Simple analogue pantry transmitter ?

[Goldieoldie]

Hi,
Quite a few posts with pantry tx using chips etc
How about since our hobby is mainly analogue radios for an analogue simple tx
Using bits from the junk box
I'm thinking perhaps using the osc from an old transistor radio.The varible cap ( or use a trimmer ) and coil is already there !
Just need an output transistor and a means of modulation
I know crystal lock is better but don't really think drifting on MW is really an issue
Come on guys thinking caps on !
Cheers Pete

[Al (astral highway)]

Originally Posted by Goldieoldie

I'm thinking perhaps using the osc from an old transistor radio.The varible cap ( or use a trimmer ) and coil is already there !
Just need an output transistor and a means of modulation

Hey Pete,

The local oscillator in an AM radio is tuned to the IF (455KHz) plus the frequency of the station you're listening to...

So are you thinking of retuning the IF strip from an old radio, or what exactly, more explicitly?

Generating a carrier wave isn't at all difficult. Generating a stable carrier wave and modulating it the right amount, and then radiating it just the right amount, into a matched antenna, while dealing with harmonics, is much more of a challenge.

I don't think there is anything radical about using any kind of oscillator under the sun to make a carrier wave - it's been done for a century now. Quite a few folk simply use an old signal generator as a modulator.
But there's nothing to stop you having fun and using your imagination, and your kind of creativity is very welcome on this forum, so why not go for it and post results?

The designs that have been posted and discussed most recently are characteristically rather noteable for their stability and for the attention that has been paid to the quality of modulation, as well as to getting rid of unwanted harmonics.

Worth looking up current thread on Pantry transmitter, as well as some of its predecessors!

[G6Tanuki]

The simplest I ever made was indeed an old transistor radio.

Ground the base of the frequency-changer [where the aerial coil of the ferrite rod went] through a 0.1uF capacitor.

Remove the capacitor across the primary of the first IFT so it becomes essentially just a RF choke.

Attach a short wire to the 'hot' side of said IFT to act as an aerial.

Disconnect the loudspeaker.

Rewire the 'cold' side of the first IFT so it gets its DC supply from the collector of one of the two push-pull output transistors.

Inject your program-material to the wiper of the volume-control.

Tune the thing until you pick up the repurposed-LO output on your target receiver (which should be tuned somewhere above 1MHz/300M).

It won't give brilliant audio quality - directly modulated oscillators are full of nonlinearities - but it's only medium-wave so you probably won't notice 10% distortion...

[Goldieoldie]

G6
Best so far !

[TonyDuell]

Probably no help, but some of the educational kits from the 1970s/1980s (the 'Denshi Block' EX series, those <n>-in-1 kits with pre-mounted components and spring terminals etc) had 'wireless microphone' projects that were very low power AM transmitters. They used the normal ferrite rod aerial and tuning capacitor that were used for the radio projects. I think there were even designs using just one transistor.

[G6Tanuki]

This has sort-of inspired me to start experimenting again with "the cheapest possible top-band transmitter" circuits.

July 1961 Practical Wireless has a circuit - I've built these in the past. Adding some bias to the base of the audio-amp transistor helps a bit.

[and yes I'm licensed to operate on top-band].

[Fishall]

That's interesting!

I wonder do you have a better scan of the page? Bit difficult to make out the text.

[paulsherwin]

If you zoom into the image it is quite legible.

A minimal pantry transmitter is just a single transistor oscillator with a coil to determine the frequency. You modulate this with a reverse connected output transformer, using a headphone output as a source. The result isn't very clean or stable, but it does work, and the simplicity makes the circuit operation obvious to a beginner.

There are much better designs with only a few more components though.

[dragonser]

Hi,
with regard to the July 1961 circuit, I am assuming that originally tr1 would have been germanium and would have some leakage to bias the base connection. Adding input bias resistors should give more stable operation.
I think I read somewhere that it is better to have a two stage rf circuit, with the second amplifier stage modulated rather than the Oscillator stage.

[Andrew2]

The main problem with modulated VFO's is that you get quite a lot of FM. Tuning in on a nearby radio you will find a 'hole' in the middle of the signal (where you would normally tune) and for best fidelity you need to off-tune a few KHz.

[Argus25]

I think there are a number of folks wanting a simple solution yet still wanting to have a hand in building the transmitter, more than designing one. AES in the USA sells valve transmitter kits on good old fashioned wooden breadboards ! One of them was copied off a 1939 Zenith circuit:

https://www.tubesandmore.com/product...ss-transmitter

Probably they have 110V primary mains transformers, but AES can more than likely supply them for 230V. They look like reasonable value for all the parts you get.

