Transcript LNA Technologies and Topologies

SKA LNA Technologies and
Topologies
Saswata Bhaumik
PhD Student
Dr Danielle George
The University of Manchester
SKADS 06.11.09
LNA Technologies and Topologies
Overview
• LNA design work carried out around the World
• Purpose of presentation is to bring together this
work
• Contributions from most major players within SKA
• Highlight some of the key research and
development
• ONLY MEASURED DATA INCLUDED
SKADS 06.11.09
LNA Technologies and Topologies
Contributors
•
•
•
•
•
•
ASTRON & GIF
CSIRO
California Institute of Technology - TDP
France - OPAR
University of Calgary
University of Manchester
SKADS 06.11.09
LNA Technologies and Topologies
ASTRON - APERTIF
1e iteration LNA
Currently installed LNA
Frequency
(MHz)
Fmin
(dB)
Rn
(Ohm)
Γopt
mag.
Γopt
phase
Fmin (dB)
Rn
(Ohm)
Γopt mag.
Γopt
phase
1000
0.99
4
0.1750
-123.17
0.52
3
0.0978
-7.3361
1200
0.99
4.8
0.1981
-94.80
0.52
3
0.1230
62.4014
1400
0.98
6.5
0.2245
-65.90
0.5
2.5
0.1372
122.7336
1600
0.94
8.5
0.2791
-29.66
0.52
2
0.1866
158.6415
SKADS 06.11.09
LNA Technologies and Topologies
ASTRON - APERTIF
1e iteration LNA
Currently installed LNA
Gain
40dB
42dB
NF50
[email protected]
35K across band
Linearity
OIP3: 25dB
OIP3: 25dB
Transistors
COTS: stage1: ATF54143,
stages 2and 3: MGA53543
COTS: same type of components
SKADS 06.11.09
LNA Technologies and Topologies
ASKAP - CSIRO
•
•
•
•
COTS components - ATF-35143 PHEMTs
700-1800 MHz
300 Ω input impedance, differential
50 Ω output
Noise Calibration Procedure
• Te of selected LNAs with liquid nitrogen.
• Y-factor of the LNAs on noise test fixture.
• Calculate Thot and ENR of the noise test fixture.
• Calibration valid for same “series” LNAs .
Reference-Acquired with permission from CSIRO
SKADS 06.11.09
LNA Technologies and Topologies
University of Calgary
LNA Details
•
•
•
•
•
90nm CMOS
0.8 to1.4GHz
Differential 50 Ω system
25K NF(50)
15dB gain
Differential Noise Measurement
•
•
•
•
Four possible combinations of two input ports
and two output ports
Single-ended F31, F32, F41, and F42 and G31,
G32, G41 and G42 with a noise figure analyzer.
Unused ports terminated in 50.
Determination of input-referred available noise
powers.
Differential noise figure determination through
theoretical calculation.
Reference- Acquired with permission from Dr. Leonid Belostotski. Copyright
belongs to IEEE.
SKADS 06.11.09
LNA Technologies and Topologies
France - OPAR
LNA Details:
• QuBiC4G: 0.25μm SiGe
• 2 stages Differential amplifier
• 100 Ohms impedance
• 72.6mW power dissipation
• -1 dBm OP1dB
Reference- SKADS-wiki
SKADS 06.11.09
LNA Technologies and Topologies
France - OPAR
LNA Details:
• QuBiC4X: SiGe:C
• Single ended 50 Ohm
• 300Mhz-1Ghz
• 69.3 mW power consumption
• -2dBm OP1dB
Reference- SKADS-wiki
SKADS 06.11.09
LNA Technologies and Topologies
University of Manchester
InP Development
20
10
0
-10
-20
-30
-40
-50
-60
0
S(2,1)
S-Parameters (dB)
S-Parameters (dB)
1μm gate-length InP process
LNA measured data
• Gain: >10dB
• Frequency: 0.2-2GHz
• Power dissipation: 45mW
Simulated
Measured
S(1,2)
S(1,1)
-5
-10
S(2,2)
-15
Simulated
Measured
-20
-25
0
1
2
3
4
5
0
1
Frequency (GHz)
Reference- SKADS-wiki
SKADS 06.11.09
LNA Technologies and Topologies
2
3
Frequency (GHz)
4
5
Caltech - TDP
• Models developed at 12K
• Resistive feedback highly
effective with the SiGe
technology for broadband
design.
• Noise Measure is more
important than Tmin.
• New noise measurement
procedure being experimented
with.
SKADS 06.11.09
CIT
Design
Type
Frequency
Range
Noise
(June’09)
NEUT
SiGe differential
feedback LNA.
IBM 8HP process
0.1 to 15
<10K
@ 15K
WBAL2
InP differential
LNA
0.5 to 11
Meas
<18K
@ 60K
P5T4
SiGe differential
feedback LNA.
IBM 8HP process
0.5 to 4
<10K
@ 60K
WBA13
InP LNA
0.5 to 12
Meas
<10K
LNA Technologies and Topologies
Caltech - TDP
LNA Details
Measured data of 2 LNAs with ST and commercially available NXP transistors at 1.4GHz.
