A field strength meter is a very handy tool for locating a
transmitter, but a
sensitive field strength meter by itself has some limitations as it will respond to practically
any RF signal that enters
its input. This has the effect of limiting the effective
sensitivity of the field strength meter, as any nearby RF source
(or
even
ones far away, if the meter is sensitive enough...) will effectively
mask
the desired signal if it is weaker than these "background" signals.
 |
Figure 1:
The modified Icom IC-2A/T HT with a broadband
field strength meter paired with the AD8307-based field
strength meter mentioned and linked in the article, below.
Click on the image for a larger version. |
This property can be mitigated somewhat by preceding the input of the meter with
a simple tuned RF stage and, in most cases, this is adequate for finding
(very) nearby transmitters. A
simple
tuned circuit does have its limitations:
- It is only broadly selective. A simple, single-tuned filter
will
have a response encompassing several percent (at best) of the operating
frequency. This means that sensitive meter preceded by a 2 meter filter will respond to
nearly
any signal near or within to the 2 meter band.
- A very narrow filter can be tricky to tune. This isn't
usually
too
much of a problem as one can peak on the desired signal (if it is close
enough to register) or use your own transmitter (on the same or nearby frequency)
to provide a source of signal on which the filter may be tuned.
- The filter does not usually enhance the absolute (weak signal) sensitivity unless an amplifier is
used with it.
An obvious approach to solving this problem is to use a receiver, but while many FM receivers
have
"S-meters" on them, very few of them have meters that are truly useful
over a very wide dynamic range, most firmly "pegging" even on
relatively modest signals, making them nearly unusable if the signal is any stronger than "medium weak". While an adjustable attenuator
(such as a step
attenuator
or offset attenuator) may be used, the range of the radio's S-meter
itself
may be so limited that it is difficult to manage the observation of the
meter and adjusting the signal level to maintain an "on-scale" reading.
Another possibility is to modify an existing receiver so that an
external signal level meter with much greater range may be connected.
Picking a receiver:
When I decided to take this approach I began looking for a 2 meter
(the primary band of interest) receiver with these properties:
- It had to be cheap. No need to explain this
one!
- It had to be synthesized. It's very helpful
to
be able
to change frequencies.
- Having a 10.7 MHz IF was preferable. The
reasons for
this will become apparent.
- It had to have enough room inside it to allow the
addition of
some
extra circuitry to allow "picking off" the IF signal.
After
all, that's the entire point of this exercise.
- It had to be easy to use. Because one may
not
use this
receiver too often, it's best not to pick something overly complicated
and would require a manual to remind one how to do even the simplest of
tasks.
- The radio would still be a radio. Another goal of
the modification was that the radio had to work exactly
as it was originally designed after you were done - that is, you could still
use
it as a transceiver!
Based on a combination of past familiarity with various 2 meter HTs
and looking at prices on Ebay, at least three possibilities sprang to
mind:
- The Henry Tempo S-1. This is a very basic 2
meter-only
radio and
was the very first synthesized HT available in the U.S. One
disadvantage
is that, by default, it uses a threaded antenna connection rather than
a more-standard BNC connector and would thus require the user to
install
one to allow it to be used with other types of antennas. Another
disadvantage is that it has a built-in non-removable battery.
It's
power supply voltage is limited to under 11 volts. (The later
Tempo S-15 has fewer of these disadvantages and may be better suited for this application, but I am
not too familiar with it.)
- The Kenwood TH-21. This, too, is a very basic 2
meter-only
radio.
It uses a strange RCA (e.g. phono) like threaded connector, but this mates with
easily-available
RCA-BNC adapters. Its disadvantage is that it is small enough
that
the added circuitry may not fit inside. It, too, has a distinct
limitation
on its power supply voltage range and requires about 10 volts.
- The Icom IC-2A/T. This basic radio was, at one time, one of the
most
popular
2 meter HTs which means that there are still plenty of them around. It can operate directly on 12-15 volts, has a standard
BNC antenna connector, and has plenty of room inside the case for the
addition
of a small circuit. (The "T" suffix indicates that it has a DTMF numeric keypad. The "non-T" version such as the IC-2A is a bit less common, but would work just fine for this application.)
