IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. IM-21, NO.

4, NOVEMBER 1972 421

FREQUENCY GHz
Fig. 14. Frequency response of RF probe.

Fig. 16. General Microwave model 435 radiation hazards meter.

Fig. 16 shows a view of the probe with the instrumentation

package, which is battery operated. The full-scale ranges ex-
tend from 0.2 to 200 mW/cm2, with an overload capability in
excess of 1 W/cm2. Due to the remarkable stability of the
TIME t/To
probe, meaningful measurements can be made at levels as low
Fig. 15. RF probe response to step of ambient temperature.
as 1 0 MW/cm2 . The measurement time is approximately 1 s.
REFERENCES
measurement is in excellent agreement with the sensitivity [1] E. E. Aslan, "Electromagnetic radiation survey meter," IEEE
measured at the RF frequencies and thus can serve as a con- Trans. Instrum. Meas, vol. IM-19, pp. 368-372, Nov. 1970.
venient method of calibration of the probe. In addition, it is [2] S. Hopfer, N. H. Riederman, and L. A. Nadler, "The properties of
thermo-electric elements as microwave power detectors," 1962
seen to confirm the plane-wave approach to the determination IREInt. Conv. Rec., pt. 3, pp. 77-84.
of cross sections at frequencies above f, [3] K. Fletcher and D. Woods, "Thin-film spherical bolometer for
measurement of hazardous field intensities from 400 MHz to
Fig. 15 shows the response of the probe to an ambient step 40 GHz," Non-Ionizing Rad., Sept. 1969.
of 40°F, together with a plot of (22). Although the latter is [4] N. Marcuvitz, Waveguide Handbook, vol. 10, M.I.T. Rad. Lab. Ser.,
p. 284, eq. (lb).
based on the simplified assumption of a two-time constant [5] A. T. Adams, Syracuse Univ., private communication.
model, it is seen to be a good approximation of the phys- [6] H. P. Schwan, "Microwave radiation: Biophysical considerations
and standards criteria," IEEE Trans. Biomed. Eng., vol. BME-19,
ical situation. pp. 304-312, July 1972.

Broad-Band Isotropic Electromagnetic

Radiation Monitor
EDWARD ASLAN

Abstract-A probe using thin-film techniques to construct an array of r HIS electromagnetic radiation probe responds equally to
extremely lossy dipoles makes possible an electromagnetic radiation ]L all polarizations and combinations of polarizations. It
monitor, having a 3-dB bandwidth from 300 MHz to 18 GHz, isotropic has a spherical response patter, measuring equally energy re-
response, and independence of poarization. The measurement range is cie f an reton eept oeasmalls ane tha
from 20 AW/cm2 to 20 mW/cm2. ceived from any direction except over a small solid angle that
centers on the handle. The probe has 3-dB bandwidth fre-
quencies from below 300 MHz to greater than 18 GHz. The
Manuscript received June 1, 1972- revised June 28, 1972. MW/C2 average
The author is with the Narda Microwave Corporation, Plainview, full-scale power density measurement is 20 mW/cm
N.Y. 11803. with a 20-W/cm2 peak power capability and the dynamic

Authorized licensed use limited to: Tubitak Ve Birimleri. Downloaded on February 19,2021 at 13:03:12 UTC from IEEE Xplore. Restrictions apply.
422 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, NOVEMBER 1972

