Beruflich Dokumente
Kultur Dokumente
Precision Antenna
Measurement Guide
September 2017
S P O N S O R E D B Y
Table of Contents
3 Introduction
Pat Hindle
Microwave Journal, Editor
10
Basic Rules for Anechoic Chamber Design,
Part Two: Compact Ranges and Near Field
Measurements
Vince Rodriguez
MI Technologies
Application Profile
22
As the number of devices featuring wireless connectivity
grows, ensuring their performance specifications while staying
within regulatory requirements becomes even more important.
Characterizing and Tuning Antennas Antenna pattern measurement is a critical step in the design
process of antennas and wireless devices. Compact antenna
measurement systems combined with a high performance
25
is mounted to the positioner’s RF Rotary Joint and the reference
range and reduced noise by reducing the IF bandwidth but
antenna is connected then positioned at the appropriate
the analyzer speed is greatly reduced, a single sweep can take
distance (typically 1 or 3 meters). Now that the antennas are
many seconds to complete. PC-based VNA’s offer the benefit
Accurate Testing
The full DAMS setup the VNA should show a strong profile with little
DAMS Positioner / AUT trace noise. Depending on the type of antenna being
measured, the antenna can be manually positioned
to a point of lowest signal to confirm that the signal
Scalar or Vector* Calibration
Clayton Karmel, Pdicta Corp.;
is still above the noise floor and the trace noise is
(* Requires VNA)
acceptable for testing requirements. After the setup is
verified, the DAMS antenna measurement software is
Ben Maxson, Copper Mountain used to make a measurement. The software indexes
the antenna to every physical point and executes a
transmission sweep on the VNA. Optionally, VSWR
data can be automatically collected at every point.
Precision RF Cables Once complete, the entire data set can be viewed,
DAMS Antenna processed and exported.
Measurement Studio
3
Basic Rules for Anechoic
Chamber Design,
Part One: RF Absorber
Approximations
Vince Rodriguez
MI Technologies, Suwanee, Ga.
T
he task of adequately specifying of range for different antenna types and
performance for an indoor anechoic frequencies of operation. Rules of thumb
chamber without driving unneces- are provided to select the best approach
sary costs or specifying contradictory re- for the required test or antenna type. The
quirements calls for insight that is not al- article concentrates on rectangular cham-
ways available to the author of the speci- bers. Simple approximations are used for
fication. While there are some articles and absorber performance to generate a se-
books1-3 that address anechoic chamber ries of charts that can be used as a guide
design, a concise compendium of refer- to specify performance and appropriate
ence information and rules of thumb on facility size.
the subject would be useful. This article The ability to measure an antenna is an
intends to be a helpful tool in that regard. important design requirement in determin-
It starts by recommending the proper type ing if energy is radiating properly and in the
desired direction, as well as how much en-
ergy is traveling in undesired directions. To
measure antennas (like many other devices
that are being measured), there is the de-
sire to have the antenna unaffected by its
surrounding environment. This is where the
anechoic chamber becomes a viable solu-
tion. The anechoic chamber provides an
environment free of echo or other radiated
signals to reduce the effects of these unde-
sirable signals.
This article covers applications where an
antenna is radiating or receiving a given
signal, and its performance as a function of
s Fig. 1 General geometry of an indoor range - two antennas are locat-
direction is being measured.
ed in the range (one for transmitting and one for receiving).
4
This equation can be used when indoor measurements at those low
1000
the antenna is under one wave- frequencies. Indoor ranges can be
500 length in electrical size. From Equa- built, but the antenna size should
300 tion 1, the far field distance can be be kept less than 2l; which limits
200
100
plotted as a function of the electri- the far field distance to 8l (24 m).
50 cal size of the antenna, as shown in This distance is close to the 10l
Figure 2. given by Equation 2. Table 1 pro-
r/
30
20
10 As a rule of thumb for indoor vides an approximate guide for
5 ranges, the far field illumination the different antenna sizes and fre-
3 techniques are better suited for quencies of operation.
2
1 antenna sizes under 10l. This rule The values in Table 1 are gen-
2 4 6 8 10 12 14 16 18 20 is related to the electrical antenna eral guidelines. Spherical near field
D/
size. Frequency of operation adds (SNF) ranges can test antennas as
s Fig. 2 The far field distance plotted another factor that will influence small as l/2. But for such a small an-
related to the wavelength. the type of range. An antenna with tenna it may be a better approach
to use a far field illumination range
TABLE I as it relates to the typical electrical
FREQUENCY RANGES AND ANTENNA SIZES FOR THE DIFFERENT INDOOR ANTENNA size of the AUT.
MEASUREMENT APPROACHES When creating an anechoic
Indoor Ranges Antenna Size in Wavelengths chamber, the goal is to obtain a
Frequency Far Field Near Field Compact Range
volume in the chamber where any
Illumination reflected energy from the walls of
the range (ceiling and floor) will be
100 MHz <2 >2 Not ideal
much lower than any of the features
500 MHz <2 >2 Not ideal of interest on the radiation pattern.
1 GHz <5 >5 >5 This volume is known as the quiet
2 GHz <10 >10 >10
zone (QZ). Figure 1 shows that as
one antenna transmits, it illumi-
≥ 4 GHz <10 >10 >10 nates the receive antenna and all
the walls and surfaces of the range.
RANGE TYPE SELECTION The energy incident onto these sur-
The general range geometry a size of 10l will have a far field faces will be reflected towards the
is shown in Figure 1. There are distance of 200l, making the test QZ.1,3 The level of reflected energy
several methods of measuring the distance 20 times the size of the an- must be a given number of deci-
radiation patterns of antennas in- tenna. At some microwave frequen- bels below the direct path between
doors: far field illumination, near cies this may be a test distance of the transmitting and receiving an-
field measurements and compact 200 inches (5 m) so an indoor range tennas.
range. While they all present pros may be easy to implement. How- As the antenna being measured
and cons, there is not a single so- ever, note a 20l antenna will have is rotated (see Figure 3), its main
lution that is ideal for all types of a test distance that is 800l. beam will illuminate different sur-
antennas and situations. The type For example, consider an 18 inch faces of the chamber. The range
of range most suitable for a given dish used by a popular satellite TV antenna will measure the level of
type of antenna is driven by two service. This satellite service oper- field radiated by the AUT along
parameters: frequency and elec- ates at 18.55 GHz. The dish anten- the direct path between the two
trical size of the antenna under na is 28.29l in size. The far field is antennas. However, the range an-
test (AUT). The far field condition at approximately 1600l or 25.86 m tenna will also receive the reflected
given by the following equation (84.84 ft). Clearly, for such an elec- energy from the walls, ceiling and
drives the selection: trically large antenna, a far field il- floor. If the reflected energy level
lumination approach indoors is not is higher than the energy radiated
2D2 economically feasible. For this an-
r≥ (1) along the direct path between the
λ tenna, a compact range or a near two antennas, then the radiation
The parameters mentioned field approach is more suitable. pattern in that direction cannot
above are embedded into the far Conversely, a 10l antenna at 300 be measured accurately. In Figure
field equation. D is the largest MHz, which is 10 m in size, would 3, the measuring antenna, (also
physical dimension of the antenna. be extremely difficult to manipulate known as range antenna or source
Wavelength is l, which is related to at a test distance of 200 m. For this antenna) is pointing at a null, but it
the frequency of operation on the case, the best solution would be an is also receiving the reflected sig-
antenna. For smaller antennas the outdoor range. nal from the wall that is illuminated
far field range length, r, can be ap- In general, for frequencies be- by the main beam of the AUT. The
proximated by:4 low 100 MHz, an outdoor range is a range antenna is receiving the re-
better approach. Current absorber flected signal in a direction of 30°.
r ≈ 10λ (2) technology does not support some In that 30° direction, the gain of
5
The coefficients in this equa-
tion are functions of the thickness.
