Do we need ground plane for SKA-low Radio Telescope? · antenna performance. ―Accurate characterization sensitivity of antenna (array) ―Helpful in calculation of power budget

SKA Technical Progress Meeting: October 22-23, 2012, Bologna, Italy

S. K. Padhi, M. W. Waterson, Budi Juswardy, Aziz Jiwani, Adrian Sutinjo and Peter Hall

ICRAR/CIRA, Curtin University, Australia

Do we need ground plane for

SKA-low Radio Telescope?

SKA Technical Progress Meeting, October 22-23, 2012, Bologna, Italy


Outlines:

Introduction

Soil characterization

Measurements:

Ground plane option:

Sensitivity with ground plane

Antenna noise calculation with mesh ground

Antenna noise on dispersive soil

Antenna Measurements:

Antenna performance

Conclusions


Introduction:

― Soil modelling

― Very important to understand the environmental influence on

antenna performance.

― Accurate characterization sensitivity of antenna (array)

― Helpful in calculation of power budget

― Ground plane choice

― Cost

― System level performances

― Mesh-grid design (for improved performance)

― Sensitivity of antenna

― Antenna metrology

― Challenges in fabrication of Spiral antenna fabrication.

― measurements


Soil Characterization

― We classify the morphology of soil as two

layers structure.

― Upper layer: mixture of gravels, sands, silt

and clay.

― Lower layer is granite base

― Thickness of upper layer: 30 cm to 80 cm.

― 5 samples were collected during sunny day

― 4 samples were taken from each corner of

200 m length square area.

― The 5th one collected from the centre of

square.

― The samples were sieved and oven dried

before measurements.

Lower layer

Upper layer


Soil Measurements:

• Chemical analysis + mineral composition

• Measured relative permittivity of the soil

samples, in CSIRO facility.

• Permittivity measurements with different

moisture contents.

• Nicholson-Ross-Weir algorithm is used in

the calculation of permittivity.

• Debye 2 parameter model is used for

fitting data (model).


0.0 0.2 0.4 0.6 0.8 1.00

2

4

6

8

10

Imag(r):

measured

fitted

Real(r):

measured

fitted

Per

mit

tivi

ty (R

e/Im

)

Freq (GHz)

2 % moisture contents:

Soil: measurements:

0.0 0.2 0.4 0.6 0.8 1.0 1.210

20

30

40

Soil data(10% moisture):

' meas.

' model

'' meas.

'' model

Freq (GHz)

' (mea

s. a

nd m

odel

)

1

10

100

'' (mea

s. a

nd m

odel

)

0.0 0.2 0.4 0.6 0.8 1.0 1.2

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

Measurements:

real(r)

imag(r)

Freq(GHz)

Per

mit

tiv

ity

(re

al)

0.0

0.5

1.0

1.5

2.0

Per

mit

tivit

y (

imag

)


Do we need a ground plane??

― Ground planes are critical components of antenna.

― GP influence (depends on various types of antenna)

― Impedance match

― Gain/sensitivity of antenna ― GP needed for mechanical support or robust placement/fixing

antenna

― Proper grounding of circuits etc..

― Cost is important for GP size ??

― Alternate option:

― Mesh grid (reducing cost (and improve performance))


Mesh grid model:

d

Infinite soil box

(A,B,C)

Phase center

GP/mesh grid

Soil C (infinite size)

Mesh size:

Pitch: 300 mm

Dia: 2.5 mm

Size of grid: 10m x 10m

Media: copper


Noise Temperature (Tmg+soil) VS pitch size:

50 100 150 200 250 300 350 400 4500

50

100

150

200

250

Tan

t_gn

d [o K]

Freq(MHz)

Meshgrid + Soil C (d=0.0 m)

300 mm

600 mm

900 mm

1200 mm

spiral LPDA

50 100 150 200 250 300 350 400 4500

50

100

150

200

250Mesh grid+Soil A (d=0.1 m):

300 mm

600 mm

900 mm

1200 mm

Tan

t_gn

d [o K]

Freq (MHz)

50 100 150 200 250 300 350 400 4500

50

100

150

200

250

Vivaldi (meshgrid+Soil A): (d=0.1 m)

300 mm

600 mm

900 mm

1200 mm

T ant_

gnd [o K

]

Freq (MHz)

vivaldi

~100 MHz

~150 MHz

~150 MHz


Sensitivity with/without mesh grid

LPDA

50 100 150 200 250 300 350 400 450-160

-120

-80

-40

0

20

Vivaldi (mesh only): d=0.1 m

300 mm

600 mm

900 mm

1200 mm

T

ant_

gnd[o K

]

Freq (MHz)

50 100 150 200 250 300 350 400 450-100

-80

-60

-40

-20

0

10

T

ant_

gnd[o K

]

Freq (MHz)

Meshgrid only:(d=0.1 m)

300 mm

600 mm

900 mm

1200 mm

50 100 150 200 250 300 350 400 450-160

-140

-120

-100

-80

-60

-40

-20

0

LPD (meshgrid only: d=0.1 m):

300 mm

600 mm

900 mm

1200 mm

T

ant_

gnd[o K

]

Freq (MHz)

spiral

vivaldi


Comparison of Tant_gnd (Soil-A)

Freq. Spiral Antenna Vivaldi Antenna LPD Antenna**

ASR-V1.4 FEKO ASR-V1.4 FEKO ASR-V1.4 FEKO

70 MHz 134.0 153.1 209.8 223.5 127.5 198.7

150 MHz 2.12 8.53 65.2 85.0 12.7 27.0

300 MHz 4.00 8.92 40.6 54.5 na 8.1

450 MHz 1.2 7.5 29.2 42.3 0.8 8.25

**: compared with d=0.16 cm data


Do we need a ground plane? Sensitivity of isolated sheet spiral at SKA site – infinite extent back surfaces

Spiral and ground plane 11

• Spiral with a metallic ground plane has higher sensitivity than SKA-low goal

• Spiral over the soil does not meet goal below 110 MHz, but is close enough to warrant detailed performance and cost trade-

off

• Highly-directive SKA-low antennas (like the conical spiral) may not need a metallic ground plane

• Or use only very wide mesh


Antenna metrology:


Conductor patterns:

- 2 identical spiral racks made from 13 sections - Water-jet cut from brass mesh (or copper foil) - low-temperature solder joints, epoxy to cone


Conclusions

― Electrical and chemical parameters of Soil are characterized in

details.

― Debye model is used to fit the measured data. The comparison

between the two are reasonable close.

― Ground plane effects are characterized for 3 antennas using MoM

solver.

― Mesh grid might be a better option and cost effective compared to

metallic ground plane. Tant is sensitive to lower end only .

― And for high requency end (> 150MHz) Tant is insensitive to mesh

grid/GP.

― With Meshgrid Tant is reduced to ~100 K (@70 MHz) for all three

antennas.

Documents

Do we need ground plane for SKA-low Radio Telescope? · antenna performance. ―Accurate characterization sensitivity of antenna (array) ―Helpful in calculation of power budget