Each of these radios is a thumbwheel-switch tuned,
synthesized,
plain-vanilla radio. I chose the Icom IC-2AT
(it is also the most common) and obtained one on Ebay for about $40
(including
accessories) and another $24 bought a clone of an IC-8, an 8-cell alkaline battery
holder
(from
Batteries
America) that is normally populated with 2.5 amp-hour NiMH
AA cells. With its squelched receive current of around 20 milliamps I will often use this radio as a "listen around the house"
radio since it will run for days and days!
"Why not use one of those cheap Chinese radios?"
Upon reading this you may be thinking "why spend $$$ on an ancient radio when you can buy a cheap chinese radio that has lots of features for $30-ish?"
The reason is that these radios have neither a user-available "S" meter with good dynamic range or an accessible IF (Intermediate Frequency) stage. Because these radios are, in effect, direct conversion with DSP magic occurring on-chip, there is absolutely nowhere that one could connect an external meter - because that signal simply does not exist!
While many of these "single-chip" radios do have some built-in S-meter circuitry, the manufacturers of these radios have, for whatever reason, not made it available to the user - at least not in a format that would be particularly useful for transmitter hunting.
Unfortunately these radios can overload quite easily when near a strong signal. This property - and the fact that they use audio processing that can vary based on signal quality - make many of these radios generally unsuitable for direction-finding! |
Modifying the IC-2A/T (and circuit
descriptions):
This radio is the largest of those mentioned above and has a
reasonable
amount of extra room inside its case for the addition of the few small
circuits needed to complete the modification. When done,
this
modification
does not, in any way, affect otherwise normal operation of the
radio:
It can still be used as it was designed!
An added IF buffer amplifier:
This radio uses the Motorola MC3357
(or an equivalent such as the
MP5071)
as the IF/demodulator. This chip takes the 10.7 MHz IF from the
front-end
mixer and 1st IF amplifier stages and converts it to a lower IF
(455
kHz)
for further filtering and limiting and it is then demodulated using a
quadrature
detector. Unfortunately, the MC3357 lacks an RSSI (Receive Signal
Strength Indicator) circuit - which also explains why this radio
doesn't
have an S-meter. Since we were planning to feed a sample
of the IF from this receiver into our field strength meter, anyway,
this
isn't too much of a problem.
 |
Figure 2:
The source-follower amplifier tacked atop the IF amplifier chip.
Click on the image for a larger version. |
We actually have a choice to two different IFs: 10.7 MHz and
455
kHz. At first glance, the 455 kHz might seem to be a better
choice
as it has additionally amplified and it is at a lower frequency - but
there's
a problem: It compresses easily. Monitoring the 455 kHz
line,
one can easily "see" signals in the microvolt range, but by the time
you
get a signal that's in the -60 dBm range or so, this signal path is
already
starting to go into compression. This is a serious problem as -60 dBm is about the
strength
that one gets from a 100 watt 2 meter transmitter that is clear line-of-sight
at
about 20 miles
(about 30km) distant, using unity-gain antennas on each end. What this means is that if we were to use this signal tap, we might still be a fair distance away from the transmitter we were seeking when its signal saturated the meter.
The other choice is to tap the signal at the 10.7 MHz point,
before
it goes into the MC3357. This signal, not having been amplified
as
much as the 455 kHz signal, does not begin to saturate until the input
reaches about -40 dBm or so, reaching full saturation by about -35
dBm. Given our example, above, -35 to -40dBm is roughly equivalent to a line-of-sight 100 watt 2 meter transmitter at 1-3 miles
(approx. 1.6-5km) - which means that we'll get much closer before the signal path saturates - but we can easily deal with that as we'll discuss shortly.
One point of concern here was the fact that at this point, the signal
has
less filtering than the 455 kHz, with the latter going through a
"sharper"
bandpass filter. While the filtering at 10.7 MHz
is a bit broader, the 4 poles of the crystal filter
do
attenuate
a signal 20 kHz away by at least 30 dB - so unless there's another
very
strong signal on this adjacent channel, it's not likely that there will
be a problem. As it turns out, the slightly "broader" response of
the 10.7 MHz crystal filters is conducive to "offset tuning" - that is,
deliberately tuning the radio off-frequency to reduce the signal level
reading when you are nearby the transmitter being sought and it starts to saturate the IF stages.
To tap this signal without otherwise affecting the
performance
of the receiver requires a simple buffer amplifier, and a JFET
source-follower
does the job nicely
(see figure 6,
below for the diagram). Consisting of only 6 components
(two
resistors,
three capacitors and an MPF102 JFET - practically any N-channel JFET will do) this circuit is simply
tack-soldered
directly onto the MC3357 as shown in
figures 2 and 3.