When the thickness of the absorb-
Antenna Being Measuring
er is such that 0.25 t 2, the
Tested (Transmitting) Antenna (Receiving) coefficients of Equation 4 are giv-
en by the following polynominals:
2
A (t) =
A11 (t) = 1.5252
1.5252
3
−
− 4.8243t
4.8243t
4
+
+ 6.9479t
6.9479t 2
−3.8332t 3 +
−3.8332t + 0.7333t
0.7333t 4 (4a)
(4a)
2
A (t) =
A 22 (t) =− 0.0754 +
−0.0754 0.24782t −
+ 40.24782t − 0.3984t
0.3984t 2
+0.2285 −
+0.2285 − 0.0442t
0.0442t 4 (4b)
(4b)
2
A (t) =
A 33 (t) = 0.0016
0.0016
3
−
− 0.00502t
0.00502t
4
+
+ 0.00938t
0.00938t 2
−0.00577t 3 +
−0.00577t + 0.001155t
0.001155t 4 (4c)
(4c)
−5 −5
A (t) =
A 44 (t) =− −1.58
1.58
−4 2
⋅⋅ 10
10−5 + 4.91−5⋅⋅ 10
+ 4.91 103 −5 tt
−1.015 ⋅⋅ 10
−1.015 10−5−44tt 2 + 6.58 ⋅⋅ 10
+ 6.58 10−5 tt 3
−1.35 ⋅ 10
−1.35 ⋅ 10 t −5 t 4 (4d)
(4d)
−8 −7
s Fig. 3 An indoor range showing one of the reflected paths and the direct path
A (t) =
A 55 (t) = 5.84
5.84
−7 2
⋅⋅ 10
10−8 − 1.78−7⋅⋅ 10
− 1.78 103 −7 tt
between the AUT and the source antennas. +4.02 ⋅⋅ 10
+4.02 10−8−74tt 2 − 2.71 ⋅⋅ 10
− 2.71 10−7 tt 3
the range is lower than in the di- 0.25 t 20. This approximation +5.7 ⋅ 10
+5.7 ⋅ 10 t −8 t 4 (4e)
(4e)
rect path (boresight) to the AUT. can be used to get a conservative
The reflected energy is a number reflectivity value of an absorber of a When the thickness of the treat-
of dB lower, for example, 20 dB. given thickness. Most manufactur- ment is such that 2 t 20,
Let us assume that the gain in the ers provide the information in their then the coefficients are given by
30° direction is 10 dB lower than datasheets. the set of polynominals:
the boresight. The signal received Figure 1 shows that some of
by the antenna on that direction the absorber in the range is not A1 (t) = 0.1751 + 0.149t −
will be -30 dB compared to the located in the normal incident 0.0119t 2 + 0.00028t 3 (4f)
energy received when the main wave direction, but rather in an
beam of the AUT was pointing oblique incidence. For oblique A 2 (t) = −0.0105 − 0.00824t +
to the range antenna. If the null incidence, the main reflectivity 0.0007t 2 − 1.61 ⋅ 10−5 t 3 (4g)
is less than -30 dB, the measured of the absorber is in the bi-static
pattern will have errors.5 direction. Backscattering occurs A 3 (t) = 0.00029 + 0.000123t
when the distance between the −1.13 ⋅ 10−5 t 2 + 2.57 ⋅ 10−7 t 3 (4h)
RF ABSORBER tips of the pyramids is l. 7 Hem-
A key design item for an an- ming 1 provides plots that show A 4 (t) = −1.69 ⋅ 10−6 − 4.77 ⋅ 10−7 t
echoic chamber is the RF absorber. the estimated bi-static reflectivity +5.08 ⋅ 10−810−8 t 2 − 1.14 ⋅ 10−9 t 3 (4i)
The absorber treatment must be of absorber at oblique incidence.
such that the reflected energy has A series of polynomial approxi- A 5 (t) = 0 (4 j)
a small or negligible effect on the mations, together with Equation
measured data. A typical RF ab- 3, provide a general description The domain of Equation 4 is lim-
sorber is a lossy material shaped to of the performance of pyrami- ited by those angles of incidence
allow for incoming electromagnetic dal absorbers of different thick- where 0° 85°and where
waves to penetrate with minimal nesses and at different angles of =0° is normal incidence. Addi-
reflections. Once the electromag- incidence. These are conserva- tionally, the domain is limited by
netic (EM) energy travels inside the tive approximations. That leaves the domain of the coefficient poly-
material, the RF energy transforms a margin of error to account for nomials. Hence Equation 4 is valid
into thermal energy and dissipates things like lights, doors, position- when 0.25l t 20l. The range
into the surrounding air.6 The ing equipment and edge diffrac- of Equation 4 should also be lim-
electrical thickness of the material tions from treatment discontinui- ited to -55 R(dB) 0. For an ab-
determines how much energy is ties. sorber thickness larger than 20l,
absorbed. The reflection level at The absorber performance in dB the reflectivity can be approximat-
normal incidence can be approxi- is given by the following polyno- ed using the results for a 20l thick
mated by the following equation: mial: absorber. Figure 4 shows the bi-
R o (t) = −13.374 ln(t) − 26.515 (3) static performance as a function of
Rθ (t, θ) = R o (t) + A1 (t) ⋅θ + A 2 (t) ⋅θ2 angle for a series of different elec-
where t is the thickness in wave- + A 3 (t) ⋅θ3 + A 4 (t) ⋅θ4 + A 5 (t) ⋅θ5 (4) trical thickness of the absorber.
lengths. The equation is valid for
6
suitable for EMC applications as
Bi-static Reflectivity of Pyramidal Absorber
0
their normal incidence absorption
Thickness in
0.25
is typically limited between 25 and
–5 0.5 35 dB.
1
1.2
1.4 RECTANGULAR FAR FIELD
–10 1.5
1.6 CHAMBERS
1.8
–15 2 Sizing the range begins with
2.5
3
rectangular far field ranges that
–20
4 have a test distance is determined
5
6 by Equation 1. It is common to find
sources stating the rules of thumb
Reflectivity (dB)
7
–25 8
9 for sizing a rectangular anechoic
10
–30 15 chamber for far field illumination.
20 Generally, the width and height of
–35
an anechoic chamber should be
three times the diameter of the
–40
minimum sphere that contains the
largest antenna being tested. It
–45
is important to check that a mini-
mum spacing of 2l between the
–50
AUT and the tips of the absorber
is maintained to avoid loading of
–55
the AUT. The far field distance is
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 given by:
Angle of Incidence (°)
r
s Fig. 4 Estimated reflectivity of RF absorber as a function of angle of incidence.
