1j37b

Hello fellow Radiophiles,

Following the RadioMuseum tradition of home-made AM transmiters listed at the end, I have come up with this simple single tube version.

I based the design of this simple transmitter on Gammatron/Gridless-Audion operation of the 1Zh37B Russian Subminiature Tube.

The most useful characteristic of Gammatron/Gridless-Audion operation is the ability of the plate to draw current at low voltages.

Keep in mind that the "plate" is the second control grid rod G1B. The remaining electrods, G2,G3 and A, are also connected to G1B because they contribute an additional 10% to total "plate" current

This transmitter requires an AA filament battery, preferably of the NiCd or NiMH kind to better meet the rated 1.2V filament operation, and an optional high voltage Bplus source for the plate.

The Bplus source is optional because the oscillator operates with Bplus between 0V and 18V, if the filament is  operated with a 1.5V cell and the plate supply is referred to the positive cathode terminal.

The transconductance with 0V at the plate is still a useable 40uS with a 500uA plate current.

When the plate voltage is zero volts with respect to the positive terminal of the cathode, it is still up to 1.5V above the negative terminal of the filament supply.

The commercially made Ferrite coil was construted with two windings: One with many turns and 740uH for tuning, and a small winding with a few turns to couple to a low impedance transistor input in a solid state radio. The turns ratio is about 7:1 and serves to step up the plate voltage from plate to grid. This high step-up ratio is necessary to overcome the intrinsic voltage gain mu=0.25. The net loop gain during the short bursts of plate conduction is 7*0.25=1.75.

The modulation scheme is a very conventional plate modulation that is injected with a step-up 10x 8R-1K audio transformer at the plate. This transformer is only needed to achieve enough modulation at higher supply voltages. The transformer takes signal from an audio earphone output, not from a high impedance line level output.

The Audio transformer bandwidth exceeds 15kHz with a 1kOhm secondary load, but the transmitted bandwidth is limited by the undriven Q of the resonant tank circuit.

The Ferrite antenna is ideal to transmitt to radios with a ferrite stick or a loop antenna that is sensitive to the magnetic field. A short wire tied to the plate output can be used to transmitt to radios with a wire antenna that are more sensitive to the electric field.

As will be shown below, this transmitter works best with modulation up to 50% for the higher frequencies, and up to 90% for the bass frequencies.

Over-modulation, results in the shutdown of the oscillator with each negative trough of the audio input, and a delayed startup when the audio signal goes positive again. The high Q and narrow bandwidth of the tuned circuit is one of the characteristics that keeps the RF envelope from following the audio signal at high audio frequencies and high modulation.

Prof. Dr.-Ing. Dietmar Rudolph has posted a good article on the problems with startup of triode oscillators.

Construction

The picture that shows the batteries includes a 0V battery for Bplus in the form of a shorted battery clip that was recycled from a dead 9V battery. The 1.5V cell you see with a 9V clip on it is used to run Bplus at 1.5V.

The lid is kept closed with an embedded pair of neodymium magnets on one of the front corners of the box and lid.

The box is made of soft Bass wood and was bought ready-made at a local crafts store. I added the two wooden posts to support the ferrite stick, and removed the original brass clasp to fit the variable capacitor, after chiseling out the wood.

A matched capacitor and ferrite stick could have been recovered from a discarded AM transistor radio. This would give a fairly accurate frequency indication. In my case, the ferrite and capacitor came from different sources.

Little knurled brass nuts always look very cute, so I used for the audio input.

I have found it a good idea to add a small schematic to the finished project. Amazing how much you forget after a few months or years.

Measurements and operation

The operating range is about 2 Feet (65cm) with Bplus set to zero volts. This a good range to demo an am radio. Rasing Bplus to 1.5V gives clean coverage for a table full of radios. If you want to cover a room or small apartment, you will need the 9V battery at Bplus.

The first scope photo shows the RF envelope at the plate and after it is broadcast and picked up with a single loop. The 400Hz audio waveform is essentially the power supply to the oscillator. The second photo shows the classic XY display with audio fed to the horizontal axis and the modulated output is at the vertical axis. This display is shows non-linearities, percent modulation and other characteristics more clearly. 400Hz at 50% modulation shows no distortions. Note that the 20Vp-p RF envelope present at the plate on the left photo corresponds to a 140Vp-p RF envelope at the grid because of the 7x feedback stepup ratio of the ferrite coil.

Audio modulation with 400Hz

Increasing modulation to 95% at 400Hz starts to show some distortion on the negative troughs that is only apparent if looking closely at the left end of the second picture

Increasing modulation to 110% shows that it takes some time for the oscillator to restart after it was extinguished with the negative audio peak.

This is usually not a problem if the oscillator runs at constant amplitude and drives a separate modulator.

Audio Modulation with 4kHz

Increasing the modulation frequency to 4kHz produces a lot of distortion that is visible in the following two photos. The most distorted part of the RF envelope is in the regrowth portion of the cycle.

