Microwave Power Protectors

Attenuators and Limiters

Khalifa ECHCHAKHAOUI1, Elhassane ABDELMOUNIM1, Hamid BENNIS2, Mohamed
LATRACH3
1
ASTI Laboratory, FST of Settat, Hassan 1st University, Morocco
2

LITEN, FPK/FST of Settat, Hassan 1st University, Morocco

Microwave Group ESEO ANGERS, France

ABSTRACT
In this chapter, microwave power attenuator and limiter theory and technological realization are
presented. The chapter is divided in two sections, first section is dedicated to attenuator circuits and the
second section is dedicated to power limiters circuits.
Authors describe, in first section, principles characteristic and fundamentals of attenuator and detail of
the most common topologies such as T-attenuator, PI-attenuator and bridged-attenuator. After a
presentation of important equations needed to calculate attenuation rate provided by each of these
previous cited topologies, authors present the variable attenuator based on active component (PIN diode,
Transistors).
In second section, authors present power limiter characteristic and fundamentals. Afterward, they present
a state of arts of technological solution to design power limiter based on solid state components such as
PIN diode and Schottky diodes.
Keywords: Attenuator, Limiter, microstrip, PIN diode, Schottky diode, attenuation rate, limiting rate, Tee
attenuator, PI-attenuator, insertion loss, impedance characteristic.

INTRODUCTION
Since the appearance of RF & microwave systems, the flow of electromagnetic waves of high power
presents a serious threat to sensitive electronic components such as low noise amplifiers (LNA), radar and
space communications (D. Shiffler, 2005). Below, few examples of damaging effects of high power
microwave flows:
Destruction of electronic components: LNA (Low Noise Amplifier) components are
sensitive to high power microwave. Failure to control the reception chain may damage these
components permanently.

Saturation of radio-receiving elements: in the case where the received power exceeds the
sensitivity threshold of the receivers, the performance of radio receivers is not linear and
therefore, these receivers cannot filter the useful signal.

Generation of interference: mobile cellular telecommunications systems such as CDMA,

WCDMA and LTE systems are based on the power control signals to prevent mutual
interference between different mobile devices covered by the same cells. Mobile devices
around the radio cell must reduce their emission power in order to transmit low power signals

compared with the powers of signals to be emitted by devices located in a remote area of the
cell.
Therefore, there is a permanent interest to integrate into the RF receiver chains Power control circuit, in
order to avoid equipment malfunction or destruction of the sensitive elements of the receiving chain.
In literature, the microwave device protection against high power signal is provided by two kinds of
circuits: Power attenuator (Sun, Choi, & Weide, JUNE 2005) and power limiter circuits (Maloratsky,
2004). The main difference between the two circuits lies in the fact that the attenuator reduces the power
signal by a predetermined ratio while the power limiter tends clipping incident signal below a threshold
power.
This Chapter is divided on two sections. First section is dedicated to attenuator circuits and second section
is dedicated to Power limiter circuits. In each section authors present necessary definition of power
control circuits and the main concerns about design and conception of these circuits. At the end of each
section, examples are presented and discussed to point out best practice to design and optimize power
attenuator and limiter circuits.

SECTION I: ATTENUATORS
Attenuators are equipment and circuits introduced between electrical and
microwave power source and receiver or load in order to reduce the signal power by
a predetermined ratio (Sundararajn & Peterson, 1999). The attenuation ratio is
expressed in decibels by the ratio of the output power (Pout) to the incident power
(Pin) as shown in equation below:
Attenuation = 10 log10 (Pin / Pout).
The attenuator should not introduce reflection to the power source. Consequently
the attenuator input impedance and output impedance must be matched on the
main line.
Attenuators are designed with lumped circuit and distributed circuits. They may be
in the form of transmission line, microstrip, stripline, waveguide component. The
most used method to introduce attenuation on a transmission line is to place
resistors in the electric field centre. Thanks to the electric field, there is an induced
current causing a loss of power on the line (Sundararajn & Peterson, 1999).
There are several circumstances where it is necessary to insert an attenuator to
reduce the power and the level of signals (current and voltage) as:

Reduce the signal level to avoid saturation of the systems.

In order to adapt the output of a circuit to the impedance of the load.
To measure the gain or loss of two ports.
To provide isolation between components of a circuit.
To extend the power range capacity of equipments such as measurement
instruments. For example, if a measurement device supports only 100mw, it
is possible to measure the power level of 200 mW if an attenuator 3dB or
more is inserted at the input of the this device.
To balance the power received from several sources

The attenuators can be classified into different types according to the nature of the
circuits used, the configuration and method of attenuation. Thus there are:

Passive and active attenuators

Reflective and absorptive attenuators
Variable and fixed attenuators

ATTENUATORS CHARACTERISTICS
The attenuators are designed according to the specifications required by their
use. Different parameters are considered in the design of attenuator
(Sundararajn & Peterson, 1999), namely:

Attenuation rate range: it is the ratio of output power and input power.
According to this feature, there are variable attenuators and fixed
attenuators.
Frequency band: The attenuation rate is more accurate in the operating
frequency band of the attenuator.
Frequency Sensitivity: it is the maximum variation (peak to peak) of the
attenuation rates across the frequency band.
Power range: this is the maximum power supported by the attenuator
without compromising its functionality and performance.
Power sensitivity: this characteristic represents the variation of attenuation
rate as a function of power. This variation can be measured by dB / W.
Operating temperature: the temperature range in which the attenuator
operates in full power.
Temperature sensitivity: the variation of the attenuation as a function of
the temperature dB / (dBx C).
Input Reflection rate: this is the level of signal reflected back to the source
when the load impedance is matched to the source.
Output Reflection rate: the level of the reflected signal to the load when
the load impedance is matched to the source.

