you ever heard of a 'nuvistor?' It didn't seem familiar to me right
away until after I looked it up. Nuvistors were high mu (high gain)
tubes, manufactured originally by RCA, used in sensitive receiver
front ends. They came in about a dozen different varieties. This
particular NASA-136 receiver for satellite reception uses a 6CW4
Wikipedia, "Most nuvistors are basically thimble-shaped, but
somewhat smaller than a thimble, and much smaller than conventional
tubes of the day. Triodes and a few tetrodes were made. The tube
is made entirely of metal and ceramic. Making
nuvistors requires special equipment, since there is no intubation
to pump gases out of the envelope. Instead, the entire structure
is assembled, inserted into its metal envelope, sealed and processed
in a large vacuum chamber with simple robotic devices."
Table of Contents
Wax nostalgic about and learn from the history of early electronics. See articles
published October 1954 - April 1985. All copyrights are hereby acknowledged.
Trivia for the day: A nuvistor was
used in the front end of the
HP3400A True RMS Voltmeter (thanks to Michael M. for that).
See all articles from
The NASA 136
By Tom Lamb, K8ERV
to listen in on the satellites? This little Nuvistor-powered converter
pulls then in on any 15-meter receiver
With the ever-increasing
interest in space science, more and more experimental satellites
are being blasted aloft each year. The general public can only marvel
at these accomplishments, but those who have appropriate receiving
equipment are more fortunate. They can actually listen in to the
radio "voices" of the satellites.
The little "NASA-136" converter
described on the following pages is designed to receive on the 136-137
mc. band now used by the National Aeronautics and Space Administration
for satellite telemetering. Employing a Nuvistor r.f. stage, the
unit has a sensitivity and signal-to-noise ratio more than adequate
to pull in signals from milliwatt powered transmitters orbiting
thousands of miles away. Use it with any communications receiver
tuning 15 meters, and you'll have many hours of fascinating listening.
But remember: this project is only for experienced builders.
About the Circuit.
The 136- to 137-mc.
signal from the satellite passes from the antenna to triode V1 (a
6CW4 Nuvistor), which is connected as a neutralized r.f. stage.
From V1, the amplified signal is coupled to the control grid of
V2, a triode-connected 6AK5 which serves as a mixer.
screen grid of crystal oscillator V3, another 6AK5, is tuned (by
coil L11 and capacitor C9) to the 38 2/3-mc. fundamental frequency
of crystal X1. Coil L10 and capacitor C8 tune the plate circuit
of the tube to 116 mc., the third harmonic of the crystal frequency.
This 116-mc. signal, like the 136- to 137 -mc. signal from
V1, is injected into the control grid of mixer V2. In V2, a third
signal is produced whose frequency is the difference between those
of the first two. The third signal, which ranges in frequency from
20 to 21 me. (depending on the frequency of the signal from V1),
appears across output jack J1.
Power for the converter is furnished by a separate supply, and an
octal output socket (SO1) on the supply chassis mates with a matching
input plug (P1) on the chassis of the converter. Transformer T1
provides heater power and a source of line-isolated plate voltage.
A single selenium diode (D1) is connected as a half-wave rectifier
and its output passes through a pi-network filter.
The completed converter and power supply chassis plug together
to make one efficient, integrated unit.
Photograph of power supply's exterior.
Pictorial diagram of power supply interior shows all essential
Drawings of the converter's major coils are shown here.
Coil L1/L2 is wound exactly as illustrated (L2 is solid
winding; L1 is dotted 2·piece winding). In coils L4/L5,
L6/L7, and L8/L9, make L5, L6, and L9 (the heavy windings)
as shown but refer to parts list for exact number of turns
on L4, L7, and L8 (the fine windings). Numbers on the coil
terminals are keyed to corresponding numbers on the schematic
Top view of converter. For exact parts placement, see Detail
Location of components on top of converter chassis is critical,
and the dimensions given in this illustration should be
Side and top views of shield and V1 socket show how shield
is bent, soldered to the socket. Components C11 and R1,
which have been moved aside for a better view, should actually
be positioned flat against shield.
Crossed leads running between L6/L7 and L4/L5 in converter
chassis (below) are shown separated for convenience; they
should be twisted together, routed in front of C3.