This
circuit
very effectively isolates the
(more or less) 50 ohm load of the field strength meter from the high-impedance 10.7 MHz input to the
MC3357
and it does so while only drawing about 700 microamps, which is only
3-4% of the
radio's
total current when it is squelched.
 |
Figure 3:
A wider view of the modifications to the radio.
Click on the image for a larger version. |
As can be seen from the pictures (
figure 2 and 3) all
of the required connections were made directly to the pins of the IC
itself,
with the 330 pF input capacitor connecting directly to pin 16. The supply
voltage is pulled from pin 4, and pins 12 and/or 15 are used for the
ground
connection.
A word of warning: Care should
be
taken when soldering directly to the pins of this
(or any) IC to avoid
damage. It is a good idea to scrape the pin clean of oxide and
use
a hot soldering iron so that the connection can be made
very
quickly.
Excess heat and/or force on the pin can destroy the IC! It's not
that this IC is particularly fragile, but this is care that should be
taken.
Getting the IF signal outside the radio:
The next challenge was getting our sampled 10.7 MHz IF energy out of
the radio's case. While it may be possible to install another
connector
on the radio somewhere, it's easiest to use an existing connector -
such
as the microphone jack.
One of the goals of these modifications was to retain complete
function as if it were a stock radio, so I wanted to be sure that
the
microphone jack would
still work as designed, so I needed to
multiplex
both
the microphone audio
(and keying) and the IF onto the tip of the
microphone
connector as I wasn't really planning to use the signal meter
and a remote microphone at the same time. Because of the very large difference in frequencies
(audio
versus 10.7 MHz) it is very easy to separate the two using capacitors
and
an inductor: The 10.7 MHz IF signal is passed directly to the
connector
with a series capacitor
(such as 100pF) while the 10.7 MHz IF signal is blocked from
the radio's internal microphone/PTT line with a small choke: Anything from 4.7uH to 100uH will work fine.
 |
Figure 4:
The modifications at the microphone jack.
Click on the image for a larger version. |
The buffered IF signal is conducted to the microphone jack using some small
coaxial cable: RG-174 type will work, but I found some slightly
smaller
coax in a junked VCR. To make the connections, the two screws on
the side of the HT's frame were removed, allowing it to "hinge" open,
giving
easy access to the microphone connector. The existing microphone
wire connected to the "tip" connection was removed and the choke was placed in series with it, with the
combination
insulated with some heat-shrinkable tubing.
The coax from the
buffer
amp was then connected directly to the "tip" of the microphone
connector.
One possible coax routing is shown in
Figure 4 but note that
this
routing prevents the two halves of the chassis from being fully opened in the
future unless it is disconnected from one end. If this bothers
you,
a longer cable can be routed so that it follows along the hinge and
then
over to the buffer circuit.
Note: It is
important
to use shielded cable for this connection as the cable is likely to be
routed past the components "earlier" in the IF strip and instability
could
result if there is coupling.
Interfacing with the Field Strength meter:
Using RG-174 type coaxial cable, an adapter/interface cable was
constructed
with a 2.5mm connector on one end and a BNC on the other. One
important
point is that a small series capacitor
(0.001uF)
is required in this line somewhere as a DC block on the microphone
connector:
The IC-2A/T
(like most HTs) detects a "key down" condition on the
microphone
by detecting a current flow on the microphone line and this series
capacitor
prevents current from flowing through the 50 ohm input termination on
the
field strength meter and "keying" the radio.
Dealing with L.O. leakage:
As soon as it was constructed I observed that even with no signal,
the field strength meter showed a weak signal
(about -60 to -65 dBm)
present
whenever the receiver was turned on, effectively reducing sensitivity
by
20-25 dB. As I suspected when I first noticed it, this signal was
coming
from two places:
- The VHF local oscillator. On the IC-2A/T, this oscillator
operates
10.7 MHz lower than the receive frequency. In other words, tuned to 146.520 MHz, the local oscillator is running at 135.82 MHz.
- The 2nd IF local oscillator. On the IC-2A/T this oscillator
operates
at 10.245 MHz - 455 kHz below the 10.7 MHz IF as part of the conversion to the second IF.
