6.776 High Speed Communication Circuits and Systems Lecture 12 Low Noise Amplifiers

advertisement
6.776
High Speed Communication Circuits and Systems
Lecture 12
Low Noise Amplifiers
Massachusetts Institute of Technology
March 15, 2005
Copyright © 2005 by Hae-Seung Lee and Michael H.
Perrott
Narrowband LNA Design for Wireless Systems
Zin
From Antenna
and Bandpass
Filter
ƒ
PC board
trace
To Mixer
Package
Interface
Zo
LNA
Design Issues
- Noise Figure – impacts receiver sensitivity
- Linearity (IIP3) – impacts receiver blocking performance
- Gain – high gain reduces impact of noise from components
that follow the LNA (such as the mixer)
- Power match – want Z = Z (usually = 50 Ohms)
- Power – want low power dissipation
- Bandwidth – need to pass the entire RF band for the
intended radio application (i.e., all of the relevant channels)
- Sensitivity to process/temp variations – need to make it
in
o
manufacturable in high volume
H.-S. Lee & M.H. Perrott
MIT OCW
Our Focus in This Lecture
Zin
From Antenna
and Bandpass
Filter
ƒ
ƒ
ƒ
ƒ
ƒ
PC board
trace
Zo
To Mixer
Package
Interface
LNA
Designing for low Noise Figure
Achieving a good power match
Hints at getting good IIP3
Impact of power dissipation on design
Tradeoff in gain versus bandwidth
H.-S. Lee & M.H. Perrott
MIT OCW
Direct Input Termination of CS Amplifier
ƒ
ƒ
ƒ
ƒ
For power match, terminate the input of CS amplifier
Voltage gain is halved
Termination resistor adds thermal noise
More appropriate for broadband than narrowband
(maybe)
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor (Low Frequency)
ƒ
Noise factor at low frequencies (thus ignoring gate
current noise) looking at short-circuit output current
ƒ
ƒ
Noise factor too high even at low frequencies.
Generally not an acceptable circuit for LNA
H.-S. Lee & M.H. Perrott
MIT OCW
Common-Gate Amplifier
Offers the possibility of power match without an explicit
termination resistor
For power match, make
At low frequencies, it can be shown
Better than the CS, but noise factor degrades
significantly at high frequencies
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor Analysis of CG Amp (Low Freq.)
ƒ
Equivalent Circuit
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor Calculation
ƒ
Noise factor at low frequencies (thus ignoring gate current
noise) looking at short-circuit output current
First calculate the output noise due to the source resistance
only (en) using voltage divider:
The next step is finding the total output noise due to en and
id. We can use superposition as before.
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor Calculation, Continued
First, let’s compute output noise due to id only. We’ll need to
use the instantaneous value of id for now:
The instantaneous value of the
output noise due to id
The mean-square value of
the output noise due to id
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor of CG Amplifier
For power match, 1/gm=Rs
This may be o.k. for broadband amplifiers. For
narrowband amplifiers, we can improve the noise
factor considerably while maintaining power match:
inductors!
H.-S. Lee & M.H. Perrott
MIT OCW
Inductor Degenerated CS Amp
Source
inout
Rs
ens
Lg
Zgs
Cgs
gmvgs
indg
Ldeg
Source
Rs
vgs
Iout
Lg
M1
vin
Ldeg
Vbias
ƒ
Same as amp in Lecture 10 except for inductor
degeneration
- Note that noise analysis in Tom Lee’s book does not include
inductor degeneration (i.e., Table 12.1 (11.1))
H.-S. Lee & M.H. Perrott
MIT OCW
Recall Small Signal Model for Noise Calculations
iout
D
Iout
G
D
G
S
Zg
Zg
Zgs
vgs
Cgs
gmvgs
indg
Zdeg
S
Zdeg
H.-S. Lee & M.H. Perrott
MIT OCW
Key Assumption: Design for Power Match
Zin
Iout
Lg
vin
Rs
Vbias
M1
Ldeg
Real =RT
ƒ
Input impedance (from Lec 9)
ƒ
Set to achieve pure resistance = Rs at frequency ωo
Adding Lg gives additional degree of freedom to set ωo and RT
independently
H.-S. Lee & M.H. Perrott
MIT OCW
Key Assumption: Design for Power Match
Zin
Iout
Lg
vin
Vbias
ƒ
Rs
M1
Ldeg
The total impedance of the RLC circuit (including the
source resistance)
Thus Q is then
H.-S. Lee & M.H. Perrott
Rs
MIT OCW
Process and Topology Parameters for Noise Calculation
Source
inout
Rs
ens
Lg
Zgs