range of the instrument is 20 or 30 dB below these full-scale

values, dependent upon instrumentation.
The probe consists of three mutually perpendicular lossy CI
probe elements. Each probe element provides a dc output
signal that is proportional to the square of the electric field
E-Leff
strength tangential to the element. The summation of the dc
signals from the three orthogonal probe elements provides a Li
measure of the total energy or power density, independent of
direction or polarization of the RF signals. The detectors for
the dc signal generation are thin-film thermocouples that pro-
vide true square-law outputs. The dc signal is proportional to
RI
the power dissipated in the thermocouple elements and the
average energy density in the volume in which the elements
are contained.
The broad-band characteristics are obtained by distributing
R2
resistive thermocouple dipoles along the length of the element
at spacings that will permit no resonant lengths over the range Fig. 1. Lumped equivalent circuit of the probe with single elements rel-
ative to the total probe. R2 representing one of the distributed
of frequencies within which the probe is intended to operate. thermocouples.
The spacing is less than IX of the highest frequency to be
measured. The probe may be viewed as a group of series-
connected small resistive dipoles or as a very low Q resonant In the high-frequency region the relative output,
circuit.
A lumped equivalent schematic representation of a probe ele- fS2)
ment is shown in Fig. 1, where R2, C2 represent a terminal
load of one of the probe elements (center-fed dipoles). Ll,
C1 are the lumped equivalent inductance and capacitance of [(R1R2)2 + (R1 +R2)2 (OoC2)-2]
the probe element determined from the average characteristic
impedanceZ0 [1]. [R1R2 + C2] + [(R1 + R2) (o2 C2)Y - W2L1R2]2
ZO = 120 [ln (2HIa) - 1], The 3-dB bandwidth frequencies for the probes are less than
300 MHz and greater than 18 GHz. The calculated value of
where S1 at 300 MHz is 0.59 and this was confirmed by the mea-
H half length of the probe element, sured value for six probes, which averaged 0.575, with the in-
a equivalent radius. dividual probes ranging from 0.52 to 0.63. The calculated
value of S2 is 0.67. The average of the measured values of
Since the elements are thin-film strips, a is equal to 4 of the
S2 on the same six probes was 0.63, ranging from 0.56 to 0.67.
average width of the probe element [2]. Over the midband region from 1 to 12 GHz, there is a ± 2 -dB
C1 = 2(7r2fOZO)1 L 1 = Zo (8fo)-1. ripple on the frequency-sensitivity curve. The I-dB bandwidth
frequencies are at 850 MHz and 15 GHz. A typical frequency-
fo is the resonant frequency of a dipole of the same length as sensitivity curve is shown in Fig. 2. The maximum variation
the element. R1 equals the total resistance of the probe ele- in indicated power density, resulting from variation of the po-
ment less R2 . C2 is the shunt capacitance across a small dipole sition or attitude of the probe in a plane-wave field, is less
and can be determined from the geometry of Fig. 3 [3] as than ± 4 dB. The probe also responds to all polarizations with
a maximum deviation of less than + dB.
C2 re [/7rLmn4D-_W\ L The use of two orthogonal dipoles to obtain a probe inde-
pendent of polarization was previously described by the
author [4] and the use of three orthogonal dipoles to obtain
The dc output of the small thermocouple dipole is propor- an isotropic response was previously described by Bowman [5].
tional to the power dissipated in it. The dc output is the low- However, the three orthogonal probe elements used in the
frequency region, relative to the dc output in midband is S1. present probe show little similarity to the prior disclosure.
They are resistive thin-film elements that are folded back or
paralleled upon themselves with the output terminals at one
SI, (fo) end, as illustrated in Fig. 3. The three probe elements may be
contained within a 2-in spherical surface, that may in turn be
[(R 1R 2)2 +(R1 +R2)2 (CAoC2)2] contained within a 4-in-diameter sphere of foamed poly-
[R 1R2 -+(C1 C2)12 + [(R +R2)(CO1C2)_1 -R2AW.C1)']2I styrene with the apparent center of the probe element array at

Authorized licensed use limited to: Tubitak Ve Birimleri. Downloaded on February 19,2021 at 13:03:12 UTC from IEEE Xplore. Restrictions apply.
 .3 .5 .75
t\ ~ ~_ t~ ~ ~ ~ ~-.25
ASLAN: ELECTROMAGNETIC RADIATION MONITOR

i
, . .

2
,

4
.

6
.

8 lo
, l, ..~~-
14 18
dB
423

FREO GHz
Fig. 2. Typical frequency response characteristics.

[- Fig. 4. Relationship of the three mutually orthogonal probe elements.

Probe elements are contained within a 5-cm sphere.

iSr 6~ )
k4cojS

Fig. 3. Antenna probe element with the distributed thermocouple film

elements. Tapered film leads reduce interaction between leads and
elements. Geometry in the section view determines C2.

the center of the larger sphere. This establishes a 5-cm spacing

from the outer surface of the sphere to conform to the Bureau
of Radiological Health specification for the measurement of
leakage from microwave ovens. As shown in Figs. 3 and 4, the
ends rather than the centers of the three probe elements inter-
sect at a point. This orientation makes the apparent center of
the array shift approximately 1.5 cm as the angle of incidence
of the energy varies from zero to 450 with the axis of the Fig. s. Two of the three film elements may be seen in this internal view
array. The handle lies along the axis of the array and is at a of the probe.
mutual angle of 54.70 to each of the elements. The orienta-
tion is such that the 5-cm spacing to the center of the array is The spacing between the cold junctions is a small fraction of
maintained when the incident energy is either aligned with the a millimeter. The close spacing permits little zero drift due to
axis or normal to it. The variation in apparent center is of no ambient temperature since a very small temperature gradient
consequence when the probe is remote from a source. will occur due to variation in ambient temperature. Variation
The probe elements are specifically composed of thin films in the ambient temperature will cause variation in sensitivity
of overlapping antimony and bismuth deposited upon a thin that is less than 0.05 percent/°C and does not degrade the
plastic substrate. The geometry creates altemate cold and hot basic accuracy over wide temperature ranges.
junctions (see Figs. 3 and 5). The hot junctions are formed at The leads that carry the dc outputs from the probe elements
the center of the narrow strips having relatively high resistance, to the metering instrumentation are high-resistance films. Un-
thereby allowing for the dissipation of energy and the resul- like those of [5], they are tapered in the region where they
tant increase in temperature. The wider sections have a low make contact with the probe elements. The tapered sections
resistance and thus function as cold junctions, the low resis- present a higher resistance near the probe elements resulting in
tance allowing little energy to be dissipated within these sec- lower currents near the probe elements and less interaction of
tions. In addition, the broad area distributes the energy and the currents in leads with those in the probe elements. The
conducts heat rapidly into the substrate so that very little total effect is to approximate the condition of the probe ele-
temperature rise occurs. The resultant dc output voltage is ments being suspended in space.
directly proportional to the energy dissipated in the resistive Because of the extremely light coupling into the field, very
portion of the thermocouple. little perturbation due to scattering is caused. As an indication

