Sziklai Pair based Small Signal Amplifier with BJT-MOSFET Hybrid Unit at
180nm Technology
SachchidaNand Shukla1, Syed Shamroz Arshad2,
Kavita Thakur3, Geetika Srivastava2
1Pt. Ravishankar Shukla University, Raipur, Chhattisgarh and
Department of Physics and Electronics (on lien), Dr. Ram Manohar Lohia Avadh
University, Ayodhya, UP, India
2Department of Physics and Electronics, Dr. Ram Manohar
Lohia Avadh University, Ayodhya, UP
3School of Studies in Electronics, Pt. Ravishankar Shukla University,
Raipur, Chhattisgarh
Sachida.shukla@gmail.com
Abstract
Two
circuit models of Small Signal amplifier, constituted with BJT-MOSFET hybrid
unit under Sziklai pair topology are designed and analyzed using ‘PSpice’ and ‘Cadence
Virtuoso and Spectre simulation tool (at GPDK 180nm technology)’ respectively. First
amplifier (Circuit-1) uses PSpice user-defined model of BJT and MOSFET whereas the
second amplifier (Circuit-2) consists of transistors available at GPDK 180nm
technology. Circuit-1 can amplify the AC signals of 1mV-1nV range with optimum
voltage gain 389.532, 137.570 current gain, 14.464MHz
bandwidth and 2.43% THD. However, Circuit-2 can amplify AC
signals of 0.1mV-10nV range with 164.018 voltage gain, 32.775 current gain, 11.906 MHz bandwidth,
and 13.608E-6% THD. Both the proposed amplifier circuits remove narrow band
problem and generate better results than earlier announced small signal Sziklai
pair amplifier with BJT-MOSFET hybrid unit in respect of voltage and current
gains, bandwidth, THD, and power consumption. Proposed amplifiers successfully
address the problem of poor frequency response of small
signal Darlington pair amplifier in higher frequency range and narrow bandwidth
limitations of small-signal PNP Sziklai pair amplifier. Dependency of the
proposed amplifiers at various biasing resistances and performance with
temperature variation, noise variation, DC supply variation, and phase
variation are also discussed herein. Proposed Circuits display strong
dependency over ideal maximum forward
beta ‘β’ of NPN transistor, Transconductance ‘VTO’ of P-MOS
transistor and additional biasing resistances ‘RA’. Layout of
Circuit-2 is found to cover 96.3898µm2 area with 11.32µm length and
8.515µm breadth. Minor percentage variation between pre-layout and post-layout
simulation results of Circuit-2 validates the proposed design at GPDK 180nm
technology. Monte Carlo and Process Corner analysis are also performed to test
the robustness and insensitivity of Circuit-2 against mean value of the
parameters and process and mismatch variations respectively. Performance summary of
the proposed circuits and comparison with the recently reported designs shows
effectiveness of the proposed circuits in terms of power gain, THD, voltage
gain, current gain, input referred noise and power gain. Qualitative analysis of the
proposed Circuits recommends its
usability
as Low Noise Amplifier in the portable RF noise measurement system.
Key Words: Sziklai pair, Small signal amplifier,
compound pair
Introduction
For a variety of electronics and communication system
applications, the preferred device configurations are the Darlington and
Sziklai Pairs [1]-[4]. Due to the identical ranges for input impedance, output
impedance, voltage gain, and current gain factor ‘β’, Darlington and Sziklai
pairs are regarded as complementary to one another in many applications [4]-[5].
But now-a-days, Sziklai pair gradually replacing the Darlington pair in
small-signal amplifiers, power amplifiers and digital circuits because of its
half base turn-ON voltage, low power dissipation, and better switching speed
than Darlington pair [5].
In order to combine the required qualities of JFETs and
BJTs, Aina et al. (1993) first ever used a JFET-BJT hybrid unit based
small-signal amplifier under Darlington pair topology [6]. They concurrently
obtained high input impedance and high current gain. These coordinated efforts
have resulted in the usage of a number of devices in hybrid combinations under
Darlington pair and Sziklai pair topology by many researchers [6]-[13]. The BJT
and MOSFET hybrid unit based Sziklai pair amplifier had been reported by Shukla
et al. in 2015 which crops high voltage gain, moderate current gain with
low THD [7].
This manuscript reports modified version of two
small-signal amplifiers circuits with BJT-MOSFET hybrid unit under Sziklai pair
topology. These circuits not only resolve the narrow band problem found in [7]
and PNP Sziklai pair Small Signal Amplifier but also resolves the poor response
problem of small signal Darlington pair amplifier at higher frequency [12],
[14]-[15]. The key merits of the reported circuits are their high voltage and
current gains, wide bandwidth, low power consumption, and low input referred
noise.
Circuit Details
Circuit
designs of Proposed amplifiers, as sketched in Fig.1(a) and Fig.1(b), are
referred herein as Circuit-1 and Circuit-2 respectively. Circuits of Fig.1(a)
and Fig.1(b) accommodate NPN type BJT at driver position and P-type MOSFET at
follower position under Sziklai pair topology. Respective circuits of proposed
amplifiers are analyzed with the aid of PSpice (Student Version 9.2), and
Cadence Virtuoso and Spectre Simulation (at GPDK 180nm technology) tools
[16]-[17]. Proposed amplifier (Circuit-1) accommodates user defined PSpice
model of BJT (QMODN with β=300) at driver position and
P-type MOSFET (PMOSD with VTO=-2) at follower position under Sziklai
pair topology [16].
