Convert Odd Tube Circuit to Transistor

I would like to convert this tube amplifier to solid state. IOW a transistor circuit that uses the same three stage format ... a "faithful reproduction", if you will. Not something "better".

Can anyone please recommend suitabletype of circuit and parts?

formatting link

Of course, it will be powered from 12VDC, not the HV of the original tubes.

Ben Hartley

I posted this earlier, but my ISP is having usenet issues so i don't think anyone saw it. Switched to APN now.

Reply to
Ben Hartley
Loading thread data ...

Nothing remarkable about it. Fairly high pass, depending on what the coils were wound on. Self capacitance will come into play; a transistor equivalent won't need as many turns on the coils, but will require capacitance in parallel to tune the coils to the same frequency (whereas the original coils may've been wound with this in mind). Strange that the plate coils have large series resistors and there is no grid bias. Also strange that there's no rectifier; whatever it does, it does it only sixty times per second.

Do you happen to know how much gain, frequency response, input/output impedance and voltage are actually required of the application?

Note that literally recreating it would require accommodating the same input and output voltage and current characteristics; at clipping, output voltage may easily peak over 100V, which will be more challenging to achieve from a

12V supply.

If it's constructed without shielding, there's enough gain here to self oscillate. Modulated at 60Hz, it would make one hell of an AM radio buzzer. Hmm.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
 Click to see the full signature
Reply to
Tim Williams

Thank you for your expert analysis. Yes, I thought it was a strange one, and wanted to replicate out of interest.

The circuit is part of an old transmitter, but I am uncertain of the original application or specs. It does not appear to be for communication.

The coils formers are 1/4" old bakelite 1" long, air core. What would this make the bandpass?

With regard to the grid bias, here is a previous version in which they have been deleted for some reason. Presumeably not required.

formatting link

I also found this transistor version, but it uses obscure parts. Does it help explain in any way what was supposed to happen?

formatting link

Any further ideas would be greatly appreciated.

Ben Hartley

Reply to
Ben Hartley

coils

the

plate

ange

per

nput

age

m a

uzzer.

Thanks to the newfangled "yahoo search", I now know you're talking about building a Hieronymus machine for the detection of Eloptic Energy:

formatting link

I don't think denizens of sci.electronics.design can help you duplicate a Hieronymus machine. There's a feller out on the intertubes selling instructions for making your own. I'd try alt.electronics.design, if one exists.

Reply to
spamtrap1888

I am actually just interested in the circuit dynamics, not any supposed fringe projects.

Just couldn't figure out what it would do, and that it might be fun to build one solid state and see.

Ben Hartley

Reply to
Ben Hartley

So you mean the tube circuit can be replaced by three components??

What would be the expected characteristics of the FET circuit?

Ben Hartley

Reply to
Ben Hartley

I don't think I saw any oscillator one the posted solid state schematics. Just terminals. I suppose AC could be connected there. boB

Reply to
boB

Well, it was supposedly designed by an engineer, so I would vote for number three. After all, 60Hz carriers are everywhere aren't they, at least in the US? The output coil also suggests this.

Which raises the question of what might be the intended input signal frequency. From the limited info I have, the input antenna was seven loose turns of wire on a 2" air core former.

From that I get 268MHz. The only reference I could find was that it is the NMR of Sodium salt.

However, I don't understand which part of the circuit does the mixing.

The output is presumeably inverted, but how much gain would be provided?

Ben Hartley

Reply to
Ben Hartley

People looking at a schematic try to make sense our of what they're seeing. So if you present them with a schematic that doesn't do anything sensible, the viewer will be hard pressed to explain the circuit dynamics. The simplest thing, because you have the circuit and the desire to get to the bottom of it, might be to acquire the tubes and a transformer, and build it. Trace the signal path with an oscilloscope. Your describing it as part of a transmitter didn't help, especially because it was built with receiving tubes.

Reply to
spamtrap1888

Stick with the transistor version, then. One of the buffs who commented online found germanium pnps at (presumably) Al Lasher's Electronics in Berkeley, CA.