Parts placement shown in this photograph of the converter's
interior, and in the pictorial diagram on the previous page,
should be carefully studied and duplicated quite closely.
that, in case you want to use the power supply for other purposes,
its full high-voltage output is available at pin 3 of SO1. No connection
is made to the corresponding pin of converter power-input plug P1.
Building the NASA-136. Start construction by putting together
the power supply unit, which is housed in a 4" x 2 1/4" x 2 1/4"
aluminum utility box. Generally speaking, neither the parts placement
nor the wiring is at all critical Be sure, however, to mount output
socket SO1 in the exact center of one of the box ends and to position
the holes for the socket mounting screws so that the alignment key
faces the bottom of the box. It's necessary to take this care in
the positioning of SO1 because the socket must mate with P1, which
is similarly placed on the converter chassis.
When the power
supply is completely wired up, temporarily jumper the remote-power-switch
terminals (1 and 2) of SO1 and plug in the line cord. Use a multimeter
to check for filament voltage (about 6.3 volts a.c.) between terminals
4 and 5 of the socket, and for plate voltage between terminal 4
and terminals 3, 6, and 7, respectively. The latter three readings
should all be roughly the same (about 150 volts d.c.) since there
is almost no load on the supply and, consequently, no, appreciable
voltage drop across resistors R4 or R5. If the supply passes these
tests, disconnect it from the line, install the cover, and temporarily
set the unit aside.
With the power supply taken care of,
turn your attention to the construction of the converter itself.
A logical first step is to wind the coils (L1-L11), specifications
for which are given in the Parts List for the converter.
Three of the coils (L3, L10, and L11) are wound on resistors.
The leads of each of these coils are cut short and soldered across
the resistor leads at points not far from where the latter enter
the body of the resistor (be sure to carry out the soldering as
quickly as possible to avoid heat damage). The resistor leads will
then be used to wire the coils into the circuit.
L4/L5, L6/L7, and L8/L9 are wound on commercial slug-tuned forms.
Diagrams of these coils (Detail "A") are given to supplement the
information in the Parts List and should be followed as closely
as possible. The forms for L1/L2, L4/L5, and L6/L7 are all of the
same type. But the form used for L8/L9, though almost identical
in appearance with the first three, is different. Be careful not
to get them confused.
The converter is housed in a 5" x
2 1/4" x 2 1.4" aluminum utility box. All of the parts, except power
switch S1 and plug P1, are mounted on the top of the box. Parts
placement is critical, and the dimensions given in Detail "B" should
be closely adhered to.
Plug P1 and switch S1 are mounted
on the ends of the box. Center P1 and position its alignment key
to match that of SO1 on the power supply chassis. A retaining-ring-mounted
plug has been specified rather than a screw-mounted type so that
P1 can be twisted, if necessary, to make it line up exactly with
SO1. Switch S1 should be mounted slightly below center to insure
enough clearance between it and the antenna terminal lugs.
When all the mounting holes have been drilled, install the Nuvistor
(V1) socket. This socket has two slots (one wider than the other)
to accommodate the alignment keys on the base of the Nuvistor. Be
sure to place the socket so that the wider slot is positioned as
shown in Detail "B." The copper shield, which is formed and bent
as shown in Detail "C," is placed over the Nuvistor socket and soldered
to pins 8 and 10 (refer to Detail "C" and the pictorial diagram).
Before mounting any other parts, make all the necessary
connections to the Nuvistor socket. Then, as you proceed with the
parts installation and wiring, be sure not to block component terminals
before you have a chance to solder to them. Try to orient all components
exactly as shown in the pictorial diagram and photos, and note the
positioning of the V2 and V3 sockets as shown in Detail "B."
A few points in the construction need special comment. First,
capacitor C7 is nothing but a turn of insulated hookup wire wrapped
around the body of capacitor C8. Next, resistors R3 and R5 have
nothing to do with the actual functioning of the circuit. They serve
only to isolate the grids of V2 and V3, respectively, for test purposes.
One end of each of these resistors is soldered to the appropriate
grid; the other end, cut very short, is left free. Finally, do not
connect the lead from pin 7 of P1 to the junction of capacitor C3
and coil L4 (point "X" in the schematic diagram). It must be left
off temporarily in order to disable the r.f. stage during the initial
steps in the adjustment procedure.
