Sunteți pe pagina 1din 99


This circuit is sensitive to low frequency electromagnetic radiation and will detect for example
hidden wiring or the field that encompasses a transformer. Pickup is by a radial type inductor,
used as a probe which responds well to low frequency changing magnetic and electric fields.
Ordinary headphones are used to for detection. The field that surrounds a transformer is heard as
a 50 or 60Hz buzz. The circuit is below:-

I threaded a length of screened cable through an old pen tube and soldered the ends to a radial
type can inductor. I used 1mH. The inductor fitted snugly into the pen tube. The opposite end of
the cable connects to the input of the op-amp. Any op-amp should work here, possibly better
results may be achieved with a low noise FET type such as the LF351. The 2M2 potentiometer
acts as a gain control and the output is a pair of headphones. Stereo types can be used if they are
wired as mono. I used an 8 ohm type, but the circuit should work equally well with higher
impedance types. The probe (shown below) may be connected via screened cable and a 3.5mm
stereo plug and socket.

The sensitivity of this circuit is good. Mains wiring buried an inch in plaster can be detected with
precision. A small load on the electric supply is all that is needed; a 20 watt desk lamp or similar
will suffice. The hum field surrounding a transformer can be detected oat over 7 inches.
Domestic appliances such as videos and alarm clocks all produce interference which can be
heard with the probe. The electric field surrounding a loudspeaker or earpiece can also be heard.
Try lifting a telephone and place the probe near the earpiece. A telephone pickup coil can be
used in place of the inductor if desired. I will make an improved version of this circuit with a
meter output later.


Natural Radio
"Natural Radio" is a fancy name for radio noise with natural origins, mainly
lightning. At first, you might think that listening to lightning crackles is pretty
uninteresting but as it turns out, electromagnetic radiation from lightning
can travel great distances and undergo strange modifications along the way.
The frequencies of the original pulse can be spread out in time (a process
called "dispersion) because the higher frequencies travel a little faster than
the lower. The result is that the short impulse from a lightning strike in South
America can sound like a chirp in Texas. Slower sweeping tones are called
"whistlers" and they are a bit of a mystery. The energy from a lightning bolt
streams out into space into a region called the "magnetosphere",
magnetized plasma created by the interaction of solar wind with the earth's
magnetic field. The lightning pulse is reflected or "ducted" back down to
earth after a very long trip during which time the frequencies are spread out
by a dispersion-like process. Short whistlers might be due to dispersion, but
some whistlers last five seconds so ordinary dispersion is probably
inadequate an explanation. A radio wave can travel a million miles in five
seconds so to accumulate that much difference in arrival times, the signal
would have to travel hundreds of millions of miles, assuming a pretty steep
dispersion curve. More likely, the whistler is an emission from the
magnetosphere triggered by the lightning pulse. (Read Alysson's emails
regarding whistlers.) When conditions are just right, numerous lightning
strikes combine with numerous reflections to give an eerie chorus that
sounds a bit like a flock of geese.
The ever-present power line hum makes listening to these signals difficult
near power lines but modern computers are fast enough to digitally clean up
the signal, making natural radio listening a practical home-based activity. I
am delighted with a tremendous freeware offering from Wolfgang Buescher
(DL4YHF) called "Spectrum Lab" that works in conjunction with the sound
card. In addition to oscilloscope and spectrum analyzer functionality, it also
provides digital filtering, including a highly effective hum filter module
contributed by Paul Nicholson. It does take a pretty quick computer, by the
way. This probably is a job for your best computer and not the old one in the
closet. Karen, a reader in the U.K. developed the "Humnuller," a simple
application that also does the trick. It has the nice feature that you can click
a button after an interesting sound and capture the last several seconds
(user selected).
Here is what it can do: My antenna without the filter With the filter
To give you an idea how much work the filter is doing, below is a 'scope view
of the hum. It is nearly clipping! (Max is +-32,768.)

I just discovered another effective notch filter in the Filter Control Window
called "Auto-notch". Set the speed to 0.05 and the FFT size to 4096 or higher
and turn off the hum killer. This filter "surges" a bit on big static crashes as it
"gates in" noise, making a sound a bit like eating celery, but it doesn't have
nearly as much ringing or echo. Here is another very short sound file with
another chirp recorded with this filter. Try playing it in a continuous loop.

Electric Pickup
I have an unused 900 MHz antenna sticking out
of my chimney that has an ungrounded vertical
element. I replaced the 12 inch element with a
two yard aluminum rod (see photo). The antenna
is only 75 feet from the power lines so I didn't
have much hope for this antenna to work. The
lead-in coax is long so its capacitance (about
1,500 pF) forms a voltage divider with the
capacitance of the antenna so the signal is
significantly attenuated. It is a big gain hit but
gain is cheap, as long as noise doesn't become a
problem. The amplifier needs to be high enough
in input resistance to give a sufficiently long time
constant to handle the lowest frequency of
interest but with 1,500 pF, an input impedance of
1 meg is sufficient to get a flat response below a
couple of hundred Hz. One should expect to run
into troubles trying to amplify a remote high
impedance antenna with a line-powered amplifier
and then feeding that signal to another line-
powered computer! The possibility for noisy
ground loops is mind-boggling. But ground loops
are easily avoided by simply not grounding the
antenna cable at the mast. Now, it would be a
safer antenna if it were grounded! A spark gap
lightning arrestor for the center conductor AND
the braid is a good idea for safety. A single-point
ground prevents ground currents from flowing in
the shield but it also directs a lightning strike right
to your receiver inside the house! At a minimum,
disconnect the antenna when storms are nearby
or when you leave.
Don't worry about ground plane,
impedance matching, cable impedance or
other RF considerations; it isn't a radio
antenna; it's just a voltage probe shielded
by the braid in the coax. A high antenna
seems best (now that I've played a bit).
On the other hand, locating the antenna
near ground would be more safe but try to
stay away from other objects. After playing
with a few random antennas, I must say
that some locations just don't work well for
no apparent reason so try several
locations until you can easily detect those
transmitters between 17 and 24kHz.
Unfortunately, a location where the hum is
low is usually a location where the desired
signals are low, too.
Try your antenna with a local ground and
lightning arrestor first; the ground loop
that results might not be a problem,
especially if the ground goes straight down
to earth and your antenna will be more
safe. You don't want to do a "Ben Franklin"
number on your expensive computer (or
yourself, for that matter).

Remember to add a neon lamp or, better yet, a Lumex gas tube transient
suppressor from the antenna to ground to bleed off static charge.

Here is a "catch-all box that goes between the antenna and the amplifier. It
has a Lumex gas discharge tube right across the antenna connector and a
0.01uF, 1,600 volt coupling capacitor in series with the signal. There is also a
10 mH inductor in parallel with an 8.2k resistor at the amplifier connector to
block radio stations. Be warned that when the Lumex fires, a couple of
hundred volt transient will be presented to your amplifier! If you have a high
value resistor, say 50 megohms, add it across the antenna connector to
bleed off charge. A lower value is fine if you are using a long cable.

Disconnect the antenna when lightning is nearby or when not in

use, regardless of these precautions!
Here is an amplifier that includes a high-pass filter to help reduce the
fundamental line frequency component:

The high-pass frequency is set by the 500k and 1meg resistors, change them
keeping the same ratio to change the response, higher values giving a lower
frequency response. The values shown will reduce 60 Hz by about a factor of
6 but don't hesitate to use lower values for more attenuation. Another way to
reduce the hum if you are willing to sacrifice some low frequency response is
to change the output capacitor from 10 uF to 0.047 uF (adds another roll-off
at about 350 Hz). The overall gain is set by the 10k and 180 ohm; increase
the 10k to 22k if hum permits. The selected capacitor at the antenna is
chosen to give a total of about 2000 pF including the cable capacitance but
this value is not critical if the hum is low enough and the gain is high enough.
The power supply is an ordinary bench supply with the negative terminal
grounded to a good earth ground. Surprisingly, no isolation was needed
between the amp and computer once the earth ground was connected!
The more I play with this, the less important the low frequencies seem to be.
My version of this amp replaces the 10 uF output cap with the suggested
0.047 uF. The plot below shows how low the line hum is before filtering. The
system can now handle really big pulses without running out of headroom,
as you can see.
Note the little on-off squiggles near the end of the trace. Those are from a
transmitter in Hawaii!
Try your amp with just the good old line power ground first but don't be
discouraged by trouble. My amplifier happens to be right next to a ground
wire that connects directly to an underground copper water pipe. Connecting
the shield of the coax to the earth ground at the amplifier's input completely
eliminated grounding problems. I can't emphasize the benefits of a good
ground at the amplifier input enough! If you don't have a cold water pipe
handy try running a ground wire out a window and attach it to a ground rod
or an outside water faucet near the water main. Such a good ground is
handy for all sorts of radio projects. In this case we are providing a ground
for audio frequencies so the main consideration is the resistance of the wire
and quality of the ground; bends, turns and length are not important.

Computer-Powered Receiver
Note: You may need to tie together the "tip" and "ring" of the stereo plug for
most soundcards. Some cards apply the microphone bias to the tip which is
also the input but most connect the bias to the pin. Also mute the
microphone and possibly the line inputs in the volume control menu. You
may find a microphone boost check box that will give you more sensitivity if

Here is a receiver for a laptop or PC that requires no power supply! The

receiver is powered by the microphone jack and makes a great dedicated
receiver for fixed locations like my outdoor antenna described above. The 10
megohm at the antenna bleeds off any accumulated charge and the Lumex
tube hopefully catches high voltage spikes. (Use a bigger resistor for short
antennas, maybe a couple of 22 megohms in series.) The voltage rating on
my 100 pF capacitor is 30 kilovolts which is unusually high but a 1 kV part
will suffice. The 270 pF capacitors and high value resistors form a notch filter
for 60 Hz and the frequency may be tweaked slightly by varying the 540 pF.
My final value was a 510 pF mica type that gave a deep null at 60 Hz. For 50
Hz environments, increase the values of the two top capacitors to about 330
pF and experiment with a value near 620 pF for the one to ground. The 330
mH choke and 100 k resistor (to kill the Q) block RF signals, as does the 100
pF to ground. My 330 mH choke is a surplus pot core style but most styles
will work fine and the value may be somewhat lower, perhaps as low as 100
mH. Much below that and there will be a resonant peak near the broadcast
band. The silicon diode may be anything similar to a 1N914 and it just
provides a current path in the event the Lumex tube fires with a negative-
going pulse.

This is my third design and this time I modeled it - hence all the changes! I
used a 1 volt source behind a 10 pF capacitor to simulate a short antenna
with no coax and I assumed that the computer contains a 5k resistor
connected to 9 volts which is on the high side. Expect something like a 2.5k
connected to 5 volts but the results will be similar. (I also used two 2N4401's
to form the darlington since I didn't have the model for the darlington
This is just the response I like for this receiver. It begins rolling off below
1kHz with a deep notch at 60 Hz and on the high side there is a slight
peaking at 90 kHz followed by a nice roll-off, reaching over 35 dB roll-off by
the bottom of the broadcast band. A 100 mH choke substituted for the 330
mH is flat out to 200 kHz and down only 20 dB by 500 kHz but that might be
adequate for most environments. I haven't heard a peep from the radio
stations with my 330 mH.
If you would like to flatten the low frequency end a bit, split the 22 megohm
bias resistor for the darlington into two, 10 megohm resistors and add a
220pF capacitor to ground at the center point. The response will be flat down
to 1kHz and 10 dB down at 400 Hz. Notice that the overall gain works out to
1X with the simulated short 10 pF antenna. With my 1,500 pF antenna
system and the bypassed bias resistors, the gain is about +16 dB and the
response rolls off a bit sooner:
This is perfect for my fixed antenna setup! Notice how the bypassing the
darlington's bias resistors flattens the lower end of the response. (Your ears
won't notice.) On a typical day, the noise of the receiver is about 20 dB
below the atmospheric noise at 20 kHz even with 1500pF of additional cable
capacity on the input. One can do better but the law of diminishing returns
sets in quickly! With a short 2 foot antenna directly connected to the box and
held near the ground, the atmospheric noise is only a few dB above the noise
floor so this circuit is best with a longer antenna or an antenna/cable
The unit is built in a tiny minibox with terminal strips and point-to-point

The box is mounted on the wall behind the computer and a ground wire
connecting directly to a cold water pipe was added to the coax by cutting
through the insulation and soldering a short wire to the braid. This extra
ground eliminated all the local hum and only signals picked up by the
antenna are present. I only tried this receiver with my outdoor antenna with
the long cable that has an estimated 1500pF capacitance. The convenience
of no additional power supply is hard to beat. The lack of power supply
means there is one less ground loop to add hum. Here is a typical spectrum
made with an earlier FET version of the receiver showing the low frequency
response and the various transmitters at the high end and this version looks
similar only with significantly more gain.

A note on soundcards: I had a bit of trouble setting up my soundcard, mainly

in finding all the needed controls and figuring out how to set them. For
example, I have the microphone input "muted" even though I'm using it. This
keeps the sound from going straight from the mic. to the speaker instead of
through Spectrum Lab first. I also had to select a setting called "1
microphone boost" buried in an advanced menu. I have no idea how other
cards are configured but you may need to poke around a bit.

Portable E-field Receiver

Here is a tiny receiver built into an Altoids tin suitable for use with a laptop
computer or portable tape recorder. The prototype antenna is only 1 foot
long when fully extended and the receiver fits into a shirt pocket.

The 2SK117 is an unusually low noise JFET but other types will suffice. Pick
one with an Idss of at least a few mA. If oscillation occurs, try adding a
capacitor to ground across the 100 megohm resistor, perhaps 22 pF and
add a 100 ohm resistor in series with the gate.