The magnitude of each of these signals was about the same, roughly -65 dBm or
so. The VHF local oscillator would be very easy to get rid of - A
very simple lowpass filter
(consisting of a single capacitor and
inductor)
would adequately suppress it - but the 10.245 MHz signal poses a problem
as it is too close to 10.7 MHz to be easily attenuated enough by a very
simple L/C filter without affecting it.
 |
Figure 5:
The inline 10.7 MHz bandpass using filter using a ceramic
filter. The diagram for this may be seen in the upper-right
corner of Figure 6, below.
Click on the image for a larger version. |
Fortunately, with the IF being 10.7 MHz, we have another
(cheap!)
option:
A 10.7 MHz ceramic IF filter. These filters are ubiquitous, being
used in nearly every FM broadcast receiver made since the 80s,
so if you have a junked FM broadcast receiver kicking around, you'll
likely
have one or more of these in them. Even if you don't have junk
with
a ceramic filter in it, they are relatively cheap
($1-$2) and readily
available
from many mail-order outlets.
This filter is shown in the
upper-right
corner of the diagram in Figure 6, below.
The precise type of filter is not important as they will typically have a bandpass that is between 150 kHz and
300 kHz wide
(depending on the application) at their -6 dB points and
will
easily attenuate the 10.245 MHz local oscillator signal by at least 30
dB. With this bandwidth it is possible to use a 10.7 MHz filter
(which, themselves, vary in exact center frequency) for some of the
"close
- but not exact" IF's that one can often find near 10.7 MHz like 10.695
or 10.75 MHz. The only "gotcha" with these ceramic filters is
that
their input/output impedances are typically in the 300 ohm area and
require a
(very simple) matching network
(an inductor and capacitor) on
the input and output to interface them with a 50 ohm system. The
values used for matching are not critical and the inductor, ideally around 1.8uH, could be
anything
from 1.5 to 2.2 uH without much impact of performance other than a
very
slight change in insertion loss.
While this filter could have been crammed into the radio I was
concerned
that the L.O. leakage might find its way into the connector
somehow, bypassing the filter.
Instead, this circuit was constructed "dead bug" on a small scrap of
circuit
board material with sides, "potted" in thermoset
("hot melt") glue and covered with electrical tape, heat shrink tubing or "plastic
dip"
compound, with the entire circuit installed in the middle of the coax
line
(making a "lump.") Alternatively, this filter could have been
installed
within the field strength meter itself, either on its own connector or
sharing the main connector and being switchable in/out of the circuit.
 |
Figure 6:
The diagram, drawn in the 1980s Icom style, showing the modified circuity and details of the added source-follower JFET amplifier (in the dashed-line box) along with the 10.7 MHz bandpass filter (upper-right) that is built into the cable.
Click on the image for a larger version. |
With this additional filtering the L.O. leakage is reduced to a
level
below the detection threshold of the field strength meter, allowing
sub-microvolt
signals to be detected by the meter/radio combination.
Operation and use:
When using this system, I simply clip the radio to my belt and
adjust it
so that I can listen to what is going on.
There's approximately 30 dB of processing gain between the antenna to
the
10.7 MHz IF output - that is, a -100 dBm signal on the antenna on 2
meters
will show up as a -70 dBm signal at 10.7 MHz. What this means is
that sub-microvolt signals are
just detectable at the bottom
end
of the range of the Field Strength meter. From a distance, a simple gain antenna
such as a 3-element "Tape Measure Yagi"
(see the article "Tape Measure Beam Optimized for Direction Finding - link) will establish a bearing, the antenna's gain providing both an effective signal boost of about 7dB
(compared to an isotropic) and directivity.
While
driving about looking for a signal I use a multi-antenna
(so-called)
"Doppler" type system with four antennas being electrically rotated to
get the general bearings with the modified IC-2AT being the receiver in
that system. With the field strength meter connected I can hear its
audio tone representing the signal strength without need to look at it. As I near the signal
source and the strength increases, I have both the directional
indication and the rising pitch of the tone as dual confirmation that I
am approaching it.
The major advantage of using the HT as tunable "front end" of the
field
strength meter means that the meter has greatly enhanced selectability
and
sensitivity - but this is not without cost: As noted before, this
detection system
will begin to saturate at about -40 dBm, fully saturating above -35 dBm
- which is a "moderately strong" signal. In "hidden-T" terms, it
will "peg" when within a hundred feet or so of a 100 mW transmitter
with
a mediocre antenna.
When the signals become this strong, you can do one of several
things:
- Detune the receiver by 5, 10, 15 or even 20 kHz. This will
reduce
the sensitivity by moving the signal slightly out of the
passband of the 10.7 MHz IF filters.