Reference-“Matched Wideband LNAs for radio astronomy” ,S. Weinreb, J. Bardin, H.Mani, G. Jones. Published in Review of Scientific
Instruments,2009
SKADS 06.11.09
LNA Technologies and Topologies
LNA Details
University of Manchester
GaAs mHEMT
•70nm OMMIC GaAs mHEMT
•Two stage single ended MMIC LNA 50Ohm system
•Gain: 23dB
•35K NF(50)
•Frequency: 0.7-4Ghz
SKADS 06.11.09
LNA Technologies and Topologies
University of Manchester
Transistor characterisation
Several wafer samples of mHEMTs
and pHEMTs of both GaAs and InP
have been measured.
1.6
1.5
1.4
1.3
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.05
1.5
1.4
1.3
gm (mSiemens/mm)
gm (mS/mm)
Temperature stabilisation may be
important
process 1
process 2
process 3
process 4
process 5
process 6
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.5
-60
-40
Power (mW)
SKADS 06.11.09
-20
0
20
Temp (Celcius)
LNA Technologies and Topologies
40
60
Summary
Measured
LNA Data
Table
ASTRON
*IGN
CSIRO
University OPAR
of Calgary
University of
Manchester
(M&N)
California
University of
Institute of Manchester
Technology (MACS)
Gain
42dB
28dB
15dB
*27dB
**24dB
10dB
>30dB
25dB
NT
35K
*35K
40K
14K
25K
*65K
**56K
N.A.
*55K
**10K
35K
Frequency
0.9-2 GHz
*0.3-1
GHz
0.7-1.8
GHz
0.8-1.4 GHz
*0.3-1.9
**0.4-1
GHz
0.2-2GHz
**0.5-11
GHz
0.7-4 GHz
Technology
GaAs
pHEMT
GaAs
pHEMT
90nm
CMOS
0.25µm
SiGe HBT
1µm InP
pHEMT
SiGe HBT &
InP
70nm GaAs
mHEMT
Topology
SE &
*Diff
Diff
SE & Diff
Diff* &
SE**
SE
Diff & SE
SE
Impedance
50Ω
*150
300Ω
85Ω
50Ω
*100Ω
**50Ω
50Ω
270Ω diff
50Ω
Temperature
RT
RT
RT
RT &
Cryo
(22K)
RT
*RT &
**Cryo (17K)
RT
SKADS 06.11.09
LNA Technologies and Topologies
Conclusions (i)
• GaAs pHEMT, GaAs mHEMT, InP pHEMT, SiGe HBT and CMOS technology
has been researched.
• Measured data of 6 SE and 5 differential LNA has been shown.
• 14K CMOS SE LNA at RT BY University of Calgary.
• 25K NT CMOS differential 0.8-1.4GHz LNA has been measured with 15dB
gain at RT in University of Calgary.
• Cryogenic (15K ambient) SiGe Differential LNA with 10K NT and decade
bandwidth has been measured in CalTech (TDP).
• OPAR has measured 10K NT and 20dB gain single ended LNA @ 22K
ambient temperature.
• 2 promising and comprehensive differential noise measurement techniques
by CSIRO and University of Calgary.
SKADS 06.11.09
LNA Technologies and Topologies
Conclusions (ii)
• GaAs LNA developed by ASTRON has OIP3 of 25dBm. And impressive
results from Apertif.
• Temperature stabilisation is necessary to maintain consistent NT and
gain. How cost effective is it?
• State-of-the-art GaAs devices can exhibit lower power consumption
characteristics than InP devices to achieve same transconductance.
Linearity issues with low power?
• 5K NT possible to achieve with cooling down to 15K ambient
temperature. The associated power also decreases significantly. But
is it cost effective?
• Not only Tmin temperature but also Noise Measure is very (if not the
most) important aspect of LNA.
• Backs up argument of having LNA test verification facility in Europe
(Dr George suggests ASTRON as base!)
SKADS 06.11.09
LNA Technologies and Topologies
Appendix-1 (Caltech)
80
75
70
65
300K, 1.8V, 50mA
Noise Temp(K)
60
55
50
45
40
35
30
195K, 1.8V, 44mA
25
20
77K, 1.2V, 20mA
105K, 1.2V, 20mA
15
10
60K, 1.2V, 20mA
5
15K, 1.2V, 20mA
0
0
1
2
3
4
5
6
7
8
9
10
11
12
Frequency (GHz)
Reference- “Cryogenics Feasibility and LNA Options, Arecibo Focal Phased Array Workshop, Cornell University, July 21, 2009”
SKADS 06.11.09
LNA Technologies and Topologies
13
14
15
Appendix-1 (Caltech)
6K noise, 20mW power, and -10 dB input return loss are feasible specs
Noise at 20K
20
1.2V 2.8mA
1.3V 4.3mA
1.5V 7.2mA
1.8V 11.7mA
2V, 14.7mA
18
16
Noise, K
14
12
10
8
6
4
2
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Frequency, GHz
GAIN at 20K
45
40
35
Gain, dB
30
25
1.2V 2.8mA
1.3V 4.3mA
1,5V 7.2mA
1.8V 11.7mA
2.0V, 14.7mA
20
15
10
5
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Frequency, GHz
Reference-“Matched Wideband LNAs for radio astronomy” ,S. Weinreb, J. Bardin, H.Mani, G. Jones. Published in Review of Scientific
Instruments,2009
SKADS 06.11.09
LNA Technologies and Topologies