Besides your receiver (which should be equipped with an
S-meter), you'll need two test instruments to carry out the adjustment:
a d.c. meter with a range of approximately 0-100 microamperes (or
a VTVM with a range of about 0-3 volts), and a signal generator
which can be set at 136.5 mc. If the latter is unobtainable, a 2-meter
ham transmitter tuned to 136.5 mc. will probably do the job.
If you must use the transmitter, be sure that it radiates only
the minimal signal required for adjustment purposes. Connect a dummy
load across the antenna terminals and, if possible, leave the final
off. If the final must be on, be sure that it draws minimum power.
A signal radiated into space on this frequency is not only illegal,
but it could easily interfere with vital government satellite telemetering.
A word to the wise is sufficient.
All adjustments may be
made with the chassis covers removed. Begin by plugging in the power
supply to the converter and to the line, inserting the crystal and
tubes, and checking to see that the lead to L4 and C3 is disconnected
at point "X." The negative lead of your test meter should be connected
to test point 1 and the positive lead to ground.
power switch S1 and, after the tubes have warmed up, adjust capacitor
C9 for a maximum meter reading (this will probably occur somewhere
near the minimum-capacity setting of C9) . Now change the negative
meter lead to test point 2 and adjust capacitor C8 for maximum reading
(once again, this will probably occur near the minimum-capacity
setting). At this point, without changing the meter connection,
capacitor C9 should be repeaked.
Typical final readings
at test point 1 are -2 volts (read on a VTVM) or 32 µ.a. (read
on a microammeter). At test point 2, the readings should be about
-1 volt or 12 µ.a. If the reading at the latter test point
is a little low, try making the loop of wire (C7) around capacitor
C8 a bit tighter.
This done, remove the meter and connect
the antenna input of a receiver set at 20.5 mc. to the converter's
output jack (J1); use a length of RG-58A/U coaxial cable. The receiver
r.f. gain should be full on and the S-meter operating. Now adjust
the slug of coil L8/L9 for maximum receiver noise.
the output of a signal generator (or transmitter) tuned to 136.5
mc. to the converter's antenna input. If you're using a coaxial
output cable, connect it between one of the antenna terminals and
ground. You should now hear the generator's signal at the 20.5-mc.
receiver setting (or near it, if the receiver's calibration is slightly
off). A fairly strong signal will be needed from the generator,
since the converter's r.f. stage is disabled.
With the signal
tuned in, slowly adjust the spacing between the turns on neutralizing
coil L3 for minimum S-meter reading. Use a plastic tool for this
adjustment and be sure the receiver always stays tuned to the signal.
Next, temporarily solder to place the lead left off of point "X,"
reduce the generator's output, and tune the slugs of coils L1/L2,
L4/L5, L6/L7, and L8/L9 for maximum S-meter reading.
the lead from point "X" and readjust L3 for minimum reading. Then
connect the lead again and readjust L1/L2, L4/L5, L6/L7, and L8/L9
for maximum readings. Repeat the procedure until there's no further
change in the maximum and minimum readings. Finally, secure the
turns of L3 with coil dope, permanently wire in the lead to point
"X," and install the box covers. The converter is ready to go.
It may be, however, that your receiver is a "ham bands only"
model, and you would prefer to set the converter's output in the
21-22 mc. band. In this case, just substitute a 38 1/3-mc. crystal
for the 38 2/3-mc. unit specified for X1 and follow the identical
procedure outlined above. The only difference is that the receiver
should be set at 21.5 mc., rather than at 20.5 mc., during the adjustments.
Wire the converter's
output to your receiver's antenna input as described in the "Adjustment"
section. Then connect a TV antenna, a 2-meter beam, or a 41"-long
folded dipole to the converter's antenna input. If the lead-in is
300-ohm line, connect it across the two antenna terminals; if it's
a coaxial cable, connect it between one antenna terminal and the
Assuming that the receiver and crystal calibrations
are accurate, the frequency of the received signal will be the receiver
dial reading plus 116 mc. (115 mc. if you're using a 38 1/3-mc.
crystal at X1). In other words, the 136-137 mc. satellite band will
be tunable between either 20 and 21 mc., or 21 and 22 mc., depending
on which crystal you use.