Fig.1(a). Proposed Amplifier (Circuit-1)
under PSpice Simulation tool
Fig.1(b). Proposed Amplifier
(Circuit-2) under Cadence Virtuoso and Spectre Simulation tool at GPDK 180nm
technology
Table
2: Model Parameters of P-type MOSFET for Circuit-1 and Circuit-2 amplifiers
|
Model Parameters
for MOSFETs
|
PMOSD
(Circuit-1)
|
pmos
(Circuit-2)
|
|
LEVEL (Model type
1,2, or 3)
|
3
|
3
|
|
L (Channel Length)
|
100μm
|
2μm
|
|
W (Channel Width)
|
100μm
|
1.9μm
|
|
VTO (Threshold
Voltage)
|
-2V
|
-3.67V
|
|
KP
(Transconductance)
|
4.54 Amp/V
|
10.15
μ Amp/V
|
|
PHI (Surface
Potential)
|
0.6V
|
0.6V
|
|
RD (Drain ohmic resistance)
|
5Ω
|
60.66mΩ
|
|
RDS (Drain Ohmic
Shunt Resistance)
|
1E+6Ω
|
444.4KΩ
|
|
RS (Source Ohmic
Resistance)
|
--
|
70.6mΩ
|
|
IS (Gate p-n
saturation current)
|
10E-15A
|
52.23E-18A
|
|
PB (Gate p-n potential)
|
0.8V
|
0.8V
|
|
CBD (Bulk-drain zero bias p-n
capacitance)
|
5E-12F
|
2.141nF
|
Table 1: Model Parameters of NPN BJT for Circuit-1 and
Circuit-2 amplifiers |
|
Model Parameters for BJTs
|
QMODN
(Circuit-1)
|
npn
(Circuit-2)
|
|
IS (p-n saturation current)
|
200E-21A
|
3.26E-16A
|
|
BF (Ideal maximum
forward beta)
|
300
|
100
|
|
NF (Forward current
emission coefficient)
|
1
|
Default
|
|
BR (Ideal maximum
reverse beta)
|
1
|
6
|
|
NR (Reverse current
emission coefficient)
|
1
|
Default
|
|
RB
(Zero-bios (maximum) base resistance)
|
5Ω
|
Default
|
|
RC (Collector ohmic
resistance)
|
1Ω
|
1Ω
|
|
RE (Emitter Ohmic
resistance)
|
---
|
Default
|
|
TF (Ideal forward
transit time)
|
200E-12S
|
25E-12S
|
|
TR (Ideal reverse
transit time)
|
5.00E-12S
|
Default
|
Similarly, at GPDK 180nm technology under Cadence
Virtuoso and Spectre simulation tool, proposed amplifier (Circuit-2) uses
lateral type of BJT (npn with β=100) at driver position and
P-type MOSFET (pmos with VTO=-3.67) transistor at
follower position under Sziklai pair topology (Fig.1(b)) [18]. Moreover,
amplifier with commercial BJT Q2N2222 (β=255.9) at driver position
and P- type MOSFET IRF9140 (VTO=-3.67) at follower position in the
design of Fig.1(a) is referred throughout the present manuscript as Reference
Amplifier [7]. Model parameters used in PSpice User-defined and Cadence
system defined, BJTs and MOSFETs, are listed in Table-1 and Table-2
respectively.
|
Qualitative performance of the proposed
amplifiers is observed with 1mV, 1KHz AC input signal source. However, the Circuit-1
and Circuit-2 produce undistorted output for 1mV-1nV and
0.1mV-10nV range of AC input at 1KHz respectively.
Results And Discussions
A.
Performance Parameters
Comparative
values of the performance parameters of the Proposed amplifiers (Circuit-1 and
Circuit-2) with Commercial transistor based BJT-MOSFET Sziklai pair amplifier
(Reference Amplifier) are listed in Table-3. Circuit-1 describes the PSpice
user-defined BJT-MOSFET based Sziklai pair amplifier whereas Circuit-2 designates
the similar amplifier design with Cadence system defined transistors at GPDK
180nm technology [17]-[18].
Table 3: Qualitative features of Proposed Amplifiers
|
Performance
Parameters
|
Reference
Amplifier
|
Circuit-1
|
Circuit-2
|
|
Maximum Voltage Gain (AVG)
|
347.995
|
389.532 (51.81 dB)
|
164.018 (44.29 dB)
|
|
Unity Gain Bandwidth, (BU)
|
8.1438 MHz
|
2.438 GHz
|
1.7957 GHz
|
|
Maximum Current Gain (AIG)
|
71.519
|
136.570
|
32.775
|
|
Band Width (BW) Corresponds to AVG
|
47.926KHz
|
14.464 MHz
|
11.906 MHz
|
|
Lower Cut-off Frequency (fL)
|
300.787Hz
|
2.5038 KHz
|
1.595 KHz
|
|
Higher Cut-off Frequency (fH)
|
48.227KHz
|
14.467 MHz
|
11.908 MHz
|
|
Power gain (PW in Watt)
|
24888.25
|
53198.385
|
37.3043
|
|
Device Current Gain (AIGD)
|
143.099
|
20701
|
108.636
|
|
Device voltage Gain (AVGD)
|
367.213
|
389.632
|
172.659
|
|
Peak Output Voltage (VRL)
|
336.537mV
|
145.812 mV
|
91.845 mV
|
|
Peak Output Current (IRL)
|
33.654μA
|
14.581μA
|
9.15μA
|
|
Input Current across RSS
|
456.009nA
|
109.714nA
|
271.076nA
|
|
Output Phase Difference θo
|
-163.479O
|
-112.173O
|
-123.245O
|
|
Total Harmonic Distortion THD
|
1.33%
|
2.43%
|
13.608E-6%
|
|
Total Power Consumption, (PC)
|
55.2 mW
|
55.2 mW
|
47.865 mW
|
|
Phase Margin of Voltage Gain (θM)
|
---
|
---
|
52.4371O
|
|
Slew Rate of Output Voltage, (SR)
|
---
|
---
|
5.06 V/μs
|
|
Power Spectral Density, (PD)
|
---
|
---
|
26.705 pV2/Hz
|
|
Transfer Function, (TF)
|
---
|
---
|
164.011 V/V
|
Fig.2. Distribution of Voltage gain on
Frequency Scale
Refer Table-3. Circuit-1(with PSpice
user-defined BJT and MOSFET) carries significantly improved voltage and current
gains, wider bandwidth, wider unity gain bandwidth, higher power consumption,
and higher device voltage and current gain than Reference Amplifier on the cost
of enhanced THD [18]. However, Circuit-2 (at GPDK 180nm technology) generates
wider bandwidth, wider unity gain bandwidth, lower power consumption, and lower
THD, than Reference Amplifier with the compromise over voltage and current
gains. In spite of this, phase difference of the Circuit-1 and Circuit-2 is
found lower than the Reference Amplifier. Moreover, due to typical CE-CD
configuration, nearly 180O phase reversal output is observed for all
the amplifier circuits under consideration [19].
