Best way to combine two dissimilar channels into PI?

[dehughes]

Hey, maybe this idea is de facto the same as each triode sharing the same parts...but is it possible to eliminate the negative aspects of parallel triodes if each triode IS paralleled, but with their own plate and cathode resistor and bypass cap, and the signals are summed after the coupling caps off of each plate?

Does that make sense?

[Wayne]

There are a lot of ways to go with this - separate everything like you said above, separate cathodes and shared plate, separate everything with slightly different values, one cathode bypassed & one not. I'll bet there are subtle differences between them all.

W

[dehughes]

There are a lot of ways to go with this - separate everything like you said above, separate cathodes and shared plate, separate everything with slightly different values, one cathode bypassed & one not. I'll bet there are subtle differences between them all.

W

Sure...and I wonder if there is a way to get paralleled triodes without the negative aspects of higher odd order harmonics. Hence my curiosity as to summing two separate triodes after the coupling cap....with each triode having it's own plate and cathode resistors.

Is that feasible? Might that eliminate the drawbacks of paralleled triodes sharing the same Ra and Rk?

[Wayne]

Certainly feasable, as long as they are driven in phase. As to what it would do for harmonic content, I'll leave that to the experimenters with better test gear than me.

Another possibility I left off the list above - if you're paralleling two triodes with completely separate support components the two triodes wouldn't have to be the same, would they?

W

[dehughes]

Certainly feasable, as long as they are driven in phase. As to what it would do for harmonic content, I'll leave that to the experimenters with better test gear than me.

Another possibility I left off the list above - if you're paralleling two triodes with completely separate support components the two triodes wouldn't have to be the same, would they?

W

EXACTLY! I don't know what I'd do differently with the triodes, though, as I'm trying to get as much gain out of each triode as I can...

[rhinson]

There are a lot of ways to go with this - separate everything like you said above, separate cathodes and shared plate, separate everything with slightly different values, one cathode bypassed & one not. I'll bet there are subtle differences between them all.

W

Sure...and I wonder if there is a way to get paralleled triodes without the negative aspects of higher odd order harmonics. Hence my curiosity as to summing two separate triodes after the coupling cap....with each triode having it's own plate and cathode resistors.

Is that feasible? Might that eliminate the drawbacks of paralleled triodes sharing the same Ra and Rk?

sort of like the first 12ax7 in the john mcintyre bluesmaker schematic from a guitar player issue way back in 80's or 90's i believe

http://www.blueguitar.org/new/schem/mis ... er_sch.gif

rh

[dehughes]

There are a lot of ways to go with this - separate everything like you said above, separate cathodes and shared plate, separate everything with slightly different values, one cathode bypassed & one not. I'll bet there are subtle differences between them all.

W

Sure...and I wonder if there is a way to get paralleled triodes without the negative aspects of higher odd order harmonics. Hence my curiosity as to summing two separate triodes after the coupling cap....with each triode having it's own plate and cathode resistors.

Is that feasible? Might that eliminate the drawbacks of paralleled triodes sharing the same Ra and Rk?

sort of like the first 12ax7 in the john mcintyre bluesmaker schematic from a guitar player issue way back in 80's or 90's i believe

http://www.blueguitar.org/new/schem/mis ... er_sch.gif

rh

BINGO!!!!! I KNEW someone here would have a point of reference for that idea. This place is golden...

So, for the mixer resistors on the two preamp stages, could I theoretically use any of the same values? Like 10k each? I ask because I'm mixing parallel 6SN7 stages and don't want to loose any more signal than I have to...

[Andy Le Blanc]

I really like the "preamp balance" control, I haven't tried that one yet.

I like the sound and feel of ye-olde pre from an old twin 5E8-A.
Its fairly easy to include grid stops at the end of some shield cable.
I liked it because it really does give you two gain choices, a single stage
or a parallel stage.

[dehughes]

Very cool.... So I'll just run the mixer resistors together after the coupling caps on the 6SN7 plates, as the volume control (in my case) would provide the ground reference before the next stage. PERFECT!

[Wayne]

Funny this parallel triode thing should come up now. I've never met a Matchless Chieftain in person, but I had a look at it's schematic today.

http://www.schematicheaven.com/newamps/ ... ieftan.pdf

This may not be news to some here, but all of the preamp stages are parallel 'AX7s. According to everything we've learned here, shouldn't they sound awful? Am I missing something here? I find it odd that the shared Ra & Rk values look pretty typical of those for a single section.

W

[dehughes]

Funny this parallel triode thing should come up now. I've never met a Matchless Chieftain in person, but I had a look at it's schematic today.

http://www.schematicheaven.com/newamps/ ... ieftan.pdf

This may not be news to some here, but all of the preamp stages are parallel 'AX7s. According to everything we've learned here, shouldn't they sound awful? Am I missing something here? I find it odd that the shared Ra & Rk values look pretty typical of those for a single section.

W

Too funny...you're right...it SHOULD sound awful...

Another amp with a parallel triode setup (at least the first stage) that sounds awful(ly good) is the Top Hat Club Royale. Amazing sounding amp...one of the best I've ever heard (still own mine...going on 11 years now).

So, I'm wondering if this rise in the higher order harmonics in paralleled triodes has to do with (noobie hat on here) lowering Ra and that being a path to AC ground. That is, in the "data sheet" posted earlier, there was no chart showing what the harmonics were doing if the Ra was left at 100k, the triodes were paralleled, and Rk and Ck were adjusted instead, with Rk being halved in value. I think I'm going to try this out for myself....

[Firestorm]

I think (from Steve's original analysis) that this is the one configuration that should sound better than a single triode because of the increase in 2nd-order harmonics across all voltage ranges.

I will be curious to see how fiddling with the cathodes while leaving the plates at stock value affects the sound in paralleled tubes.

[dehughes]

I think (from Steve's original analysis) that this is the one configuration that should sound better than a single triode because of the increase in 2nd-order harmonics across all voltage ranges.

I will be curious to see how fiddling with the cathodes while leaving the plates at stock value affects the sound in paralleled tubes.

Cool. So, I'm running each 6SN7 triode with a different set of plate caps and cathode caps....but I'm kinda hesitant to put mixer resistors in the signal path to join the two triodes' signals, as the 6SN7 is such a low gain tube as is. Do I HAVE to use mixer resistors? Could I just join the two plate caps to the input lug on the volume control???...or would my amp explode?

[Wayne]

Well, nothing *should* explode (notice how I'm hedging my reply - I think we've all seen stuff explode for no apparent reason )

You'd certainly get output from the stage(s), but both triodes would be working into the same AC plate load. Assuming that your coupling caps are suitably transparent at frequencies of interest, that would be Rp//other triode's Rp//Rvol. If one cathode was fully bypassed and the other only partially you'd get a treble boost, but IMO it should work the same as both plates sharing the same resistor.

W

[dehughes]

Interesting. The setup would be:


Triode One: Ra(100k) - Plate cap \
-----------------------------------------Volume
Triode Two: Ra(100k) - Plate cap /


That shouldn't do anything odd, right? I mean, I suppose I'm wondering what purpose mixer resistors serve, or would serve, when the signal is the same albeit coming from two different triodes?