Parallel-to-single V1 stage

[tubeswell]

FWIW I had a few hours up my sleeve so I tried this circuit (from Ch 4 in Merlin's 1st book) yesterday on one of my amps. There is quite a noticeable gain difference between each input.

[Chuck H]

Interesting then is the common advice that parallelstages DON'T increase gain. It could be that because of the lower impedance output the tone stack doesn't load the signal as much where a straight fed triode (coupling cap, 1M pot) might be less notable. I'd have to do math and stuff for that though.

[kleuck]

It's necessarily increasing gain here, as the load is the same resistor for the two triodes. 1974X power Actually, i made a little mixer/buffer/splitter with two fets -same schematic as this one- and when using the two inputs, there's more gain and a "fatter" sound too. Crappy drawing, and i did not draw the mono-parallel switching, but :

[Chuck H]

[QUOTE=kleuck;244519]It's necessarily increasing gain here, as the load is the same resistor for the two triodes. [\QUOTE]

Ah. Good catch. I wonder then what the tone/gain effect would be if each triode had it's own load resistor?

[tubeswell]

[QUOTE=Chuck H;244549]

It's necessarily increasing gain here, as the load is the same resistor for the two triodes. [\QUOTE]

Ah. Good catch. I wonder then what the tone/gain effect would be if each triode had it's own load resistor?

I did a 5F6ish style amp where I connected the grids of the first two V1 stages. With either vol pot cut down the amp is quieter than if you have both pots up at the same rotation, but its probably a function of the AC load from the two pots right?

[MWJB]

Tubeswell, have you tried different 12AX7s, does the gain remain the same from tube to tube?

[Chuck H]

That's a good Q. Some 12ax7's are very dissimilar side to side. It could be the sort of amp that really makes a difference from one tube to another for tonal variation. It might even be fun to drop a 12dw7 in there (If the low mu triode happens to engage in parallel mode).

[tubeswell]

Yeah I've tried different tubes in that V1 position in the '5F6B+' amp, and it makes no difference - the gain is always higher if you have both pots up at the same rotation, than if you have one pot cut. Mind you I have only tried this with the pots going up to about 7/10ths of full rotation - it gets too damn loud for my ears. It seems to make the most volume difference when you do this on the lower settings of the pots (But is this because of the difference in differing AC loads?, or because you hear vol differences more at lower power?)

Surely the voltage swing in the '5F6B+' amp V1 stage is not much different in either triode, and when you put two of the same voltage swings together, the voltage swing remains the same as a single voltage swing.

Now on the other hand, if you put twice as much current through a single plate resistor that is connected to two parallel stages, then the voltage drop across the resistor must double, and if the HT is still the same (as it was when you only had 1/2 as much current through the plate resistor) then the plate-to-cathode voltage must decrease somewhat, but does this add up to nil overall gain? I'd need to do a load line to see.

But whatever the answer is to that it seems to me that the neat thing about Merlin's circuit is that when you run both inputs in parallel you are getting more gain than when you are running the single stage input, because of the effect that the idling triode has on pulling the active stage's load line down somewhat when in single stage input mode (as well as the output impedance difference between single and parallel modes I guess).

[kleuck]

I think that the most important thing is the internal resistance of the tube, which you halve when paralleling the triodes.

[Tom Phillips]

Remember that the tube by itself does not amplify the audio signal. You need to compare the effective before and after values of the anode resistor and the cathode circuitry to determine the gain difference between the single and parallel triode circuits. Tubeswell - Since you are using a 100k plate resistor and a 1k cathode resistor in your parallel triode circuit I would expect that you would get more gain than a single triode using common circuit component values for a Fenderish type input stage. The theory is explained in detail in the same section of Merlin's book that you referenced in the OP.
Regards,
Tom

[mooreamps]

The last two posts are both correct...

-g

[hasserl]

You're also halving the input grid resistance from 33k to 16k5.

[kleuck]

Does not impact gain, nor practically the Miller effect, as the capacitances are twice compared to a single triode.

[hasserl]

It impacts the perceived gain, as the strength of the signal at the grid will be stronger, causing an increase in current across the tube.

[Chuck H]

It impacts the perceived gain, as the strength of the signal at the grid will be stronger, causing an increase in current across the tube.

But there's very little grid current and those grid resistors are (typically) located AFTER the grid 0V reference. Because they're located after the 0V reference there is no voltage division and because there is minimal current there is no voltage drop. IMHE there is such a small difference in tone with different values in that circuit that I've wondered if I'm just imagining it. Almost certainly no notable difference in gain.