PASSIVE ATTENUATORS
At low frequency, Passive attenuators are made of pure resistances. The
arrangement of resistances between them and their values are determined
according to the desired attenuation and the circuit topology. In high frequency,
Microstrip attenuators are implemented using resisting layers, which are made
by etching thin-film resistive coating on an isolating substrate (Zagorodny,
Voronin, Yunusov, Goshin, Fateev, & Popkov, 2012). In the literature, there are
several attenuator topologies known according to the arrangement of
resistances. The most popular formats are: PI, T, T-bridged, L topology,
rectangular (Micke, 1990) and cruciform (Fournier & Boillot, 2007), interdigitated
structure (Veteran, 1991).

T attenuator:
This configuration is consisting of three resistors in T format as drawn in the
figure below. This attenuator can be symmetric if R1 = R2 or asymmetrical in the
general case (ie R1 R2). Where R1 R2, the T-attenuator will serve also for
matching impedances between the source and the load.

Figure 1: T-attenuator topology

Calculation of R1, R2 and R3 resistance values

If we set the attenuation k = Pin / Pout, the following equations give the values of
R1, R2 and R3:
R1=

R2=

Zink +1
ZinZout k
2
k1
k1

(1)

Zoutk +1
ZinZoutk
2
k 1
k1

(2)

R3= 2

ZinZout k
k 1

(3)

Important special cases:

Symmetrical T-attenuator
If the input impedance and the output impedance of the circuit are equal, then Zin
= Zout and the equations (1), (2) and (3) can be simplified as follows:
R1=R2=Z*

k 1
k +1

(4)

et

R3=2*Z* k1

(5)

L-pad attenuator
If R2 = 0 and ZinZout, T attenuator circuit is transformed into L-pad attenuator
circuit. Equations (1), (2) and (3) become:
R1= Zin *

R3=

k1
k +1

Zoutk +1
k 1

Zin k +1
=
Zout 4 k

(6)

(7)

(8)

Calculating the power dissipated in the T-attenuator system:

It is proposed to calculate the power dissipated in the resistance in a circuit
symmetrical T attenuator as specified in the Fig below. The calculation is
necessary to predict the power handling of resistors used in a T attenuator
circuit and to dimension the corresponding heat generation.

Figure 2 symetrical T-attenuator

We target in this exercise to calculate P(R1), P(R2) and P(R1) with:

P (R1): dissipated power at the source side serie resistor R1

P (R2): dissipated power at the parallel resistor R2
P (R1 '): dissipated power at the load side serie resistor R1

Calculation of P(R2)
P (R2) = (1/2) * R2* Ur22 = (1/2) * R2 * [(R1 + ZL / R2] * IL (8)

We replace IL with 2 * Pout / Zl and we get the following formula:

P (R2) = [(R1 + Zl) / (Zl * R2)] * Pout

(9)

In formula (9) we replace R1, R2 according to their formula in (4) and we replace
Pout by Pin/k, the final Power dissipated in R2 will be:

2
k 1
k

P(R2)=
Pin
k +1

(10)

For the power dissipated in the series resistance at output side, P (R1) = 1/2 *
R1 * Il
Knowing that Pout = 1/2 * Zl * Il and we replace IL in P (R1 ') we obtain P (R1) =
(R1 / Zl) * Pout
Using formula (4) and replacing Pout by Pin/k, the equation of the power
dissipation at Output series resistor becomes:

1
k1
k
P(R1)=
Pin
k +1

(11)

The same calculation done for P(R1) will be applied at input series resistor if we
replace load impedance by source impedance and we replace Pout by Pin, the
formula of P(R1) becomes:
P(R1)=

k 1 Pin
k +1

(12)

In an adapted circuit, The system shows no reflection to the source,

consequently, conserving power law is checked for the system. we obtain that:
Pin = P (R1) + P (R2) + P (R1 ') + Pout. (13)

PI attenuator
This is an attenuator composed of three pure resistors forming a circuit of the
form of the Greek letter as shown in the figure below:

Figure 3 PI-attenuator

Calculation of resistance values R1, R2 and R3

The values of R1, R2 and R3 are given by the following formulas:

R 1=

Zink 1

Zin
Zout

(14)

Zoutk 1
Zin
k +12 k
Zout

(15)

(k 1) ZinZout
2 k

(16)

k +12 k

R 3=

R 2=

Where

K=

Pin
Pout

Important special case Zin = Zout:

In the particular case where Zin = Zout, we have the following equations:

R 1=R 3=

Z k +1
k 1

(17)

et

R 2=

Z(k 1)
2 k

(18)

Calculation of the power dissipated in a symmetrical PIattenuator Circuit

R3 et Zout are parallel, PR3=

Zin
PinZout
Pink + 12 k
Zout
=
kR 3
k(k 1)

R1 et Zin are parallel, PR1=

Zin
PinZin
Pink + 12 k
Zout
=
R1
( k1)

Applying the law of conservation of power: PR2= Pin-Pout-PR1-PR3

(19)

(20)

(21)

T-bridged attenuator
This is a T-attenuator circuit to which a bridge resistance were added in series
between the source and the load as presented in figure 4 below.