Reducing modulation at 4kHz to 25% makes the distortion disappear, but some linear modulation delay is still apparent in the XY photo on the right.

Going Further

I have found that moderating the modulation level produces produces a very good sounding result. This is in part caused by most of the audio content being concentrated at the lower frequencies.

Perhaps an RM member would like to try adding a modulation stage to this oscillator, while keeping the low voltage operation. If so, please share your results. If a modulation stage is added, the oscillating coil should be shielded to avoid radiating unmodulated energy.

In keeping with an elegant design orientation, my favorite external modulator approach would be that posted by RM officer Konrad Birkner at Simplest diode modulator.

Another alternative is the choice of tube. Other tubes with a low mu may be good candidates for this circuit, when used in reverse triode operation, with the input at the plate and the output at the grid. The rule of thumb in selecting an alternative tube is to look for a mu less than 10. Remember that the step-up ratio for the oscillating coils must overcome the inverse of the mu, or up to 1/10. Candidates for reverse triode operation could be the 6AS7, 01A, 201A, 45, 2A3, 1G4, DM70/1M3. The DM70 even glows green. Perhaps European members could contribute European part numbers.

If your goal is single supply operation, then the tubes with the higher filament voltage at 3V, 5V or 6.3V have an advantage over those that run at 1.5V.

One thing to consider in the selection of a common triode/tetrode/pentode for reverse triode operation is the magnitude of the grid current in forward bias. A simple sweep with a power supply will show at what point the grid current becomes excessive. O good upper limit is the rated max cathode-plate current in the data sheet.

Solid state builders may like the low voltage operation, while still using a real, if not very special, 1Zh37B tube.

Regards,

-Joe

Other AM transmitters

Examples of transmitters posted by RM members can be found at:

Dynatron tetrode transmitter

1 Watt AM Transmitter (solid state)

Home made AM transmitter with dual gate mosfet

A survey of AM modulator/transmitters

Another survey of AM modulators/transmitters

Am tube modulator

Simplest diode modulator

Granco-ARC60 AM Modulator/transmitter runs off 12.6VDC.

 High Performance modulator with Resistance stabilized oscillator

Hello radiophiles,

This post was stimulated by Christian Bruckner's inquiry about Magnetic transmission.

More related content was also posted by Konrad Birkner about the use of ferrite antennas.

I used Google-Translate to read these German language posts

Reception of Magnetic fields

Many AM tube radios have inputs for an external wire antenna. Many have a loop antenna as an alternative, or in addition to the external antenna input.

In later years, the loop antenna came to dominate AM tube radio design. Nearly all solid state AM radios used loop antennas.

The loop antenna has two important advantages over wire antennas:

1-Loop antennas are usually smaller than the equivalent wire antenna. This is true for air loops, and even more so for ferrite loop antennas.

2-Most man-made noise in the AM band tends to be concentrated in the Electric field, so an antenna that derives it's signal from the Magnetic field is less likely to pick up noise. Some of the better AM radios even employ a grounded sheild over the ferrite loop to prevent pickup of the electric field by the ungrounded high impedance end of the loop. When you bring your hand near a loop antenna and  you notice an increase in noise, you are probably injecting Electric field noise.

Natural random noise power from atmospheric discharges is cut in half by the reception pattern of a loop, as it would be by a dipole, because a reduced amount of random noise power is received near or at the notches in the reception pattern. This -3dB improvement in SNR (Signal-to-Noise-Ration) is not dramatic, but worth having.

The two reception notches that occur perpendicular to the axis of the loop virtual cylinder can also be used to eliminate reception from a distance station on the same frequency, or from a nearby point source of man-made noise.

Transmission of Magnetic fields

The polarization (orientation) of the magnetic field of modern AM transmitters creates magnetic lines of force that are parallel to the earth. This orientation usually changes after reflections by the ionosphere. The typical AM transmitter antenna consists of one or more vertical radiating masts with a vertical reciprocating current flow at the transmission frequency. This current flow then creates concentric magnetic lines of force that are parallel to the earth.

The horizontal magnetic field lines of this polarization, are the perfect match for receiving loops with a horizontal axis that are universal in AM radios.

Given that most AM radios only need the magnetic field to work, why do AM transmitters radiate Electric field at a ratio to the Magnetic field that approaches the ideal free space impedance of 377 OHms?

The answer is that you can't cover more than a few wavelenghts of distance with just Magnetic field or Electric field. It is necessary to radiate Electric and Magnetic fields at the antenna in the ratio of 377 Ohms, so more energy can be radiated at reasonable drive currents and votlages.

The following graph shows how an Electric field or a Magnetic field emission becomes an emission with both fields at a ratio of 377 Ohms after a few wavelengths. You can see the full page scan with text, by clicking on the graph, but I no longer know the source of the scan. I think I got this scan from my colleague, and former Delco AM radio designer, Tom Hack.

