6av6 question, would this work?

[AZJimC]

The thought I had with the 6av6 and dual diode is this.

Suppose I use a resistor to send a portion of the ac signal coming from the plate, post coupling cap, back to the diodes. It looks like this would/could lift the voltage of the cathode, (subject to the cathode resistor size) in response to increased output. Is it possible that this property could be used to create a simple one tube compression effect? Would this be simply a local nfb? Or would that depend on how loose the cathode resistor is?

The diodes say they're 2.0ma each at an input of 10vdc (above cathode level I assume).

[jjasilli]

I've wondered that myself, but can't find any useful info or schematics.  This is the best I've found:  http://www.ax84.com/bbs/dm.php?thread=61729


Maybe find an old radio schematic.  The tube is used as a radio detector.  Not sure if its gain control function acts on the incoming radio signal, or on the resultant audio signal. 

[sluckey]

The diode cathodes are connected to the triode cathode. This may pose a problem for what you are thinking about.

[AZJimC]

The diode cathodes are connected to the triode cathode. This may pose a problem for what you are thinking about.

It's that connection that I thought of using, i.e. manipulation of the cathode bias voltage via the signal, possibly an inverted signal from a stage following the av6, if needed? seems that the plate signal from that triode could be used to raise cathode voltage thus nudging the triode's output down based on a higher level of current draw through the diodes, and resulting raise in cathode voltage.

[jjasilli]

I found some schematics which actually make use of the diodes in the 6AV6 by Googling "vintage radio schematic 6AV6" and looking in "Images".

[AZJimC]

I found some schematics which actually make use of the diodes in the 6AV6 by Googling "vintage radio schematic 6AV6" and looking in "Images".

I looked at a few of those with the 6av6-12av6 type. It looks like most of those are using the diodes as detector, their design purpose, to strip the AM carrier from the audio signal.

The amplifier usage I've seen attempted so far seems to be as a clipping diode, and that might be all I could end up with. My thought would be to adjust the cathode resistor so that the ac peaks of audio signal could pull the cathode voltage upwards smoothly, in a more linear way than clipping, and result in a smooth compression. The specs say that a 10vdc on the diode plates can draw 2ma each, so 4ma could lift cathode voltage slightly given the right Rk. The problem I see is that it would only effect the positive swing of the ac signal, and result in an imbalance possibly. On the other hand, if the upswing were lowered, that should lower the total peak to peak voltage.

The other issue is, once the Rk is sufficiently "loose" for a <4ma current to change the bias level, the triode may be incapable of much, if any amplification. Unity would be okay, since the effect is what I'm after, but it may not even be capable of unity. I wish I could sim the whole thing, but I have no sim software, and have yet to learn how to configure one if I had it.

[HotBluePlates]

The problem I see is that it would only effect the positive swing of the ac signal, and result in an imbalance possibly.

You mean "half-wave rectification"  which is what the 6AV6 normally does as a detector.

Remember that the rectified signal has to be converted to a d.c. control voltage to do "automatic volume control" or compression. That takes a C and an R. Once you make C & R big enough to steady the rectified signal to d.c., the R is too big to act as a cathode resistor anymore.

[AZJimC]

Once you make C & R big enough to steady the rectified signal to d.c., the R is too big to act as a cathode resistor anymore.

I think you're saying what I think you are. But, consider no rectification, (even an unbypassed cathode resistor) just straight AC audio sent to the diodes. Every AC upswing would cause the diodes to conduct. This could have two effects. One, the AC signal being sent would load some, against the cathode resistor. This could work toward trimming signal amplitude some, on a wave by wave basis. Second, provided a loose enough cathode resistor could keep the triode at least unity gain, the up to 4ma current could pull the cathode voltage ever so slightly upward. Hmmm, the thing may oscillate, due to the phase delay of the coupling cap though....... I like it when I think myself into a corner like this. Is there a way out?

[sluckey]

Study the use of that tube as an audio detector/AVC circuit. The diodes rectify the IF signal and a cap strips the IF signal leaving only the audio. But this rectified signal is also sent to another filter that produces a dc voltage that is proportional to signal strength. This dc voltage is sent to previous IF gain stages to automatically control the gain. Strong signals produce a larger AGC voltage that will reduce the gain of the previous IF stages. Weaker signals will produce a smaller AGC voltage that will not reduce gain of the previous IF stages as much. The net results is that a weak radio station will produce a comparable audio signal as a strong radio signal.

The above operation can be compared to the operation of a compressor circuit. With some tinkering, you may be able to come up with a workable compressor/sustainer circuit for guitar. I'd like to see what you come up with.

[PRR]

> signal coming from the plate, post coupling cap, back to the diodes. It looks like this would/could lift the voltage of the cathode

Plate current and cathode current are equal.

This means the plate can not over-power the cathode. It will "lift" but very slightly. IMHO, not enuff to make any real difference. Meanwhile the actual output signal will be severely distorted by diode load into low cathode impedance.

Go ahead and try it. It isn't an all-day wiring job.

Don't fret the diode current ratings. I suspect they could take much more in musical bursts. I suspect the voltage drop will get huge before any damage happens. (10V at 2mA is only 0.020 Watts, and even those teeny plates can stand more.) I suspect they really picked that low number to "force" you to design-in a bigger (more price more profit for tube-maker) tube for heavy work. And it isn't like 6AV6 is big money.

[kagliostro]

I was looking to the Gibson Service Manual that Doug has on the archive and I just discovered the GA-100 Bass Amp circuit

where, I think, there is something about the use of the diodes that can be of your intereset



Franco

[AZJimC]

K,
Thank you for your thought, and yes, that is of interest. I see those tubes don't have the diodes using the same cathode as the triode does. I guess that allows some trickery possibly not available on the 6AV6. I'll have to digest this scheme later, when I have time to "become the electron", but a tube compression that's rather simple is definitely of interest to me.

Thanks, Jim

[kagliostro]

Ciao AZJimC

I remembered that long time ago I've find this compressor / expansor schematic



(I think the author was an East Europe guy who build some bass amp, if I remember well)

The author told you can change from compressor to expansor simply inverting one diode (the diode labelad as "D")

the tube (ech84 / 6jx8) can be found on the cheap

https://tubedepot.com/products?utf8=✓&keywords=ech84

http://www.thetubestore.com/Tubes/6J-Types/6JX8-ECH84

I hope this can be of your interest, unfortunately I've no more info about this circuit

Franco