unbypassed cathode resistor

[rongon]

I've been experimenting recently... (Uh ohhhhh.....)

I have a 6N30P line amp, common cathode amplifier, 15k plate load resistor, Va = 85V, Ia = 18mA, Vg = 3.6V. Using two series-connected red LEDs in the cathodes to get the grid bias.

I could use less gain. So... What if I change the LEDs to an unbypassed 200 ohm cathode resistor?

- Gain should go down from +25dB to about +20dB.

- 6N30P ra will go up from under 2k ohms to maybe 6k ohms. But I have a 15k plate load (B+ is +400V), so that should be OK with no changes necessary.

I've been reading all sorts of internet posts and articles about unbypassed cathode resistor. The conventional wisdom seems to be that it will lower distortion, possibly change distortion spectra to less 2nd HD, relatively more 3rd HD. Some say it's a bad thing to do, others think it's a good thing to get rid of the bypass cap.

I'm going to try it and see what happens, but I was wondering if there are theoretical disadvantages to reducing gain this way. Or is it the kind of thing where it's a good idea in some situations, not so good in others?

 

[SY]

Or is it the kind of thing where it's a good idea in some situations, not so good in others?

Bingo.

Easy to try, so go for it.

 

[bob91343]

If there is no feedback, this should work. But if there is a feedback loop it's not such a good idea. If the gain reduction is too great, bypass part of the resistor or use fewer LEDs and a resistor.

 

[Wavebourn]

Cathode resistor actually introduces a feedback loop by current that increases resulting tube's internal resistance.

 

[kavermei]

If you replace the diodes with a potentiometer, and connect a capa between the wiper and ground, you could "dial in" the sound and gain you like best. Should be interesting

 

[Arnulf]

Um, isn't almost a twofold decrease due to cathode degeneration just a tad overoptimistic ? I didn't put the numbers together but it sounds like huge change (the difference is usually 30% or thereabouts, not 50%).

 

[rongon]

Didn't get to try the unbypassed cathode resistor last night. But I did make headway on updating the power supply in my pp 2A3 amp.

isn't almost a twofold decrease due to cathode degeneration just a tad overoptimistic ? I didn't put the numbers together but it sounds like huge change (the difference is usually 30% or thereabouts, not 50%).

Do you mean I'll get only 30% less gain? If so, that would be OK. I ran the numbers and that's what I got, but I could have done it wrong...

Cathode resistor actually introduces a feedback loop by current that increases resulting tube's internal resistance.

That would decrease the transconductance and mu too, right?

What I don't understand is how the result would be lowered distortion... but that's what it sounds like when I tried it last (using an ECC88 with about 10mA going through it).

In this case a reduction in gain would be a good thing, so I'd be able to open up my volume pot past 1 o'clock for a normal listening level.

No global feedback loop in this circuit. Just a plain-jane common cathode amplifier with a selector switch and a pot.

-=|=-

 

[Arnulf]

Do you mean I'll get only 30% less gain? If so, that would be OK. I ran the numbers and that's what I got, but I could have done it wrong...

Yes, it depends on how you run your tubes though (quiescent point).

 

[DF96]

An unbypassed cathode resistor reduces transconductance but leaves mu unchanged. This is why it reduces gain less than you might expect. If a triode has a CCS load then the gain could be virtually unchanged.

As it is a form of negative feedback it reduces total distortion, although there may be some shuffling of the order of distortion from low to high. Reducing gain will reduce the Miller effect, but how this affects HF roll-offs will depend on the source resistance feeding the stage and the load capacitance after the stage. It could go either way.