Authorized licensed use limited to: Tubitak Ve Birimleri. Downloaded on February 19,2021 at 13:03:12 UTC from IEEE Xplore. Restrictions apply.
424 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, NOVEMBER 1972

of the degree of coupling to the field, a term "effective vol-

ume" may be applied. This can be defined in a manner similar
to that of effective area. The ratio of the power in the termi-
nating impedance to the power density of the field, is defined
as effective volume. Based upon knowledge of the efficiency
of the thermocouple elements, the effective volume of the
probe is approximately 1 cm3. The probe elements are con-
tained within an actual volume of 44 cm3. The probe is a
relatively inefficient device. However, its inefficiency is an
indication of its ability to measure a field without causing
perturbations to the field. The probe is an E-field instrument,
the probe elements of which couple very lightly into the field
causing extremely small perturbation to the field, and it may
be used in the near field to make accurate measurements of
equivalent plane-wave power density.
The metering instrumentation is a high-input-impedance
Fig. 6. Probe and metering instrumentation. Larger unit affords addi-
voltmeter calibrated to indicate power density directly. The tional facilities such as recorder output and audible alarm, with the
calibration circuits are contained within the probe providing same instrumentation accuracy.
for the interchangeability of probes and meters. Fig. 6 ex-
hibits the metering instrumentation. A fail-safe mechanism 2-wave dipole and calibrated crystal are used to establish a
inherent within the operational amplifiers of the metering in- field approximately 40 dB below the calibration level, the
strumentation will, in the event of a failure of the probe, cause transmitted power is then accurately measured and increased
the meter to read off scale. The larger instrument has the
the necessary 40 dB to establish the calibrating field.
added feature of a presettable audible alarm, which will sound The described radiation monitor is the subject matter of US
if the power density exceeds the threshold settable on the patent 3 621 239 and other pending patent applications.
panel control. The control may be set from 0 to 100 percent
of full scale. This unit also has a recorder output for remote REFERENCES
metering. The unit may be operated from an ac line or from
its rechargeable internal battery. The smaller unit is powered [1] E. C. Jordan, Electromagnetic Waves and Radiating Systems. New
from a 9-V transistor-type battery. Both units have a 20-dB York: Prentice-Hall, 1950, pp.452-461.
[21 E. Hallen, "Theoretical investigations into the transmitting and re-
dynampic range consisting of 2 full-scale ranges of 20 and ceiving qualities of antennae," Nova Acta Regiae Soc. Sci. Upsal.,
2 mW/cm2. An additional 30-dB dynamic range unit utilizing ser. IV, vol. 11, no.4, pp. 1-44, 1938.
[3] J. D. Ryder, Networks Lines and Fields. New York: Prentice-Hall,
a chopper amplifier provides an increased measurement range. 1955, p. 206.
The standard field method and the standard antenna method [4] E. Aslan, "Electromagnetic leakage survey meter," J. Microwave
Power, vol. 6(2), June 1971.
[6] are both used to calibrate the probes. From 750 MHz [5] R. Bowman, "An isotropic electric energy-density probe for high-
through 18 GHz, the standard field method is used. Standard level fields," presented at Internat. Union Radio Science, Spring
Meeting, Washington, D.C., Apr. 8-10, 1971.
gain horns with corrections for near zone gain [7] are used to [6] Measuring Field Strength in Radio Wave Propagation, IEEE Stan-
establish a standard field. An anechoic sled is used to average dard 291, May 1969.
[7] R. Bowman, "Near zone calibrations for antennas and field inten-
out multipath interference including scattering from the an- sity meters," RADC-TR-415, Jan. 1970.
echoic material of the receiving pedestal. Below 750 MHz, the [8] H. Taggart and J. Workman, "Calibration principles and proce-
dures for field strength meters 30 MHz to 1 GHz," Nat. Bur.
standard antenna method is used for calibration [8]. A cut Stand., Tech. Note 370, Mar. 1969.

Authorized licensed use limited to: Tubitak Ve Birimleri. Downloaded on February 19,2021 at 13:03:12 UTC from IEEE Xplore. Restrictions apply.