= 2n 2 (5)
λ
flectivity — higher by about 10 dB. Where n is the number of wave-
1 Equation 4 The manufacturer specifications lengths in size of the AUT. The QZ
2 Equation 4
1 Computed TL Approximation were only provided from 45° to 80° must be large enough to encom-
2 Computed TL Approximation and normal incidence. Computed pass the AUT. Hence the QZ is
1 Manufacturer A Specs
2 Manufacturer A Specs results were obtained only at a few nl. Figure 6 shows a typical rect-
0 angles. For the 1l thick absorber, angular range geometry. From
–5 the different methods follow simi- the geometry, an equation for the
–10 lar trends, with the polynomials distance x can be derived. The
–15 providing the most conservative distance x is the distance from the
Reflectivity (dB)
9
Basic Rules for Anechoic
Chamber Design, Part Two:
Compact Ranges and Near Field
Measurements
Vince Rodriguez
MI Technologies, Suwanee, Ga.
The task of adequately specifying performance for an indoor anechoic chamber without
driving unnecessary costs or specifying contradictory requirements calls for insight that is
not always available to the author of the specification. Although there are some articles
and books1-3 that address anechoic chamber design, a concise compendium of reference
information and rules of thumb on the subject would be useful. This second part of the series
intends to do that, concentrating on the sizing of compact ranges and chambers for near field
systems. As was done in part one, simple approximations are used for absorber performance
to generate a series of equations that help specify performance and size of facilities.
P
art one of this series identi- COMPACT RANGES
fied limitations in using far Although barely mentioned in
field chambers, mainly re- the IEEE standard test procedure
lated to the electrical size of an- for antennas,4 the compact range
tennas that can be tested. As (CR) has become an important tool
was shown, an 18” dish used by for measuring electrically large an-
a popular satellite TV service will tennas. The CR uses a parabolic re-
be almost impossible to test in a flector to create a plane wave illu-
far field chamber. The satellite ser- mination at the location of the an-
vice operates at 18.55 GHz, the tenna under test (AUT). This plane
dish antenna is 28.29 wavelengths wave simulates the field distribu-
(l) in size, so the far field is at ap- tion that the antenna experiences
proximately 1600 l or 25.9 m (84.8 in the far field. Figure 1 shows a
ft). Clearly for such an electrically parabolic reflector illuminated by a
large antenna, far field illumination source located at the focal point of
indoors is not economically fea- the parabola. The plane wave be-
s Fig. 1 Simulated results of a parabol- sible. For this antenna, a compact havior can be seen a short distance
ic reflector, showing plane wave behavior range or near field measurement is from the reflector. The reflector sys-
on the right. more suitable.
10
TABLE 1 height of the chamber. The length
of the chamber will be affected by
COMMERCIALLY AVAILABLE COMPACT RANGE REFLECTORS the focal length of the reflector.
QZ Size Overall Length of Frequency of Focal Length fl The distance from the vertex of the
(Length and Reflector Size Serrations Operation (cm) reflector to the quiet zone (QZ) is
Diameter) (Including (cm) (GHz) given by the following rule:
(cm) Serrations)
(cm) 5
r= fl (1)
61 216 × 188 38 4 to 200 182 3
122 432 × 335 76 2 to 200 366 where fl is the focal length of the
182 488 × 416 76 2 to 200 366 reflector. Referring to the satellite
244 864 × 670 152 1 to 200 732
TV antenna, requiring a far field
distance of 25 m to test, one might
366 975 × 833 152 1 to 200 732 expect long chambers and large
distances for CR testing. However,
tem is the controlling factor when est frequency of operation. Table 1 Table 1 with Equation 1 indicates
sizing the range. The reflector must provides a typical list of reflectors, the test distance for a 61 cm QZ is
be large enough to provide a plane showing their overall size and fre- 3 m. This is sufficient to test the sat-
wave that illuminates the entire an- quency ranges. Note that as fre- ellite TV antenna.
tenna being tested, and the reflec- quency increases, the reflector be- As a rule, the length of a CR
tor should be properly terminated. comes more efficient. While some chamber is given by the following
The purpose of the termination is to reflectors can operate well into the equation:
reduce the effects of the terminated millimeter wave range, extra care
paraboloid on the illumination. The should be taken during manufactur- 5 1
L = R clr + f l + QZ + ( 2 + t ) λ (2)
two most common ways of termi- ing and surface finishing, as surface 3 2
nating a reflector are serrations and imperfections will affect the perfor- where Rclr is the reflector clearance.
rolled edges.6 In the case of ser- mance. This includes the mechanical struc-
rated edge reflectors, serrations can Reflector size is the determining ture to support the reflector, which
be between 3 l and 5 l at the low- factor when sizing the width and ranges from 60 cm to 2 m, depend-
90
ing on the overall reflector size. In
120 60 general, the wall behind the reflec-
tor has a small absorber, usually
150 30 l/2 in thickness, and only covers
the perimeter of the wall. The pa-
rameter t is the thickness of the end
180 wall absorber. For a CR, this is the
–60 –40 –20 0
most critical wall and should have
the lowest reflectivity; it is recom-
–150 –30 mended the value of t be no less
–120 –60
than 3 to 4.
–90 The width of the chamber is cal-
culated using:
W = CR w + ( 4 + 2t ) λ (3)
90
120 60 where CRw is the overall width of
the reflector. There is an additional
2 l from the tips of the serrations
150 30 to the absorber tips on each side of
the reflector, although in some cas-
es the spacing can be as small as
180 one wavelength on each side. The
–60 –40 –20 0
final item determining the width of
the range is the thickness of the ab-
sorber.
–150 –30
While for far field ranges the ab-
sorber on the ceiling, floor and side
–120 –60 walls should be thick enough to
–90 provide good bistatic reflectivity at
oblique angles, in the CR the side
s Fig. 2 A typical compact range layout showing the reflector pattern, side (a) and wall absorber does not need to be
top (b) views. At 2 GHz, the energy incident on the side walls, floor and ceiling is more
than 40 dB down.
as thick. Figure 2 shows a typical
11
2.5 2.5 5
4
2.0 2.0
3
1.5 2
1.5
Y (m)
Y (m)
1
90
1.0 0 60
1.0
–1
0.5 0.5 –2
–3 30
–4
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 –5
(a) X (m) (b) X (m) –60 –40 –20 0
2.5 2.5 5
4
2.0 2.0 –30
3
2
1.5 1.5
Y (m)
Y (m)
1 –60
1.0 1.0 0 –90
–1
0.5 0.5 –2
–3
–4
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 –5
(c) X (m) (d) X (m)
s Fig. 3 Wave propagation from the source horn vs. time – 6.6 ns (a) 10.4 ns (b) 11.3 ns (c) 15.1 ns (d) – compared to the far
field pattern.