Figure 4: T-bridged attenuator

The typical design of this circuit is to take the series resistors R1 equal to the
impedance characteristics of the source and load, and calculate R2 and R3
according to the desired attenuation by the following formulas:

R 3=Zo( k 1)

R 2=

Z0
k1

(22)

(23)

The table below shows the values of the resistors according to the desired
attenuation rate (k = Pin / Pout) for three attenuator format, the characteristic
impedance at the input and output is considered at the value Z0.
Table 1: important attenuator equations

T
atten
uator

R 1=

Zo k1
k +1

R 2=

2Zo k
k1

power :
P(R1) =

Att=10*

k 1 Pin
k +1

P(R2)=

log ( k )
= 20*log(

2
k 1
k
Pin
k +1

R 1+ Z 0

Z 0R 1
Z02=2*R1*R2+R
1

P(R1)=

1
k1
k
Pin
k +1
PI
atten
uator

R1=

Zo k +1
k 1

k 1
R2=Zo* 2 k

Power :
P(R1) =

k 1 Pin
k +1

P(R2)=

2
k 1
k
Pin
k +1

Att=10*log(k)
=
20*log(

R 1+ Z 0

R 1Z 0

P(R1)=

1
k1
k
Pin
k +1
Bridg
ed T
atten
uator

R1=

R2=

Zo( k1)
Z0
k 1

Att=10*log(k)
=
20*log(1+

R 1
R 2

Z0=

R 1R 2

VARIABLE ATTENUATOR:
Variable attenuators are used to adjust the attenuation level according to the needs.
Variable attenuator are useful in automatic gain control (AGC) circuits and power
leveling applications [6]
Thus, the design of variable attenuator consists of using attenuator resistive circuit
and replacing fixed resistor with active component having the ability to change its
resistance with a control input (current or voltage control) such as PIN diode. The
variable attenuator design will be simple if there is no need for a matched
attenuator circuit (the reflection signal to the source is tolerated). However,
elaborated and more complex circuits are designed for matched variable attenuator.
Variable attenuators can present continuous attenuation rate or discrete reduction
rate (step by step).
1. Step by step attenuation: To construct discrete attenuator, a series of
fixed attenuators are arranged to provide mechanically variable attenuation
rate step by step. For each level of attenuation, a switch is used to select the
appropriate attenuator. This arrangement allows discrete attenuation values
in each position. Resistive attenuators networks provide a theoretically
unlimited bandwidth (resistor characteristics limit in reality this assumption),
but require many attenuator stage to provide desired attenuation while the
aggregation of several stages will present insertion loss and reflection due to
mismatch impedance. In addition, switchs may introduce noise and distortion
[1].
2. Continuous attenuation: The attenuation rate can also be controlled by a
current or voltage and automatically adjusted according to the needs using
active components such as Pin Diode, MESFET transistors.

PIN DIODE ATTENUATOR

PIN Diode Attenuators are a subset of variable RF attenuators and are useful for
circuits requiring continuously changing attenuation levels [1].
Pin diodes are used in many microwave communication systems because of their
high breakdown voltages, fast switching characteristics, and their variable
resistance characteristics with bias. They provide circuit functions in antenna
switches, phase shifters and attenuators for automatic gain control (ALC) or RF
level control applications[].
PIN diode attenuators can take many forms: from a simple diode connected in
series or parallel, acting as a lossy reflection device, to more complex structures
that maintain a mached input impedance over the full attenuation rate capacity
of the attenuator[Emmanuel Gatard].

When the PIN diode is used in the attenuator, the attenuation characteristics are
controlled by the forward bias current through the PIN diode. This is compatible
with the fact that the RF resistance of the PIN diode is determined by the forward
bias current[].

PIN diode characteristics

A PIN diode is a semiconductor device formed of of three layers:

positively doped layer P +

lightly doped intrinsic layer and very small width W
negatively doped layer N +

Figure 5: PIN diode layer presentation

Forward bias operation :

At low frequencies, the PIN diode behaves as an ordinary PN junction diode, but
at high frequencies it behaves as a resistor whose value can be controlled by
current. Fig. 5 shows a PIN diode high frequency equivalent linear model where:

CI: is the constant capacitance, Which depends on the geometry of the

intrinsic layer.
RI is the variable resistance which depends on forward current passing
through the diode.
Lp and Rp represent parasitic packaging inductance and resistance.