The 1k resistor in the source is only a starting value and it should be

selected to achieve about 5 volts on the drain (400 uA). With some JFETs
with high Idss, this value may be considerably higher than 1k!

The 100 megohm bias resistor may be several 22 megohm resistors in

series and the value is not critical. A higher value (or longer antenna) will
give better low frequency response but that only invites line frequency
troubles. Stand reasonably still when listening or changing electrostatic
fields will overload the amplifier, especially if you are wearing rubber sole

The 50k pot may be a higher value if desired but don't drop much below
25k or gain may suffer. If the op-amp is fast enough, more gain may be
had by increasing the 100k feedback resistor. My brief experiments
indicate that the gain is plenty high as-is.

The 1uF capacitor in the source and negative op-amp input may be
reduced to give the circuit more of a high-pass characteristic to reduce
overload from line frequency. The 22uF in the output may also be reduced
for the same purpose. In extreme cases, try 0.47 uF caps for these three
caps. (The output cap is presently large to accommodate low impedance

The op-amp may be just about any type that will work with a 9 volt power
supply including most CMOS types, "single-supply types, and low power

There are two capacitors indicated across the power. Place the 1 uF near
the op-amp and the 47 uF near the FET. If your circuit has stability
problems, break the line between the two capacitors and add a small
resistor, perhaps 100 ohms. (Keep the op-amp on the battery side of the

The circuit will drive a "crystal" ceramic earphone but the problem with
crystal earphones is that they just aren't very loud with a 9 volt signal and
outdoor noises can make listening difficult. I had good luck adding a 2k to
10k transformer across the output to boost the voltage but better results are
achieved by plugging the circuit into a portable recorder and using the
recorder's output earphone jack or speaker for real-time listening. You can
just make out the tiny transformer near the output jack. This part will be
removed soon and is not shown in the schematic.
I painted my tin with gray hammertone paint and affixed a label to the lid.
The circuit is built on a piece of polished copper clad circuit board sprayed
with a light coat of clear Krylon to keep it shiny. All the ground connections
are made directly to the copper and a couple of terminal strips hold most of
the components. A battery compartment is formed with two pieces of
copper clad board soldered in place as walls and a piece of foam rubber on
the lid secures the battery. Just about any assembly technique should work
fine so don't feel compelled to copy this approach.
I just upgraded my recorder to an M-Audio Microtrack 24/96 and it is
amazing; it easily outperforms my soundcard! But there is a problem (as
usual). The device emits a huge amount of hum and buzz and a receiver
can't be anywhere near it. A lot of it sounds just like 60 Hz line noise, too,
making it more confusing. I came up with a solution that completely
eliminates the interference:
I formed a shield from perforated aluminum, using aluminum eyelets to hold
it together. The "trick" is to add a short 1/4" dia. brass standoff on the inside
that lines up with one of the 1/4" jacks so that the shield is grounded when
the standoff plugs in (second photo). Keep the standoff short, maybe 3/8", so
that it doesn't reach the input terminals in the socket. I can use the Super-
Tiny receiver below without significant interference, although the gain is a bit
low. The Altoids receiver above is a better choice. I cut a hole for the record
button since that is the only button needed when "in the field". Here are two
spectrum plots with then without the shield with the Altoids receiver about 2
feet away from the recorder. I have the volume down so that the unshielded
test doesn't overload the recorder. The difference is night and day (Paul
Nicholson's filter is ON in these plots, removing 60 Hz and harmonics):
There is interesting stuff above 24 kHz that I have never seen:

If you want to analyze the files yourself, here are the two files down-sampled
to 44.2 kHz: bad.wav good.wav They are still pretty big files, about 1.2
meg, but you won't see the high stuff. You can hear a radio station in the
good one so my Altoids receiver needs a little RF filtering. One of these days
I'll add a 10mH choke or resistor.
Here is a spectrum from my magnetic receiver below with this recorder
without the shield but positioned for minimum interference:

It certainly isn't suffering from a lack of gain or bandwidth!

My simple computer-powered darlington receiver with the rooftop antenna
(see above) is my favorite:
No worry about interference from the recorder here; it's in the basement!

Super-Tiny VLF Receiver

Here is a VLF receiver built into an earphone plug! I recently acquired a

digital voice recorder to replace the microcassette recorder above and
discovered that it is extremely sensitive, easily amplifying the signal from
the FET version of the computer-powered receiver above. This new receiver
consists of a very low current JFET, a bias resistor (2, 22 megohms in series)
and a capacitor; that's all! The recorder is designed to power electret
microphones so no battery is needed.

The drain of the FET connects to the center pin of the plug and the source
connects to the ground pin. Two, 22 megohm resistors are connected in
series from the gate of the FET to the ground pin and a 560pF capacitor
connects from the gate to a flexible wire antenna about 14 inches long (not
critical at all). The capacitor value isn't critical and a smaller value will work
fine. I selected it for its mechanical strength! For the antenna, use wire that
is stiff enough to hold its shape but bends easily enough to protect your
recorder's jack. The JFET has an IDSS of only about 125uA so power
consumption is quite low. (These are the JFETs that I have in quantity. If you
want a couple, let me know at This thing really works!
I just walked out to the end of my sidewalk and made a recording. A car
passed near the end and the electric fields produced by the tires make quite
a roar. There was no lightning for hundreds of miles but the spherics are
easily heard. The hum was removed by Spectrum Lab and the file was down-
sampled using dBpowerAMP to make the file small. (I could have used
Spectrum Lab for that, too but there is more to the story. See the note

As is often the case with quickie projects like this, a problem arose. The darn
thing picks up the LCD mux frequency. I tried shielding the connector and
lower portion of the antenna and it helped greatly but it didn't look very nice.
The solution is quite simple; just use an earphone extension cord. The cord
lets you easily hold the antenna over your head while keeping the recorder
comfortably low, anyway. Yes, that's the ticket; its a feature... Or, find a
metal box that the recorder fits inside and connect the box to the recorder's
ground, perhaps at the earphone jack.
This setup will easily catch whistlers and other audible phenomenon, it fits in
your pocket easily, and the Olympus is very sensitive, but the recording
quality doesn't hold a tiny birthday candle to the Microtrack.
Note: This recorder only makes WMA files and a converter is required to use
the data with Spectrum Lab. After a little panic attack, I found With the addition of a CODEC, this program can
convert WMA files to WAVE format (see the website for instructions). It's easy
to use, once you get it installed and running; just right click on the file and
select "Convert To", a new entry in the menu. All sorts of options are
available. The program is free except that the mpeg portion expires after 30
Another Note: The audio input on these recorders can be "tricky". Mine looks
at the impedance of what is plugged into the earphone jack to decide
whether it is a microphone or earphone and whether it has stereo capability.
It seems to do a test at the instant of initial connection and remembers the
results until the plug is removed.

Dual-Mode VLF Whistler Receiver

See the note above for connecting to a computer soundcard.
This portable receiver may be completely powered by a recorder or other
device designed to use button microphones or it can use its internal battery
to provide headphone listening. I decided to see if I could increase the values
on the notch filter like that used in the computer-powered receiver above by
nearly a factor of 10. Since this receiver will be able to run completely on the
power supplied by a soundcard or recorder microphone jack, there isn't any
power to spend on a source-follower. So, the notch filter isn't buffered and
must operate at a very high impedance to avoid loading the antenna. This
circuit probably pushes practical values to the edge, however! Although the
very high value resistors exhibit a correspondingly high noise voltage, the
improved sensitivity from this impedance increase "outruns" the noise and
the performance is pretty good. (The noise voltage of a resistor only
increases as the square-root of the resistance.) This is about it, however; the
capacitors are already getting close to the capacitance of even a short
antenna. A 32" antenna seems to bring the atmospheric background noise
above receiver noise adequately, even on a fairly quiet day. (The circuit
noise was observed by substituting a 15 pF capacitor for the antenna.)
The resistors were made by hand-selecting 22 megohm resistors and
soldering 2 or 4 in series to get precise values. (The resistors are in white,
green, and red heatshrink tubing. The green and red ones are the carefully
selected values for the filter.) 30 pF disk capacitors were the perfect value
for these resistors to get a deep 60 Hz notch and, surprisingly, no tuning was
needed. The 300 mH consists of three, 100 mH chokes in series with a 270k
resistor to kill the Q. Just about any value of choke from 100 mH to 1 Hy will
work here for most purposes but such chokes aren't particularly common.
The wires on these chokes is extremely fragile and bending the leads near
the body can easily break the fine wire.
The response is flat to above 100 kHz but is down over 20 dB by the bottom
of the broadcast band and dropping fast. The simulated response is shown
Instead of the darlington transistor used earlier, I tried a JFET that is intended
for such a job. They aren't easy to find, however! Look for a JFET that has an
Idss of a few hundred microamperes max and a pinch-off voltage near one
volt or lower. Not too many FETs fit the bill. I included the Lumex tube in the
event I use the receiver with my rooftop antenna which can pick up a charge.
It probably isn't needed for portable whip antenna use. The silicon diode on
the gate could be left out, too since it is only there to prevent a high reverse
voltage when the Lumex tube fires.
Note: FETs with a higher pinch-off voltage, say 2 volts, will work with
soundcards that supply a higher bias voltage like 5 volts. The soundcard
voltage and current may be directly measured with a DC multimeter. My
laptop supplies 2.25 volts and only 0.75 mA but one desktop I measured
supplies 5 volts at 2 mA. The bottom line is that you want the drain to be at
least a volt above the source.
The receiver can also run on an internal battery and an audio amplifier is
included to drive headphones. Notice that when a recorder (or soundcard) is
plugged into the top jack, the internal 10 k resistor is disconnected and the
JFET gets its power from the recorder. Most earphone jacks have that little
switch built in. The internal headphone amplifier may be operated with or
without a recorder connected and the power switch may be left in the "off"
position to save the internal battery if the recorder or computer has an
adequate sound output capability. The amplifier has more gain than is
needed and a resistor may be added in series with the 10 uF between pins 1
and 8 to reduce the gain a bit (see the manufacturer's data sheet).
Construction is mostly point-to-point
wiring using terminal strips and isn't
particularly critical. The headphone
amplifier is built on a little piece of
perfboard that is supported on one end
by a short length of PCB card guide glued
to the side of the case. A little drop of
wax keeps the board from sliding out. It's
just typical prototype construction. A PCB
would be nice!
I like this receiver. It is fairly small (4.5"
x 2.3"), works with my computer and
various recorders without using the
internal battery, operates as a stand-
alone receiver, and works well with a
short antenna or my rooftop whip. It
probably isn't a project for everyone, however. The FETs are hard to find and
I don't have my typical endless supply. The one I used is a house-numbered
part that has the right Idss and gate cut-off voltage determined by
experimentation! (The electrometer FETs that I have in the hundreds will
work but the gain is just a bit low for the typical recorder or soundcard.)
Possible number include: 2N4338, 2N4302, E201, 2N3370, 2N3438, 2N3460,
and maybe a VCR4N. The chokes aren't much easier to find ( I used three in
series) and the resistors are a bit of a pain in the neck to match. The
matching may not be necessary but I wanted a really deep null and I wasn't
disappointed. The 60 Hz is virtually invisible.
Other Possible E-field Approaches
Note : A reader has recommended a lower input current op-amp (like the
OPA363) for the following two circuits in place of the OP27 used above for
less input current noise. For antenna systems with a lot of cable capacitance
as above, the low noise bipolar amp is best above a couple of kilohertz but
for antenna systems with low capacitance (no cable) a CMOS or JFET amp is
best. (The capacitance where the two types of op-amps perform equally is
about 500 pF at 3 kHz.) Even more capacitance than the 2000pF used above
doesn't hurt with the bipolar amp since the noise current is still the limiting
factor; the noise and signal drop together. It takes about 5000pF before the
high audio frequencies are degraded (if I did that quick calculation right). So,
use as long a cable as you need! But for short or no cables between the
antenna and amp, use a FET input op-amp.
If you place the amplifier at the antenna, you avoid all that cable capacitance
but your amplifier must have a correspondingly higher input impedance to
maintain flat gain at the lower frequencies. For a typical 10 pF antenna
(about a meter long) you would want a 100 megohm input impedance to
have flat gain down to 150 Hz! One simple solution is to add some shunt
capacitance across the antenna, eating into the gain but giving a lower
frequency response. Some hand-held designs have a 10 megohm input
impedance along with a 50 to100 pF shunt capacitor in parallel. Although the
shunt capacitance works great, another elegant approach is to bootstrap a
10 megohm bias resistor so that it has an effective impedance of 100
megohms, or more at the frequencies of interest. The 10 megohm will still
discharge static charges quickly, a desirable feature for a portable receiver.
These following experimental receivers will probably need a 10 mH choke or
some sort of filtering to eliminate broadcast band radio signals.
Experimental circuit (not tested):
This circuit amplifies the voltage on the antenna by a factor of 11. The
output signal is divided back down by the 10k and 890 ohms to about 0.9
times the input voltage. This voltage is applied to the bottom of the 10
megohm bias resistor through the .0033 uF capacitor, keeping the voltage
across the resistor down to about 1/10 the antenna voltage. Since 1/10 the
voltage is across the resistor, it loads the antenna 1/10 as much, as though it
were 100 megohms. The feedback signal drops off below a couple of
hundred hertz and the resistor begins to look like 10 megohms near DC.
Additional gain will be required after this amplifier but too much gain in one
box may lead to instability. Make the next stage an inverting stage with a
gain of 100 or less and keep the output leads short and away from the
antenna. It would be a good idea to use shielded cable to connect the op
amp to the output connector. Here is a possible implementation (also not
This circuit could drive an earphone or speaker amplifier but I would highly
recommend completely enclosing a speaker amp in its own metal box or
metal compartment within the chassis. Otherwise, the large voltage swings
will be picked up by the antenna and oscillation will result. This circuit might
make a great portable whistler receiver but I'm still sold on the passive
antenna/coax approach for permanent installations.
As I mentioned earlier, I'm not convinced that flat audio response down to a
couple of hundred hertz is necessary or even desirable. Try adding a switch
in series with the bootstrap 10k resistor (the one that goes to the .0033 uF
cap). With the switch open, the input impedance will drop to 10 megohm and
the frequency roll-off will jump up to a couple of kilohertz. You will still hear
lower frequencies, just somewhat attenuated. I suspect that switch position
will be your favorite when in town. (I haven't tried this circuit yet, by the

Here are a few recordings:

Loud Lightning Crash
That motor starting (Not as loud as the magnetic pickup, but still clear.)
Little Chirps from distant lightning.
There are some interesting carriers that come and go, too. A narrow carrier
just appeared at 17.8 kHz, dropped in amplitude, then reappeared as a
spread-spectrum signal. I caught a second burst of the spread-spectrum
signal starting up (and ending as I type):
Here's a signal starting on 21.4 kHz that may be from Hawaii. It is big
enough to see on the 'scope, despite the AC hum. Now that tells me the
antenna/amp are working!
The plot below shows that there is really a lot of activity right now.
The 17.8 kHz signal has retired but several others are quite strong.
A quick web search suggests that the 24 kHz signal on the right is from a transmitter
in Maine.