CR chamber. The radiation pattern where CRh is the overall height of approaches, the field (amplitude
of the CR reflector has been super- the reflector. The spacing between and phase) radiated from the AUT is
imposed over the chamber draw- the tips of the reflector and the measured on a surface, and the far
ing. The reflector in the figure pro- tips of the ceiling absorber is 2 l. field behavior is derived mathemati-
vides a 3.66 m × 1.82 m elliptical The parameter K provides a factor cally from this measurement. Three
QZ. The depth of the QZ is 3.66 m. for the spacing between the floor different near field techniques —
The important aspect of the CR is and the reflector. For the floor ab- planar (PNF), cylindrical (CNF) and
that it has a very directive pattern, sorber, we want a larger separation spherical (SNF) — represent the sur-
with directivities in excess of 25 between the edge of the reflector face where the data is measured.7-9
dBi. As Figure 2 shows, the energy and the tips of the floor absorber. The most basic near field measure-
incident on the absorber on the This reduces the angle of incidence ment approach is planar scanning,
side walls is already 40 dB below at the specular point between the where the field radiated from the
the direct path. A 1l thick absorber reflector feed and the reflector to antenna is scanned on a single
will provide 10 dB of absorption at minimize the impact of the floor re- plane. This is a good technique
over 60 degrees of incidence (see flection on the reflector illumination for high gain antennas, as there is
Figure 4 in part one, published (see Figure 4). Equation 4 includes a very small amount of energy ra-
in January 2016). Combining the K wavelengths of space between diating to the back of the antenna.
reflectivity with the difference in the tips of the floor absorber and Cylindrical scanning is where the
magnitude between the direct ray the serration tips. K should be large field is measured on the surface of a
and the reflected ray, results in a enough to provide sufficient space cylinder excluding the top and bot-
reflected energy level of approxi- for the feed positioner supporting tom surfaces. This is ideal for long
mately -50 dB. The reflector is be- the feed antenna that illuminates antennas that are omnidirectional
ing used in the near field while the the reflector. As was the case with or have a wide beam on one of the
radiation pattern of the reflector is the side walls, the absorber on the principal planes but a narrow beam
a far field concept. However, this is floor and the ceiling can be 1l in the perpendicular plane. Spheri-
an acceptable approximation, as it thick. Special consideration must cal scanning is a more general mea-
provides a method for estimating be given to the floor absorber be- surement approach. Here the field
the level of energy that radiates tween the feed and the reflector, is measured on a sphere that con-
from the reflector in the direction which may be 2l thick. In general, tains the entire antenna. In general,
of the walls. As Figure 3 shows, the absorber electrical thickness at the test distance for PNF measure-
the reflector will send some energy the lowest frequency can be t # 1.2 ments is between 3l and 10l. For
towards the side walls, estimated and t $ 0.75 for the side wall and SNF, the probe can be further away.
from the far field pattern of the re- ceiling treatments, respectively. The same equations developed
flector. for far field chambers can be used
The height of the chamber has a NEAR FIELD RANGES for SNF with the exception of the
similar equation for calculating the Different techniques are used test distance. In general, the equa-
size: for performing near field measure- tion is given by:
ments; they align with the type of
H = CR h + ( 2 + K + 2t ) λ (4) antenna being measured. With all L = d pp + ( n + 6 + 2t e ) λ (5)
12
be treated with absorber, such as
those behind the AUT. The test dis-
tance, as stated above, is between
3l and 10l. The dominant factor
for sizing a PNF range is the scan-
ner, where the scan size is given by:
–60
–90
(
L x = n + 2k tan ( θs ) λ ) (9)
–30 us is the maximum angle for ac-
–120 curate far field and nl is the electri-
0 cal size of the antenna being tested
(see Figure 5). The variable k is
–5
– 10
– 15
– 20
–150 the test distance in wavelengths;
– 25
30 hence, 3 < k < 10. The physical
scanner will usually be slightly larg-
–180 er than the scan plane. Typically, 2l
60 is the separation to the absorber
150
tips.
90 The width of the range be-
120
comes:
s Fig. 4 Absorber on the floor between the feed positioner and the reflector is criti-
cal to reduce the reflected energy from illuminating the reflector. (
W = n + 2k tan ( θs ) + 4 + 2t s λ + )
Δ scn (10)
where dpp is the depth of the probe Taking the limit as n → `, u <
(measuring antenna) and its posi- 63.4 degrees. Using the absorber which can be written as:
tioner. The variable n is the diame- approximations presented in part
ter in wavelengths of the minimum one of this series, we can estimate W=L x +( 4 + 2t s ) λ + Δ scn (11)
sphere that contains the AUT. The that ts < 2te. To do this, we check
absorber on the two end walls will the reflectivity of the end wall ab- where Dscn is the additional space
have a thickness of tel, where te is sorber at normal incidence and se- required for the scanner structure,
the thickness, in wavelengths, of lect the thickness of the absorber and ts is the thickness of the ab-
the end wall absorber. As is cus- that will provide similar reflectivity sorber.
tomary, 2l is added between the for the 63.4 degree incident angle. The length of the range is given
minimum sphere and the absorber The ceiling and the floor will have by the following equation:
tips. Finally, 4l is estimated to be the same absorber as the side
the distance between the probe walls. L = Sclr + A d + ( 4 + k + t ) λ (12)
and the sphere containing the an- The chamber height can be es- Where Sclr is the scanner depth,
tenna. timated using the following equa- which should include the spacing
The width of the SNF chamber tion: to the absorber, if any (the scan-
is given by: ner can be placed very close to
W = ( n + 4 + 2t s ) λ (6) H = h p + ( n + 4 + ts ) λ (8) the tips), and the probe length.
Ad is the depth of the AUT and
In this case, ts is the thickness, where the variable hp accounts for the support structure for align-
in wavelengths, of the side wall the height of the positioning equip- ing that antenna with the scanner.
absorber. This is a rough approxi- ment. In a typical roll-over azimuth The 4l in Equation 12 is the space
mation. For both Equations 5 and positioner used in SNF measure- between the back of the AUT and
6, a minimum of 1 meter should ments, hp should include the height the range wall. For very high gain
be added to prevent the position- of the floor slide, the azimuth po- antennas, this wall does not need
ing equipment from hitting the sitioner and the offset slide. The absorber treatment. If absorber is
probe as it rotates the antenna be- positioning equipment in the far desired, the thickness of absorber
ing measured. The chamber also field chamber equations or the CR for this wall can be as small as l/4.
should provide room for people equations (except for the feed po- The thickness of the absorber on
to work inside to set up the mea- sitioning) is not an issue because the wall behind the scanner takes
surement. This is more critical for other dimensions are so dominant
higher frequencies (above 2 GHz), in these ranges (i.e., the far field test Lx
where the 4λ separation may not distance or the reflector size).
be enough for the positioner to PNF systems use a planar scan- s k s
clear the probe. ner to measure highly directive
The angle of incidence onto the antennas (i.e., gain > 20 dB). The
side absorber is: high gain of the AUT benefits the n
14
Extending the Quiet Zone
Using an RF Lens on a
Conical Tapered Chamber
to 18 GHz
V. Rodriguez
ETS-Lindgren L.P., Cedar Park, Texas
S. Matitsine
Matsing Pte. Ltd. and Temasek Laboratories, National University of Singapore
T.T. Chia
DSO National Laboratories and Temasek Laboratories, National University of Singapore
P. Lagoiski, L. Matytsine and M. Matytsine
Matsing Pte. Ltd.