While the three tuned circuits
between the converter's antenna terminals and mixer grid tend to
eliminate image responses, you may still pick up an image from a
local FM station. Should this be the case, try installing a stub-type
wave trap. If you have a coaxial lead-in, use an 18 1/2"-length
of coax cable with one end open and the other connected, at the
converter, in parallel with the lead-in. If you're using a 300-ohm
lead-in, the stub is a 24"-length of 300-ohm line connected in the
TRACK THESE SATELLITES WITH THE NASA-136
As the payload weight of American satellites increases,
so does the power output of the satellite transponders and beacon
transmitters. More power and better antennas on the satellites enable
SWL's to pick up these signals with greater ease.
beacon transmitters were originally scheduled to operate around
108.0 mc., but because of the number of satellites the United States
has launched, the band between 136.0 and 137.0 mc. has been set
aside to give each satellite frequency room. All future satellites
will carry a beacon transmitter or transponder operating in this
frequency band, which is now used in countries outside the Iron
Curtain for satellite tracking.
Four satellites are transmitting
as this is being written, and two more are likely to be launched
and transmitting before this article is in print.
Schematic diagram of converter. Lead from
pin 7 of P1 to point "X" is not connected
at "X" until initial
adjustments are finished (see text).
PARTS LIST FOR CONVERTER
600-volt ceramic capacitor
C3, C 10-0.001-µƒ., 500-volt,
silver-mica button capacitor (Erie 370-FA-102K or equivalent)
C4-100 µµƒ. 600-volt ceramic disc capacitors
C5-10 µµƒ. 600-volt ceramic disc capacitors
C6, C11-0.001 µƒ. 600-volt ceramic disc capacitors
C7-1 turn of insulated wire around C8-see text
C8, C9-0.5-5 µµƒ.
tubular trimmer capacitor (Erie 532-A or equivalent)
phono jack (Smitchcrajt 3501FP or equivalent)
L1-5 turns of
#24 enameled wire, center-tapped; bifilar wound with L2
turns of #24 enameled wire, wound in center of a Cambridge Thermionic
L3-25 turns of #30 enameled wire, close-wound
on a 1-megohm, 1-watt resistor
L4-4 1/4 turns of #24 enameled
wire, close-wound near top of a Cambridge Thermionic PLS6/2C4L/D
L5-1 turn of insulated hookup wire wound on L4
turn of insulated hookup wire wound on L7
L7-3 3/4, turns of
#24 enameled wire, close-wound near top of a Cambridge Thermionic
L8-23 turns of #32 enameled wire, close-wound
in center of a CTC PLS6/2C4L/O coil form
L9-2 turns of insulated
hookup wire wound on L8
L10-11 turns of #24 enameled wire, close-wound
on a 1-megohm, 1/2-watt resistor
L11-37 turns of #32 enameled
wire, close-wound on a 1-megohm, 1/2-watt resistor
octal plug (Amlphenol 86-CP8 or equivalent)
R1-47,000 ohms 1/2-watt
R2-220,000 ohms 1/2-watt resistors
R3, R4, R5-100,000
ohms 1/2-watt resistors
S1-S.p.s.t. toggle switch
tube (RCA Nuvistor)
V2, V3-6AK5 tube
crystal (International Crystal Type FA-5)
1-5" x 2 1/4" x 2
1/4" aluminum utility box (Bud CU-3004-A or equivalent)
socket (Cinch-Jones Type 5NS or equivalent)
tube sockets, wafer-type
1-Socket for X1 (International Crystal
150-109 or equivalent)
Misc.-Scrap copper for tube shield, wire,
2-1ug terminal strip (screw type), length of coax cable, hardware,
|PARTS LIST FOR POWER SUPPLY
C1a/C1b-20/20 µƒ., 150-volt electrolytic
capacitor (Sprague TVL-2415 or equivalent)
130-volt (r.m.s.) selenium rectifier (Sarkes Tarzian Type
50 or equivalent)
R1-470-ohm, 1-watt resistor
R3-1000-ohm, 1/2-watt resistor
R5-3300-ohm, 1-watt resistor
socket (Amphenol 77MIP8 or equivalent)
primary, 117 volts; secondaries, 125 volts at 15 ma., 6.3
volts at 0.6 amp. (Stancor PS-8415 or equivalent)
x 2 1/4" x 2 1/4" aluminum utility box (Bud CU-3003-A or
Misc.-Terminal strip, grommets, line cord
and plug, hardware, solder, etc.
Schematic diagram of the power supply, which
utilizes a simple half-wave diode rectifier.