Magnetic Pickup
There are two approaches for receiving the signals, a loop or coil antenna
that picks up magnetic fields or a whip or wire antenna that picks up the
electric field. (Nature manages to make electromagnetic waves at these
frequencies but you cannot make an "electromagnetic" or tuned antenna, it
would be too big.) The prototype magnetic pickup coil shown below is made
with a common 120 volt valve coil. Its 1/2" dia. hole perfectly accommodates
the larger Amidon ferrite core (R33-050-400). (These removable coils slip
over an enclosed solenoid and are commonly used in industrial equipment
valves. Make sure to take them out of the metal housing and remove any
metal parts.)
The core is simply secured with a cable tie on each side of the coil. The
resulting inductance is about 900 mH. The loopstick is hot-melt glued into
the case as is the circuit board and battery holder. The front panel was
designed with a CAD program, printed on glossy report cover stock and
sprayed with a clear coating. Spray adhesive holds the label to the Bakelite
cover. The second op-amp is not visible in the photo below; it was added
"dead-bug style" near the pot.
The spectrum exhibits plenty of activity down to 60 Hz and up to the top of
the hearing range but, presently, I am not sure of the actual bandwidth. The
black coil probably wasn't wound with picking up submarine communications
signals in mind! I suppose I could disconect it to determine the resonant
frequency but it seems plenty adequate for the task. (See the spectrum chart
made with my new Microtrack above.)

This is a spectrum with the antenna on my desk in a less-than-perfect

orientation. Despite all the noise in this room, the VLF transmitters can be
seen at the high end (along with a few computer monitors). Let me reiterate
that this environment is awful!
The 2.35 meg resistor is made from two 4.7 meg resistors in parallel. Most
sound cards and recorders will not need the additional gain provided by the
second op-amp and the volume control will need to be set very low but the
extra gain is nice for listening with a crystal earphone.
My version is slightly different than the schematic. My second op-amp is a
CA3130 with a 56pF compensation capacitor between pin 1 and pin 8. Also,
pin 3 of mine connects to the wiper and pin 2's 10k connects to the bottom
of the pot. These differences shouldn't matter. I just walked out to the end of
the sidewalk with the antenna and my laptop and made a short recording
with the antenna sitting on the ground. It leaves a lot to be desired
compared to my e-field whip but I have strong magnetic fields out there; a
fair test will require some effort. Notice the predominance of little chirps. The
magnetic pickup seems to pick up less of the local static crashes and more of
the distant signals than the whip antenna. I think this thing will be a real
contender away from power lines.

Old spectrum snaps:

Typical Day
Lots of activity! What's that at 19,460?

I think I'll unplug my antenna now!

Look at the amplitude and bandwidth of this signal!


Here is a spreadsheet that can be used to compare various op-amps for use
as VLF amplifiers:

The spreadsheet allows the user to adjust the antenna length, add shunt
capacitance, vary the bias resistor and noise voltages and currents for a
particular amplifier. The OP27 and OPA363 are the default amplifiers. The
default setup is similar to my 2 meter antenna with the long coax cable.
Warning: I just threw this spreadsheet together and there may be an error or
two! If you try to figure it out, I converted noise sources to Norton
equivalents to make the math easier (except the op-amp noise voltage which
just adds).
Note that the OP27 is a good choice for my long cable length but it would be
less than optimum for a short cable. Also, try a JFET like the older LF355. It is
only 11dB worse than the OP27 at the floor with my long cable and better
through the audio hearing range for as much as 100pF shunt capacitance.

Alysson sent a couple of compelling emails regarding whistlers that I have combined
into one below. Although I have no scientific basis for my thoughts, her explanation
seems on target. It is easy to imagine that a batch of high-velocity ions from the sun
arriving near a pole will spiral in the earth's magnetic field and will radiate energy in the
process. As they slow down it makes perfect sense that the frequency will drop. Diffuse
whistlers might be due to a spread in arrival times for larger clouds of ions. I personally
don't see the connection to lightning (that's not saying much) since the ions will travel in
spirals simply due to their kinetic energy and the magnetic field. Also, since all the
"whistlers" I've picked up in Texas have been motors, I wonder if the radiation is more of
a near-field phenomenon. Anyway, enough ranting by an uninformed hack; on to the
excellent emails:
I was just browsing your site, and found your VLF Whistler Detection article
In that page, you state that:
Short whistlers might be due to
dispersion, but some whistlers
last five seconds so ordinary
dispersion is probably
inadequate an explanation. A
radio wave can travel a million
miles in five seconds so to
accumulate that much
difference in arrival times, the
signal would have to travel
hundreds of millions of miles,
assuming a pretty steep
dispersion curve. More likely,
the whistler is an emission from
the magnetosphere triggered
by the lightning pulse.
You may be interested to know that when I was at University College of Wales,
Aberystwith in the late 1970’s, the Physics department research budget was spent
almost entirely in the pursuit of this phenomenon. The research was led by my tutor, Dr
A.D. Maude at that time (sorry, no citation, just my memory of his lectures, but you
might like to try
hl=en&q=Aberystwyth+Ionospheric+Whistler&meta= ). What was found was that these
were generated by the entrapment of high-energy ions in the Earth’s magnetic field. The
ions originate, reasonably enough, at the solar surface.
At the relatively low field densities of the temperate latitudes in the high ionosphere,
these particles will penetrate deep into the field before losing enough energy to become
trapped. They are then deflected along the field, but due to their momentum, travel in a
corkscrew orbit. This acceleration of the charge results, quite naturally in the emission
of radio frequency radiation.
In shedding further energy, the helical orbit decays – becoming more linear (a longer
orbital period), and thus a change in the frequency of the emitted RF. The period of
decay for these high energy particles is of the order of 5 to 12 seconds, but the VLF
signals become difficult to detect without a high-altitude detector. (We sent up rockets
I went to look at what remains of my old notes last night …. From what Dr Maude said,
the lightning whistlers are probably caused by ions generated within the storm itself
undergoing the same process, but since their energies are only a few MeV and in
(relatively) low atmosphere, the paths tend to be short (maybe 2-3 seconds only).
The best whistler times are during periods of intense sunspot activity – when large
volumes of solar ions are being pumped out with energies in the GeV and possibly TeV
ranges. My own theory (unsupported) is that the thunderstorm RF discharges cause a
resonant oscillation in the solar ions which, as you say, will kick them into their helical
orbit. The solar whistlers are, though, observable at any time albeit rather less
Dr Maude’s rockets were built using (then) the highest sensitivity receivers available (it
was a very generous budget), the whistlers only being detectable from stratospheric
altitudes or above. I find it amusing that it is now possible to put together a receiver
sensitive enough to make observation of this phenomenon for “pocket money” prices.
I hope this is of interest.
Alysson Rowan

My site contains a few low power transmitters of one type or another, but until now
no receiver. This circuit can be used to "sweep" an area or room and will indicate if a
surveillance device is operative. The problem in making a suitable detector is to get its
sensitivity just right; too much and it will respond to radio broadcasts, too little
sensitivity and nothing will be heard.

This project has few components, can be made on veroboard and powered from a 9
volt battery for portability. My prototype shown below, worked OK on a
Circuit operation is simple. The inductor is a moulded RF coil, value of 0.389uH and
is available from Maplin Electronics, order code UF68Y. (See my links page for
component suppliers.) The coil has a very high Q factor of about 170 and is untuned
or broadband. With a test oscillator this circuit responded to frequencies from 70 MHz
to 150 MHz, most of the FM bugs are designed to work in the commercial receiver
range of 87 - 108 MHz. The RF signal picked up the coil, and incidentally this unit
will respond to AM or FM modulation or just a plain carrier wave, is rectified by the
OA91 diode. This small DC voltage is enough to upset the bias of the FET, and give
an indication on the meter. The FET may by MPF102 or 2N3819, the meter shown in
the picture is again from Maplin Electronics, order code LB80B and has a 250 uA full
scale deflection. Meters with an FSD of 50 or 100 uA may be used for higher

In use the preset is adjusted for a zero reading on the meter. The detector is then
carried around a room, a small battery transmitter will deflect the meter from a few
feet away.


An AM voice transmitter with variable tuning. The antenna circuit is also tuned and
transmits via a long wire antenna. Please Note. It is illegal to transmit on the AM
wavebands in most countries, as such this circuit is shown for educational purposes

Please read the disclaimer on this site before making any transmitter circuit. It is
illegal to operatea radio transmitter without a license in most countries. This circuit
is deliberately limited in power output but will provide amplitude modulation (AM) of
voice over the range 500kHz to 1600kHz with values shown. You can input values in
the calculator below, remember to change drop down box to picofarads for
capacitance and microhenries for the coil. The coil is fixed at 200uH, the capacitor
values can be varied and resonant frequency found by using the calculator below.

Tuned Circuit Resonant Frequency Calculator

Top of Form
Capacita 6 Microfarads

Inductanc 9 Henries


Bottom of Form

Coil Data
If winding your own coil then you may find Martin E Meserve page very helpful:
Single Layer Air Core Inductor Design

An alternative is to use a toroid core of appropriate material. Toroid's come in

different sizes and colours, see the sample below.

A T130-2 core requires approximately 137 turns of 36 SWG wire.

Mike Yancey has a very useful Toroid calculator on his webpage, link below:
Toroid Calculator

Circuit Notes
The circuit is in two parts, a microphone pre-amplifier built around Q1 and an RF
oscillator circuit (Q2). The oscillator is a standard Hartley oscillator which is tunable.
Tank circuit L1 and C1 control frequency of oscillation, the power in the tank circuit
limited via emitter resistor R1. The transmitter output is taken from the collector, L2
and C2 form another tuned tank circuit and help match the antenna. L1,L2, C1 and
C2 may be salvaged from an old AM radio if available. The antenna should be a
length a wire about 10 feet or more. In the schematic I have shown coaxial cable to
be wired to the "longwire" antenna, the outer coax shield returned to ground.
Ground in this case is a cold water pipe, however even without a ground and coax
cable a signal should still be possible.
L2 and C2 not only help match the antenna to the transmitter, but also help remove
harmonics and spurious emissions in the transmitter circuit caused by non linearity
in the transistors.

Q2 needs regenerative feedback to oscillate and this is achieved by connecting the

base and collector of Q2 to opposite ends of the tank circuit which is achieved by
C4. C3 ensures that the oscillation is passed from collector, to emitter, via the
internal base emitter resistance of the transistor, back to the base again.

Emitter resistor R1 has two important roles in this circuit. It ensures that the
oscillation will not be shunted to ground via the very low internal emitter resistance,
re of Q2, and secondly raises input impedance so that the modulation signal will not
be shunted.

Q1 is wired as a common emitter amplifier, C7 decoupling the emitter resistor and

realizing full gain of this stage. Bias of this stage is controlled by R4,R5 and R3. The
microphone is an electret condenser type microphone, R7 setting operating current
of the ECM and C6 providing DC blocking. The amount of modulation is controlled
by the 10k preset resistor PR1 which is also the collector load. The preamp stage is
decoupled by R6, C8 and C10. This ensures no high freqency feedback from the
oscillator gets into the audio stage. Some electrolytics capacitors have a high
impedance at radio frequencies, hence the use of C10, a 10n ceramic to bypass any
oscillator frequencies.

WARNING: Transmitting on the UK Domestic AM band is illegal in the UK, please see
the general disclaimer. In some countries however low power may be permissible,
but consult with the appropriate authorities first. This circuit is shown for
educational purposes only.