Fig.2 represents the voltage gain of all the
amplifiers with respect to frequency. It is evident from Fig.2 that the Proposed
amplifiers (Circuit-1 and Circuit-2) removes the narrow bandwidth restrictions
of the Reference Amplifier and PNP driven Sziklai pair amplifier [7], [12].
Theses amplifiers are also found free from the poor-response-problem of
small-signal Darlington pair amplifier at higher frequencies. Performance
parameters observed in Table-3 suggests that proposed amplifiers may be used to design Cascadable gain blocks for radio and TV receiver stages and 1KHz-14MHz frequency range power sources [15], [20].
It must be mentioned that the performance of
Circuit-1 strongly depends on Ideal maximum forward beta ‘β’ of user-defined
BJT and Transconductance ‘VTO’ of user-defined MOSFETs in PSpice
[14]. It is found that voltage gain increases with increasing value of β and
becomes saturated at higher β value whereas it decreases with increasing value
of VTO. In addition, meaningful amplification with β is received for
4 ≤ β ≤ 500 whereas for VTO, the range for faithful amplification is
-12 ≤ VTO ≤ +3. The voltage gain of
Circuit-1 increases with rising β because increasing β of the NPN transistors beyond β=4, causes increment in Small Signal transconductance,
gm and collector current IC which perhaps
increases the voltage gain of Circuit-1. In contrast, voltage gain of Circuit-1
decreases with increasing VTO because small signal transconductance
gm decreases when the value of VTO is varied
beyond -12 to +3 range.
Proposed amplifiers
may be used as Low Noise amplifier (LNA) in RF noise measurement system, shown
in Fig.3 [21]. In this set-up, a biconical antenna is fed to bandpass filter
through coaxial wires. The output of this filter is, then, applied to the LNA
(having gain
30dB-60 dB in radio frequency region). The recording of the noise phase and
quadrature data is performed
with the help of spectrum analyser.
Fig.3. Block Diagram of Portable Noise
Measurement System
B. Small Signal AC Analysis
Small-signal AC equivalent circuit of Proposed amplifiers is depicted in
Fig.4(a). With RX=ro||RA and RY=rd||RS
and RB=R1||R2, the reduced version of Fig.4(a)
is sketched in Fig.4(b).
Fig.4(a).
Small-signal AC Equivalent of the Proposed amplifier
Fig.4(b).
Small-signal AC Equivalent of the Proposed amplifier
Based on simulation results of Circuit-1, BJT of the Sziklai unit
consists base-emitter resistance rπ=3.69KΩ, collector-emitter
resistance ro=1x1012Ω, AC current gain factor β=300
whereas MOSFET consists gm=87.7x10-3, drain-source
resistance rd=4.19KΩ [15].
Analysis of Fig.4(b) suggests that
Now voltage across register RX
would be
Hence,
Now, the voltage at input half
(3)
Now by Equation (2) and Equation (4)
Now, by Equation-5, small-signal AC voltage
gain of the proposed amplifier may be expressed as-
Now by Equation (6)
Small-signal AC current gain of the proposed
amplifier may be defined as-
C. Performance Without RA
Performance
of the proposed circuits significantly depends on RA which is to be necessarily included in circuit
structure to retain amplification status [3], [16]. Table-4 summerizes the
status of the parameters of the proposed circuits without RA.
Table 4: Qualitative features of Proposed Amplifiers without RA
|
Performance
Parameters
|
Reference Amplifier
|
Circuit-1
|
Circuit-2
|
|
Maximum Voltage Gain (AVG)
|
0.208
|
0.447
|
0.463
|
|
Maximum Current Gain (AIG)
|
0.009
|
0.120
|
0.076
|
|
Band Width (BW) Corresponds to AVG
|
5.2762 MHz
|
---
|
13.5523 Hz
|
|
Lower Cut-off Frequency (fL)
|
1.0334 MHz
|
---
|
1.0943 Hz
|
|
Higher Cut-off Frequency (fH)
|
6.3096 MHz
|
---
|
14.6466 Hz
|
|
Power gain
(PW in Watt)
|
-27.447
|
-12.705
|
-14.559
|
|
Device
Current Gain (AIGD)
|
0.697
|
15.103
|
0.260
|
|
Device
voltage Gain (AVGD)
|
0.850
|
0.838
|
0.559
|
|
Peak Output Voltage (VRL)
|
4.7249 μA
|
86.874 μA
|
42.302 μV
|
|
Peak Output Current (IRL)
|
472.490 pA
|
8.6892 nA
|
4.2309 nA
|
|
Input Current across RSS
|
2.4944 μA
|
3.2383 μA
|
1.7732 μA
|
|
Output Phase Difference θo
|
-61.640O
|
-83.300O
|
-86.912O
|
|
Total Harmonic Distortion THD
|
1.75%
|
2.47%
|
22.13E-15%
|
|
Total Power Consumption, (PC)
|
23.8 mW
|
26.2 mW
|
38.44 mW
|
|
Slew Rate of Output Voltage, (SR)
|
---
|
---
|
0.347 V/us
|
|
Power Spectral Density, (PD)
|
---
|
---
|
11.036 nV2/Hz
|
|
Transfer Function, (TF)
|
---
|
---
|
463.085 mV/V
|
Refer Table-4. Exclusion of RA causes
reduction in voltage and current gains of the proposed circuits (both AVG
and AIG goes below unity). However, power consumption of both the
proposed circuit reduces in the absence of RA. In addition, THD of
Circuit-1 remains constant whereas for Circuit-2, THD significantly reduces in
the absence of RA. This happens because of the fact that when RA
is detached from the Circuit-1, drain current ID and drain-to-source
voltage VDS of P-type MOSFET decreases which consequently decreases
the voltage and current gain across RL.
Fig.5. Effect of RA on
Voltage Gain
Fig.5 portrays the status of
voltage gain of the proposed circuits as a function of RA.
