The AM Forum

Some new stuff.

 Specification Sheet for 4-1000 Triode Connected Grounded-Grid Service

 Tube Limter from 1957 Radio & Television News, tnx to WQ9E


http://www.amwindow.org/tech/htm/gen.htm

Thanks for updates Steve; I and a lot of others have always enjoyed your page.

It's not my page, it's your page. I just put the stuff up there. TNX just the same.   ;D

Amen.  I've learned a lot from AM Window.  Kudos to Steve for maintaining it.

Very nice! This is a new application for me to use the grids themselves for their own AGC voltage sources. the tube in question should be plentiful, being an old TV set tube. Hopefully.

One thing I do not get in the article is the statement (page 50, principles of operation) is that "Distortion produced by the non-linear operation characteristic of the variable-mu tubes is predominately even-order harmonic and is, therefore, phase canceled in the audio output transformer T2."

When I look at the diagram, consider the cathode bias to be (example) 2V, and think about a non-sinusoidal waveform that has a (example) 3 volt positive peak and a 0.5V negative peak being applied equally and out of phase to the two grids in the push pull stage, I can see the 'top' tube section with the wannabe 3V peak making a negative charge in its grid blocking capacitor thereby reducing its own gain and the 'bottom' tube with the 0.5V negative peak not making a charge and therefore still running full gain. The gains of the + and - sides of the amplifier are different and I do not see how they would cancel the distortion via the output transformer.

Would the cancellation take place via the cathode bias resistor, being unbypassed, providing a phase inverting function to the opposite grid-cathode circuit to thereby cause the bottom section to reproduce (ignoring for the moment the -0.5V wave applied to that other section) the inverted version of the top tube's waveform at its plate? Is this what the author means?- that the distortion cancellation is eventally done in the OPT due to the signal impressed upon the common cathode resistor of the stage?

Another question about this, as well:

As each stage gives 10dB compression, the two stages making 20dB, I wonder if a third stage (30dB range) would be worthwhile or if it would be so difficult to implement (noise floor wise) that it would not be worth it (attachment). It is stated that any number of stages can be used.

Based on the circuit, if another stage were added in the middle as shown, would the range be extended to 30dB without the distortion becoming too high due to the first stage being well overdriven?

Is there a point in having more than 20dB of consonant amplifier action, or would any such need be better handled by a slower acting gain rider or the "operator in attendance" mentioned in the article?

Once things get sorted out here a little better, I would like to try this circuit as-designed and see how well I can get the results the OM did. But I may punch a couple extra hoes in the chassis just for fun.

I would value the opinions of any audio experts on the questions above though. This one has got me thinking too hard.