Figure 6: Linear Equivalent Circuit Model for PIN diode

The equivalent impedance of PIN diode is calculated from its equivalent circuit as

R
1+ jRiCiW . When the current through the diode is below a threshold value,

the diode acts as a capacitor of low value and a high resistance; When the
current exceeds the threshold (the threshold of the diode is defined by the width
of the intrinsic region and the doping of the P + layer and N +), electrical
charges (electrons and holes) are pushed toward the intrinsic region and
therefore the diode becomes equivalent to a low resistance value. In a low
frequency, If the reactance value of CI is much larger than variable resistance,
the PIN diode can be considered as a pure resistor (when the parasitic package
inductance and resistance will be ignored).
The variable resistance of PIN diode depends on bias current according to the
following equation[4] :

Rs=

W2
( p+ n )If

Where :

R: The resistance characteristic of a PIN diode

W: the I-region width
: carrier lifetime
p, n: the hole and electron mobility respectively

This equation is valid for frequencies higher than the transit time of the I-region: f >
1300/ W2 (f in MHz and W in microns)[5].
For a PIN diode with an I-region width of typically 250 m, carrier lifetime of 4 s,
and n of 0.13 m2/s, p of 0.05 m2/s, Figure 3 shows the Resistance vs current
characteristic. At 0.01mA, the resistance is about 8.6K . This resistance fall to 4
at If=20mA.

Figure 7: Typical Diode Resistance vs. Forward Current

PIN DIODE ATTENUATOR TOPOLOGY

The main use of PIN diode in attenuator circuit is to provide a tool to control
attenuation rate with a current signal taking in consideration the
resistance/forwarding current characteristic.

T-attenuator designed with two PIN diode:

In Fig 7. T-attenuator is designed using two PIN diodes. According to equation
presented in table 1:

R+ Z 0

Attenuation= 20*log(
Z 0R
Z02=2*R*R2+R
This equation shows that impedance charactersitic changes with diode resistance.
Consequently, the attenuator circuit will not be matched all times.
If we have the same PIN characteristic presented in previous section, the table
below presents the attenuation and impedance charaterestic of the circuit versus
forwarding current:
Table 2: attenuation and impednace characterestic of T-attenuator designed with two PIN
diodes

If (mA)

R(ohm)

R2 (ohm)

Z0(ohm)

1
2
3
4
5
10
15
20
25

87
43
29
22
17
9
6
4
3

50
50
50
50
50
50
50
50
50

127.34
78.89
61.08
51.39
45.14
30.71
24.74
21.28
18.95

Attenuation
(dB)
14
11
9
8
7
5
4
4
3

C2

C3

L2

L1
V+
DC supply

Pin_ Diode1

P_ nTone
PORT1
Source

Pin_ Diode2
C
C1
C=100.0 pF

DC_Block
DC_Block1

DC_Block
DC_Block2

Term load
Term2
Num=2
Z=Zl Ohm

R2

Figure 8:T-attenuator with PIN diode

T-bridged attenuator designed with PIN diodes

The most appropriate for matched broadband attenuator applications, especially
those in the RF bands from HF Band through UHF Band, are the Bridged TEE & PI
circuits.
In Fig 8. T-bridegd attenuator is designed using two PIN diodes. According to
equation presented in table 1:
Attenuation=20*log(1+

Z0=

R1
R2 )

R 1R 2

Where R1 is the resistance of diode D1 and R2 is the resistance of diode D2.

These equations show that attenuation rate depends on the ratio of R2 to R1 while
impedance characteristic of the circuit Z 0 depend on their product. R. D. Joos
proposes in reference [6] procedure to bias PIN diode in order to respect matching
condition of matched Bridge TEE Attenuators for Low Frequency Distortion.

Figure 9: T-bridged attenuator designed with two PIN diodes

PI attenuator designed with PIN diodes

In Fig 9. Presents a PI attenuator designed with three PIN diodes. the biasing circuit
involved with this design is unbalanced. Ray Waugh designed a blanced PI
attenuator using four PIN diode as presented in Fig 10.

V+
DC_Block
DC_Block1

R
R3
R=6.81 kOhm

DC_supply
DC_Block
DC_Block1

L
L1
L=1.0 nH
R=

L
L1
L=1.0 nH
R=

DC_Block
DC_Block5

T erm
T erm1
Num=1
Z=50 Ohm

P IN_diode
P inDiode2

PIN_diode
PinDiode1

DC_Block
DC_Block2

R
R3
R=6.81 kOhm

R
R1
R=450 Ohm

P IN_diode
P inDiode3

DC_Block
DC_Block3

DC_Block
DC_Block7
R
R2
R=1.47 kOhm

Term
Term2
Num=2
Z=50 Ohm

DC_Block
DC_Block5

Term
Term1
Num=1
Z=50 Ohm

PIN_diode
PinDiode1
PIN_diode
PinDiode4
R
R4
R=1.47 kOhm

R
R1
DC_Block
R=450 Ohm
DC_Block2

PIN_diode
PinDiode2

R
R5
R=450 Ohm

PIN_diode
PinDiode3

DC_Block
DC_Block3

R
R2
R=1.47 kOhm

DC_Block
DC_Block7

Term
Term2
Num=2
Z=50 Ohm

Control_current

Figure 10: PI attenuator designed with three PIN Figure

diodes11: PI attenuator designed with four PIN
diodes

If D2 is replaced by two diodes, as shown in Fig 10, several benefits result:

Since the maximum isolation of the network is set by the capacitive

reactance of the series diode(s), the use of two diodes in place of one will
increase the maximum attenuation or double the upper frequency limit for a
given value of attenuation.
The twin diodes which occupy the position of the series resistor are physically
set up 180 out of phase, resulting in the cancellation of even order distortion
products.
The resulting attenuator network is symmetrical and the bias network is
substantially simplified.