Reply to
spamtrap1888

I am curious Mike. Briefly, what is your reasoning that it has no purpose?

From the responses so far, I gather the circuit is prone to oscillation, the output is 60Hz pulsed, and there there are design "irregularities".

Assuming a low level input, what is your best guess what the designer had in mind?

Ben Hartley

Reply to
Ben Hartley

The fact that their own later CK722 amplifier provided a totally different transfer function verifies that the whole project was a worthless scam.

The typical output of the tube circuit would be a half wave 60 Hertz modulated waveform of a noisy input signal likely in the low megahertz range.

Which might under some circumstances feel "warm and fuzzy".

The engineering, of course, was not even wrong.

<
formatting link
>
--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
 Click to see the full signature
Reply to
Don Lancaster

The mixing is done by the fact that the tubes turn on and off at a 60 Hertz rate. When the plate is negative, the gain is zero. When the plate is positive, the gain is some ( really stupid ) function of the half sinusoid line voltage.

The output ( if this was a "real" circuit ) would be the input synchronously demodulated against the power line. With bunches of nonlinearity thrown in.

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
 Click to see the full signature
Reply to
Don Lancaster

The designer was obviously either a blatant scamster or else a few chips shy of a full board.

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
 Click to see the full signature
Reply to
Don Lancaster

The output will be mainly composed of rf modulated at 50/60 Hz with significant harmonics due to the non-linear and discontinuous current flow. The control grid is at cathode potential such that a modest anode current will be flowing, also grid current will flow on +ve input excursions. The frequency range of interest must be in the high MHz range given the air core transformers with just a few turns.

I have no idea what the designer had in mind apart from possibly disrupting/jamming radio reception.

--
Mike Perkins
Video Solutions Ltd
 Click to see the full signature
Reply to
Mike Perkins

Hmmm, wouldn't it be better to see that actual circuit work with the tubes ??

What would you have to compare against ?

Ebay has these tubes available at about $20 each.

If your going to spend the money, Build the original.

hamilton

Reply to
hamilton

If you need less than 50 % duty cycle, so why bother with mains rectification :-).

In a CW/FM transmitter chain that would not be too unusual, but then there would also be capacitors across the coils to make resonant circuits.

Class B or C in a transmitter chain ?

This might to be some "black box" building block that might serve some usable purposes. Adding some external components might make this a usable system building block for example:

  • with a frequency selective feedback, this could be used as a Q-multiplier with some superregenerative features in which the ne negative half cycle is used to kill any oscillation
  • Used between a chopper and a synchronous detector could be part of a low drift DC amplifier
  • Fed with some high frequency carrier, this would produce comb spectrum with sidebands every 50/60 Hz.

The comb generator is of course usable as a jammer, but also part of a frequency synthesizer. If you can generate a comb spectrum with carriers with say every 50 Hz, quite accurate frequencies can be generated, if the desired sideband can be selected.

A classical method of selecting a single tone from the comb spectrum would be to use a VFO, mix down and use a BPF filter with 10-20 Hz bandwidth to select the particular tone and then mix back up the only signal that passed through the BPF with the _same_ VFO back to original frequency.

Since the same VFO is used in both down mix as well as up mix, the (long time) stability of the VFO does not matter, but microphonics might still be a problem.

Reply to
upsidedown

If this is really a part of a transmitter, why bother running the stages in linear class A, more likely some class B or C stages could be used, if the final output stage contains some LC resonant circuits.

Reply to
upsidedown

But someone could sell hundreds of them on Ebay. ;-)

--
You can't have a sense of humor, if you have no sense.
Reply to
Michael A. Terrell

It looks like junk. The input is referenced to one side of the AC (!), but the grid bias is referenced to earth? V3's input looks odd too, it's referenced to earth, not the rest of the circuit. So there's something funky going on there.

Just download LTspice and simulate it. Cheaper, safer, faster.

formatting link

If you're still not satisfied, then you can consider the expense of actually building this. I would assume the weird earth thing might be the key to this thing.

Reply to
a7yvm109gf5d1

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.