Note that the wavelenght at 1MHz is 300m (980ft). This very long distance makes it very reasonable to assume that all distances inside a home have negligible wave effects, and the fields can be understood as if they were static DC fields. Keep in mind that this is no longer true if you live in a 100m long castle!

Personal Magnetic transmission

The arguments just outlined can be reversed for the personal use of small AM transmitters:

The small AM transmiter should be purely an Electric field transmitter, or a Magnetic field transmitter. I chose the Ferrite antenna for the 1.5V AM tube transmitter to generate the magnetic field that nearly all my AM radios were designed to receive.

The magnetic field transmission will also decay much more rapidly, with 1/distance^3, than a transmission with a 377 Ohm ratio of Electric to Magnetic fields with a decay of 1/distance . This is beneficial to reduce the reach of your transmission outside the  region of interest, while maintaining a comparatively strong magnetic field locally.

The little 1.5V AM tube transmitter includes a 10pF wire antenna hookup for radios that only use a wire for reception, such as this little 4 tube Kent set from 1940. This wire came with a soldered 8 foot wire antenna that is surprisingly sensitive, once it is streched out.

The grid input end of the the tuned coil in the 1.5V AM tube transmitter swings on the order of 140Vp-p with a 9V battery at Bplus and a few volts with zero Bplus. The electric field in the vicinity of this end of the coil is enough for the little Kent radio to pick up if it is sitting nearby. The plate of the 1.5V AM tube transmitter swings 20V p-p with a 9V Bplus and about 0.5V p-p with zero volts Bplus. This is sufficent to drive several feet of wire through 10pF without significant detuning to the low impedance plate circuit.

E-field transmission over short distances can be understood as the coupling of a weak capacitor, while M-field transmission over short distances can be understood as loosely coupled inductors.

One extreme case of attempting to mismatch transmission impedance away from the 377 Ohm E to M ratio of free space is being exploited to transmit power over short distances, while avoiding radiating it to great distances. It is fascinating that impedance matching and impedance mismatching can be used to great advantage, depending on application: mismatching is ideal for local power transmission, while matching is ideal to maximise transmission over long distances.

In conclusion: Pure E-field or pure M-field radiation is best for personal AM transmitters to strengthen the local "near-field" reception, while weakening distant "far-field" reception.

Regards,

-Joe

p.s.: The formal letter to represent the magnetic field is H. I used M to avoid confusion for readers that are not familiar with electro-magnetic theory. Konrad Birkner uses the formal H in his post listed at the top.

Small radios often use a loop antenna or a ferrite coil antenna. In the case of realigning such a receiver, the signal has to be coupled into its loop or ferrite coil antenna.

This clearly is a near field radiation or even nearly a magnetic coupling. Realigning this kind of receivers may be faciliated by placing the test loop on the test bench as suggested by the next figure.

Lit.: Ghirardi, A.A.: Receiver Troubleshooting and Repair, Rinehart, 1955

Regards,

Dietmar

Also known as 1ZH37b-r, 1sh37b-r and properly 1Ж37Б

Useful information on these kinds of sub-miniature Rod Pentode tubes (valves):

Russian Subminiature Tubes

The 1j37b is  still (2011) readily available to purchase as New Old Stock (NOS).

Nützliche Informationen über diese Art von Sub-Miniatur-Rod Pentode Röhren:

Russian Subminiature Tubes

Die 1j37b ist immer noch (2011) leicht zugänglich zu kaufen (2011) leicht zugänglich wie Neu Alte Aktien Kauf (NOS)

1j37b  1Ж37Б

Cross-section showing single filament feed on top
Glass (in grey) about 8mm diamenter

Fellow radiophiles:

Measurements comparing 1j24b vs 1j37b filament efficiency have been added to the series on Russian Subminiature Tubes.

Regards,

-Joe

Fellow Radiophiles,

Sometimes different versions of a tube data sheet reveal different details. I have two loose data sheets that came with a ebay purchase of the 1ZH37B.

I scanned these sheets and performed OCR of the Cyrilic characters with ABBYY express software. The original print quality was poor, however enough words were recognized to produce a useable translation by uploading the OCR result to Google-Translate.

1zh37b_sheet.pdf two page scan with OCR Russian text

1zh37b_sheet_EN.pdf Google translate output with English text, but no pictures diminished formatting.

1zh37b_3pages.pdf three page scan with OCR Russian text

1zh37b_3pages_EN.pdf Google translate output with English text, but no pictures diminished formatting

The Google-Translate output is very difficult to save. So far, the only sucessful method I found is to print the output frame to a single very tall pdf page. The text can then be copied and pasted readily from this output. The two files above with EN in the file name have just the Google-translated English text.

Regards,

-Joe

MIT Electrical Engineering Student Dimitri Turbiner has kindly translated the 1Zh37B data sheet.

The translation quality is excellent and we are all very grateful for Dimitri's contribution to the understanding of the Russian Subminiature Tubes with rod construction.

Thank you Dimitri.

-Joe