SECTION II: POWER LIMITER

A power limiter is a power attenuation device inserted between the source and
receiver to prevent incidents signals above a maximum power to pass to the
receiver while allowing signals below a given threshold to pass to the receiver with
minimum loss. Powers attenuators induce attenuation independently of receiving
power signal value.
Usually, power limiters exploit the functionality of the impedance variation of a
component according to the incident signal. Thus, the component used to limit
power is inserted in the circuit in anti-parallel form shunted between the
transmission line and the common ground. When the power increases, the
component impedance decreases, and starts to absorb some of the received power.
Solid state Power limiters based on semiconductor components are mainly based on
the reflection and absorption of a portion of the stream of a transmission line by
using many topologies. Among these topologies, we find the topology presented in
Fig.1, the diode used to limit the power level is inserted in the circuit in anti-parallel
position (shunted between the transmission line and the common ground). When
the power increases, the diode impedance decreases, and starts to absorb some of
the received power and a rest of the power will be reflected.
RF IN
IN

RF Out
DC_Block
1

TL1

TL2

DC_block2

OUT

D1

Figure 12: Classic Power Limiter Toplogy

According to current supply of the circuit, power limiter can be distinguished in two
types: active power limiters that require external current to operate and passive
power limiters that dont demand any bias. In addition to these two modes limiters
(reflective and absorbing limiter), there is a third mode of operation where the
excess power is redirected to a dedicated circuit for analysis (for military
applications, for example).

KEY PARAMETERS OF A POWER LIMITER

Limiters are specified by a number of Key Parameters:

Power range
Operating frequency band.
Low insertion loss in the power range of the protected circuit
Limiting rate for signals that exceed the performance limit tolerated by the
circuit to be protected.

Response time (on the order of nanoseconds) upon arrival signals exceeding
the threshold power limit.
Recovery time to normal operation as soon as the incident power is below
accepted threshold.
Noise and distortion induced by the limiter and attenuator
Dissipation of temperature.
Powers limiters operating configuration.

Power range
Power range is the most important characteristic that impact the choice of a
solution to limit or attenuate Microwave Power. For power limiter, we distinguish
three thresholds:
-

Activation of limiting mode: signal having power under this threshold are
called small signal. In this case, the power limiter must introduce a
minimum insertion loss. Signal having power above this threshold are
called large signal.
Power limiting capacity: if the incident power exceeds the capacity of the
limiter, the limiter cannot continue to clip the signal. A portion of the
source power is transmitted to the load.
Maximal power supported by the limiter: the power threshold of the limiter
without damaging its active components

Operating frequency band:

The ideal Power Limiter or attenuator must have same behaviour in all frequency
bands. However this kind of power limiter doesnt exist and the choice of active
component and circuit design depends on operating frequency band.
For high frequency, the component chosen to operate as Power limiter must have a
small capacity. If this condition cannot be met, the circuit will exhibit low impedance
and will present a high insertion loss. Varactor are made from materials that change
their dielectric according to power signal applied to it. But this component is note
suitable for high frequency because of the presence of the capacity.
In semiconductor component, PIN diode present two equivalent circuits depending
on operating frequency [Applications of PIN diode (AN922), November 1999]. In
frequency is below f0.1*f c, the PIN diode acts as a simple diode PN, but in
frequency f10*fc, PIN diodes acts as variable resistance.
In very high frequency, intelligent materials such as Dioxide Vanadium are more
suitable for power limiter.
Insertion loss:
This parameter defines the small signal throughput loss (S21) of the Limiter.
Insertion loss is defined over a dynamic range up to the input limiting range. To
avoid signal distortion, Limiter should exhibit a constant insertion loss over the
operating frequency band.

Limiting rate
The ratio of difference of output power to difference of input power over the limiting
input power range.
Response and recovery Time
Power limiters are required to respond quickly to large signals and to recover quickly
from limiting events to get back on line (ns requirement) (100 - 3500 ns for PIN
diode limiters). This is a pulsed condition and is defined as the time between the
50% point of the trailing edge of the high power pulse to the time where the output
reaches 90% of the final small signal level.[22]
Noise and distortion
Limiter must not introduce noise or distortion to incident signal in operating
frequency range. Thus, the insertion loss and the limiting rate should be identical
for the frequency range.
Temperature dissipation
As the limiter absorbs a portion of incident signal power, the limiter design must
implement solutions to handle temperature dissipation. Also the behaviour of the
limiter must remain constant over the temperature range
Operating configuration
Power Limiters operate in two configurations:

Active power limiters require external DC bias to perform limiting function .

Passive (self-bias) power limiters are turned on by the RF signal itself. A

passive limiter is a self-activating switch that is activated by high-level
incident power.