This is usually a very simple and effective technique, although heavy
modulation
can cause the signal strength readings to vary.
- Add attenuation to the front-end of the receiver. The
plastic
case
of the IC-2A/T is quite "leaky" in terms of RF ingress, but it is good
enough for a step attenuator on the antenna lead to work nicely and will thus
extend
usable range to at least -10dBm dBm. I use a switchable step attenuator for this and I have found that I can drive to the location (house, yard, park) where the transmitter is located and still have sufficient adjustment range.
- When you are really close (e.g. 10s of yards/meters) to the transmitter being sought you can forgo the
receiver
altogether,
connecting the antenna directly to the field strength meter!
If you want to be really fancy, you can build the 10.7 MHz bandpass
filter
and add switches to the field strength meter so that you can switch 20 dB of attenuation in and out as well as routing the signal either to
the receiver, or to the field strength meter using a resistive or
hybrid
splitter to make sure that the receiver gets
some signal from
the
antenna even when the field strength meter is connected to the
antenna.
What to use as the field-strength meter:
The field strength meter used is one based on the Analog Devices AD8307 which is useful from below 1 MHz to over 500 MHz, providing a nice, logarithmic output over a range that goes below -70dBm to above +10dBm. It is, however, broad as the proverbial "barn door" and the combination of this fact and that its sensitivity of "only" -70dBm is nowhere near enough to be useful with weak signals - especially if there are
any other radio transmitters nearby - including radio and TV stations within a few 10s of miles/kilometers. The integration of this broadband detector with the narrowband, tuneable receiver IF along with its gain makes for a complete system useful for signals that range from weak to strong.
The description of an audible field-strength meter may be found on the web page of the Utah Amateur Radio club in another article that I wrote, linked here:
Wide Dynamic Range Field Strength Meter - link. One of the key elements of this circuit is that it includes an audio oscillator with a pitch that increases in proportion with the dB indication on the meter, allowing "eyes-off" assessment of the signal strength -
very useful while one is walking about or in a vehicle.
There are also other web pages that describe the construction of an AD8307-based field strength meter
(look for the "W7ZOI power meter" as a basis for this type of circuit) - and you can even buy pre-assembled boards on EvilBay
(search for "AD8307 field strength meter" or "AD8307 power metetr"). The downside of most of these is that they do
not include an audible signal strength indication to allow "eyes off" use, but this circuit could be easily added, adapted from that in the link above.
Another circuit worth considering is the venerable NE/SA605 or 615 which is, itself, a stand-alone receiver. Of interest in this application is its "RSSI"
(Receive Signal Strength Indicator) circuit which has both good sensitivity, is perfectly suited for use at 10.7 MHz, has a nice logarithmic response and a wide dynamic range - nearly as much as the AD8307. Exactly how one would use
just the RSSI pin of this chip is beyond the scope of this article, but information on doing this may be found on the web in articles such as:
- NXP Application note AN1996 - link (see figure 13, page 19 for an example using the RSSI function only)
Additional comments:
- At first, I considered using the earphone jack for interfacing to
the
10.7
MHz IF, but quickly realized that this would complicate things if I
wanted
to connect something to the jack (such as pair of headphones or a Doppler unit!) while
DFing.
I decided that I was unlikely to be needing to use an external
microphone while
I was looking for a transmitter!
- I haven't tried it, but these modifications should be possible
with the
222 MHz and 440 MHz versions (the IC3 and IC4) of this radio - not to mention other radios
of this type.
- Although not extremely stable, you can listen to SSB and CW
transmissions
with the modified IC-2A/T by connecting a general-coverage/HF receiver to
the 10.7 MHz IF output and tuning that receiver to about 10.7 MHz, +/- a few kHz. Signals may be slightly
"warbly"
- but they should be easily copyable!
Finally, if you aren't able to build such a system and/or don't mind spending the money and you are interested in what is possibly the best receiver/signal strength meter combination device available, look at the
VK3YNG Foxhunt Sniffer - link. This integrates a 2 meter receiver (also capable of tuning the 121.5 MHz "ELT" frequency range) and a signal strength indicator capable of registering from less than -120dBm to well over +10dBm with an audible tone.
Comment: This article is an edited/updated version of one that I posted on the Utah Amateur Radio Club site (link) a while ago.
[End]
This page stolen from "ka7oei.blogspot.com"