These "ion chambers" are nothing more than a bare wire stuck through a hole into a
metal can! No special gas or sealing is required. For best performance it is probably
a good idea to add a desiccant to the inside of the can to keep the humidity low. (I
didn't!) Build one; its really simple!
When ionizing radiation (ultra-violet light, x-rays, etc.) pass through a gas,
collisions with the gas molecules produces ion pairs, typically charged
molecules and free electrons. If an electric field is present, the ions will move
apart, each moving in opposite directions along the electric field lines until
they encounter the conductors that are producing the electric field.
An ion chamber is an extremely simple device that uses this principle to
detect ionizing radiation. The basic chamber is simply a conducting can,
usually metal, with a wire electrode at the center, well insulated from the
chamber walls. The chamber is most commonly filled with ordinary dry air
but other gasses like carbon dioxide or pressurized air can give greater
sensitivity. A DC voltage is applied between the outer can and the center
electrode to create an electric field that sweeps the ions to the oppositely
charged electrodes. Typically, the outer can has most of the potential with
respect to ground so that the circuitry is near ground potential. The center
wire is held near zero volts and the resulting current in the center wire is
The voltage required to sweep the ions apart and to the center wire or outer
can before a significant number of them recombine or stick to a neutral
molecule is usually under 100 volts and is often just a few volts. In fact, if the
voltage is above a couple of hundred volts, the speeding electrons will
produce additional ion pairs called "secondary emissions" giving an
enhanced response. Geiger tubes operate at even higher voltages with a
special mixture of gasses and exhibit a sudden and very large discharge for
each ionizing particle. But below 100 volts the only current is the ions
produced by the radiation. The resulting current is extremely low in most
situations and detecting individual x-rays is difficult, especially with ordinary
air at atmospheric pressure. Usually the capacitance of the electronics
connected to the center wire smoothes the individual pulses too much for
detection even when feedback is used to greatly reduce the time constant.
These room-pressure chambers therefore respond to the average level of
ionizing radiation and do not provide "clicks" like a Geiger counter tube.
Sensitive homemade ion chambers for detecting nuclear radiation are fairly
easy to build but the circuitry is tricky and should only be attempted by
"seasoned" experimenters - the currents are likely to be well below 1 pA
unless there is a serious nuclear war in progress! (The simple version is
"beginner friendly"!) Special electronics is needed at the front end, typically
called an "electrometer" circuit, which produces an output voltage in
proportion to the input current. The electrometer must have a very low bias
or leakage current to avoid masking the desired signal and the intrinsic
impedance of the amplifier must be extremely high. The input impedance of
the electrometer may be fairly low, however, using feedback to convert the
tiny current into a usable voltage.
Older designs used special electrometer tubes like the 5886 which requires
only 10 mA at 1.25 volts for the filament and about 10 volts for the plate.
These tubes are great for the experimenter because they are relatively
immune to static discharge and they consume about the same amount of
power as a typical transistor stage. Some electrometers use vibrating
capacitors or mechanical choppers to convert the tiny DC currents into AC
before amplification to avoid DC bias and leakage problems. Newer circuits
typically use MOSFETs or Electrometer grade JFETs in the front-end. MOSFET
op-amps usually contain protection diodes which can be responsible for
several picoamperes of leakage at room temperature and a fairly steep
increase in leakage as the temperature increases but in some ion chamber
applications this extra leakage is tolerable. Non-protected MOSFET front-
ends are easily damaged by static electricity and special low-leakage
protection diodes are usually added. Low current JFETs like the 2N4220 give
respectable performance and the types intended for electrometer
applications like the 2N4117A are quite impressive, exhibiting leakage well
below 1 pA. They have the added benefit of being significantly less sensitive
to static electricity than unprotected MOSFETs. Full ESD precautions must be
observed with any of these approaches!
Big Resistors!
As mentioned earlier, most electrometer circuits use feedback to reduce the
effective input impedance and to direct the tiny input current through a very
large feedback resistor such that a reasonable voltage is produced at the
output. The feedback resistor must be quite large, however. If the input
current is 1 pA and the feedback resistor is 100 megohms, the output
voltage will only be 100uV. Special resistors measuring in the millions of
megohms are available but are usually difficult for the experimenter to
obtain (see , for example).
Recently, I stumbled upon a quantity of 13,000 megohm resistors - send me
an email if you want one. That same 1 pA would give 13 millivolts which is
easily amplified with ordinary op-amp circuitry.
An Experimental Circuit
An extremely sensitive circuit was desired that didn't require special
resistors and that didn't fail every time a slight ESD mistake was made and
the result is the experimental circuit shown below. It uses a 2N4117A as the
input amp and another as the feedback resistor. If one studies the tiny
curves supplied in the data books and uses a little "extrapolation" and
imagination, the leakage of the 2N4117A with the drain and source
connected together can be seen to have a slope equivalent to about 75
million megohms! There is unit-to-unit variation and it is necessary for the
input JFET to have lower leakage than the feedback JFET so the circuit is not
for everyone. (If the input JFET leakage is higher, the output voltage will be
very low. Simply swap the JFETs.) The FETs are easily damaged, too, which
can lead to frustration when the "best one" gets zapped. The circuit will not
be particularly accurate since the actual feedback resistance is not known
(nor linear) but the experimental ion chamber is not easily characterized
anyway. Despite the circuit's shortcomings, it is extremely sensitive and
surprisingly stable. A "good one" might drift only 0.1 fA in a day if ambient
conditions are relatively constant. (That corresponds to about 10 mV drift on
the output.) The short-term variation is below 1 mV which corresponds to
0.01fA! If the ion chambers really work, this circuit should be able to see the
The input FET (the one on the right) and the two transistors form an "error"
amplifier that attempts to maintain the drain voltage at 10 volts (set by the
resistor divider in the emitter of the NPN). If current flows out of the ion
chamber causing the gate voltage to rise, the drain will begin to drop and
the voltage on the collector of the NPN will go up. This rise will decrease the
current in the PNP and thus lower the output voltage. The voltage drop
across the first "resistor" FET will increase and more current will flow through
it - nearly all of the ion chamber current, in fact. There is sufficient loop gain
that the input voltage does not change very much and most of the ion
chamber current flows through the feedback FET. The zener in the source of
the input FET moves the gate voltage operating point up above ground so
that dual polarity supplies are not needed. The output voltage should be a
few volts, perhaps 3 or 4, depending upon the relative leakage of the FETs. If
the voltage gets too near 6 volts, the sensitivity will drop and the response
will become more logarithmic (which might be useful for some applications).
If the voltage is too low, the circuit might "bottom out" and loose control.
There isn't much that can be done to set the operating point expect swap out
FETs! The glass around the FET leads must be VERY clean. Use a good
solvent to remove any contaminants.
Ion Chambers
The first experimental ion chamber was
made with a zinc can from a D-cell battery
and an old 8-pin glass-to-metal header as
seen in the photo below. The two FETs
were mounted inside the chamber with
the theory that this would eliminate the
problem of connecting the extremely high
impedance probe to the outside world
without creating leakage paths to ground.
The problem with the concept is that the
transistor bodies and leads compete with
the wire for the free ions! Carefully
painting the transistor bodies and legs with conformal coating helped but the
circuit will not tolerate the coating around the base of the transistor - it is too
conductive! (In retrospect, the transistors should reside in their own can with
the sense wire passing though a hole into the ion chamber which is what the
schematic shows.) The pickup wire should be thin and near the center of the
can to keep the capacitance low so the response time is as short as possible.
When power is first applied, it can take a very long time, maybe 20 minutes,
before the circuit settles out to a steady reading. At first, about 150 volts
was applied to the can but it was soon discovered that only a few volts are
adequate and a 9 volt battery was used instead. The 15 volt power supply
voltage should be fine for most ion chamber sizes. If the voltage is too low,
the readings will be low as the ions have time to recombine before being
swept to the electrodes.

To test the chamber, a 1.3" diameter disk of radioactive material (a

calibration disc from an old Geiger counter) was leaned against the can. The
voltage changed a few 10s of millivolts but I quickly lost interest in this
chamber when the lid slipped and zapped the FETs. I had already spotted an
old mint can at the back of the workbench which I liked better for a chamber.
(See pictures below.) An audio connector was added to the center of the 3"
dia. tin can and the FETs were mounted directly to the pins. A ring of
wire was used for the center electrode. The insides were washed well with a
solvent and then dried with a hot air gun before the base was added and
tack-soldered. (I should have removed the plastic coating on the inside of the
Little feet were added to the bottom of the can so that I could easily slip my
radiation disk underneath without disturbing the chamber. This ion chamber
gave gratifying result: the little radiation source gave an output voltage
change of about 70 mV which was very large compared to the meter wander
of about 2 mV.
At this point in the festivities, I decided to try a crude calibration. Really
crude. My calibration reference was a Heathkit Geiger counter which has a
meter that reads counts per minute and mR/ hour. The scales are a little
suspicious since the CPM scale is an exact power of 10 bigger than the
mR/hr scale. (0.3 mR/hr = 300 CPM on the X1 scale, for example.) It is
entirely possible that the Geiger tube dimensions were selected to achieve
just this result. In the past I had compared this Geiger counter against
another "bomb shelter" type and obtained surprisingly close readings -
maybe within 10%. The radiation disk gives 1500 CPM when held directly
against the Geiger tubes mylar window and 500 CPM when the lid of a mint
can is placed in between (to simulate the ion chamber walls). The
background radiation measured about 13 CPM. Now here is where the
calibration gets a bit "iffy". The disk is large compared to the Geiger tube
window but it is small compared to the diameter of the ion chamber. To
make a long story short, the ion chamber will read low by some factor -
maybe 4. What I think it all means is that the ion chamber gives about 6 mV
for a radiation level that causes about 10 CPM in the Heathkit unit
corresponding to 0.01 mR/hr. The background radiation should give a
reading just above 2 mV which is about how much the readings wander from
minute to minute. This calibration may be within one order of magnitude!

But now I am hooked. I want an ion chamber that can easily see the
background radiation. To make a bigger chamber, I chose a 4.5" by 4"
diameter peanut can (see below). I also decided to move the FETs into their
own compartment. For this compartment, I used a steel wheel from the
center of an electronic component reel. Any small can would work here, but
this piece fits nicely on the bottom of the peanut can and it had a hole
perfect for the 8-pin header that I happen to have in large quantities. A hole
was drilled in the bottom of the peanut can for the electrode. No insulator
was used - just the air gap. You can see the wire sticking up from the FETs in
the first picture and the wire is visible passing through a hole into the
chamber in the second picture. The end of the peanut can was sealed with
aluminum foil to keep out air currents and electric fields but to allow less
energetic or larger particles in.
The voltage on the outer can was increased to 22.5 volts by using an old 'B'
battery on the theory that a larger chamber would need a larger field to
sweep out the ions quickly enough.
After power was applied and
sufficient settling time went by
(about 15 minutes), the meter
reading was seen to be significantly
more jumpy. The FETs were the same
ones from the previous chamber and
great care was taken to keep
everything clean so I immediately
suspected that I was seeing
individual ion trails. When I slipped
the radiation disk under the
aluminum window, the reading
climbed to a whopping 1.5 volts! And
what a coincidence! The Heathkit gives a count of 1500 CPM for this same
source when covered with aluminum foil. (Actually, the foil hardly attenuates
the radiation coming from the disk.) So now I have a direct readout of mR/hr:
1 volt = 1 mR/hr. Unfortunately, I have not yet corrected for the much larger
detector area of the new chamber but it works out to be nearly 10! So the
sensitivity of the new chamber is 1 volt per 0.1 mR/hr which is pretty
sensitive! The radioactive element from a smoke detector was held up to the
Geiger counter and the count soared to about 22,000 CPM but placing a
piece of aluminum foil in between dropped the count to 200 CPM. The ion
chamber gave a reading of 200 mV which is in perfect agreement. But I
didn't expect agreement since this source is small relative to the Geiger
tube, also. The mylar window on the Geiger tube blocks the alpha particles
some and this may account for the agreement. These calibrations are really
coarse! By using the Geiger counter to measure the background radiation it
was determined that the ion chamber should be indicating 13 mV but since
the zero setting is arbitrary, it was hard to confirm this level. Reversing the
polarity on the outer can caused a shift of about 30 mV (after several
minutes of settling) which is about what is expected if the background is
near 15 CPM (plus 15 to minus 15 is a total of 30). The experimental setup is
shown below:
More Experiments:
I tried a long chamber made from a section of air duct with a thin wire
stretched between two Styrofoam plugs:

The internal wire was soldered

to both ends, the plugs taped in
place and the tube stretched to
tighten the wire.

The finished detector was pretty awful! First of all, the

electrostatic shielding on the ends was inadequate, causing a
huge 60 Hz signal. And, the wire would vibrate with the
slightest bump causing wild swings in the output. Maybe this
thing might be the start of a seismometer, but it stinks as an
ion chamber!