The quasi-active limiter in solid state implementations based on diodes uses

a high-power Schottky barrier diode detector that supplies rectified DC
current to the limiter diode. The Schottky diode rectification process is much
faster than limiter diode self bias. The limiter diode is turned on primarily by
the external bias current from the Schottky detector diode, which is sensitive
to the incident signal.

OPERATING STATE OF A POWER LIMITER

A Microwave Power limiter operates in three states:

Small Signal: the incident power is below the trigger level of the dynamics of
power limiter. In this case, the limiter should only insert a small loss at output
signal.

Large signal: forward power exceeds the dynamic threshold of the limiter. In
this case, the output power tends to remain constant with increase of input
power.
Very large signals: forward power exceeds the capacity of the limiter. The
limiter cannot continue to clip the incident power. Part of the power source is
transmitted to the load.

An ideal limiter is assumed to operate in two states:

Small signal: no insertion loss

Large signal: the power is clipped regardless of the incident power

The below figure illustrate the three state operating modes of an ideal limiter and a
practical limiter:

Figure 13: operating mode of power limiter

TECHNOLOGICAL SOLUTIONS TO ACHIEVE POWER LIMITERS:

Several technological solutions have been used to make limiter circuits and power
attenuators. Below is a summary of these solutions with their key features
Semiconductor-based limiters (solid state limiter)
These limiters are most prevalent in domestic telecommunications systems (mobile
phones) and are made primarily based on PIN diodes, Schottky diodes, MESFET
transistors.
Different topologies are cited in state-of-art literature to perform power limiter
based on semiconductor components mounted in microstrip circuit and in
monolithic technology. Below some technical realizations are given as illustration:

The Limiter design incorporate a network of PIN or Shottky diode mounted

across a planar 50 transmission line as described by Leo G. Maloratsky in
Passive RF&Microwave integrated circuits, Newnes, 2004, Elsevier Inc

By using a microstrip lines coupled with a resonator circuit as proposed by

Nikolai V. Drozdovski&Lioudmila M. Drozdovskaia in Microstrip and
waveguide passive power limiters with simplified construction, Journal of
Microwaves and Optoelectronics, Vol. 1, No. 5, December 1999

Power limiter based on MESFET mounted across the lines transmission as

proposed by Nikolai V. Drozdovski in Microwave passive power limiters based
on MESFETs, Journal of Microwaves and Optoelectronics, Vol. 1, No. 2, April
1998.

a Zero Bias Broadband Microstrip Power Limiter based on two Shottky diodes
mounted on ring line inserting a 180 shift phase to the main line

The main characteristic exploited in these devices is the change in their resistance
as a function of the received power. A low power, these devices have a very high
resistance when in great power they have high resistance.
Superconductors and smart materials
The super-conductor based limiters using superconducting materials that change
from conducting state to the semiconductor state when its temperature rises. The
operating principle is as follows: A super-conducting layer is inserted in series with
the main transmission line. When the power is below a threshold, the temperature
loss of the circuit is low. The super conductor has a very low resistance. When the
temperature increases, the superconductor behaves like a semiconductor and
blocks the incident signal. This type of limiter has advantages in terms of time
response but is difficult to realize as it requires cooling mechanisms and control of
the ambient temperature to avoid that an external warming induces activation of
the limiter and therefore a loss of useful signal
Smart materials are materials that exhibit varying electrical characteristics based
on external effects (temperature, light ...). The Vanadium Oxide VO2 is an example
of material that passes from the semiconductor state to conducting state according
to a threshold temperature.
This characteristic is exploited to use this material as a power limiter or as
-disjoncteur switch. If the incident power exceeds a threshold, then the VO2 layer
heats up and begins to short-circuit current to ground.
Materials type Metal Oxide Varistors (eg ZnO) have a variable resistivity as a
function of the applied electric field. The power limiters based on these materials
withstand high power but presents limitation at high frequency because of their
high capacity.
The MEMS components (Micro Electro Mechanical Systems)
MEMS are electronic components consisting of a mechanical system deformable by
an electrostatic or magnetic field. Thus, electrostatic and mechanical force acts on
electrical characteristics of the MEMS component (its impedance) and especially on
its capacity. The use of these components in the power limiters exploits the

possibility of varying the capacitance of the component as a function of the

received power.
Ferroelectric material
The ferroelectric material has the property of having a permittivity which depends
on the electric field. As the value of the capacity depends on the permittivity of a
medium, these materials can be used to mismatch a transmission line and get the
power limiter function.
This type of power limitation has high threshold power (100 W). Therefore they are
not suitable for use in telecommunications systems. In addition, and given the
capacity utilization, this type limiter will present in high frequency a significant
insertion loss even in low power.
Ferromagnetic material.
Some materials (eg Yttrium Iron Garnet, lithium ferrite) are saturated in the
presence of a magnetic field. This property is exploited to design power limiter
based on these materials. When the incident power is sufficiently high, the
magnetic field causes the saturation of the material and excess power is dissipated
by the material in the form of heat. These limiters also have limitations in terms of
frequency bandwidth (a few MHz), low power handling and a high response time.
Power limiters based on gas/plasma spark.
The power limiters based Gas discharge tubes are designed for protection
applications against high electrical powers (lightning arrestor for example). The
basic principle is based on the ionization of a gas between two metal plates where a
high electric field is applied to these plates. The spark plasma-based power limiters
exploit the same plasma ionization phenomenon but the powers involved are
relatively low [47 dBm] compared to gas discharge. Vacuum diode based power
limiters exploiting the phenomenon of electron emission from a metal in vacuum in
the presence of an electric field.
Comparison with state-of-art technique
Thetablebelowpresentscomparisonsbetweensomeimplementationsolutionscitedinthescientificliteratureof
powerlimiters.
Table 3. Comparison of microwave power limiter.
Design
Microstrip Passive limiters
usingdiscreteSchottkydiodes
Passive limiters using PIN
diodeanddetectordiode
PlanarSchottkydiode
Planar Schottky diode and
MESFETbasedlimiter
PassivelimitersusingDiscrete
MESFETandSchottkydiode