Permissible range of RA for Reference amplifier for meaningful
amplification is 470Ω<RA<5KΩ
[7]. For Circuit-1 and Circuit-2, Voltage gain increases with increasing value
of RA up to RA=5KΩ, and starts decreasing thereafter. It
is also to note that both the proposed circuits produce distortion at RA
≥15KΩ. Hence, the purposeful range for amplification of Circuit-1, and
Circuit-2 is 1KΩ<RA<15KΩ
and 100Ω<RA<15KΩ
respectively. Observed phenomenon is found in accordance with the deduced Equations (6) and (7).
D. Effect of Baising Resistances
Fig.6. shows the variation of the voltage gain of the
proposed circuits with respect to Souce resistance RS [9].
Fig.6. Effect of RS on
Voltage Gain
The Voltage gain of Circuit-1 elevates with rising
value of RS and reaches its peak value at RS= 15KΩ, and
acquire decreasing trend at higher value of RS. Similarly, in case
of Circuit-2, voltage gain goes on increasing with increasing value of RS. However, this circuit produces distortion at
RS≥25KΩ. Therefore, meaningful range of amplification for Circuit-1
and Circuit-2 is 100Ω≤RS≤100KΩ and 100Ω≤RS≤25KΩ respectively. For Reference Amplifier, AVG increases non-linearly
with source resistance RS and reaches its maximum value at
RS=12KΩ. Thus, purposeful response is received in 1KΩ<RS<12KΩ
range [7]. This generally happens because VDS
of P-type MOSFETs increases with increase in RS (Upto 15KΩ) which in-turns
increases the voltage gain of Circuit-1. However, beyond RS≥15KΩ,
drain to source voltage starts to reduce which reduces the voltage gain.
Fig.7. Effect of RD on
Voltage Gain
Fig.7 shows the variation of the maximum
voltage gain with respect to drain resistance [12]. For Circuit-1, voltage gain
rises with rising value of RD and reaches to its peak value at RD=2KΩ,
thereafter starts decreasing non-linearly. It is also to note that this circuit
produces distortions at RD≥10KΩ. Hence, meaningful range for
amplification with RD for Circuit-1 is 100Ω<RD<10KΩ.
Similarly, voltage gain for Circuit-2 increases non-linearly with increasing
value of RD and attains its maximum value at RD=2KΩ, and
starts decreasing beyond this value. Therefore, meaningful range of RD
for Circuit-2 is 1KΩ≤RD≤40KΩ.
However, for Reference Amplifier, voltage gain decreases with increasing values of drain resistance RD up to 7KΩ and beyond this, amplifier performance becomes poor
[7]. This usually happens because beyond 2KΩ, drain current and drain to source
voltage both starts to decrease which is probably responsible for the deterioration
of voltage gain.
It is also worth mentioning that
the current gain of both the proposed amplifier circuits remain more or less unaffected
with the variation of RSS (Figure not shown) [19]. Instead, it is
found that the voltage gain increases with decreasing value of RSS
for both the proposed circuits. However, for the Reference Amplifier, the
voltage gain AVG receives its maximum value at RSS=10Ω
and minimum value at RSS=80KΩ [7].
E. Effect of DC Supply Voltage
Supply
voltage scaling causes significant impact on the proposed amplifiers’
performance [13]. Fig.8 depicts the variation of voltage gain of the proposed
circuits with respect to Supply Voltage VDC. Respective observations
are recorded up to VDC=50V and beyond this limiting value, the
proposed amplifiers appear with distorted outcome.
Fig.8. Effect of VDC on Voltage Gain
Fruitful range for amplification for
Circuit-1 and Circuit-2 are observed to be 25 Volt ≥VDC ≥ 5 Volt and
35 Volt ≥VDC ≥10 Volt respectively. In addition, voltage gain for
both the proposed amplifiers increases with increasing VDC. However,
Reference Amplifiers produces meaningful amplification with VDC in
the 7-40V range [7].
The factor which is responsible for this
behavior is the significant enhancement in drain to source voltage of the of
P-type MOSFET at increasing values of VDC which in-turns increases
the effective voltage gain of the amplifiers.
F. Phase Variation
Phase
difference of output voltage with respect to frequency is shown in Fig.9 for
the proposed amplifier circuits [18], [22].
Fig.9. Phase
Variation
At 1KHz of operating frequency, Circuit-2 produces higher phase reversal
output (-123.624O) than Circuit-1 (-112.173O). However,
Reference Amplifier generates phase reversal output close to 180O (i.e.
-163.479O) [7]. With the increasing value of frequency, output phase
difference of Circuit-2 initially decreases up to 100 Hz and undergoes sudden
increment at 1KHz frequency, and thereafter increases non-linearly. Similarly,
for Circuit-1, output phase difference changes in zig-zag manner up to 1KHz
frequency, and increases non-linearly beyond this frequency.
G. Temperature Dependent Performance
Table-5(a) and Table-5(b) refers the
performance of the Circuit-1 and Circuit-2 with the variation of temperature in
-20OC ≤ T ≤ +50OC range [23].
Table-5(a):
Temperature effect on Circuit-1 with PSpice User defined Transistors
|
Environmental
Temperature (OC)
|
Voltage Gain (AVG)
|
Current Gain (AIG)
|
Bandwidth (BW)
|
Total Power
Consumption (PC)
|
|
-20
|
457.520
|
138.401
|
16.516 MHz
|
55.0 mW
|
|
-10
|
441.102
|
138.006
|
16.122 MHz
|
55.0 mW
|
|
0
|
425.842
|
137.614
|
15.507 MHz
|
55.1 mW
|
|
10
|
411.617
|
137.225
|
15.340 MHz
|
55.1 mW
|
|
27
|
389.532
|
136.570
|
14.464 MHz
|
55.2 mW
|
|
50
|
336.227
|
135.696
|
13.725 MHz
|
55.3 mW
|
Table-5(b):
Temperature effect on Circuit-2 with Transistors Available at GPDK 180nm Technology
|
Environmental
Temperature (OC)
|
Voltage
Gain (AVG)
|
Current
Gain (AIG)
|
Bandwidth
(BW)
|
Total
Power Consumption (PC)
|
|
-20
|
197.205
|
34.5065
|
10.666
MHz
|
47.2631
mW
|
|
-10
|
189.251
|
34.14
|
10.957
MHz
|
47.389
mW
|
|
0
|
181.818
|
33.77
|
11.212
MHz
|
47.516
mW
|
|
10
|
174.863
|
33.403
|
11.400
MHz
|
47.645
mW
|
|
27
|
164.018
|
32.775
|
11.632
MHz
|
47.865
mW
|
|
50
|
151.006
|
31.9228
|
12.489
MHz
|
48.166
mW
|
Refer Table-5(a) and Table-5(b). Voltage and
Current gain reduce whereas total power consumption increases for both the
proposed amplifier circuits with rising temperature. However, bandwidth
decreases with increasing temperature for Circuit-1, whereas for Circuit-2, it
increases with temperature elevation. In addition, for reference amplifier,
voltage gain goes down but current gain goes high with rising temperature [7].