But then:
Another chamber was constructed with a large cookie tin similar to the peanut can
design above. The performance of this much larger chamber was excellent. A single
Coleman lantern mantle nearly "pegged" the output. The background radiation gives
about 4 mV (400 mV after amplification) which corresponds to 40 fA current. (Some
CMOS opamps have input current below 40 fA like the LMC6001 and would work fine
without the JFET.) Even though the circuit was given a low frequency response to
reduce 60 Hz response, the meter jitters in response to individual ion trails. (The
superior shielding of the cookie tins would probably allow for a faster response, if
desired, but watch out for circuit instabilities.)
This tin measures about
10.5" across and about
6" tall (a "regular" height
tin should work as well).
The center portion of the
lid was cut out with
scissors to make a frame
to hold the aluminum foil
window. The circuit is
housed in a smaller
cookie tin tack-soldered
to the bottom.
Connections are made
via a 5-pin audio

The electrode is a 5" dia. wire ring mounted to a Teflon standoff

and a short piece of stiff telephone wire connects the electrode to
the circuitry. The wire passes through a large hole to reduce the
chances for leakage currents. The circuitry is a modified version
of the first schematic featuring a resistor for the feedback and an
op-amp for boosting the output signal. The transistor circuit was
also modified to increase the loop gain and improve the stability
(see ckt. desc. below).
The electrode looks like a
A word of caution: the
metal sure looks like
ground and a person
(um, like me) might start
soldering components
that go to ground to it.
The can will actually be
connected to +45 volts
or more so the "ground"
connections are made
above the metal. The
only components that
connect to the can in the
photo are a large yellow
cap and a couple of white
caps used to support my
elevated ground buss.
The new circuit includes several improvements. The feedback FET is replaced with a
Victoreen 100,000 megohm resistor which is the long glass tube in the photo. A zener
diode was added to the emitter of the 2N4401 to increase the loop gain and a .01 uF
Miller capacitor was added to reduce the amplifier frequency response (for stability and
to reduce 60 Hz gain). An op-amp (OP-07) was added to boost the output by a factor of
100. The "zero" pot is used to set the output to a few volts since the OP-07 cannot
swing below 1 or 2 volts out without a negative supply. This pot must be able to be
adjusted to the gate voltage and with some FETs the voltage may not go low enough.
The symptom will be a high op-amp output voltage. If so, just lower the 10k resistor or
add a 1k above the pot. An additional zero pot for the meter could be added as in the
first schematic to get a near-zero reading for the background radiation, if desired.
Notice that the drain resistor was reduced to 125k. This value was experimentally
determined by finding the drain current that gives the 2N4117A a near-zero temperature
coefficient. The test circuit is simple: connect a sensitive current meter from +10 volts to
the drain, ground the gate, and connect the source to ground through a 500k pot. The
current is observed at room temperature then the FET is warmed and the current
change is noted. The pot is adjusted until little or no change occurs. I heated the FET by
touching a warm PTC to the can - probably reaching about 65 degrees Celsius and the
final current change was below the current change caused by a 100 uV gate voltage
change. (Corresponds to less than 1 fA ion chamber current for 40 degrees.) Room
temperature may vary by +-4 degrees which would correspond to a wander of 0.1 fA
which is well below the 40 fA background current from the chamber. The bias current
that gave this wonderful temp-co was 40uA and since the drain resistor will have 5 volts
across it, the desired resistor value is 5/40 uA = 125k. "Your results may vary." Actually,
the FETs are surprisingly stable at all currents and the whole procedure may be
unnecessary; just use 125k!
Also consider the circuit used in the CDV-715 mod below. The mod is hard but the
circuit is easy. Also, my circuits use ultra-low leakage JFETs because I have about a
thousand of them but there are also op-amps that can do the job directly. Investigate the
LMC6001 (25 fA, tested!). Just leave the FET and source resistor out of the CDV-715
circuit below and connect the negative input of the op-amp directly to the sense wire.
Don't connect the 33k battery test resistor.
A 22.5 volt battery was insufficient to capture all of the ions but two batteries (45 volts)
seemed to do the job - in other words, higher voltage did not result in a higher reading.
Higher voltages may be desired for observing individual events, however, since the ions
will be swept to the electrode faster.
Parts Notes:
The 4.7uF capacitor should be a non-polar film type with a voltage rating above
the voltage used (45 volts in the schematic). A non-polar type allows the voltage
polarity to be reversed for experimental purposes.

The 100,000 megohm resistor is a specialty device which may be hard to obtain.

The 2N4117A is an unusual electrometer-grade JFET which has few substitutes.

The other components are not particularly critical.

A commercial Geiger tube is pretty hard to beat for general radiation

monitoring but the simplicity and versatility of a homemade ion chamber
that requires no special gasses or pressures makes it an attractive
alternative for many experiments.
Interesting email from readers

Ion Chamber Bias Supply (Battery Topper)

After discovering that one of my bias batteries was jumping around a few
volts, wreaking havoc with the readings, I decided to build a floating,
regulated high voltage supply. The result is a micro-power 110 volt supply
that runs on an ordinary 6V lantern battery. The circuit is similar to my
Geiger counter supply but without the additional voltage multiplier on the
output. Current consumption in a typical ion chamber setup is only 150 uA
(average) so a lantern battery will last a decade, if the shelf life doesn't get it
first. No power switch is needed.
The circuit has an output filter consisting of a 100k resistor and a .22 uF
capacitor and a typical experiment will have additional capacitance across
the chamber, too. To get rid of the last bit of wander as the circuit pulses,
increase the 100k to 10 megohm since ion chambers don't draw significant
current. I left the value lower in case this device is used to drive a heavier
load and added an external 10 meghom in series with the output with a 10
uF polyester capacitor to ground right at the ion chamber. The load can't be
too heavy, however. This circuit can just barely drive a 22 megohm resistor
with the full 110 volts and a 10 megohm multimeter will load it down to
about 85 volts. The .22 uF will still bite you when it's charged and a charged
10 uF will really get your attention! So be careful!
The circuit bolts right onto the terminals of the battery so I have dubbed this
type of circuit a "Battery Topper". Sorry, I couldn't help it! I actually like the
idea of mounting handy circuits right on a lantern battery for quick lab
circuits. Solder lugs were added to make contact with the battery terminals
and narrow nuts were screwed onto the battery first to raise the circuit up a
bit so that the hand wiring underneath doesn't press against the top of the
battery. The circuit was built on a piece of countertop laminate.

Experimenter's Ionization Chamber

Here is a truly simple experimenter's chamber made from an ordinary cookie tin:

This experimenter's chamber is made from a 4" (10 cm) diameter, 5.5" (14 cm) tall
tin with a tight-fitting lid. The inside of the can is conductive and does not appear to
have the typical plastic coating. A 5-way binding post is mounted in the center of
the can and a 4" (10 cm) wire is suspended from the post inside the can. The wire
length is short enough to insure that it doesn't touch the lid. Another all-metal
binding post and pin are installed in the bottom of the can, and a sheet of gray
insulating plastic is glued into place to keep hastily constructed experiments from
contacting the can. The electrometer circuitry will be extremely sensitive to stray
electric fields, so a shield is mandatory. Another can previously containing mints is
pressed into service:

A pin jack is soldered to the tin so that it can be plugged onto the pin on the
chamber, and tape is applied to the lip of the can so that the pin is the only
connection point. The inside surfaces of the shield are also insulated with
tape to prevent accidental contact with the circuitry. The method of
connecting the shield isn't critical, and a clip lead will also work; the main
culprits are the ever-present, low frequency, line-related electric field and
changing electrostatic fields due to movement near the chamber. The wires
from the test circuitry can simply slip between the shield and the chamber,
or a small notch may be made in the shield to make a little room for a few
conductors. The opening of the chamber may be covered by the original lid,
aluminum foil, or wire screen, depending on the experiment. Leaving the end
open will let in too much stray electric field in most environments.
Here's a simple starter experiment:
The can is connected to the positive battery voltage through a 4.7k resistor, and the
meter is connected between the collector of the transistor and the positive terminal of
the battery. The meter is on the 1 volt scale for most measurements.
The transistor is an ordinary NPN
Darlington type like the MPSAW45A.
The resistor can be any value above 1k;
it simply limits current in the event of a
short circuit. A little piece of double-
sided foam sticky tape holds the
battery in position.
When a ray passes through the
chamber, several atoms are ionized and
the positive voltage on the can attracts
the electrons. The positively charged
atoms wander to the more negative center wire and, upon contact, reclaim
their missing electrons. This process results in a current flow in the base of
the transistor which is amplified by a factor near 30,000. This higher current
flows through the 10 megohm resistance of the meter, producing the
indicated voltage. As a point of reference, a reading of 10 mV would
correspond to roughly 200,000 electrons per second, so even weak
radioactive sources produce large numbers of ions.
To observe the background and leakage level, the lid is placed on the
bottom, the top shield is added and the reading is allowed several minutes to
stabilize. The meter settles to a little over 30 mV and exhibits an occasional
jump. A camping lantern mantle known to contain radioactive thorium is
place in the lid of the chamber, and the lid is secured on the open end of the
can such that the mantle is inside the chamber. The meter reading climbs to
over 600 mV:

Placing the item to be tested inside the chamber in this manner gives the
ultimate sensitivity, but care must be taken to avoid touching the center
This very simple detector demonstrates how easily an effective radiation
sensor may be made with a minimum of effort. Below is shown another way
to build this simple circuit with even less effort. First, solder the 4" wire
directly onto the base of the transistor:
Drill a hole in the can right in the center of the bottom
and epoxy the transistor, face-down, such that the
wire protrudes into the can without touching the
sides. Make sure that no epoxy touches the center
lead of the transistor (base lead). The epoxy is too
Connect two wires to the collector and emitter leads.
The picture shows a length of solid copper telephone
cable used for the connections. The two blue
conductors are connected to the transistor legs and
the two orange wires are connected to the can. The
blue wires are given a little slack so that the cable
pulls on the can connection and not on the transistor
The resistor is connected on the other end of the
orange wires. A mint tin is tacked to the coffee can in
a couple of places to act as a shield and a wiring reminder is written on the
back of the mint tin. A little optional notch is seen near the bottom of the
mint tin that allows the phone cable to exit the tin without being pinched.
The opening of the coffee can is covered with ordinary kitchen aluminum foil
held in place by an elastic band.
Without a radioactive source, the meter reading settles to 50 mV and placing
the lantern mantle on the foil gives a reading of just under 150 mV. The
reading is lower because the foil is blocking a significant amount of the
radiation from the mantle.
This simple circuit has serious limitations. It is extremely sensitive to the ambient
temperature and a slight warming will cause a large increase in the zero reading
and the gain. A modern small-signal transistor should be chosen since older or
larger die transistors will probably exhibit too much leakage current. Despite these
limitations, the simple circuit can be used for sensing unusually high radiation
levels, observing sudden changes as when bringing a radioactive item near the
chamber window or even for making functional but silly gadgets simply for the
entertainment value! Seeing such a simply chamber work so well can provide the
motivation to build a more sophisticated circuit, too.
An Improved Transistor Circuit for the
Ionization Chamber

The single Darlington circuit could use more gain for detecting weaker
sources, but simply adding another transistor exacerbates the temperature
drift problem. By building a differential amplifier with one side not
connected, a temperature-compensated circuit results:
If the room temperature increases a little, both sides leak more current but
the voltage across the meter stays about the same. The 10k resistors are not
necessary but are included to protect the transistors from inadvertent shorts
during testing. Another 10k resistor between the battery voltage and the outer
can is probably a good idea in the event the can is shorted to the base of the transistor.
The schematic shows 12 volts but a 9 volt battery works fairly well. The higher
voltage sweeps the ions out of the chamber faster, before they have a
chance to recombine so the meter reading will be a little higher at 12 volts.
The sensitivity is good; the photo shows the response to a radioactive mantle
held about an inch away from an aluminum foil window on the end of the
chamber. Construction of the chamber is similar to the earlier devices except
the metal end cap from a spool of resistors is used for the shield. This end
cap accommodates a multi-pin header that is used to hold the components
and also acts as a feedthru. The ion chamber sense wire is soldered directly
onto the base of the transistor and is suspended in air by the transistor
without additional support. It might be a good idea to use a plastic bead to
hold the wire more securely as described above but this unit is well-behaved
as-is. The meter should read upscale slightly but, if it doesn't, try connecting
the sense wire to the other base and reverse the meter leads. If it still
doesn't read a little above zero, clean the transistors with acetone. If that
doesn't work, try new Darlington transistors since you probably have a leaky
one. Hold a radiation source near the aluminum foil window and the reading
should quickly climb. The meter can be a 1 volt digital panel meter or even a
digital multimeter set to the 1 volt scale.
Since the original construction of this meter, a couple of useful modifications
have been made. A wire shelf was added to allow items to be inserted into
the chamber through a small hole in the foil without fear the item will touch
the center pin. (The foil is peeled back to reveal the shelf in the photo to the

