency
We can see that the efficiency has to be 100% for an ideal

switch. That’s because there is no where for the DC power to
flow except to the load.
The collector voltage can be decomposed into a Fourier series
VDD
vd = (1 + s(ωt))
2

4 1 1
s(θ) = sign(sin(θ)) = sin θ + sin 3θ + sin 5θ + · · ·
π 3 5
The load current is therefore
4 VDD 2VDD
iL = sin θ = sin θ
π 2R πRL

Class D Efficiency (cont)
The load power is thus
iL2 RL 2 V2 V2
PL = = 2 DD ≈ 0.2 DD
2 π RL RL
The drain current are half-sinusoid pulses. The average
current drawn from the supply is the average PMOS current
Ip 2 VDD
IQ = = 2
π π RL
2
2VDD
PDC = IQ · VDD = = PL
πRL
As we expected, the ideal efficiency is η = 100%

Class D Reality
As previously noted, a real Class D amplifier efficiency is

lowered due to the switch on-resistance. We can make our
switches bigger to minimize the resistance, but this in turn
increases the parasitic capacitance
There are two forms of loss associated with the parasitic
capacitance, the capacitor charging losses CV 2 f and the
parasitic substrate losses.
The power required to drive the switches also increases
proportional to Cgs since we have to burn CV 2 f power to
drive the inverter. A resonant drive can lower the drive power
by the Q of the resonator.
In practice a careful balance dictates the maximum switch
size.

Differential Class D
f0
L1 C1
RL
VDC
+ +
vin −vin
− −
Differential circuit have a huge benefit in ICs. First they have

common mode rejection and also inject (ideally) less common
mode into the supplies. Makes system integration easier (e.g.
VCO pulling). Also differential allows higher swings on load
for a fixed supply, and less impact from bond-wire inductors.
Downside: Need a balun.
Differential Class D Analysis
Assume a high Q tank so that the output current is sinusoidal

(phase φ will be computed)
iout = Io sin(θ + φ)
The voltage at the secondary of the transformer is given by
vo (θ) = nVDC s(t) − n2 Io Ron sin(θ + φ)
The fundamental component of the secondary load voltage is

given by
4
vL (θ) = RL iout = nVDC sin(θ) − n2 Io Ron sin(θ + φ)
π

Class D Analysis (cont)
From the above relation, it’s clear that the voltage and
current are in phase (φ = 0)
4nVDC 1
Io =
πRL 1 + n2 RRon
L
Each transistor conducts a half sinusoid pulse (similar to

differential Class B), so it’s DC value is
nIo
id1,DC = id2,DC =
π
The DC power consumed is therefore
8n2 VDC 1
PDC = VDC (id1,DC + id2,DC ) = 2
π RL 1 + n2 RRon
L

Class D Efficiency
From the AC and DC power, we are now in a position to

calculate the efficiency
1 8n2 VDC
2
1
Pout = I02 RL =
2 πRL 1 + n2 RRon
L
Pout 1
η= =
PDC 1 + n2 RRonL
Note that the switch on resistance is a key factor. We cannot

arbitrarily increase it due to CVf 2 losses.

Estimating Switching Losses
We can estimate the switching losses by noting that the

device “on”-resistance is related to the device gds,lin = gm,sat
1 1
Ron = ≈
gds gm,sat
gm 1
ωT = =
Cgs + Cgd Ron (Cgs + Cgd )
The total input power involves switching approximately
2Cgs + 2γCgs , where γ is a technology parameter relating the
drain-bulk capacitance to the gate-source cap. Take γ ∼ 1 for
an upper bound
2 2 2 1
Ploss = Ctot VDC f = 4Cgs VDC f = 4VDC f
ωT Ron

Class D Power Handling Capability
The power handing capability of Class D is given by
vdrain,max = 2VDD
idrain,max = nIo
Pout η
PC = = ≤ 0.159
2 · vdrain,max · idrain,max 2π
This compares favorably with Class A amplifiers.

Class D−1
VDC
vs1
IDC IDC
is1
IDC
f0 L1
vs1 vs2
is1 i1 is2
C1
RL
The “Dual” Class D amplifier (interchange voltage/current ⇒

square wave current, sinusoidal voltage, parallel LCR filter)
Chokes act like current sources. ZVS by “design” but only if
there is no device capacitance to begin with.

0 0<θ<π
vs1 = 4
− π IDC RL sin(θ) π < θ < 2π

Class E

Class E Amplifier
Switching PA like Class D but can absorb transistor parasitic

capacitance into load network. This allows the switch to
become very large (low on-resistance).
The load network is design to ensure zero switching condition.
Figures: “Class-E RF Power Amplifiers,” by Nathan O. Sokal,
Design Automation, Inc
Switching RF Power Amplifier
Switching amplifier = LC network and a switch

Low Q
Source: Patrick Reynaert and Michiel Steyaert

Switching amplifier = LC network and switch

High Q


perfect Q



Output power of Class E is low

2
Vmax
Pout,E < 0.045
RLOAD
Compare to Class A
2
Vmax
Pout,A < 0.125
RLOAD
0.35um technology, Vmax = 8V ⇒ Pout < 60mW at 50Ω
Class E Waveforms
CMOS Power Amplifiers for