This usually happens because the temperature elevation accelerates majority
carrier generation process in both the proposed circuits which elevates
collector and drain currents and therefore affects AVG, AIG
and BW [24].
This probably happens because the
Drain-Source resistance of the P-type MOSFET rises with temperature which in
turn reduces the drain current and the voltage and current gains [14].
Similarly, the bandwidth of proposed amplifiers also reduces with raising
temperature for Circuit-1 because the effective drain to source capacitance of P-type
MOSFET decreases with increasing temperature and causes reduction in the
bandwidth [14].
H. Noise Sensitivity
Variation of input and output noise
with temperature (at different operational frequencies) for Circuit-1 and
Circuit-2 is listed in Table-6(a) and Table-6(b)
respectively [24]-[25].
Table 6(a):
Noise Sensitivity of Circuit-1 with PSpice User defined Transistors
|
Temp
(oC)
|
Noises at 10Hz
|
Noises at 1KHz
|
Noises at 100KHz
|
Noises at 100MHz
|
|
Out
Noise
(V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-9
|
Out
Noise
(V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-12
|
Out
Noise
(V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-12
|
Out
Noise
(V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-12
|
|
-20
|
12.140
|
6.1389
|
59.553
|
403.402
|
181.320
|
396.416
|
29.613
|
396.900
|
|
-10
|
12.275
|
6.2081
|
60.985
|
414.776
|
179.849
|
407.823
|
28.534
|
408.293
|
|
0
|
12.409
|
6.2765
|
62.392
|
426.108
|
178.868
|
419.185
|
27.539
|
419.643
|
|
10
|
12.541
|
6.3442
|
63.773
|
437.401
|
177.169
|
430.506
|
26.617
|
430.952
|
|
20
|
12.672
|
6.4113
|
65.130
|
448.657
|
175.945
|
441.788
|
25.761
|
442.223
|
|
27
|
12.762
|
6.4579
|
66.066
|
456.515
|
175.130
|
449.664
|
25.197
|
450.091
|
|
50
|
13.057
|
6.6087
|
69.057
|
482.226
|
172.666
|
475.425
|
23.525
|
475.830
|
Table 6(b):
Noise Sensitivity of Circuit-2 with Transistors Available at GPDK 180nm
Technology
|
Temp
(oC)
|
Noises at 10Hz
|
Noises at 1KHz
|
Noises at 100KHz
|
Noises at 100MHz
|
|
Out
Noise (V/
Hz) x 10-6
|
In
Noise
(V/
Hz) x 10-9
|
Out
Noise
(V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-9
|
Out
Noise (V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-9
|
Out
Noise
(V/
Hz) x 10-9
|
In
Noise
(V/
Hz) x 10-9
|
|
-20
|
2.0498
|
0.001522
|
269.22
|
2.8071
|
249.55
|
1.2655
|
26.078
|
1.2579
|
|
-10
|
2.133
|
0.001589
|
280.83
|
2.9656
|
244.65
|
1.2927
|
26.169
|
1.2838
|
|
0
|
2.2208
|
0.001662
|
293.08
|
3.1355
|
239.96
|
1.3198
|
26.243
|
1.3093
|
|
10
|
2.3074
|
0.001798
|
305.32
|
3.31034
|
235.49
|
1.3467
|
26.302
|
1.3345
|
|
20
|
2.396
|
0.001811
|
317.926
|
3.4941
|
231.211
|
1.3734
|
26.349
|
1.3593
|
|
27
|
2.459
|
0.001866
|
326.365
|
3.6283
|
228.329
|
1.3921
|
26.374
|
1.3765
|
|
50
|
2.676
|
0.002052
|
357.955
|
4.1040
|
219.445
|
1.4532
|
26.422
|
1.4320
|
Refer Table-6(a) and Table-6(b). Input noise of Circuit-1 and Circuit-2
at all the mentioned frequencies increases with increasing temperature.
However, output noise of Circuit-1 at 10Hz, and 1KHz frequencies increases with
increasing temperature whereas at 100KHz and 100MHz frequencies, it reduces
with temperature elevation. Similarly, for Circuit-2, output noise at 10Hz,
1KHz, and 100MHz increases whereas at 100KHz, it decreases with rising
temperature.
I. Layout and Post layout Simulation
Table
7: Pre-Layout and Post-Layout Simulation Results of Circuit-2
|
Parameters
|
Pre layout
|
Post layout
|
Percentage variation
|
|
Voltage Gain, (AVG)
|
164.018
|
142.23
|
14.22%
|
|
Current Gain, (AIG)
|
32.775
|
23.12
|
34.54%
|
|
Total Power Consumption, (PC)
|
47.465
mW
|
40.22
mW
|
16.52%
|
Layout of Circuit-2, depicted in Fig.10, is designed at
GPDK 180nm technology using Layout XL editor tool in Cadence Virtuoso and
Spectre simulation [26]-[27]. Layout
of Circuit-2 covers an area of 96.3898µm2
having 11.32µm length and 8.515µm breadth. In this layout, first block
replicates NPN transistor whereas second block shows P-type MOS transistor. The
P-type MOS transistor, having dimensions of 0.68μm x 0.955μm, is taken as
integrated because source and substrate terminal is connected to same wire.