Additionally, a 100k potentiometer was connected across the battery and the
wiper was connected through a 1 megohm resistor to the base of the
Darlington that was originally not connected. This potentiometer becomes an
effective zero control. The picture to the right shows the reading produced
by a 2% thoriated tungsten welding rod inserted into the chamber such that
it rests on the new internal wire shelf. The meter was zeroed with the new
potentiometer before inserting the rod. The reading climbs quickly to over
half-scale and drops slowly back to zero when the rod is removed.
Modify a CDV-715 Survey Meter for High
This project is not for the faint hearted. It's way up there on the difficulty scale and I
only give enough information here for the more advanced experimenter. The circuit
would be great for homemade chambers like the cookie tin version above.
This project endeavors to modify an ion chamber style radiation meter, the
CDV-715, so that it has useful sensitivity, taking advantage of the internal
ion chamber, high value resistor, and nice packaging. The new circuit
increases the electrical gain by 1000, converting the scales from R to milli-R.
I decided to abandon virtually all of the existing electronics for several
reasons. The chamber current is converted into a voltage by switched
resistors but the ceramic switch will simply leak too much; it can only be
used for range switching after amplification. There is no need for 1.5 volts
since there will be no filament to heat so the inverter that converts 1.5 volts
to 50 volts and 10 volts will be unnecessary. A 9 volt battery will power the
electronics and a unique version of the Cockcroft-Walton voltage multiplier
will generate the 50 volts for ion chamber bias. The electrometer tube will be
replaced by a very low leakage JFET and a micro-power op-amp to save
power. According to the manual for the CDV-715, the output should be about
14 femto-amperes per mR/hour, a pretty tiny current. Using the 220,000
megohm resistor already in the instrument, the corresponding voltage will be
about 3 mV per mR/hour which isn't completely unmanageable. If the 100X
switch setting is to become 500mR full-scale, the voltage will still only be 1.5
volts when the meter pegs. So, no range switching is needed at the input.
In order to take advantage of the instrument's calibration, the ion chamber
should be biased at 50 volts as is done by the original circuit. A modified
Cockcroft-Walton voltage multiplier driven by an astable two-transistor flip-
flop boosts the 9 volts to 50 volts. This inverter circuit draws only 100uA but
for even less overall power consumption, five 9 volt batteries could be added
in series with the main battery to get 45 volts. These batteries could be tiny
10A size batteries with no power switch since there is virtually no current
required. I potted my inverter with wax in a little plastic box:
The voltage output is a fraction of a volt below 50 volts with 9 volts power;
just right! A very high impedance voltmeter is required to accurately
measure the voltage; an ordinary 10 megohm meter will load the voltage
down to about 35 volts. As the wax was cooling, I dropped a piece of tinned
copper-clad board on top of the wax to serve as a bottom cover and I
soldering a couple of wires onto the copper that line up with some of the
holes on the PCB for mounting.
I removed all but a couple of the components on the original circuit board,
leaving only the 5.6 meg resistor and 0.1uF capacitor used for filtering the
50 volts. I don't show them on the schematic but they are between the 50
volt inverter and ion chamber. Feel free to leave them out; they aren't really
doing much. I mounted the 220,000 megohm resistor to a new terminal
bolted to one of the switch mounting holes. The other end of the resistor has
a large diameter socket for the pin of the ion chamber and a smaller socket
soldered on sideways for the JFET gate. If you decide to not use sockets for
the JFET, you must turn power off and discharge the chamber to the negative
of the battery every time or you will lose the JFET! (I lost several - bad habits
die hard.) I used two traces right below the JFET for its drain and source,
cutting them to make two isolated pads. The sockets are sticking up because
they will hit the ion chamber if they stick out the bottom of the board.
The op-amp circuit is built on a piece of tinned copper-clad board. One op-
amp is used to generate V/2 (4.5 volts) and the other is the feedback
amplifier. I used two single micropower op-amps but a dual CMOS type would
be fine. I like micropower types to give the best battery life but an ordinary
LM358 should be fine.
I used the switch section with the terminals sticking up (where the high-value
resistors were connected) for the power switch and the switch terminals
mounted in the PCB for the range switch. The three resistors that set the
range may be mounted in the holes where the cal pots were located.
However, cut away the trace on the wiper terminal located near the ion
chamber pin because that trace goes other places. Feel free to figure out a
completely different switching scheme.
If you use the chamber as-is you might have a roughly calibrated unit but if
you want maximum sensitivity to any radiation, you may want to modify
your chamber. I unsoldered the chamber holding it over the stove (after
removing the solder plug in the center to let out pressure). It took two oven
mitts and a bit of effort. The heat partially melted the internal plastic liner
and the exhaust fan was quickly turned on. I removed the liner and the metal
plate, drilled a large hole in the lid, nearly the full diameter, and replaced the
disc electrode with a wire loop:
To make the foil window I cut aluminum foil to the outside diameter so that it
folded up at the edges when inserted from the back of the can. I then added
a band of tin (length of grounding strap) to press the foil against the side and
bottom of the can and tack-soldered the tin band into place while pressing. A
little conductive silver paste insured good electrical contact and a little bead
of glue sealed the seam. In order to take advantage of the thin foil window, I
cut a 3" round hole in the bottom of the case directly beneath the ion
chamber and covered the hole with some wire mesh. Remember, chopping
up the chamber like this destroys any "calibration" but makes the unit much
more sensitive.
I reinstalled the battery holder after installing a metal 9 volt battery clip
inside. A battery snap connects the battery to the circuit. Total current is
about 400uA.
The result is a very sensitive detector! On the X10 scale a single lantern
mantle will give a reading of 1 on the meter and there are two more
sensitive ranges! (There is a mantle beneath the unit in the picture above.)
On the X0.1 scale the meter occasionally jumps up to mid-scale when a "big
one" strikes and it gives a 1/2 scale reading for a mantle 5 inches away!
There is no provisions for zeroing the meter but setting it near the bottom of
the scale on the X0.1 scale seems adequate. The zero pot is touchy on the
most sensitive scale and you may wish to tame it down by increasing the
series resistors once you know the required voltage. I soldered a wire from
the case to the 50 volts, making it long enough to allow the case to be set to
the side with the unit open. Alternately, a spring contact could be added to
make contact between the case and lid. The round metal plate that was in
the ion chamber could be used as a beta blocker by placing it between the
aluminum window and the wire mesh. It would be a good idea to add a fresh
desiccant packet to the inside of the ion chamber since humidity impacts the
sensitivity. I didn't solder the ion chamber back together since it was such a
snug fit so it is relatively easy to open for changing desiccant packets. I think
I'll add a clip on the inside of the chamber to hold one. You could just seal
the chamber really well but pressure changes might cause problems with the
aluminum window. I just had an idea: there is a little solder-covered pinhole
in the middle of the back of the ion chamber that could be used as a
breather hole. Seal the chamber really well, leaving that little hole open.
Now, somehow affix a little metal or plastic box over that hole big enough to
hold a desiccant packet. Poke a little breather hole in the box and the air will
have to pass through the desiccant to get into the chamber. I'll work on that.
I don't know if all this is worth the trouble but now we know it can be done!

The Polonium Pen

Recent events point to the need for a simple device for
testing cocktails and beers for excessive quantities of
polonium. The Polonium Pen is a pocket-sized ion chamber
with LED readout that is perfect for the job. Simply hold
the Polonium Pen over your drink and, if the LED lights up,
order something else.
The circuit is similar to the single transistor detectors
above and only requires two Darlington transistors, an
LED, and one or two resistors along with a battery, power
switch and tiny homemade ionization chamber. This
prototype was a bit difficult to build, but the pen may be
built in any number of ways as long as a couple of critical
requirements are met. The wire probe sticking into the ion
chamber must not touch anything that is ever-so-slightly
conductive, not even most glasses, and the electronics
must be surrounded by a metal package to shield it from
stray electric fields. Because the ion chamber is so small, the LED will only
light in the presence of a significant amount of radiation, most likely alpha

The prototype will light up brightly when held near the Americium bit in a
smoke detector's ion chamber and a polonium-laced drink should produce
the same reaction. But most radioactive sources the experimenter is likely to
have will not produce any response at all, placing this in the novelty category
for most of us. The LED also comes on when the button is first pressed for a
couple of seconds before fading out, hence the lit LED in the photo! This
"feature" insures that the pen is working properly.
The first prototype required a
high-value resistor (66
megohms) across the
emitter-base of the PNP
darlington to get the LED to go out but
the second unit didn't need one,
possibly due to superior insulation of
the sensor wire. A value as low as 22
megohm is fine and even lower is
probably OK for detecting really "hot"
drinks! The prototype's ion chamber is
made from a 1.5" long piece of 11/32" dia. brass tubing and any similar size
will work. The first probe wire was a pin from one of those "pin art" bed-of-
nails toys. The insulator in the photo is a glass necklace bead, but it turned
out to be far too conductive. It was replaced with a similar sized plastic bead.
The new bead was slightly too large for the brass tube, but by heating the
tube with a soldering iron while pressing the bead into the tube, the bead
was easily cut down to size. (Place masking tape on your fingers to prevent
burns!) The bead was easily popped back out with a screwdriver. A nail was
glued into the bead with a little length at the head end for making the transistor
connection and then the bead was glued into the brass tube. Another bead was
temporarily slipped into the other end to hold the nail in the center of the tube while the
glue dried. I switched to a nail for the probe because it fit in the new plastic bead better.
Tin and clean the nail before inserting in the bead, especially if steel flux is needed.
Later, solder the base lead on quickly; the plastic beads melt easily. The epoxy
technique mentioned earlier will also work well.
A fine mesh screen was wrapped over the open end of the
chamber to keep out wind currents and electric fields. Rolling
the screen between the fingers makes it conform to the
shape of the tube nicely. Solder the screen to the tube in one
spot. Some screen materials may require steel flux, but the
screen should be tinned with the flux and then washed with hot water and
detergent before placing it on the tube; that flux is highly corrosive! Copper
screen is recommended; the screen shown is plated copper and it takes
solder nicely. Make sure the sensor wire doesn't touch the screen and is
fairly well centered in the tube (not critical). The circuitry was constructed
using point-to-point wiring and is held in place by the base wire soldered to
the nail's head and the emitter of the PNP soldered to the brass tube. One
end of a battery connector was fashioned from another glass bead (long,
dark tube on the left in the photo below) and a copper-plated nail. A white
plastic bead was used to cover the end of the copper nail where the wire

Some thin Kapton tape was carefully added to insulate the circuit from the
outer tube but no tape was allowed to touch the NPN's base lead. Even
electrical tape is too conductive!
The outer tube was fashioned from 1/2" stainless steel tube, but copper
would have been much easier to machine. The assembly was placed next to
the tube to locate the hole for the LED, and a cotton swab was marked for
depth with a black marker for applying epoxy to the inside of the tube to
hold the glass bead battery terminal.

The swab was dipped in epoxy, inserted up to the black mark, and the epoxy
was swabbed along the inside wall of the outer tube. After the epoxy was
applied to the inside of the steel case, the assembly was slid into place (bead
first) and the LED was positioned to face out the hole. After the epoxy set,
conductive silver paint was gingerly applied to the edges of the screen to
hold the ion chamber in position and to ensure a good electrical connection
to the outer tube.
The other battery contact was made from a spring stolen from a AA cell
battery holder fastened to a 1" steel standoff. The screw fits inside the spring
and holds the narrow end of the spring to the standoff. The wide end of the
spring friction fits in the steel tube to make contact. It is nicely centered by
the tight fit and doesn't touch the center contact of the battery. When the
button is pressed, the head of the screw makes contact with the battery's
positive terminal, completing the circuit. The standoff "button" is held in
place by a homemade nylon bushing that screws into the steel tube.
(Threading the stainless steel was difficult! And fashioning the bushing from
a nylon rod was no picnic.) The end of the standoff that protrudes was
ground down to match the steel case after cutting off enough to eliminate
the threaded hole. The steel case was given a coarse sanded finish, too, to
hide all the scratch marks from the threading process! The brushed finish
actually looks nicer than the original shiny finish and it hides fingerprints.
This isn't the easy way to build such a pen but it gets the idea across.

A tiny 10A 9 volt alkaline battery was installed and the threaded nylon
bushing was tightened. A little clear epoxy fills the LED hole and the pen is

Every spy and "diplomat" should have one!

Note: It has been suggested that this pen might not detect polonium in a
drink at concentrations that are lethal. Polonium emits copious quantities of
alpha particles, really a mind-bending amount, but alphas cannot travel
through a liquid. They barely make it through a few inches of air. So, the pen
will only receive alphas from the very surface of the liquid. Polonium is
ridiculously toxic and it doesn't take much to kill. On the other hand,
polonium really pours out alphas. Bottom line is I don't really know if it works
and I'm fresh out of polonium to try.
High-Value Resistor from Neon Lamp

(One reader reports no conductivity with his neons so 'your results may vary'.)
I was intrigued by the idea of using an ion chamber with a bit of radioactive
material inside as a resistor-like circuit element for very high impedance
circuits. But as I was getting ready to try it, I remembered various
discussions about NE-2 and similar neon bulbs containing a radioactive
element to help the bulbs start. So it seemed that a simple neon lamp might
be a nice resistor. I connected a bulb to a leakage meter and measured the
current for various voltages across the lamp. The bulb was kept in the dark
during the experiment. The "resistance" measured about 300 gigohms with 4
volts across the bulb. To make sure I wasn't measuring leakage along the
glass, I snipped off the top of the bulb to "let out the radioactive gas". To my
surprise the resistance dropped in half! It turns out that the electrodes in the
bulb are slightly radioactive, probably due to added thorium. My Geiger
counter jumped from 10 CPM to 60 CPM when the elements were held near
the mylar window of the tube. The air must provide a better "target" than the
rarified neon gas for the radiation to make ions which is why the resistance
dropped when I snipped off the top of the bulb.
With a low voltage across the neon lamp, most of the ions have time to
recombine before reaching the electrodes and this gives the bulb a resistor-
like characteristic curve. I would expect the current flow to level out at some
potential as all the ions reach the electrodes. So this resistance isn't
particularly linear and probably varies from bulb to bulb but many projects
just need a really big resistor to pull some low voltage node toward a known
potential. For example, you might want to pull a gate of an FET to ground
with a really big resistor to maintain a very high input Z and a neon lamp like
this might by perfect.
To prepare the bulb, clean it with a powerful solvent like lacquer thinner or
acetone and keep it in the dark when in use. I have successfully painted a
bulb with black enamel which helps but I had to mask off the area near the
leads because the paint is too conductive. A black plastic box would be a
good idea. Close the box on the leads and then heat the leads with a
soldering iron so that they melt into the plastic, allowing the box to close
fully. Most plastics are excellent insulators. Also, keep the inside of the entire
project's case dark.
Oh, yes. There was one last experiment to perform. I wiggled the electrodes
of the neon bulb back and forth until they broke off. Now, all that was left
was the leads and the glass base. I measured the resistance and got a
reading in excess of 20,000 gigohms so the surface of the glass isn't a
significant player, at least for that bulb. The resistance might be even higher
than that; I just didn't have the patience to wait for the slowly drifting meter!
The leakage meter really slow down when the resistance gets that high.
I checked a few more lamps and found great variation between them. They
seem to vary between a couple of hundred gigohms to over 5000 gigohms.
That's a lot of variation within a batch of identical-looking lamps! Keep these
variations in mind. After testing some other types of lamps, I've concluded
that there is a wide variation in the amount of radioactive element in the
lamps. I was somewhat surprised to be able to easily measure the
radioactivity from those tiny electrodes of the first lamp I tested! That bulb
was unusually hot! More info: I tested the two tiny electrodes individually
and found that only one of them is really hot, probably due to manufacturing
So, there you have it; a box of neons with radioactive electrodes is actually
an assortment of resistor values!
I have a lot of the bulbs in the picture. ( (The series
resistor is insignificant and it was left in place.)