vgs
Cgs
gmvgs
indg
Ldeg
ƒ
Process parameters
ƒ
Circuit topology parameters Zg and Zdeg
- For 0.18µ CMOS, we will assume the following
H.-S. Lee & M.H. Perrott
MIT OCW
Calculation of Zgs
Source
inout
Rs
ens
Lg
Zgs
vgs
Cgs
gmvgs
indg
Ldeg
H.-S. Lee & M.H. Perrott
MIT OCW
Calculation of η
Source
inout
Rs
ens
Lg
Zgs
vgs
Cgs
gmvgs
indg
Ldeg
= Rs
H.-S. Lee & M.H. Perrott
MIT OCW
Calculation of Zgsw
ƒ
By definition
ƒ
Calculation
H.-S. Lee & M.H. Perrott
MIT OCW
Calculation of Output Current Noise
ƒ
Step 3: Plug in values to noise expression for indg
H.-S. Lee & M.H. Perrott
MIT OCW
Compare Noise With and Without Inductor Degeneration
Source
inout
Rs
ens
Lg
Zgs
vgs
Cgs
gmvgs
indg
Ldeg
ƒ
From Lecture 10, we derived for Ldeg = 0, ωo2 = 1/(LgCgs)
ƒ
We now have for (gm/Cgs)Ldeg = Rs, ωo2 = 1/((Lg + Ldeg)Cgs)
H.-S. Lee & M.H. Perrott
MIT OCW
Derive Noise Factor for Inductor Degenerated Amp
Source
inout
Rs
ens
Lg
Zgs
vgs
Cgs
gmvgs
indg
Ldeg
ƒ
Recall the alternate expression for Noise Factor derived in
Lecture 11
ƒ
We now know the output noise due to the transistor noise
- We need to determine the output noise due to the source
resistance
H.-S. Lee & M.H. Perrott
MIT OCW
Output Noise Due to Source Resistance
Source
inout
Rs
ens
Lg
Zin
Zgs
vgs
Cgs
gmvgs
indg
Ldeg
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor for Inductor Degenerated Amplifier
Noise Factor scaling coefficient
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor Scaling Coefficient Versus Q
Noise Factor Scaling Coefficient Versus Q for 0.18 µ NMOS Device
7
Achievable values as
a function of Q under
the constraints that
1
= wo
(Lg+Ldeg)Cgs
gm
L = Rs
Cgs deg
Noise Factor Scaling Coefficient
6.5
6
5.5
5
4.5
c = -j0
4
3.5
3
2.5
1
Note:
1
Q=
2RswoCgs
c = -j0.55
c = -j1
1.5
H.-S. Lee & M.H. Perrott
2
2.5
3
Q
3.5
4
4.5
5
MIT OCW
Noise Factor Comparison
ƒ
Inductor Degenerated CS
Noise Factor scaling coefficient
ƒ
Non degenerated (From Lecture 10)
Noise Factor scaling coefficient
H.-S. Lee & M.H. Perrott
MIT OCW
Noise Factor Scaling Coefficient Comparison
Noise Factor Scaling Coefficient Versus Q for 0.18 µ NMOS Device
Noise Factor Scaling Coefficient Versus Q for 0.18 µ NMOS Device
7
8
Achievable values as
a function of Q under
the constraints that
1
= wo
(Lg+Ldeg)Cgs
gm
L = Rs
Cgs deg
6
5.5
5
7
Noise Factor Scaling Coefficient
Noise Factor Scaling Coefficient
6.5
4.5
c = -j0
4
3.5
3
2.5
1
Note:
1
Q=
2RswoCgs
c = -j0.55
6
c = -j0
c = -j0.55
5
Achievable values as
a function of Q under
the constraint that
1
= wo
LgCgs
c = -j1
c = -j0
4
c = -j0.55
3
Minimum across
all values of Q and
1
Note: curves
meet if we
approximate
Q2+1 Q2
2
1
1.5
2
2.5
3
Q
3.5
4
With degeneration
H.-S. Lee & M.H. Perrott
4.5
5
0
1
2
3
c = -j1
LgCgs
c = -j1
4
5
6
7
8
9
10
Q
Without degeneration
MIT OCW
Achievable Noise Figure in 0.18µ with Power Match
ƒ
Suppose we desire to build a narrowband LNA with
center frequency of 1.8 GHz in 0.18µ CMOS (c=-j0.55)
- From Hspice – at V
gs =
1 V with NMOS (W=1.8µ, L=0.18µ)
ƒ measured gm =871 µS, Cgs = 2.9 fF
- Looking at previous curve, with Q ≈ 1.4 we achieve a
Noise Factor scaling coefficient ≈ 3.25
H.-S. Lee & M.H. Perrott
MIT OCW
Component Values for Minimum NF with Power Match
ƒ
Assume Rs = 50 Ohms, Q = 1.4, fo = 1.8 GHz, ft = 47.8
GHz
-C
-L
-L
gs
calculated as
deg
g
calculated as
calculated as
H.-S. Lee & M.H. Perrott
MIT OCW
Id
Vgs
M1
1.8µ
W
=
L
0.18µ
gm and gdo (milliAmps/Volts) and Vgs (Volts)
Have We Chosen a Good Bias Point? (Vgs = 1V)
4.5
Vgs , gm , and gdo versus Current Density for 0.18µ NMOS
Chosen bias
point is Vgs = 1 V
4
3.5
3
2.5
Lower power Higher power
Higher ft
Lower ft
Lower IIP3
Higher IIP3
gdo
gdo
Lower g
Higher g
m
m
gdo
2
Vgs
1.5
gm
1
0.5
0
0
100
200
300
400
500
600
700
Current Density (microAmps/micron)
ƒ
Note: IIP3 is also a function of Q
H.-S. Lee & M.H. Perrott
MIT OCW
Calculation of Bias Current for Example Design
ƒ
Calculate current density from previous plot
ƒ
Calculate W from HSpice simulation (assume L=0.18 µm)
ƒ
- Could also compute this based on C
ox
value
Calculate bias current
- Problem: this is not low power!!
H.-S. Lee & M.H. Perrott
MIT OCW
Download