Yellow and Violet wires represent
poly and metal wires respectively. The λ rule for connecting the wires is well
abided to avoid wire mismatch and overlapping [19], [28]. Minor % variation
between pre-layout and post-layout simulation, recorded in Table-7,
authenticate the design of Circuit-2 at GPDK 180nm technology.
|
Fig.10. Layout of Circuit-2
J. Monte Carlo Analysis
For 51
samples, Monte Carlo simulation is done to check the robustness of the proposed
amplifiers for Voltage gain, and total power consumption against process and
mismatch [29]. Statistical representation of such variations are shown in Fig.11,
and Fig.12 respectively.
Fig.11. Monte Carlo Simulation of Voltage gain
Fig.12. Monte Carlo Simulation of Total
Power Consumption
Table 8: Monte Carlo
Simulation of Circuit-2
|
Parameters
|
Mean (µ)
|
Deviation (
)
|
|
Voltage Gain
|
72.2460
|
71.7907
|
|
Total Power Consumption
|
48.4902mW
|
445.118μW
|
Table-8 shows the deviation of the Voltage gain, and total power
consumption from their mean value. It is found that the measured
parameters shows small deviation against their respective mean value which
shows the robustness against the process and mismatch variations [11].
|
K. Process Corner Simulation
Fig. 13, and Fig.14 represents the five-corner simulation (TT, FF, SS,
FNSP, SNFP) of voltage gain and total power consumption at 27oC,
where TT, FF, SS, FNSP, and SNFP stands for typical-typical, fast-fast,
slow-slow, fast-NPN-slow-PMOS, and slow-NPN-fast-PMOS [30].
Fig.13. Corner Analysis of Voltage gain
Fig.14. Corner Analysis of Total Power
Consumption
The SS corner simulation produces
the largest Voltage gain, measuring 195.21 whereas the SS consumes lowest power
consumption, measuring 47.1mW. In addition, all the corners have almost similar
values of parameters at different corners which shows that the Circuit-2 is
insensitive against the process and mismatch variations.
L. Tuning Performance of Circuit-1
The Load capacitor CL
is included in Circuit-1 as an essential circuit component which help the
amplifier to tune at a specific frequency to match the frequency of a
particular channel [30]. It is achieved under two conditions, first by keeping
the CL constant at 1nF and varying CD (Table-9(a)), and
second by varying CL and keeping CD constant at 10μF (Table-9(b)).
Table-9(a): Performance
parameters at Varying CD and keeping CL constant
|
CD (CL=1nF)
|
AVG
|
AIG
|
FL
|
FH
|
BW
|
|
1μF
|
288.243
|
132.514
|
18.476 KHz
|
19.602 MHz
|
19.599 MHz
|
|
10μF
|
389.532
|
136.570
|
2.5038 KHz
|
14.467 MHz
|
14.464 MHz
|
|
100μF
|
403.718
|
136.990
|
263.770 Hz
|
14.030 MHz
|
14.029 MHz
|
|
1mF
|
405.194
|
137.031
|
30.758 Hz
|
13.943 MHz
|
13.942 MHz
|
|
10mF
|
405.342
|
137.036
|
7.6096 Hz
|
13.935 MHz
|
13.934 MHz
|
|
100mF
|
405.357
|
137.036
|
5.1210 Hz
|
13.936 MHz
|
13.935 MHz
|
Table-9(b): Performance
parameters at Varying CL and keeping CD constant
|
CL (CD=10μF)
|
AVG
|
AIG
|
FL
|
FH
|
BW
|
|
1nF
|
389.532
|
136.570
|
2.5038 KHz
|
14.467 MHz
|
14.464 MHz
|
|
10nF
|
288.059
|
132.392
|
1.8522 KHz
|
1.9548 MHz
|
1.9544 MHz
|
|
100nF
|
79.903
|
101.340
|
515.633 Hz
|
703.398 KHz
|
702.883 KHz
|
|
1μF
|
9.714
|
30.293
|
65.026 Hz
|
582.178 KHz
|
582.113 KHz
|
Refer Table-9(a) and Table-9(b).
Tuning with CD is achieved in 10μF-100mF range, while the range of
tuning with CL is 1nF-10nF. Fig. 15 shows two combinations of CL
and CD which tune at a specific frequency with varying bandwidth.
Fig.15. Tuning Performance of Circuit-1
For Reference Amplifier, bandwidth
varies with drain capacitor CD and load capacitor CL in
10μF–100μF and 0.001μF–.001F variation range respectively. However, in
Circuit-2, CL is not connected to the circuit structure hence no
such type of tuning effect is recorded [9], [14].
M. Performance Summary and Comparison
Table-10 compiles the comparison of proposed
amplifiers, Circuit-1 and Circuit-2, with the BJT-MOSFET hybrid unit based small
signal Darlington pair amplifier and small signal amplifier with BJT-MOSFET hybrid
unit under Sziklai pair topology [7], [11].
Table 10: Performance Summary and Comparison
|
References ®
|
Shukla et al. [7]
|
Pratima et al. [11]
|
This Work
|
|
Design
Strategy for small signal amplifiers
|
BJT-MOSFET in Sziklai pair Topology
|
BJT-MOSFET in Darlington pair Topology
|
BJT-MOSFET in Sziklai pair Topology
(Circuit-1)
|
BJT-MOSFET
in Sziklai pair Topology
(Circuit-2)
|
|
PSpice Commercial Transistors
|
At GPDK 180nm Technology
|
PSpice User Defined Transistor
|
At GPDK 180nm Technology
|
|
Year
|
2015
|
2023
|
2022
|
2022
|
|
Technology (nm)
|
--
|
180
|
--
|
180
|
|
Supply Voltage (Volt)
|
+18
|
+15
|
+18
|
+18
|
|
Maximum Voltage Gain (AVG)
|
347.995
|
281.331
|
389.532
|
164.018
|
|
Maximum Current Gain (AIG)
|
71.519
|
33.698
|
136.570
|
32.775
|
|
Band Width (BW)
|
47.926KHz
|
22.562 MHz
|
14.464 MHz
|
11.906 MHz
|
|
Power gain
(PW in Watt)
|
24888.25
|
9480.292
|
53198.385
|
37.3043
|
|
Input Referred Noise, IRN
|
---
|
23.167 nV/√Hz
|
456.515 pV/√Hz
|
3.6283nV/√Hz
|
|
Output Referred Noise, ORN
|
---
|
5.5523 pV/√Hz
|
66.066 nV/√Hz
|
326.365 nV/√Hz
|
|
Total Harmonic Distortion THD
|
1.33%
|
5.296E-6%
|
2.43%
|
13.608E-6%
|
|
Total Power Consumption, (PC)
|
55.2 mW
|
165.061 mW
|
55.2 mW
|
47.865 mW
|
Refer Table-10. Proposed
Circuit-1 generates higher current and voltage gains, higher power gain, and
lower THD than the amplifiers of Shukla et. al. [7] and Pratima et.