Nuclear War Detector

Being caught off guard by a nuclear war can
be inconvenient. This project should alert the
utterly unobservant individual to any
significant nuclear exchange or radiation
spill in the immediate vicinity, giving the
owner time to kiss various body parts
goodbye. The detector consists of a small
ionization chamber as described above with
the single Darlington transistor amplifier and
a sensitive level detector that activates a
flashing LED when the chamber current
reaches a sufficient level.
The device is built into two empty spice
cans, one serving as the ion chamber and
one as the circuitry shield and battery case.
The sensitivity is set so that the lamp will
begin to flash with approximately 3 pA of
chamber current, about the amount one
might expect if two lantern mantles were
stuffed into the can. The mantles would be
providing mostly alpha radiation in close
proximity to the sense wire but in a nuclear
exchange, the chamber would be responding to beta particles and gamma
rays. The volume of the ion chamber is about 1/8 that of an old CDV-715
survey meter and there are no elements within the chamber to flatten the
response for different energy particles. A wild guess for the sensitivity is
between 2 r/hr and 10 r/hr for lamp flashing. In other words, the lamp will
start flashing when the old CDV-715 is reading on the second or third scale.
Suffice it to say, a blinking light isn't a good situation.
The schematic features a front-end like the experimental chamber above but
without the current-limiting resistor connecting the can. It should be a less
likely to accidentally touch the base of the transistor to the can since the unit
is sealed and connecting the can directly to the battery makes wiring the
rest of the circuit a bit simpler. A 1 megohm resistor can be used in place of
the wire to connect the base to the pickup loop to protect the transistor, if
desired. Instead of a meter, the MPSW45 pulls current from the base of a
2N4403 which further amplifies the current and turns on the BS170 by
raising the gate voltage. When the FET is on, the LED with the built-in flasher
begins to flash.

When the unit isn't flashing, it draws virtually no current, so the battery will
last its normal shelf life. The unit may be tested by warming the chamber
with a hot air gun or hair dryer. Heat it gently and give time for the heat to
reach the transistors. The heat will cause leakage within the transistors,
simulating radiation and the lamp will flash until the transistors cool again.
The prototype circuit was built right onto the bottom of the chamber can with
the help of a small piece of PCB material. The can is at the positive battery
potential and the PCB is at the negative potential. The base of the Darlington
is connected to a black insulated wire (see picture below) that passes
through a generous hole to a Teflon standoff inside the can. (That black wire
could be a 1 meg resistor as mentioned previously.) The standoff also holds
a bare wire loop inside the chamber for collecting the ions. The side of the
loop isn't critical, and straight wire a couple of inches long is also fine.

After the pickup loop and wire are installed, the can is closed with a piece of
tinned PCB material soldered in a few places. Other solderable metal will also

(The views above are the same can but the top photo is after paint.)
The Teflon standoff may be replaced by a homemade plastic terminal. Those little
beads for making necklaces make great insulators although they do have the tendency
to melt during soldering. One way to limit the tendency to melt is to secure the electrical
terminal in the bead with epoxy putty. The putty insulates the plastic from the soldering
heat. Choose a terminal that doesn't conduct heat very well such as a copper-clad steel
nail (sold in most home improvement stores). Brass nails also work. Make sure the
epoxy doesn't get on the outside of the bead. A wipe with a little alcohol on a paper
towel will remove any epoxy residue. To mount the bead, drill a hole in the metal slightly
smaller than the bead and run a soldering iron around the hole to heat it on all sides.
Press the bead into the hot hole and the plastic will melt. With a little effort, the terminal
will be securely mounted with plastic insulation between the nail and metal inside and
out. The photos below show the raw materials and a finished terminal.
The nail can be pushed all the way through the epoxy if a feedthru is
desired. For this project, push the nail all the way through and the
base of the transistor may be connected on one side and the little
pickup loop on the other side, eliminating the need for the large hole
and black wire.
Mounting a bead in this manner
without the epoxy and nail makes a
great insulated hole for wires, too.
Just melt the bead into the hole and
pass the wire through. (Most wire
insulation leaks too much for ion
chamber connections and it shouldn't be allowed to
touch metal surfaces.) This larger bead is nice in that it
is tapered and has a shoulder in the middle. Heat was
applied with a soldering iron held in the right hand while the can was rotated
with the left hand. The left index finger pushed the bead down with a steady
force, requiring a bit of dexterity. The shoulder seated nicely and the plastic
flowed into the sharp edges of the hole; the bead isn't going
anywhere! (These bead photos are to illustrate the idea but were not used in
the Nuclear War Detector prototype.)
The larger pepper can houses the battery and LED. The battery holder is
fashioned from balsa wood strips secured with epoxy and a battery door is
cut into the bottom of the can with an abrasive cutting wheel (see photo
below). The LED is held in position by its anode lead which is soldered
directly to the can. A blue plastic "mosaic" piece from the art store covers
the LED hole.

It is easier to paint the cans first then touch up the solder connections later. Try a Q-tip
dipped in a little of the spray paint. The finished unit shown below is ever-vigilant
on the shelf of the author's bomb shelter, at least long enough to make the
picture, anyway.

Super-Sensitive Ionization Chamber

Improved performance may be realized with an electrometer JFET or ultra-

low leakage CMOS op-amp in place of the Darlington transistor. This detector
features an old-fashioned electrometer JFET transistor like those commonly
used in early smoke detectors, pH meters and electrometer amplifiers. The
leakage in these FETs is remarkably low, typically below 25 fA, depending on
the package and surface contamination.

One of the op-amps in the LM358 is used to generate 1/2 the battery voltage
near 4.5 volts to be used as a virtual ground. When the meter reads zero, the
gate of the FET will also be at 4.5 volts. The source will be at a slightly higher
voltage due to the FET characteristics, and the zero pot applies a voltage to
pin 3 that is set to that source voltage. This voltage will depend on the
individual FET. As a starting point, short the large value resistor and measure
the source voltage. Select the resistor to get that same voltage on pin 3 with
the pot at its center position. Don't be concerned if you need a much larger
resistor than 5k, even approaching 82k, but the source voltage shouldn't be
much above 4.5 volts, perhaps 6.5 volts at the most with higher Idss parts.
Reduce the 121k (not a critical value, by the way) for experimenting with
higher Idss FETs to keep the source voltage below 6 volts.
Specialized op-amps are also available with similar leakage properties to
electrometer JFETs (see the National LMC6001, for example). To use such an
op-amp, simply leave out the FET and connect pin 2 directly to the 1
megohm resistor. However, extreme caution must be used to prevent the
wire from touching the can or the op-amp may be destroyed. The 1 megohm
resistor protects the rugged FET, but it may not protect many op-amps. Don't
use a high voltage on the chamber with an op-amp directly connected to the
sense wire, or the turn-on surge will likely damage the op-amp. (JFETs are
especially good about this, since the gate forward-biases during turn-on.)
With only an op-amp, it may be desirable to connect the selected resistor
from the pot to zero volts instead of 4.5 volts since the required zeroing
voltage will be near 4.5 volts. Another 100k resistor would be a good choice.
In fact, the pot may not be necessary at all; simple connect pin 3 to the 4.5
volts from the other op-amp.
In order to convert the very tiny chamber current into a reasonable voltage,
a very large resistance is needed. This prototype uses a 100,000 megohm
(10^11 ohm) glass Victoreen resistor which will convert 1 fA into 100 uV at
the junction of the 1 meg and 10k resistor (10 mV at pin 1) making it
reasonably easy to see extremely small changes in chamber current. The 1
meg and 10k between pins 1 and 7 add additional gain and that ratio may be
changed to accommodate a different high value resistor. For example, a
Victoreen version of the CDV-715 has a 220 gigohm resistor and the 10k
could be raised to 22k to give roughly the same sensitivity as the prototype.
A much lower value, say 10 gigohm, can also work, but temperature drift
from the FET may become annoying. Most of these electrometer JFETs have
a bias current at which the temperature coefficient is zero, so the advanced
experimenter might wish to vary the source resistor to find the best
operating point. (An ultra-low current op-amp without the FET will exhibit
less drift.)
It is also possible to eliminate the JFET and use an op-amp with relatively
poor bias specifications, as long as the leakage remains fairly constant and
the offset drift is low. For example, the TLC2652 chopper-stabilized op-amp
used without the FET might appear to be a poor choice, due to its relatively
large 4 pA bias current, but that current is internally compensated and
remains fairly constant for modest changes in temperature around room
ambient. This leakage will require the zero pot be adjusted to a different
voltage by only about 400 mV. The tremendous offset voltage stability of the
chopper-stabilized op-amp will allow for the use of a much smaller feedback
resistor, too. Don't bother with op-amps that have input bias currents above
10 pA (which, by the way, is most of them).
Having written all that, I must say that the rugged electrometer JFET is hard
to beat for the hobbyist. I've zapped more expensive op-amps than I care to
chamber is
made from
an empty
3.8" (10
6.8" (17
cm) tall tin
can. Removing the chocolate
cappuccino wafers from the tin
required several cups of coffee but
was well worth the effort. Seriously,
tins like these take solder easily
and make great enclosures for a variety of quick scientific and hobby
projects. The insulator is made with another large tapered plastic necklace
bead as described previously. The hole was drilled small and then widened
carefully with a tapered reamer until the bead almost rested on its shoulder
when inserted. The metal around the hole was heated with a soldering iron
and the bead was pushed into the hole until the shoulder itself started to
melt a little. The plastic oozes into the splits in the metal from the reaming
process and becomes solidly anchored. These beads have a hole large
enough for a #4 bolt so two standoffs were mounted on each side of the
hole. The interior standoff holds a stiff piece of 14AWG copper wire salvaged
from a piece of Romex house wiring. The photo to the right shows the
internal copper wire soldered to one bent standoff and the outside view of
the other standoff. The bolt was eventually put in the other way because of
the difficulty in threading the nut onto the bolt down in the chamber. This
simple insulator is quite effective and seems to perform as well as exotic
Teflon standoffs. The only drawback could be the large amount of exposed
plastic that might accumulate a charge in some instances. The insulation
properties are excellent and a thoroughly melted-in standoff is very strong. A
piece of copper PCB material was cut to fit well inside a mint can cover, and
a center hole was drilled to clear the plastic insulator. The PCB will be at zero
volts potential and the tin will be at a higher positive voltage, so it is
important to make sure the two don't touch. The mint can shown below was
coated on the inside with electrical tape (not shown) and the bottom, which
is normally concave, was pressed out to make the interior of the can larger.
The rounded end of the red handle of a large X-Acto knife makes a good tool
for pressing the metal out. Remove the blade first! Two notches are
eventually cut for the wires and access to the zeroing potentiometer. The
PCB material was secured in place with spray adhesive. The circuit is built
directly onto the copper board using point-to-point wiring.
The large glass resistor is held in
position by two more necklace beads,
the new insulator of choice in the lab.
The beads were partially melted by
holding a soldering iron tip in the hole
for a minute or so, taking care not to
touch the plastic. While the plastic was
still soft, the bead was pressed onto the
resistor. The beads were melted onto
the PCB by heating the PCB with the
soldering iron and a little solder while
gently pressing the bead down. This
connection isn't particularly strong, and a little plastic glue was added to the
now-flat side of the beads when they came loose. Keep the resistor clean
and free of fingerprints, and make sure the wire from the insulator to the
resistor doesn't touch anything. Construction is crowded simply to allow for
more circuitry in the future.
The cans were painted yellow, but provisions must be made to make good
electrical contact between the chamber can and the end cap which can be
aluminum foil, wire screen, etc. The prototype has a short length of tinned
copper shield braid soldered to the inside of the can and folded over the lip
of the opening:

A single piece of bare copper wire would also work. An ordinary pipe clamp is
used to secure the foil or screen but a rubber-band would also suffice. The
clamping action forces the foil or screen against the grounding conductor.
Alternatively, the paint could be removed from the end of the can.
The actual unit deviates from the schematic somewhat and the finishing
touches evolved. Firstly, the digital meter was replaced with an analog
current meter with a custom face. In order to achieve good sensitivity, a
range switch was added as were connections for an external meter or
computer data taker. For "normal" sensitivity, a 3k resistor is connected in
series with the meter (3 volts gives a full scale current of 1 mA). The toggle
switch connects a 220 ohm across the 3k to give a total resistance near 300
ohms (adding the meter's resistance) for a X10 scale. The large meter was
mounted directly to the top of a box that once held a pair of calipers by its
electrical terminals instead of mounting it in the conventional manner. The
chamber was mounted to the box with a couple of sheet metal screws
screwed right into the sides of the chamber.
So far, these are pretty ordinary modifications, but now the project deviates
from "normal engineering practices". In order to get a very long averaging
time for the meter, a 1 farad super capacitor was connected across the
terminals of the meter. This unorthodox technique actually works pretty well.
The meter resistance is about 100 ohms so the time constant is about 100
seconds. By keeping the resistance across the super cap low, any problems
associated with dielectric adsorption, etc. are minimized. Another RC was
added for an analog voltage output (the two banana plugs at the top of the
photo above). The capacitor is 0.22 farad and the series resistor is 500 ohms
(two 1k resistors in parallel). The capacitor has two 5 volt zeners head-to-
head across it to prevent over-voltage which will quickly damage a super