P.K. Tan
Temasek Laboratories, National University of Singapore
A
tapered chamber is traditionally con- to the turntable, the quiet zone at 8 GHz
structed using a square based pyra- is reduced to 45 cm. A solution to extend
midal shaped taper that transitions the quiet zone at high frequencies employs
to an octagon and then finally a large dielectric lens installed in front of
into a cylindrical launch sec- the turntable to improve the phase distribu-
tion. This approach is related tion of the field. A lightweight, broadband
to the manufacturability of dif- lens with a diameter of 2 m was developed
ferent absorber cuts. In this ar- and weighs just 35 kg with a focal length
ticle, we introduce a chamber of 10 m. With the lens installed, the usable
where the conical shape of the far field QZ is increased, allowing electri-
launch continues throughout cally larger antennas to be measured in the
the entire length of the tapered chamber. The use of the lens can also be
chamber. The results of free applied to traditional square cross-section
s Fig. 1 A typical tapered anechoic space VSWR measurements tapered chambers.
chamber.
at different frequencies over a
1.5 m diameter quiet zone (QZ) BACKGROUND
are presented. The conical ta- Tapered anechoic chambers have been
per appears to have a better around for almost 50 years,1,2 introduced
illumination wave front than to address issues present in rectangular
the traditional approach and chambers at frequencies below 500 MHz.3,4
better QZ VSWR levels even at At lower frequencies, high gain antennas
frequencies above 2 GHz. used in an antenna measurement range be-
As with all antenna cham- come physically large and can be difficult to
bers, when the frequency in- handle inside an anechoic chamber, so less
creases, the usable or far field directive antennas are used. These radiate
illuminated QZ is reduced. At more energy to the side walls, ceiling and
s Fig. 2 Shaping from square to oc- a 12 m distance from the feed floor of the chamber forcing it to grow in
tagonal cross-section at the feed.
15
EHP-24PCL
Absorber
EHP-12PCL A
Absorber
Scanner
EHP-24PCL
Absorber B
AUT EHP-24WW Absorber
Positioner
D
CRV-60PCL
Absorber Custom EHP-18WGCL
Absorber on Taper
EHP-36PCL
C Absorber
(a) (b)
s Fig. 3 The conical tapered anechoic range plan and elevation (a), and a picture of the taper section (b).
size in order to accommodate thick- make the transition from the coni- with attenuators at the feed to re-
er absorbers needed to reduce re- cal section to the square section as duce the effects of the high VSWR.
flections. Tapered anechoic cham- smooth as possible and to create The QZ is scanned with an ETS-
bers were introduced to solve this the illumination in the QZ.2-5 Lindgren model 3106B dual ridge
low frequency problem. Instead of The design presented in this horn. The scanning antenna and
trying to eliminate specular reflec- article introduces a conical taper source antenna are shown in Fig-
tions in the quiet zone, the specular (see Figure 3). The entire tapered ure 5.
area is brought closer to the mea- structure maintains a constant Figure 6 shows reflectivity levels
suring antenna and the specular re- angle and a circular cross section. of the QZ versus direction for hori-
flections are used to create a QZ il- The tapered section is about 10 m zontal and vertical polarizations.
lumination.2,5 Traditionally, tapered in total length. Results for the free- Results are shown for 200, 400,
anechoic chambers were built hav- space VSWR6 are presented and 800 and 1,000 MHz. All of these re-
ing a square based pyramid as the compared with similar chambers sults are measured with the source
taper (see Figures 1 and 2). To bet- employing a traditional design. To antenna at a fixed position in the
ter accommodate different feed improve performance at high fre- apex of the taper. The antenna is
antennas, the square section may quencies a dielectric lens is used to commonly moved as frequency
be gradually transformed to a cy- create a plane wave behavior.7 changes to maintain the phase
lindrical cross section taper. These center close to the reflections and
changes in cross section require a MEASURED RESULTS maintain a QZ illumination free of
lot of special cuts of absorber to The chamber QZ is scanned us- ripples.3,4
ing free space VSWR tests6 at a se- For frequencies above 2 GHz,
ries of frequencies from 200 MHz to an ETS-Lindgren 3164-05 dual
18 GHz. The chamber is lined with linearly polarized open boundary
60” (152 cm) curvilinear absorbers quad-ridge horn, rated from 2 to
on the back (i.e., receive) wall and a 18 GHz, is used. For scanning the
combination of 24” (61 cm) pyrami- QZ, a series of standard gain horns
dal absorbers and 36” (91.44 cm) are used with gains ranging from
on the sidewalls, floor and ceiling. 10 to 20 dBi. Additionally, since a
The tapered section has a specially smaller horn is used as the source,
cut wedge material that lines the it is positioned inside an extension
s Fig. 4 The tapered section as seen tapered section from the feed lo- of the conical taper. Figure 4 shows
from the feed location. cation to the QZ area. The wedges one of the two halves that make up
range from 18” (45.72 cm) at the this high frequency extension.
QZ end to 8” (20.32 cm). Figure Figure 7 shows the results of
4 shows a picture of the conical the scans at high frequencies. As
treatment. The tapered section is discussed by Rodriguez and Han-
built inside an RF shielded room to sen,5 tapered chambers are better
avoid outside interference during suited for low frequencies and care
measurements. must be taken to properly position
The source antenna is an ETS- the source antenna. However, it is
Lindgren model 3164-06 dual lin- possible to use them at these high
early polarized open boundary frequencies once the chamber is
quad-ridge horn,8 rated from 300 characterized.
s Fig. 5 The scanning antenna at the MHz to 6 GHz. In this application,
QZ viewed from a point right behind the
source antenna at the apex of the taper.
the antenna is used from 200 MHz
16
COMPARISON WITH either side of the source antenna.
200 MHz Horizontal 200 MHz Vertical
TRADITIONAL CHAMBERS These variations are not seen in the
–30 dB limit Comparison with traditional conical tapered chamber.
Measured Reflectivity Data chambers is difficult. There are no The chamber is configured with
270 two identical chambers that have the two ranges, a far field tapered range
exact same absorber treatment with and a NF-FF planar and spherical
225 315
the exception of the taper geom- range. Figure 3 shows the plan of
etry. A qualitative comparison, how- the chamber with the two ranges.
ever, suggests some advantages. The antenna under test uses the
180 0
With traditional chambers, anten- same positioner for both ranges,
–60
–55
–50
–45
–40
–35
nas with gains of 16 dBi and above and the QZ is the same as well. For
are required to achieve adequate the spherical range the probe is lo-
135 45
illumination in the QZ. It appears cated between the QZ and the pla-
90 that one of the features of the coni- nar scanner on the opposite wall.
(a)
cal taper is that lower gain antennas The planar scanner can be used for
400 MHz Horizontal 400 MHz Vertical can be used. At 10
2,000 MHz Horizontal 4,000 MHz Horizontal
–36 dB limit GHz, the source 2,000 MHz Vertical 4,000 MHz Vertical
Measured Reflectivity Data antenna has a –42 dB limit –42 dB limit
270 directivity of 12 Measured Reflectivity Data Measured Reflectivity Data
dBi,8 whereas the 270 270
225 315
conical quad ridge 225 315 225 315
horn used in many
180
traditional tapered
0
anechoic cham- 180
–60
–55
–50
–45
–40
–35
0 180 0
–80
–70
–60
–50
–40
bers has a directiv-
–80
–70
–60
–50
–40
ity of 14 dBi. The
135 45
open boundary 135 45 135 45
90 ridge horn is suc-
(b) (a) (b)
cessfully used in 90 90
the conical cham- 10,000 MHz Horizontal 18,000 MHz Horizontal
800 MHz Horizontal 800 MHz Vertical
–39 dB limit
ber design; how- 10,000 MHz Vertical 18,000 MHz Vertical
–42 dB limit –42 dB limit
ever, when used in
Measured Reflectivity Data Measured Reflectivity Data Measured Reflectivity Data
270 a traditional cham- 270 270
225
ber, a smooth am-
315 225 315 225 315
plitude taper is not
achieved (see Fig-
ure 8).