al., [11]. However, Proposed Circuit-2 holds wide bandwidth, and higher
power gain than [7], lower input noise than [11], lower THD than [7], lower
power consumption than [7], [11].
In addition, Circuit-2 emerges
with lowest power consumption, and lowest THD whereas Circuit-1 carries highest
voltage and current gain, lowest Input referred noise, and higher power gain.
Conclusion
Two Circuit designs of small signal Sziklai pair amplifier with
BJT-MOSFET hybrid unit are studied and analysed under the purview of present
manuscript.
The proposed circuits commonly use CE-CD configuration; therefore, the
entire unit produces 180O phase shift in the output. In spite of
this, proposed circuits provide solution to the poor frequency response of
Darlington pairs at higher frequencies and narrow bandwidth limitations of
small-signal PNP Sziklai pair amplifier. Additionally, these circuits also
remove the narrow band problem of Reference Amplifier with comparatively high
voltage and current gain as well as wider bandwidth.
The value of the additional biasing resistance RA, as an
essential circuit component, must be taken in the range 1KΩ<RA<15KΩ and 100Ω<RA<15KΩ for Circuit-1 and Circuit-2 respectively. In addition,
for Circuit-1, meaningful amplification with ideal maximum forward beta β is received for the range 4 ≤ β ≤ 500 whereas for transconductance VTO,
the range for faithful amplification is -12 ≤ VTO ≤ +3. It is also
to be noted that load capacitor CL acts as an essential circuit
element for Circuit-1 to provide tuning whereas Circuit-2 does not require any
load capacitor. Tuning performance of Circuit-1 can be availed with CL=1nF,
CD=1μF and CL=10nF, CD=10μF. Comparison of
proposed circuits with the similar small signal amplifiers design reported
recently suggests voltage gains, current gains, input referred noise,
bandwidth, THD, power consumption, and power gain as their prominent features.
Analysis of the proposed circuit suggests its
use as Low Noise Amplifier in Street-scale Mapping of Radio frequency Noise at
VHF and UHF region.
References
[1]
S.N. Shukla,
B. Pandey,
“Two-Stage Small-Signal Amplifier with
Darlington and Sziklai pairs”, Proceedings of 11th IEEE International
Conference on Semiconductor Electronics (IEEE-Xplore), pp.13-16, 2014, DOI:
10.1109/SMELEC.2014.6920783,
[2]
David A. Hodges, “Darlington’s Contribution
to Transistor Circuit Design”, IEEE Transactions on Circuits and Systems-I,
Vol.46, No.-1, 1999, pp.102-104 DOI: 10.1109/81.739258
[3]
B. Pandey, S. Srivastava, S. N. Tiwari, J.
Singh and S. N. Shukla, “Qualitative
Analysis of Small Signal Modified Sziklai Pair Amplifier”, Indian Journal of Pure and Applied Physics,
50, 2012, pp.272-276, URL: http://nopr.niscpr.res.in/handle/123456789/13784
[4]
Rod Elliott, “Compound pair Vs Darlington
pair”, http://sound.westhost.com/articles/cmpd-vs-darl.htm,
January 6, 2011
[5]
George C. Sziklai, “Push-pull complementary
type transistor amplifier”, U.S. Patent 2,762,870,
September 11, 1956
[6]
Olaleye A. Aina, Eric
A. Martin, “Field Effect Transistor-Bipolar Transistor Darlington pair”, US
Patent 5187110, Feb. 16, 1993.
[7]
S.N. Shukla, "Novel
circuit of small signal amplifier with BJT-MOSFET hybrid unit in sziklai pair
topology," 2015 Radio and Antenna Days of the Indian Ocean (RADIO),
pp. 1-2, 2015 DOI: 10.1109/RADIO.2015.7323394.
[8]
S.S. Arshad, G.
Srivastava, and S.N. Shukla “Sziklai pair and Darlington pair RTL inverters for
High Drive current Applications”, Journal of Rajasthan Academy of Physical
Sciences, Vol. 22, No. (1&2), 2023, pp.22-40.
[9]
S.N.
Shukla, S. Srivastava, “Qualitative and Tuning Performance of MOSFET
Based Small-Signal Darlington pair Amplifiers”, International Journal of Enhanced Research in Science Technology and
Engineering, Vol.1,
Issue 2, pp.1-6, 2013, DOI:
http://www.doi.org/ste/13.01.08
[10] Hassan Jassim Motlak, “Design of Low Voltage, Low Power
(IF) Amplifier Based-On MOSFET Darlington Configuration”, Circuits and
Systems, Vol. 4, pp. 269-275, 2013, DOI: 10.4236/cs.2013.43036.
[11] P. Soni, G. Srivastava, S.S. Arshad, A.K. Sharma and
S.N. Shukla, “High Gain Narrow Band Small Signal Hybrid unit Darlington pair Amplifier”,
Journal of Theoretical and Applied Information Technology, Vol. 101, No.
19, pp.6102-6113, 2023, https://www.jatit.org/volumes/Vol101No19/22Vol101No19.pdf
[12] Shou-Hsien Weng, Hong-Yeh Chang, Chau-Ching Chiong,
“Design of a 0.5-30 GHz Darlington Amplifier for Microwave Broadband
Applications”, 2010 IEEE MTT-S International Microwave Symposium, pp.
137-140, 2010 DOI: 10.1109/MWSYM.2010.5518254.