The power supply is a bit of an after-thought to avoid batteries. The two

black wires come from an unregulated molded "wall-wart" power supply that
supplies about 20 volts. A 390 ohm resistor and 9 volt zener form a 9 volt
supply to run the op-amp and a 10k resistor, and a 16 volt zener provide 16
volts to bias the chamber. The modified schematic is shown below:
Not shown in the schematic is a last-minute power indicator LED added in
series with the 390 ohm resistor. If the circuit will not zero on the sensitive
scale, the 9 volt power supply is changing voltage too much (due to the extra
current required to charge the super caps). In that case, switch to a three-
terminal regulator and set the ion chamber voltage a little lower to assure
the regulator has enough voltage drop across it. A 100 uA meter movement
will also eliminate the problem, allowing the meter resistors to be raised by a
factor of 10, from 3 k to 30 k and 220 ohm to 3 k.
The ionization chamber works about as well as can be expected without
significantly more care in chamber design. The long averaging time makes it
possible to see radiation from very weak sources. Below is a plot of the
response when a radioactive lantern mantle is rested up against the foil
window then removed five minutes later:
The long super cap time constant is clearly seen and the absence of noise is
refreshing. Such extra-long averaging times would be useful for testing
samples slipped directly inside the chamber, too, but the casual
experimenter may wish to shorten the response time a bit by choosing lower
value caps. Although the schematic doesn't show it, a double-pole switch
was eventually added to disconnect one leg of each super capacitor for those
times when a fast response is desired. Also, keep in mind that the circuit will
take a long time to settle after power is first applied and much longer when
the super caps are used. The meter will peg for several minutes and
adjusting the zero with the caps switched in can be tricky since it takes many
minutes to see the results of an adjustment. With experience, the speed of
the drifting can allow the zero to be set with the caps connected. Simply try
to stop the meter. Then very, very slowly adjust the needle to a low reading,
perhaps 20% of full-scale. Expect to be "lost" once or twice with such a long
time constant!
Other Ideas
Using a wire mesh instead of
aluminum foil for the window allows
alpha and beta particles to enter
the chamber more easily. The mesh
also allows air to be blown into the
chamber with an induction motor or
brushless motor fan. (Brush motors
will probably generate too many
ions of their own.) The photo shows
a fan positioned off to one side to
create a bit of airflow through the
chamber. This arrangement blows
ions in the surrounding air into the
chamber where they are separated
by the chamber potential, producing an upward meter deflection. Lighting a
match or butane lighter anywhere near the fan's input pegs the meter. Alpha
emitters such as the element from a smoke detector or a lantern mantle give
very high readings if held nearby. Turning the fan on and off might provide
some idea about the level of ionization in the room, but leaving the fan on
and blocking the air with cardboard might give more consistent results since
the electric and magnetic fields from the motor may influence the reading.
The basement lab where this technique was tried showed a consistent 100
mV increase with airflow (the super caps and computer plotter were
indispensable for the measurement). Perhaps a radon test is in order!
A wire tray was built onto the original lid and a hole was cut so that items
could be inserted into the chamber right above the center wire without
opening the chamber. Power should be removed when installing or removing
the tray to prevent shorting the outer can to the center wire. The tray is
useful for observing the radiation from long objects like thoriated tungsten
electrodes or test tubes filled with radioactive substances. If a coarse mesh
is used the circuit will respond to the slight change in electric field when an
item is inserted, so some settling time will be required. Experience has
shown that observing the change when the item is removed instead of
inserted gives better results. Heavy items will cause the tray to bend toward
the wire slightly, so it is best to flip the lid 180 degrees so that the inserted
item rests on the bottom of the can with the screen above. If the chamber is
mounted with sheet metal screws through the bottom as with this prototype,
twist the lid slightly so that inserted items miss the screws. An open paper
tray allows the observation of radiation from alpha and beta emitters. The
tray in the photo below is made from thin cardboard and is filled with
potassium chloride, a beta emitter. The low level of radiation is just
detectable using the long averaging and a computer plot, giving about a 20
mV change. The discrete steps in the plot are due to the limited resolution of
the commercial data taker that has a 10 volt, 11 bit input (about 5 mV per

Fairly weak sources can be measured with the wire mesh in place. A thick
piece of pink granite was found to give a Geiger-counter reading about 60%
higher than the background reading using a long averaging time. The slab
gives a noticeable reading increase when placed in front of the ion chamber
as shown below. The chart starts with a "hot" part of the slab in front of the
chamber, then the slab is removed, and lastly, the slab is replaced but with a
less radioactive portion in front of the chamber.

The two large spikes are from the vibration produced when moving the slab; it is heavy!
The voltage change from the high reading to background level is about 160 mV. With
these long averaging times, the chamber performs well, even when compared to a large
tube Geiger-counter. (The chart runs for 30 minutes.)

Radon Accessory
Here's an interesting accessory to the ion chambers above or other radiation
detectors that will allow for the indirect detection of radon gas. Radon is a
noble gas and is hard to detect directly; it doesn't react chemically or easily
stick to anything and it is usually present in very small amounts. It has a
short half-life of under four days so the concentration in a home is due to a
constant replenishment as the gas seeps in from the ground or the
structure's building materials. Its decay products are much easier to detect
because they readily stick to dust in the air and typically have a short half-
life, making them more radioactive. This accessory draws air through a filter,
catching dust laden with the radioactive "daughters" of the radon present in
the air.
An ordinary computer case fan is mounted in one end of a 4" PVC plumbing
coupler as shown. This particular fan is held in place by friction but any
technique is fine, as long as it pulls air through the coupler. A white poster
board doughnut is secured with some foam caulking to seal the gaps along
the edges of the fan. The fan is oriented to pull air through a filter affixed to
the other end of the coupler with tape. The filter in the picture is a piece of
3M SV-DF01 dusting cloth but other thin dusting or cleaning cloths that will
allow a little airflow will also work.

When the fan is running, radioactive dust collects on the outside of the filter.
The isotopes of interest all decay within about a couple of hours so they are
decaying as they are building up on the filter. At some point, there is enough
radioactive material on the filter that the rate of decay equals the rate of
collection. This "equilibrium point" is seen as a flattening in the increase in
The filter end of the contraption is placed very near the end of an ion
chamber but with a little room for air flow. To get consistent readings, use a
single layer of corrugated cardboard as a spacer. Remove the spacer when
the filter and ion chamber are both flush against it.
This accessory may be used in two ways. If you have a data taker for your
ion chamber, it is interesting to plot the output vs time after you turn on the
fan. The radiation will build up, exponentially reaching a steady level. Turn
off the fan and the opposite exponential will be plotted. Here's a typical plot:
The faint dotted lines are an hour apart and the vertical sensitivity
corresponds to about 10 fA of chamber current per division. The fan is turned
on at the first marker and turned off at the second marker just after the
middle of the plot. The height of the response will depend on the level of
radon in the vicinity, the velocity of air through the filter, and a host of other
variables. It would be hard to calibrate the response but it could be useful for
checking the relative levels in different parts of a house or for periodically
checking for significant level changes. Seeing the exponential helps to
validate the detection.
Another way to use the gadget is to simply place it in the room to be tested
and let it run for a few hours. Then position it up against the ion chamber
and make a measurement. It can touch in this case, since no air is flowing.
The reading will settle after a couple of minutes. The radiation level will
begin to drop after about 15 minutes so don't take too long. Since these
isotopes are all but exhausted after a few hours, the filter may be used over
and over, until it becomes dusty enough to restrict air flow. The filter should
not cause a reading increase when held against the ion chamber at the
beginning of a test before the fan has run. If it does, it may be time to
change the filter. I simply affixed mine with scotch tape.
Either way you go, a long averaging time is best for the ion chamber so the
readings are not bouncing around. My squirrel cage fan pulls a much better
vacuum and collects a much higher density of radioactive particles. The
radiation was easily detected with an ordinary Geiger counter despite the
thick walls of the Geiger tube! Where's my gas mask?
If you want to try a simple experiment to see if this accessory is worth the
trouble, try taping a little piece of dusting cloth over the opening of a squirrel
cage fan. Let it run for three hours then gently affix it to the aluminum foil
window of an ion chamber. I used more tape attached to the original tape
because the dusty side of the filter should be against the foil. These
experiments are detecting beta particles which have little penetrating

The meter reading jumped from near zero to two-thirds scale on my simple
darlington chamber. And all this time I thought my house didn't have radon!

The following is a discussion of the components used in the ionization
chambers above. Most of the parts are ordinary resistors and capacitors with
no particularly critical specifications, but a few must be chosen carefully.
The JFETs used above are "electrometer" types, typically from the 2N4117
family. They have very low IDSS, usually below 250 uA. Here are a few

Part Number IDSS

2N4117, 2N4117A, PN4117, PN4117A 30 to 90 uA

2N4118, 2N4118A, PN4118, PN4118A 80 to 240


2N4119, 2N4119A, PN4119, PN4119A 200 to 600


Intersil House Number 343A002 < 100 uA

Motorola House Number SPF3059 200 uA typ.

The plastic parts seem to have lower leakage. Mosfets are also useful for ion
chambers but the vulnerability of the gate to excessive voltage makes them
difficult to handle, especially when a long wire is connected directly to the
gate. Using a positive voltage on the outer can and adding a resistor in
series with the JFET gate helps to prevent damage. If the outer chamber is
accidentally shorted to the sense wire, the JFET diode forward-biases, and
the resistor limits the current. Larger die JFETs might make interesting front-
ends. One obvious choice is the die for the family including the 2N4220 to
2N4224, 2N3821 to 2N3824, 2N5556 to 2N5558, 2N5457 to 2N5459, MPF109
and MPF 111. These FETs will exhibit more leakage than those listed above
but leakage should still be well below 1 pA at room temperature. The higher
IDSS suggests a higher resistor in the source to set the current to a low level,
perhaps selected for a couple of hundred uA. The FET current can be chosen
to reduce the temperature coefficient, but the technique can be tricky.
Temporarily ground the gate and select the source resistance that gives a
minimum change in source voltage when the FET is gently heated with a
warm PTC or resistor. Don’t overheat the part to avoid erroneous results.
Freeze spray cools the parts too much and the condensation leaks too much,
so stick with heating.
Electrometer-grade op-amps are available with leakage currents competitive
with the best JFETs. One of the best is the LM6001, tested to 25 fA! Most
CMOS op-amps will have leakage currents low enough to be useful in ion
chamber projects, but offsets of a few pA are to be expected with many. The
op-amp used with a JFET front-end is not particularly critical, and the
ordinary LM358 dual op-amp is a typical choice. The circuits above use a
"synthetic ground" set to 1/2 the battery voltage and most op-amps that will
operate properly on only 9 volts will work well.
High-value resistors are often a problem for experimenters. Companies that
manufacture such resistors include Ohmite, Ohmcraft, SRT Resistor
Technology, IRC, IMS, Micro-Ohm, and others. High value resistors can also
be found in older equipment like the CDV-715 survey meter. Most
experimenters collect such resistors by cannibalizing old equipment or by
finding them on an auction site. The prototypes above use either carbon film
or metal film resistors for the lower values. Metal film resistors are hard to
beat, exhibiting excellent accuracy and temperature stability, but most of
the resistors aren't critical. The power dissipation is low for all of the
resistors, usually well below 1/4 watt. The 390 ohm resistor supplying the 9
volt zener dissipates a little over 300 mW.
The Darlington transistors used above are garden-variety types but very old
parts may not exhibit the amazingly low leakage of modern devices. The
gain of one MPSAW45A was verified to be over 30,000 down to picoampere
inputs and the leakage might be as low as 100 fA at room temperature.
Large non-polar mylar capacitors, perhaps 1 uF, are suitable for bypassing
the chamber voltage to the battery negative voltage ("ground"). The use of a
non-polar capacitor allows the polarity to be reversed for experimental
purposes. For example, the background radiation can be distinguished from
zero offset or leakage by reversing the voltage on the outer can and
measuring the change in the output. A microprocessor or computer could
change the voltage with a relay then measure the current change after
sufficient settling time.
The super capacitors used to filter the meter and output can be expensive if
purchased new, but they are fairly common on relatively modern surplus
circuit boards. Their application above was more of an afterthought, but it is
fun to use such components in unexpected ways. They feature very low
leakage so the time constants can be extremely long without significant
Mechanical Parts
Teflon terminals are excellent for these projects but the plastic beads work
great, even the ones with the metallic glitter inside. Other phenolic or
ceramic terminals should be avoided or at least thoroughly tested. Most
glasses have too much surface conductivity, too. The softer plastic 5-way
binding post used for the first chamber is questionable but seems to work
well enough.
The best choice for the chamber is the tin alloy typically used for cookie tins.
They are easy to work and take solder beautifully. Watch those sharp edges,
however. Many of these cans have a plastic lining that forms an insulation
layer. Several chambers with that plastic coating seem to work fine but it
might be a good idea to choose a can without it or to sand it out. The plastic
could block alpha emissions from the metal alloy but the lighter alloys
probably don't emit many, anyway. Avoid un-plated steel cans and keep
soldering to a minimum on the inside since some solders have significant
quantities of radioactive contaminants. The tin plating should block any
alphas, as does the common plastic coating. Actually, it's probably pretty
hard to find un-plated or coated steel cans since they would rust!
These simple ionization chambers are the "crystal radios" of radiation
detection. They are truly simple to build and really work. They respond to
any ionizing radiation that can get inside the chamber from 100 nm ultra-
violet light through X-rays and gamma-rays. They also respond to alpha
(helium nuclei) and beta (free electrons or positrons) radiation if provisions
are made to allow them inside the chamber. Alphas can't even make it
through a piece of paper so a wire mesh or placing the sample in the
chamber is best. The serious hobbyists can find a wealth of literature for
improving these simple designs, converting them into serious scientific