180 0 In Figure 9, a 180 0 180 0
–80
–70
–60
–50
–40
–80
–70
–60
–50
–40
–60
–55
–50
–45
–40
–35
comparison of the
reflectivity of the
135 45 conical tapered 135 45 135 45
90
chamber and a tra- (c) 90 (d) 90
(c) ditionally imple-
mented chamber s Fig. 7 Reflectivity levels in the QZ versus angle at 2 (a), 4
(b), 10 (c), and 18 GHz (d).
1,000 MHz Horizontal 1,000 MHz Vertical at 400 MHz shows
–42 dB limit
a slight difference
Ref –34 dBm Atten 0 dB
Measured Reflectivity Data
270
in back wall re- Peak
flectivity (180°), Log
225 315 but this is related 5 dB/
to differences in
absorber treat-
The different curves are from
180 0 ment between the moving the antenna back and forth
–60
–55
–50
–45
–40
–35
17
400 MHz Horizontal Planar Near Field Scanner
400 MHz Vertical 5305 mm
–30 dB limit
400 MHz Horizontal Traditional Antenna Stand for
Planar Near Field and Spherical Near Field
400 MHz Vertical Traditional
Far Field and Spherical Near Field Positioner
Measured Reflectivity Data
270
8000 mm
225 315
2000 mm
Lens
180 0 16097 mm
–80
–70
–60
–50
–40
–80
–70
–60
–50
–40
135 45
s Fig. 10 Placement of lens in tapered chamber.
ple, the tapered chamber installed high ratio of the size of the planar 52.0
48.0
44.0
at the National University of Singa- wave front to the lens diameter (a 40.0
36.0
pore (NUS) has a quiet zone of 1.4 factor of about 0.7D, where D is
m at 500 MHz but only 45 cm at the diameter of the lens). Hence, a
8 GHz. To increase the quiet zone 2 m diameter lens can produce a
at the higher frequencies, a cus- 1.4 m plane wave front. The profile
tom RF lens is integrated inside the P(x,y) of the lens is designed using
chamber. We are not aware of any the following equations from Kraus (a)
other method to increase the quiet and Marhefka:9
zone without physical alterations to Z E-Field Phase (deg)
200.0
160.0
xx == r −− y ,,
40.0
(1)
–40.0
m –200.0
where
where
y y=ρρ sin
y= sin θθ,, (2)
(2)
and
and
r P(x, y) ρρ=
(
ff εε −− 11 ) (3)
= ε cos θ − 1' (3) (b)
m ε cos θ − 1'
with
with
x
r ⎛ 1 1 ⎞
hh == r ⎜⎛ 1 −− 1 ⎟⎞
h
(4)
(4) s Fig. 12 Predicted field distribution at
⎜ sin θθm
εε −− 11 ⎝⎝ sin tan θθm
tan ⎠⎟ 8 GHz, both in magnitude (a) and phase
s Fig. 11 Lens geometry. m m⎠
(b).
18
manufactured by Matsing Pte Ltd. (corresponding to the quiet zone 10
is used. The material allows the region) on the other side of the 8
No Lens
With Lens
control of the dielectric permittiv- lens. For simplicity, the lens and the
ity to a high degree of accuracy. It dipole are simulated in free-space 6
has extremely low-loss (ε’ < 10-4). without the tapered chamber since 4
MAGNITUDE (dB)
Its low density (40 kg/m3) means the primary aim of the simulation is 2
that the 2 m lens weighs only 35 to ensure that for the given length
kg, making it portable and easily of the taper, the lens provides the 0
installed. The material is also iso- best possible illumination. Includ- –2
tropic and broadband, making the ing the chamber with its absorb-
–4
lens suitable for both vertical and ers in the simulation model would
horizontal polarizations over a wide drastically increase the problem –6
range of frequencies. size and complexity beyond the ca- –8
pability of the numerical package
–10
NUMERICAL ANALYSIS at these high frequencies. –1 –0.5 0 0.5 1
The performance of the lens is Figure 12 shows the predicted TRANSVERSE DISTANCE (m)
first evaluated using FEKO EM sim- fields (for a quadrant) at 8 GHz. 80
ulation software. A half-wavelength The circles in the plots represent No Lens
With Lens
dipole is placed at the focal length the outline of the 2 m lens. Cuts of 60
of the 2 m lens. The focal length the fields along the lens diameter
corresponds to the distance (10 m) are shown in Figures 13 and 14 40
between the feed and aperture of for 2 and 8 GHz, respectively. The 20
PHASE (deg)
the tapered chamber. The field is fields of the dipole in the absence
observed at a vertical plane at 2 m of the lens are superimposed in 0
the figures for reference. For ease
of comparison, the magnitudes –20
19
10 enhance the performance of the
10
No Lens 8 No Lens tapered chamber. Its light-weight
8 With Lens
W/Lens construction enables easy user
6
6 installation. The NUS tapered
4
4 chamber with an RF lens is now
MAGNITUDE (dB)
MAGNITUDE (dB)
–4 –4 References
1. W.H. Emerson and H.B. Sefton,
–6 –6
“An Improved Design for Indoor
–8 –8 Ranges,” Proceedings of the IEEE,
–10 –10
Vol. 53, No. 8, August 1965, pp.
0 0.3 0.6 0.9 1.2 1.5 0 0.3 0.6 0.9 1.2 1.5 1079-1081.
TRANSVERSE DISTANCE (m) TRANSVERSE DISTANCE (m) 2. H. King, F. Shimabukuro and J.
80
80 Wong, “Characteristics of a Ta-
60
No Lens 60 No Lens pered Anechoic Chamber,” IEEE
W/Lens With Lens
Transactions on Antennas and
40 40 Propagation, Vol. 15, No. 3, May
1967, pp. 488-490.
20 20
PHASE (deg)
3. L. H. Hemming, Electromagnetic
PHASE (deg)
0
Anechoic Chambers: A Fundamen-
0
tal Design and Specification Guide,
–20 –20 Wiley-Interscience, John Wiley and
Sons and IEEE Press, Piscataway,
–40 –40 N.J., 2002.
–60
4. V. Rodriguez, “Using Tapered
–60 Chambers to Test Antennas,” Eval-
–80 uation Engineering, Vol. 43, No. 5,
0 0.3 0.6 0.9 1.2 1.5 –80
0 0.3 0.6 0.9 1.2 1.5 May 2004, pp. 62-68.
