Transcript Powerpoint

Antenna, Feed, and LNA Integration
S. Weinreb, May 7, 2009, LAX
Antenna Working Group Meeting
1. Issues
2. Feed Summary
3. LNA Summary
4. Differential LNA’s
5. Feed LNA Interface
6. Noise budget
7. Summary and integration plan
Feed/LNA Interconnect Issues
Low Frequencies (< 3 GHz) at 300K
• Losses – Increase in Tsys due to small losses in the
interconnection. Difficult to keep these < 0.2 dB which adds 14K to
Tsys. The addition can be as high at 30K when the considerations
below are included.
• Differential LNA’s – Most wideband feeds have a differential
(floating with respect to ground) output with impedance in the 100 to
270 ohm range. There is limited experience with the design,
connections, and testing of these LNA’s.
• Input Reflections and Correlated Noise Out of Input – Especially
important for phased-array feeds.
• Environmental – The LNA and feed must be protected against
moisture, insects, rodents. lightning, temperature cycling and hail
• Mechanical – The feed connection points are small and difficult to
mechanically support. A robust, easily manufactured design is
needed.
Feed/LNA Interconnect Issues
High Frequencies (1.4 to 10 GHz) at < 70K
• Losses – At higher frequencies the interconnections must be
smaller (or they will radiate) and this increases the loss. However
conductivity increases with cooling and a loss of 0.2 dB gives a
tolerable ~4K noise at 70K physical temperature.
• Differential LNA’s – Same comments as for low frequency feeds..
• Vacuum – Any cooling below dew point temperature requires
sealing against condensation and vacuum insulation is usually
required. This leads to a need for microwave windows and hermetic
seals. Vacuum at 70K is more difficult than vacuum at 15K.
• Cooler – Compact, long life (>106 hours MTBF) coolers to 70K are
available for ~$2,000 in quantity of
Summary of Wideband Feeds
From Bradley and Gawande, URSI Meeting, Boulder, Jan 2009
Lindgren 3 to 12 GHz Quadridge Feed
• Measured Tsys < 35K and efficiency > 40% on GAVRT 34m
antenna.
• Patterns improve with surrounding vacuum dewar
• Have 50 ohm single-ended output so simple LNA’s can be used
• Robust mechanical structure inexpensive to replicate/
Noise vs Frequency of SiGe Transistor LNA at 3 Temperatures
ST first stage, NXP 2nd stage, tested May, 2008
Noise, K
Typical gain 35 dB, typical bias 2V, 12mA
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
LNA 300K
@ 300K Connector with LNA @ 54K
LNA 54K
LNA 20K
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Frequency, GHz
3.5
4.0
4.5
5.0
Low-Cost SiGe 0.5 to 4 GHz Cryogenic LNA
• 7K noise at 17K with $.44 NXP transistor
• With STM transistor input stage noise is
2.5K at 17K, and 7K at 55K.
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
40
35
30
25
20
15
Noise, 1.7V, 10mA
10
Gain, 1.7V, 10mA
5
Gain, dB
Noise, K
NXP BFU 725 2 stage LNA @17K
April 15, 2008
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
GHz
SiGe transistors in
2mm plastic package
on printed circuit board
Noise Temperature vs Frequency at 300K, 195K, 105K, 77K, 60K, and 15K
InP HEMT MMIC, WBA13, Tested at Caltech May, 2007
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
Frequency (GHz)
10
11
12
13
14
15
Interface with Feed at 300K and LNA at 60K
Output coax
Input Twin-Lead Lines
Active Balun (Differential) LNA for ATA
Dual Differential LNA Module
• Can house either of 3 differential LNA chips on hand: ATA InP or 2 SiGe IBM
• Module and PCB’s in fabrication, April, 2009, expect test results by Aug, 2009
• Transition from 270 ohm quadline to 135 ohm microstrips has been tested
Transition to quadlines to feed
differential impedance 200 to 270 ohms
22pF
290- 14
290-14G Pin
0-80
10
0402
0805
1000pF
0.1uF
Out V Polariz
10
22pF
50K
.052" I.D. PTH for SS quadlines
47pF
PCB ABout1 Rogers 6010LM, .01" thick
PC B ABin1 Rogers 5880, .031" thick
Output ahd bias PCB
Input
PCB
Differential LNA MMIC
Out H Polariz
Dual 270 ohm resistors
simulating feed impedance
SKA Tsys Budget – Current and Expected 2011
Component
Current Technology
2009
Noise, K,
1.4 GHz
Sky
Background + atmosphere
4
No improvement here!
4
Spillover & Blockage
15 dB edge taper + 2.5%
blockage, total 4% at 300K
12
Low spillover optics
7
Feed loss
10cm of .085”, 7K + 5K feed
loss
12
Twin-lead feed terminals
5
LNA to feed loss
10cm of 0.141 Cu coax bend
to dewar, .04 dB at 300K
3
40mm twin-lead
2
Vacuum feedthru
Glass/Kovar bead, 0.1 dB
7
Quartz/gold bead, 0.04 dB
3
10cm or .141 SS/BeCu
.09 dB at 190K
4
Air line
2
Coupler at 70K
Werlatone C7753, 0.2 dB
3
or noise lamp coupling
2
Total
Total Above
45
Total Above
25
LNA @ 300K
Commercial 0.5 to 4 GHz LNA
35
Improved LNA @ 300K
20
LNA @ 60K
Current LNA
14
Improved 70K LNA
7
Coax in dewar
Innovation Path
2011
Noise, K,
1.4GHz
Total Tsys, 300KLNA
80
Total Tsys, LNA @ 300K
45
Total Tsys, 60K LNA
59
Total Tsys. 60K LNA
32
Summary of Current Feed/LNA Status
• There are 5 candidates for wideband feeds currently
under development. Pattern and noise measurements at
a common facility are needed
• Cooled LNA’s for 0.5 to 3 GHz (< 7K noise at 60K) and 2
to 10 GHz (< 18K noise at 60K) are available.
• Room temperature LNA’s to meet the SKA Tsys < 35K
requirement are difficult for 0.5 to 3 GHz and not feasible
for > 3 GHz.
• Integration of feeds with LNA’s is in an initial phase with
much progress expected in the next year.
• A US prototype antenna that can index to 3 different
feed/LNA packages is recommended.
Integration Plan – May 2009
Feed pattern tests and
optics optimization
(by others)
Test feed integrated
with dewar at USC
Test 3 differential LNA’s
first in coaxial and
then quadline fixture
at 300K and 60K
2010
Noise and pattern test
of integrated dewar,
feed, and LNA
2011
System test on
antenna, efficiency
and Tsys
Dewar Construction
and Cryogenic Test