Insertion
loss(S21)
0.9dB

Frequency
range
2.45Ghz

Limitingrate

Reference

0.8dB

7Ghz

12dB

[8]

1dB
1dB

1Ghz
1Ghz

20dB
15dB

[10]
[10]

0.9dB

7Ghz

15dB

[13]

20dB

Design

Insertion
loss(S21)

Frequency
range

Limitingrate

Reference

EXAMPLES OF SOLID STATE POWER LIMITER DESIGN

SolidstatePowerlimiters(PL)aremainlybasedonthereflectionandabsorptionofaportionofthestreamofa
transmissionlinebyusingmanytopologies[311].Amongthesetopologies,wefindthetopologypresentedin
Fig.12,thediodeusedtolimitthepowerlevelisinsertedinthecircuitinantiparallelposition(shuntedbetweenthe
transmissionlineandthecommonground).Whenthepowerincreases,thediodeimpedancedecreases,andstartsto
absorbsomeofthereceivedpowerandarestofthepowerwillbereflected[2].
Accordingtocurrentsupplyofthecircuit,Powerlimitercanbedistinguishedintwotypes:activepowerlimiters
thatrequireexternalcurrenttooperateandpassivepowerlimitersthatdontnecessitateanycurrentsupply.
Thepresenteddesignisapassivepowerlimiterbasedontwomicrostriplines.Oneofthemisalinearlinethat
transmitsthemainsignalandthesecondlineisequippedbytwoSchottkydiodeswhichareusedtodivertaportion
ofincidentsignalwhentheamplitudeofthesignalreachesthethresholdofthediodes.ThetwoSchottkydiodes
operateasvoltagecontrolledattenuatorandamplitudedetector.

Figure 14: the structure of microstrip power limiter based on ring line

Inlowsignals,thecharacteristicimpedanceofthediodesishigh.Consequently,thesignalcannotpassthroughthe
bypasssecondline.Themainlinetransmitsthesignaltotheoutputwithalowinsertionlossgeneratedmainlyby
thetangentiallinelossesandcapacitiesofSchottkydiodesjunction.
In high signals, the received power exceeds the detection threshold of the Schottky diode. It follows that the
impedanceofSchottkydiodefallsandtheRFsignalstartstospreadonthebypassline.Sincethedifferencebetween
theelectricallengthofthebypasslineandthemainlineisequalto /2,thesignalspropagatingbetweenthetwo
lineswillhaveaphaseshiftof.Consequently,thepoweroftheresultingsignalisreduced.
Inhighpower,SchottkyDiodewillgenerateaDCrectifiedcurrent.Therefore,anantiparallelstubconnectedtothe
commongroundisinsertedtothemaintransmissionlinetoassuretheDCreturnpath.Thestubmusthaveaquarter
wavelengths(/4)inordertoprovideanopencircuitforhighfrequencyandshortcircuitforDCcurrent.
Toimprovetheperformanceofthisdesignintermsoflimitingrateandinsertionloss,thefinaldesignwillbe
composedoftwostages:

Figure 15. Microstrip Power limiter based on Schottky diode

Thesimulationresultsofthiscircuit byusingADS(AdvancedDesignSystem)from Agilenttechnologiesare

presentedbelowinFig.17.Thecircuitprovidesless1dBofinsertionlossatlowsignalover1Ghzbandwidth,and
upto20dBoflimitingpowerratewithawideoperatingfrequencyband.

Figure 16. S-parameters results versus frequency and Output Power versus input power at
2.1, 2.45, 2.6 and 2.9 GHz

FUTURE RESEARCH DIRECTIONS

Technological developments tend towards miniaturization of devices and integration
of several functions in one equipment. In this trend, recent research concerns the
use of MMIC and CMOS technologies to improve insertion loss and extend the
supported frequency band of microwave limiter and attenuator. In this trends
Peregrine has announced in April 2014 the introduction of a new limiters based on
ultraCMOS technologie. According to Peregrines press release, the new power
limiters provide a 10-100X improvement in response and recovery time; and deliver
greater than 40 dB improvement in linearity (IP3); offer a 20X improvement in ESD
(electrostatic discharge) protection [20].

Figure 17: Peregrines UltraCMOS power limiter (right) replaces discrete GaAs PIN diode
circuits (left)

Another technology trend concerns the development of power limiters for protection
against high-power microwave signals. In this category, solutions based on vacuum
diodes and microplasma are proposed to limit the high power microwave signals.