TRANSVERSE DISTANCE (m) TRANSVERSE DISTANCE (m) 5. V. Rodriguez and J. Hansen, “Eval-
s Fig. 17 Measured field distribution at 2 s Fig. 18 Measured field distribution at 8 uate Antenna Measurement Meth-
GHz. GHz. ods,” Microwaves and RF, October
2010, pp. 61-67.
sults. Note that the transverse dis- CONCLUSION 6. R.E. Hiatt, E.F. Knott and T.B.A.
tance in these figures, unlike that of This article introduces a new ap- Senior, “A Study of VHF Absorb-
Figures 13 and 14 is relative to the proach to manufacturing tapered ers and Anechoic Rooms,” Techni-
start of the measurement position anechoic chambers that provides cal Report 5391-1-F, University of
at 0 m. The plots show that the lens good QZ reflectivity results over Michigan, February 1963.
has indeed improved the phase sig- wide frequency ranges. Addition- 7. V. Rodriguez, S. Matitsine, T.T.
nificantly without adversely affect- ally, it appears to allow the use Chia, P. Lagoiski, L. Matytsine, M.
ing the amplitude. The size of the Matytsine and P.K. Tan, “A Cone
of lower directivity antennas than
quiet zone for ±10° phase variation Shaped Tapered Chamber for An-
the ones used in traditional cham- tenna Measurements Both in Near
with and without the lens is summa- bers. A lower directivity antenna Field and Far Field in the 200 MHz
rized in Table 1, demonstrating that provides smaller amplitude tapers to 18 GHz Frequency Range and
the lens has significantly improved across the QZ, reducing errors Extension of the Quiet Zone using
the phase performance of the ta- during gain measurements. With an RF Lens,” Journal of the Applied
pered chamber. Measurements are the addition of an RF lens, the Computational Electromagnetic
also done from 500 MHz to 1 GHz phase of the chamber’s quiet zone Society, Vol 28, No. 12, December
to confirm that the lens does not af- at higher frequencies (2 to 10 GHz) 2013, pp. 1162-1170.
fect the original quiet zone of the is significantly improved. The lens 8. ETS-Lindgren horn, www.ets-lind-
chamber at low frequency. gren.com/pdf/3164-05.pdf.
provides a quick and easy way to
9. J.D. Kraus, R.J. Marhefka, Anten-
nas for All Applications, 3rd edition,
TABLE I McGraw-Hill, 2001.
SIZE OF QUIET ZONE (IN CM) FOR ±10° PHASE VARIATION Vicente Rodriguez attended Ole Miss,
WITH AND WITHOUT LENS in Oxford Miss., where he received his
B.S.E.E in 1994, his M.S. in 1996 and
f (GHz) 2 4 6 8 10 Ph.D. in 1999. Dr. Rodriguez joined ETS-
Lindgren as an RF and Electromagnetics
engineer in 2000. In 2004 Dr. Rodriguez
Without lens 95 65 55 45 40 became senior principal antenna design
engineer, placing him in charge of the
With lens 140 140 140 140 140 development of new antennas. In 2006
20
Dr. Rodriguez became antenna product and technical director of Matsing Pte. position of engineer at Matsing Pte. Ltd.
manager, placing him in charge of Ltd. His research interests include His research interests include antenna
development, marketing and maintenance electromagnetic materials, metamaterials, measurement system and RF lenses.
of the antenna product line. Dr. smart materials, multi-beam antennas,
Rodriguez is the author of more than 50 antenna measurement techniques, and Leo Matytsine received his B.S. from
publications and holds patents for hybrid most recently, lightweight, large-size RF the University of Southern California in
absorber and dual ridge horn antennas. lenses. He has more than 60 publications 2009 and his MBA from the Australian
in these areas, as well as four patents. Global School of Management in 2013.
Serguei Matitsine graduated with He has been with Matsing Pte. Ltd. since
honors from the Moscow Institute of Tse-Tong Chia received his B.Eng. 2009 and currently holds the position of
Physics and Technology in 1979 and degree with first class honors in director. His interests include RF convex
received his Ph.D. in 1982. From 1982- 1986 from the National University of lenses and antenna measurement systems.
1984 he held the position of senior Singapore, and his M.S. and Ph.D. in
researcher at the Institute of Radio- 1991 and 1994, respectively, from Ohio Michael Matytsine received his B.A.
Engineering and Electronics of the State University. He has been with the from Chapman University in 2006 and
Russian Academy of Sciences. From DSO National Laboratories in Singapore his MBA from La Verne University in
1984 until 1995 he has held several since 1986, where he is currently a 2010. He has been with Matsing Pte.
positions including senior researcher, distinguished member of the technical Ltd. since 2006 and currently holds the
head of the electromagnetic laboratory staff. Chia was a laboratory head from position of director. His interests include
and deputy director at the Institute of 1995 until 2010 when he stepped down to Luneburg and convex RF lenses.
Theoretical and Applied Electromagnetics focus on research. He is currently also a Peng-Khiang Tan received his degree
of Russian Academy of Sciences. In principal research scientist in the Temasek in electronic and computer engineering
1995 Dr. Matitsine joined the research Laboratories at the National University from Ngee Ann Polytechnic in 1999 and
and development group at Singapore of Singapore. His research interests his Bachelor of Technology in Electronics
Technologies Aerospace as technical include computational methods for Engineering (second Class Honors) from
director and later moved to the position of electromagnetic scattering and installed the National University of Singapore in
chief engineer. Since 2001 Dr. Matitsine antenna performance, as well as the use of 2008. He currently works as a laboratory
has also been working at Temasek lenses for antenna applications. technologist within the antenna group at
Laboratories of the National University of the Temasek Laboratories of the National
Singapore as an adjunct senior principal Pavel Lagoiski received his B.S. E.E.
from the National University of Singapore University of Singapore.
research scientist. He is also the chairman
in 2010. Since then he has held the
24
Practical Antenna
Connection for Accurate
Testing
Clayton Karmel, Pdicta Corp, San Diego, CA and Ben Maxson,
Copper Mountain,
Indianapolis, Ind.
C
onnecting RF test and measurement advice for such test and measurement sce-
equipment to an Antenna Under narios.
Test (AUT) usually involves tradeoffs
among measurement accuracy, electrical Tools of the Trade
considerations, cost, and mechanical rug-
gedness. In this article, we describe some TESTING VIA AN RF TEST
aspects of this connection with practical CONNECTOR: U.FL VS COAXIAL
SWITCH
When provided, an antenna test port will
usually be at a 50 ohm point. If the design
includes a coax switch for inspection
and final test purposes, an
antenna test sample can
be produced or reworked
with the switch rotated
180°. This reverses the in
and out terminals so con-
necting a test cable inter-
rupts the feedline and the cable faces the
antenna instead of the device.
A test port and test cable of the appro-
priate type reduces the need for strain relief
as the cable can be connected just before
measuring. A right-angle connector at the
device end may further reduce stress and
allow easier routing through the device
housing. Note that most ultra-miniature
connectors have a mating life on the order
of a few dozen cycles, but failure will be-
SMA to RF switch, SMA to U.FL (both sexes), SMB, SMA bullet and SMA come self-evident in the measurements.
to U.FL short cable, and a Vector Network Analyzer.
25
at the right edge of the Smith
chart.
• Create a home-brew set of Cali-
bration standards. Take three
sparsely populated boards;
add 50 Ohm shunt, 100pf shunt
(short) and one left open.
• If the reference plane isn’t ex-
tended by either custom calibra-
tion standards or by manually
extending the reference plane,
then matching efforts will be ex-
tremely confused because the
Smith chart will be rotated.
This test board omitted the
WiFi IC, installing a U.FL connec-
tor, instead. We made four of these
boards, one each configured for
Load, Open, Short and Thru to the
un-matched Antenna.
