PPT - Cornell

Download Report

Transcript PPT - Cornell

Feeds and LNA’s for SKA
TDP Progress Report, June 4, 2010
S. Weinreb
Caltech
1.
2.
3.
4.
5.
6.
7.
4/13/2015
Caltech EE group
SKA-Mid receiver requirements
Five candidate feeds
Quadridge update
ATA feed update
Eleven update
QSC update (by G. Cortes)
1
Caltech EE Radio Astronomy Group, May, 2010
Ahmed Akgiray – Ph.D. student
Damon Russell – Ph.D. student
SKA feed design and integration
LNA design and testing
Hamdi Mani – Research technician
Steve Smith – Research engineer,
worked on everything
LNA’s, systems
Glenn Jones - Ph.D. graduate 2009
works on wide bandwidth processing
Zan Zhang – MSEE graduate 2010
SKA feed design and tests
Hector Navarette – Research
technician, LNA’s construction
Joe Bardin – Ph.D. graduate SiGe
Cryogenic LNA’s
SKA Mid Reflector Array – Receiver Requirements
1. Operation from 0.3 to 10 GHz
A. > 60 % Efficiency
B. Tsys < 35K for 0.8 to 10 GHz
2. Secondary but important considerations
A. Power consumption
B. Manufacturabililty
C. Maintainability
D. Cost including IP rights
E. Stabiliy for dynamic range and polarization
3. Reference design
A. Uncooled 0.3 to 1.4 GHz receiver with 0.8m diameter
B. Cooled 1.4 to 10 GHz receiver with 0.3m diameter dewar
C. Benchmark against octave band receiver (0.7 to 1.4 or
5 to 10 GHz)
Status of Candidate Feeds
QUADRIDGE – Cooled 3.5 to 14 GHz on 34m telescope with measured
efficiency and Tsys. Versions for 0.3 to 1.5 and1.4 to 10 GHz in design.
ATA – Uncooled 0.5 to 11 GHz on 42 ATA antennas with 40 to 100K Tsys.
Prototypes of cooled 1 to 10 GHz versions are under test.
QSC - Uncooled 0.4 to 4 GHz prototype constructed and tested showing
good patterns and return loss. Update by G. Cortes follows.
ELEVEN – Cooled 2 to 12 GHz prototype with measured patterns, return
loss and noise. A 1 to 10 GHz cooled prototype under construction and
will be delivered to Caltech this summer.
SINUOUS – In design and test at NRAO. Uses foam support and dewar.
Improvements to Quadridge Feed
Good return loss is achieved from 1.4 – 11GHz
Concept for Integration of LNA’s Into Dual-Polarized Quad-Ridge Feed
• Single-ended 100 ohm LNA’s are mounted on bottom side of fins with probe
coupling across slot to opposite fin.
• Coaxial output and bias lines of LNA will be routed along back of fin
• LNA will be on a bolt-in copper carrier for easy replacment
Probe coupling 100 ohm
slot line to microstrip;
LNA
5mm
LNA
LNA on 10mm base
could be repackaged
to 5mm x 20mm area
Photograph of 0.5 to 14 GHz System at Goldstone DSS28, October, 2008
Secondary (shadow)
2-14 GHz Cooled Feed
0.5-4 GHz Feed, Cooled LNA
Rotatable Tertiary
Efficiency and Tsys of Quadrige Feed Measured on 34m GAVRT Telescope
Two Concepts for Cooling of the ATA Feed
• Current ATA feed is uncooled (with LNA cooled to 65K) and covers 0.5 to
11 GHz with system noise temperature of 40 to 100K
• Two concepts to cool the feed to reduce Tsys to < 35K over the band of 1
to 10 GHz or higher are being investigated.
Cooled feed in aluminum
vacuum chamber with
Mylar window
Cooled feed in glass bulb
vacuum chamber
Coolable Prototype of ATA Feed at Caltech for Tests
of Impedance, Pattern, and Noise Temperature
• Feed made of copper (to be plated) and is ~40 cm long
• Feed tips are connected to LNA within pyramid by 93 ohm coaxial cable
as shown at right below.
• Tests will be conducted during summer of 2010
Eleven Feed - Noise of Cryogenic Feed and LNA Now Measured
(Beaudoin, Klein, Yang, Kildal, and others)
•Noise measured for feed, baluns, and LNA’s at 20K
•Measured patterns predict good efficiency
Very Low Noise Amplifier Development Summary - 2009
• InP HEMTS - Indium-phosphide high-electron-mobility transistors have been implemented
in almost all low-noise amplifiers in radio astronomy for the past 10 years with little change in
performance. They are usually cooled to 15K to reduce the noise by an order of magnitude.
The noise performance from wafer to wafer can vary by a factor of two due to unknown
reasons.
• Room Temperature Very Low Noise Amplifiers. - The cost of the large number of receivers
required by arrays could be greatly reduced if LNA’S with sufficiently low noise could be
realized at room temperature. Transistor device improvements may enable this. Current
performance is around 17K noise or 0.25 dB noise figure at 1.4 GHz; a goal is < 10K of noise
(0.14 dB NF) at 1.4 GHz.
• SiGe HBT - A promising new transistor, the SiGe HBT, is being rapidly developed for high
speed computer and communication applications. These transistors greatly improve in
current gain and transconductance when cooled. During the past year SiGe amplifiers with
noise temperatures of <5K at frequencies under 4 GHz have been built and tested at Caltech
using transistors from IBM and STM. Advantages are: 1) simultaneous noise and power match
by feedback, 2) high gain stability, 3) can be integrated with CMOS on the same chip. 4)
rapidly developing technology.
Caltech-Developed Cryogenic LNA’s
Caltech
e
EE has Delivered 400 LNA’s to Other
Research Centers During the Past 6 Years
6 Models, @ 15K
SiGe HBT
001 to 1 GHz < 6K
1 to 3 GHz < 8K
4-12GHz LNA #82D at 12K
MMIC: WBA13, CIT1 4254-065 , R8C2
Bias: Vd=1.2V, Id=20mA, Vg1=2.33V, Vg2=2.33V
20
0.5 to 11 GHz, Tn < 6K
18
45
16
40
14
35
Noise Temp(K)
4 to 14 GHz, Tn < 8K
50
12
6 to 20 GHz, Tn < 12K
30
Noise Temperature (K)
Gain (dB)
10
11 to 26 GHz, Tn < 20K
Differential InP HEMT
0.5 to 11 GHz,Tn <18K at 60K
25
8
20
6
15
4
10
2
5
0
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Frequency (GHz)
4/13/2015
S. Weinreb
13
Gain, dB
InP HEMT
Packaged Differential Cryogenic LNA
Will be used for tests of ATA, QSC, and Eleven feeds and is available
if a differential transition to a quadridge feed is desired. Optimum
generator impedance is in the 200 ohm range.
InP HEMT MMIC
LNA
200 ohm
differential
input
50 ohm
coaxial output
4/13/2015
14
Noise Temperature Measurement of Differential LNA’s
• Noise temperature of a differential amplifier can be measured with the Y factor
method by dipping a termination resistor in LN2 at 77K
• The resistance change vs temperature can be measured very accurately at
DC with on ohmmeter. The change in typical thin-film resistance is negligible.
• Shown below is a small board with two 270 ohm resistors connected to two
differential LNA’s with a gold-plated SS tubing quad transmission lines.
S. Weinreb, Jun 2009
15