Wireless Communication,
King-Chun Tsai, Ph.D.
VDDπ
Dissertation, UC Berkeley,
Lωo
2007.
Figure 6-3 Circuit configuration and characteristic

waveforms of a Class-E PA
Class E Issues
Problems with Class E include low power gain and large

voltage swings. The drain waveform can swing to more than
3.5 times the supply voltage. This means the transistors have
to be able to handle much larger voltages.
Efficiency for this class of PA is extremely good. Power
capacity PC < 0.107, much better than Class C low
conduction angle operation.

will discuss the injection locking mechanism in greater details later in this chapter.
First GSM CMOS Class E
Use standard CMOS to achieve output power of 1W

Figure 7-1 Schematic of the 1-W 1.9GHZ CMOS Differential
Class-E power amplifier
Measured efficiency > 50%
Use cross-coupled devices to boost driver capability (requires
injection locking)
149
CMOS Power Amplifiers for Wireless Communication, King-Chun

Tsai, Ph.D. Dissertation, UC Berkeley, 2007.
Output Matching
Use chip-on-board testing (avoid lead inductance)

Use a microstrip balun to convert from differential to
single-ended to drive antenna
Source: “A 1.9-GHz, 1-W CMOS Class-E Power Amplifier for

Wireless Communications,” King-Chun Tsai and Paul R. Gray

RF CMOS Class-E PA
Implemented in a 0.18 um CMOS technology


Chip micrograph

Other Classes

Class S
If the switching rate is much higher than the fundamental,

then amplitude modulation can be converted to pulse-width
modulation.
A low-pass filter at the output faithfully recreates the
envelope of the signal.
Class S Amplifier
VDD
Amplitude
to PWM
RL
Class S amplifiers are commonly used at low frequencies

(audio) since the transistors can be switched at a much higher
frequency than the fundamental.
This allows efficiencies approaching 100% with good linearity.

Pulse Density Modulation
RF Pulse Train
Mixer PA Filter
Pulse-density
modulation process
Inherent linearity
Improved efficiency in power backoff, but ...

Pulse Density Modulation Process
AM process ⇒ Extra harmonics

Tradeoff between oversampling ratio and Q
Out of band spectrum
Efficiency
Noise shaping: digital ∆ − Σ
Conclusions
No major efficiency advantage with Q< 5-10
Linearity may be the compelling factor
(almost) pure digital implementation!
Need to run PDM process *as fast as possible*

Module 6 2 PA Classes

Uploaded by

Copyright:

Available Formats

Module 6 2 PA Classes

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Module 6 2 PA Classes

Uploaded by

Copyright:

Available Formats

Berkeley

The ABC’s of PA’s

Prof. Ali M. Niknejad

Niknejad Advanced IC’s for Comm

Niknejad Advanced IC’s for Comm

Class A amplifiers have a collector current waveform with

Niknejad Advanced IC’s for Comm

The CE amplifier has the advantage of higher power gain

VCC − VCE!sat VCC

We see that to avoid clipping, we should bias the transistor at

Niknejad Advanced IC’s for Comm

The efficiency is therefore limited to 25%

It’s important to note that to achieve these optimum

Since the load resistance is usually fixed (e.g. antenna

Niknejad Advanced IC’s for Comm

If we AC couple a load RL to the amplifier, then the Q point

Niknejad Advanced IC’s for Comm

Recall that the voltage polarity across the inductor is given by

Niknejad Advanced IC’s for Comm

Since the swing is almost double, the efficiency now

Niknejad Advanced IC’s for Comm

In practice the collector inductor can double as a resonant

Niknejad Advanced IC’s for Comm

A good metric when comparing PA’s is to measure the output

Niknejad Advanced IC’s for Comm

For a current swing of 1A at 1 GHz, a typical parasitic

Niknejad Advanced IC’s for Comm

The above circuit utilizes two transistors. Each device only

Niknejad Advanced IC’s for Comm

Niknejad Advanced IC’s for Comm

The average current drawn by each transistor is given by

Where Ip is the peak voltage drawn from the supply.

Note that the total DC current draw is twice IQ since both

Niknejad Advanced IC’s for Comm

Since the collector voltage swing can be as large as VCC

Niknejad Advanced IC’s for Comm

The efficiency drops linearly as we back-off from the peak

Niknejad Advanced IC’s for Comm

A tuned Class B amplifier works with a single devices by

Niknejad Advanced IC’s for Comm

In a single transistor version, the “minus” pulse is in fact

Niknejad Advanced IC’s for Comm

Niknejad Advanced IC’s for Comm

The efficiency is therefore the same

The DC power drawn from the supply is proportional to the

Niknejad Advanced IC’s for Comm

Integrating the above expression

Niknejad Advanced IC’s for Comm

Recall the definition for PC

Supply DC-DC converter

Envelope tracking supply and dynamic class-A

Niknejad Advanced IC’s for Comm

Niknejad Advanced IC’s for Comm

Often amplifiers are characterized by their conduction angle,

Niknejad Advanced IC’s for Comm

ic (t) vc (t) ic (t)

The most optimal waveform is shown above, where a current

Niknejad Advanced IC’s for Comm