Now we can calibrate to the
entry point of the antenna with
An RF Switch allows production testing the short, open and 50 ohm load.
of the WiFi module. Turn it around to In production, there are no extra
tune the antenna. But it adds cost to components or space required.
every unit. Alternatively, if we had used
testing either the WiFi module or coaxial calibration standards (N or
SMA), we would subsequently “ex-
the antenna. tend” the reference plane by an ad-
CUSTOM CAL STANDARDS justment in the VNA, usually called
Port Extension. How far? Use the
Lumped element tuning for PCB Open and Short configurations and
antennas requires that your VNA extend the reference plane until
U.FL connector sits on a stub with the calibration is in the proximity of the these align with the right and left
same length as the trace to the WiFi next series or shunt element to be edges of the Smith Chart.
module. In production, omit the U.FL added. You have options:
connector. If your instrument and cabling
• Use a co-axial calibration stan- mass is substantial compared to
The example PCB on the left has dard, usually to an SMA inter-
a U.FL connector on a very short the counterpoise (ground-plane) of
face, and then extend the ref- the board, add a DC-DC connec-
stub. One can isolate the WiFi de- erence plane across the length
vice by omitting the 100pf cap and tor prior to calibration. When you
of your board connection. This start to match, move the reference
installing the U.FL to obtain coaxial extension may compensate for
access to the antenna. Alterna- plane (Port Extension) backwards,
an SMA-U.FL adaptor, or maybe bringing the reference plane back
tively, and as shown at right, an RF a short cable. Add reference
switch can be installed allowing for with you as you add tuning ele-
plane time until the ‘Open’ is
ments.
TUNING A LOOP
The keyfob design below was
based on a radio chip with a dif-
ferential output. The round shape
of the board and differential drive
suggested a loop antenna which is
less susceptible to de-tuning than
a monopole, dipole or folded F.
Dielectric detuning happens when
the keyfob is held by different
hands or in different orientations or
sits on different surfaces.
On the right is the production
board with differential RF feed at
Application-Specific RF test board the top of the image. On the left
26
is a test board with the same an- in design, or maybe the product Open/Short/Load values in and out
tenna and same ground-plane. requirements were just too tight. of place.
Notice tuning elements both at the Here are some coaxial cable op- 50 Ohm matches are increas-
entry of the differential drive and tions if you need to forge a reliable ingly typical at the output of mixed
at the mid-way point. L5 enables transmission line connection. signal RF SOC IC’s, but if not, re-
the loop to be extended, in case First, you will most likely want to view your options at your duplexer,
simulations (or back-of-the-hand identify a position on your board diplexer, balun or within your RF
guesses) are off when judging loop where the impedance is, or is ex- matching circuit to the antenna.
size/characteristics. pected to be, 50 ohms. This isn’t al- The ideal test point has close by ac-
Notice Y2. It’s a 1:1 balun, con- ways required, but simplifies things. cess to the antenna’s ground plane,
figured so that a single-ended RF Remember that you still face two preferably one well reinforced with
interface can drive the loop dif- calibration/reference plane options vias.
ferentially. Look up the radio IC’s – calibrating with normal standards Whether your coaxial cable at-
component output match and tune at the ‘connector’ end of your co- tached with “horizontal” or “verti-
the antenna impedance to the axial cable, then extending the ref- cal” launch, the goal is to maintain
complex conjugate of that match. erence plane, or building multiple the TEM modes that exist between
Remember to extend the reference boards and implementing a cus- the center conductor and shield of
plane to the lumped elements first. tom ‘calibration set’. the test cable the feedline microstrip
The latter option can be a prob- and its ground plane through the
TESTING WITHOUT A COAXIAL lem in this case, however, because transition between them.
CONNECTOR it requires that your coaxial cable A 50 ohm microstrip feed line
Sometimes there just isn’t an have the same electrical length may be as little as 0.03 to 0.1 mm
opportunity to have a U.Fl or Coax- on each of 3-4 boards, or that (.015” to .025”) wide if over an
ial switch. Maybe it was overlooked you have to regularly solder your embedded ground plane of a thin
circuit board assembly. Protect-
ing such a thin trace from being
pulled up by the test cable requires
a mechanically sound connection
between the cable sheath or shield
and the ground plane, preferably in
an area where the ground plane is
reinforced by vias.
HORIZONTAL LAUNCH
For a horizontal launch, to the ex-
tent possible the test cable should
access the broken microstrip feed-
Small board example RF Test Circuit line trace directly from the ground
plane and in the direction of the
feedline. Keep the feed line and
center conductor and their separate
ground return currents close by, in
proximity, parallel to each other,
and minimally different in length.
Avoid 90° connections and bends
until both currents are safely within
the test cable. Eisenhart’s “edge-
launch design” can be approximat-
Small board example, loaded as Calibration ‘Load’ ed even if the test point is not on
the edge of the circuit board.
VERTICAL LAUNCH
For a vertical launch, it may be
practical to break the feedline trace
and reflow solder a U.FL or simi-
lar surface mount connector to the
feedline at the break and the ground
plane on both sides. This gives the
ruggedness advantage of the con-
nector, albeit at the presence of
Small board example, loaded as Calibration ‘Short’ some reflection. If this isn’t practi-
27
the cable result. This can be done
by forming a 1-turn loop around
the minimum bending radius for
a semi-flexible or hand-formable
test cable, or with a ferrite bead
or two placed over the test cable
where it departs from the ground
plane. While ferrite beads are not
characterized above 500 MHz and
Small board example, loaded as Calibration ‘Open’ all of them lose performance as
frequency increases, they are in-
expensive and unlikely to impair
the test results. Type 61 ferrites are
usually recommended above 300
MHz but Type 43 is available in a
greater range of dimensions and
may work.
A bead can be selected for fit
and evaluated by placing a sample
at the shorted end of a coax stub
and extending the R54 port to that
Small board example, Measuring the Antenna raw match with extended reference point. As a useful example, the
plane. Kemet B-20L-44 RG-178 stub dem-
onstrated |Z| in excess of 58 ohms
out to 3 GHz, which includes the
GSM, UMTS, and LTE to Band 7.
It was in excess of 30 ohms out to
5GHz. Any common mode current
from the DUT can be arbitrarily re-
duced by stacking multiple beads.
Test Cable
OK to
Bend
Eisenhart
Edge Launch
Clearance
Dielectric
Hole
Sharp Flare
in Braid
29
Compact 2-Port USB VNAs up to 8.5 GHz Frequency
Copper Mountain Technologies’ popular Com-
pact Series delivers full-featured lab-grade
performance in a compact package suitable
for many research and production applica-
tions. These instruments are a great value
solution for magnitude and phase measurements from 9 kHz to 8.5 GHz and
an option for 75 Ω measurements to 3 GHz.
http://www.coppermountaintech.com/products/Ohm-50/category/11/
30
EXTEND YOUR REACH
Diamond Engineering has integrated the
CobaltFx Frequency Extension System into
their antenna measurement system.
COST-EFFECTIVE
C4209
Available options are 2- and
4-port VNAs with a maximum
MM-WAVE SYSTEM
frequency of 9 GHz or 20 GHz CobaltFx Frequency Extension System is an unprecedented
approach for millimeter-wave S-parameter measurements, suited for
antenna range measurements, automotive radar & sensor testing,
material characterization, and many other mm-Wave applications.
The system is anchored by a Cobalt Series USB VNA
Three dedicated waveguide Includes frequency extension modules and all necessary
bands, 50-75 GHz, 60-90 accessories in collaboration with Farran Technology
GHz and 75-110 GHz
In collaboration with:
Visit the websites below for
more information on the
CobaltFx System:
cpmt.link/cobaltfx or
cpmt.link/farrancobaltfx www.farran.com
Read the white paper from Frost & Sullivan about the high-performance
Millimeter Wave VNA System: http://cpmt.link/fswhitepaper
www.coppermountaintech.com