[G6Tanuki]

Originally Posted by dragonser

with regard to the July 1961 circuit, I am assuming that originally tr1 would have been germanium and would have some leakage to bias the base connection. Adding input bias resistors should give more stable operation.

Yes the original transistors as specified were Ediswan/AEI Germanium types and this circuit clearly depends on leakage for the first transistor to work. It certainly would have 'issues' if built with modern Silicon transistors, unless a squiff of forward-bias was applied to the modulator.

[Plenty of early transistor circuits only 'worked' by virtue of the leakage intrinsic to first-generation semiconductors: I suspect that as technology and production methods improved and the leakage reduced - even with devices of supposedly the same type-number - circuits which worked with the first-production transistors may not have been so successful if used with later-production ones of the same type!]

The full circuit detais are available in the appropriate Practical Wireless which can be downloaded from the americanradiohistory website which is often mentioned here.

[kalee20]

Originally Posted by Goldieoldie

I know crystal lock is better but don't really think drifting on MW is really an issue.

Slow Frequency drift isn't an issue - but if the frequency shifts as you go up and down in amplitude, it is! Amplitude modulating the oscillator is just rapidly altering the supply rail (or whatever). Due to device non-linearities, this is invariably accompanied by a small frequency shift. And so you get...

Originally Posted by Andrew2

The main problem with modulated VFO's is that you get quite a lot of FM. Tuning in on a nearby radio you will find a 'hole' in the middle of the signal (where you would normally tune) and for best fidelity you need to off-tune a few KHz.

That's because an FM deviation of, say, 1kHz is enough to get slope-detected in the receiver. A crystal set, or even a simple TRF with 2 tuned circuits may not be sufficiently selective to slope-detect, but most superhets will. And the slope-detected output will interfere with the intentional AM output, so you get funny-sounding audio.

[kalee20]

Originally Posted by Goldieoldie

G6
Best so far !

Ah, but he had to add a capacitor!

Why not a conventional 5-valve superhet? Connect AF to the volume control, after disconnecting the detector.

Disconnect the oscillator HT feed from the HT rail and connect to the output valve anode.

Connect short aerial to the oscillator section of the tuning gang.

Switch on, tune up, and enjoy! No extra components!

[Ian - G4JQT]

You could probably modify the Minimod pantry transmitter to use the LO signal from a transistor radio. The circuit is here: https://www.vintage-radio.net/forum/...5&postcount=85

Replace the VFO fet and associated components with your radio's LO injected at C3.

You might have to reduce C3 and/or add a series resistor if the LO signal is too high.

[m0cemdave]

I have always found that an AM transmitter with incidentatal FM gives an ummistakeable characteristic effect when tuning through the signal.
Rather than a hole in the middle (characteristic of a slope-detected FM signal), there is a lot more modulaton on one side of the carrier than the other.
I assume this is due to partial phase cancellation of one sideband due to the mixture of AM and FM.

[Terry_VK5TM]

Here's a pdf of the PW Tx from post#6 for those that wanted a readable version.

[kalee20]

Originally Posted by m0cemdave

I have always found that an AM transmitter with incidentatal FM gives an ummistakeable characteristic effect when tuning through the signal.
Rather than a hole in the middle (characteristic of a slope-detected FM signal), there is a lot more modulaton on one side of the carrier than the other.

I have found this, too.

Tuned properly, and with only a small degree of modulation, there will be only a bit of FM, so hopefully no slope detection if it stays within the passband. But the AM will of course be present, and that'll be detected and outputted. So, unlike a true FM receiver which includes a limiter to strip off the AM, there's no hole-in-the-middle.

Tuned off to one side, on the passband slope, any FM at all will be slope detected. FM output will be of one phase. Tuned off to the other side, the output-vs-frequency slope will be the opposite direction, so slope-detected output will be the other phase.

The intentional AM will always be the same phase, so one direction of mistuning will interfere constructively, the other destructively. That's why it will seem asymmetric!

With a lot of modulation, it's quite possible that deviation may start hitting the passband slope on both sides. Now that REALLY sounds horrid!

[ThePillenwerfer]

This one works with a range of a couple of feet — probably as well as it produces harmonics. It's output can also be connected directly to a radio's aerial socket.

Please don't ask me to justify the design as the theory is way beyond me.

[Ian - G4JQT]

Originally Posted by Terry_VK5TM

Here's a pdf of the PW Tx from post#6 for those that wanted a readable version.

There's no bias on TR1. Will it even work?