[13] S.N. Shukla, S.S. Arshad,
G. Srivastava, A. K. Sharma, S.K. Pandey, S.N. Singh, and P. Soni, “Study of Novel Small-Signal JFET Amplifiers in Sziklai
pair Topology”, Proceedings of 10th International Conference of
Signal Processing and Integrated Networks, pp. 93-98, 2023, DOI:
10.1109/SPIN57001.2023.101117320.
[14]
Motayed, S.N.
Mohammad, “Tuned Performance of
Small-signal BJT Darlington pair”, Solid
State Electronics, Vol. 45, No. 2, pp. 369-371, 2001, DOI: 10.1016/S0038-1101(00)00247-1
[15] S.S. Arshad, S.N. Shukla, A.K. Sharma, and G.
Srivastava, “Darlington Pair based Small-Signal Amplifier under Triple
Transistor Topology”, Journal of International Academy of Physical Sciences,
Vol. 26, No. 4, pp. 427-442, 2022, http://www.iaps.org.in/journal/index.php/journaliaps/article/view/960/821
[16] M.
H. Rashid, “Introduction to PSpice
Using OrCAD for Circuits and Electronics”, Pearson Education, 3rd Ed., p-255, 2004
[17] Antonio J. Lopez Martin, “Tutorial: Cadence Design
Environment”, Klipsch School of Electrical and Computer Engineering, New
Mexico State University, October 2002
[18] S.N. Shukla, S.S. Arshad, and G. Srivastava, “NPN
Sziklai pair small-signal amplifier for high gain low noise submicron voltage
recorder,” International Journal of Power Electronics and Drive Systems
(IJPEDS), Vol. 13, No. 1, pp. 11-22, 2022, DOI:
10.11591/ijpeds.v13.i1.pp11-22.
[19] S.N. Shukla, S.S. Arshad, P. Soni, G. Srivastava, A.
Dixit and A. Kumar “Small Signal Sziklai Pair Based Tuned Amplifier with Low
Power High Gain”, Journal of International Academy of Physical Sciences,
Vol. 26, No. 3, pp. 281-306, 2022, http://www.iaps.org.in/journal/index.php/journaliaps/article/view/944/815
[20] M.H. Ali, Aliyu Sisa Aminu, “Analysis of Darlington pair in Distributed Amplifier Circuit”, IOSR
Journal of Electrical and Electronics Engineering,
Vol.10, Issue 2, Ver. I, pp.77-80, 2015, DOI: 10.9790/1676-10217780
[21] D. J. Breton, C. A. Haedrich,
M. J. Kamrath, and D. K. Wilson, “Street-scale Mapping of Urban
Radiofrequency Noise at VHF and UHF”, Radio Science, Vol. 54, No. 1, pp. 1-20, 2019,
DOI: 10.1029/2019RS006893
[22] Anna Richelli, Luigi Colalongo, Zsolt Kovacs-Vajna,
Giacomo Calvetti, Davide Ferrari, Marco Finanzini, Simone Pinetti, Enrico
Prevosti, Jacopo Savoldelli and Stefano Scarlassara, “A Survey of Low Voltage
and Low Power Amplifier Topologies”, Journal of Low Power Electronics and
Applications, Vol. 8, p. 22, 2018, DOI: 10.3390/jlpea8030022.
[23] Sanna Mairaj, Suhaib Ahmed and Vipan Kakkar, “A Survey
on Emerging Technologies and Architectures of Low Power Preamplifiers for
Biomedical Applications”, International Journal of Nanoelectronics and
Materials, Vol. 11, No. 2, pp. 167-178, 2018.
[24] Ming-Da Tsai, Chin-Shen Lin, Student, “Broad-Band MMICs
Based on Modified Loss-Compensation Method Using 0.35-μm SiGe BiCMOS
Technology”, IEEE Transactions on Microwave Theory and Techniques, Vol.
53, No. 2, pp. 496-505, 2005, DOI: 10.1109/TMTT.2004.840766.
[25] Jongsoo Lee, John D. Cressler, “Analysis and Design of
an Ultra-Wideband Low-Noise Amplifier Using Resistive Feedback in SiGe HBT
Technology”, IEEE Transactions on Microwave Theory and Techniques, Vol.
54, No. 3, pp. 1262-1268, 2006, DOI: 10.1109/TMTT.2005.864097.
[26] Maria Drakaki, Stylianos Siskos, Alkis Hatzopoulos, “A
0.5–20 GHz bandwidth enhanced distributed amplifier”, Microelectronic
Engineering, Vol. 90, pp. 26-28, 2012, DOI:
10.1016/j.mee.2011.04.031.
[27] Pokryszka,
Piotr, Mateusz Wosko, Wojciech Kijaszek, and Regina Paszkiewicz. "High
performance optical shutter design with scalable aperture”, Bulletin of the
Polish Academy of Sciences: Technical Sciences, pp.1-6, 2021, DOI:
10.24425/bpasts.2021.138236
[28] Kevin W. Kobayashi, Reza Esfandiari, Majid E. Hafizi,
Dwight C. Streit, Aaron K. Oki, Liem T. Tran, Donald K. Umemoto, and Mike E.
Kim, “GaAs HBT Wideband Matrix Distributed and Darlington Feedback Amplifiers
to 24 GHz”, IEEE Transactions on Microwave Theory and Techniques, Vol.
39, No. 12, pp.2001-2009, 1997, DOI: 10.1109/22.106539.
[29] Arief Wisnu Wardhana, Yogi Ramadhani, Priswanto, “Design
and Simulation of Multistages Common-Emitter, Common-Collector, AC Voltage
Amplifier”, Journal Nasional Teknik Elektro, Vol. 11, No. 2, pp. 85-96, 2022,
DOI: 10.25077/jnte.v11n2.1009.2022.
[30] S.N. Shukla, S.S. Arshad and G. Srivastava, “Novel
Complementary Sziklai pair based high gain low noise small signal amplifier”, International
Journal of Power Electronics and Drive Systems, Vol. 14, No. 4, pp.
2283-2292, 2023, DOI: 10.11591/ijpeds.v14.i4.pp2283-2292.