CONCLUSION
Microwave attenuator and limiter are presented in this chapter. The target is to
make this chapter a reference manual for microwave power attenuator and limiter
devices. Principles Characteristics and key parameters of limiters and attenuator are
presented and fully explained.
REFERENCES
1. D. Shiffler, T.K. Statom, T.W. Hussey, O.Zhou, P. Mardahl, High-power
microwave source, in Modern Microwave and MillimeterWave Power
Electronics, pp. 691-730, 2005,Wiley Interscience. (D. Shiffler, 2005)
2. Kae-Oh Sun, Min Ki Choi, and Daniel van der Weide, A PIN Diode
Controlled Variable Attenuator Using a 0-dB Branch-Line Coupler, IEEE
MICROWAVE AND WIRELESS COMPONENTS LETTERS, VOL. 15, NO. 6, JUNE
2005
3. Leo G. Maloratsky, Passive RF&Microwave integrated circuits, Newnes,
2004, Elsevier Inc
4. Microwave microstrip attenuators for GaAs monolithic integrated circuits
5. Doherty, W. E. , and R. D. Joos. PIN Diode Circuit Designers' Handbook.
Watertown, MA: MicroSemi Corp, 1998. Print.
6. K. W. Kobayashi, A. K. Oki, D.K. Umemoto, S. Claxton, and D. C. Streit
GaAs HBT PIN Diode Attenuators and Switches, IEEE 1993 Microwave
and Millimeter-Wave Monolithic Circuits Symposium.
7. Byung-Jun Jang, In-Bok Yom and Seong-Pal Lee, An Enhanced PIN Diode
Model for Voltage-Controlled PIN Diode Attenuator, Microwave
Conference, 2003. 33rd European (Volume:1 ).
8. G. Hiller, Design with PIN diodes MA-COM Application Note AG312
9. R. D. Joos, Selection of PIN Diodes in Designing Matched Bridge TEE
Attenuators for Low Frequency Distortion, Microsemi Corp. MPD 103, 199

10.Waugh, R. W., "A Low Cost Surface Surface Mount PIN Diode PI
Attenuator," Microwave Journal, vol. 35, no. 5, May 1992, pp 280-284.
11.B. M. Coaker, Radar receiver protection technology, Microwave Journal:
Military Microwaves Suppliment, August 2007, pp. 8-14.
12.B. M. Coaker, D. M. Dowthwaite and N. E. Priestley, High-Power multifunction radar receiver protection, Proc. European Radar Conference,
(EuRAD), Manchester, September 2006.
13.V. V. Volkov, V. P. Ivanova, Yu. S. Kuzmichev, Yu. V. Solovev, Design and
technology of monolithic GaAs pin diode limiters for the millimeter
wavelength range TECHNICAL PHYSICS LETTER Vol. 31 No. 7, 2005.
14.D. J. Seymour, D. D. Heston, and R. E. Lehmann X-Band and Ka-Band
monolithic GaAS PIN diode variable attenuation limiters 1988 IEEE MTT-S
Digest.
15.N. Roberts, A Review of Solid-State radar receiver protection devices,
Microwave Journal, Vol. 34, No. 2, February 1991, pp. 121125.
16.Nikolai V. Drozdovski & Lioudmila M. Drozdovskaia Microstrip and
waveguide passive power limiters whith simplified construction, Journal of
Microwaves and Optoelectronics, Vol. 1, No. 5, December 1999.
17.Bahl I.J., 10W CW broadband balanced limiter/LNA fabricated using MSAG
MESFET process, International Journal of RF and Microwave ComputerAided Engineering, vol. 13, Issue 2, pp. 118127, March 2003.
18.Nikolai V. Drozdovski Microwave passive power limiters based on
MESFETs, Journal of Microwaves and Optoelectronics, Vol. 1, No. 2, April
1998.
19.Sandeep Chaturvedi, G. Sai Saravanan, Mahadeva Bhat K., Sangam Bhalke
"MESFET process based planar Schottky diode and its application to
passive power limiters" 978-1-4799-2501-8/13/$31.00 2013 IEEE.
20.Bera, S., Basak, K., Jain, V., Singh, R., and Garg, V.: Schottky diode based
microwave limiter with adjustable threshold power level, Microw. Opt.
Technol. Lett., 2010, 52, pp. 16711673.
21.Peregrine press release BEIJING EDI CON April 7, 2014
http://www.psemi.com/newsroom/press-releases/772717-peregrineintroduces-ultracmos-rf-power-limiters-the-industry-s-first-monolithicalternative-to-discrete-pin-diode-limiters
22.Narda Microwave-East company Pin diode control product Application
Note
23.Mini-Circuits Application Note, High Power Limiter Modules, AN-42-001
Rev.: OR M124030 (08/27/09).
24. Sundararajn, R., Peterson, E. & Nowlin, R. 1999. Attenuators. Wiley Encyclopedia of

Electrical and Electronics Engineering. .

25.Emmanuel GATARD, doctoral thesis Analyse des phnomnes physiques
dans les diodes p-i-n : Contribution la modlisation lectrothermique
pour les applications de puissance RF et hyperfrquences 2006,
UNIVERSITE DE LIMOGES