balanced tube ccts

hey guys... have there been any (successful/established, i guess) tube mic pre circuits that are balanced throughout, as in separate amplification for side of the signal? Is this worthless/overthetop/ridiculous?

I guess tube doesnt necessarily matter. But I'm just wondering if anything is ever designed essentially like a stereo unit, but where pin 2 goes to one side, and pin 3 goes to the other, CMR never takes place, and it counts on a later stage to perform CMR.

All things being more or less matched, wouldn't it drop the noise of the entire stage as low as the CMRR of the following stage will allow?

I guess even a stereo unit could do this, if used unbalanced. Maybe i can just finish this g9 and this NYD 1bottle and try it out for myself. Just wondering if its been done much.

billy

That´s not what he is talking about. Disa has common cathodes.

where pin 2 goes to one side, and pin 3 goes to the other, CMR never takes place, and it counts on a later stage to perform CMR.

Click to expand...

What´s the point of this? Never seen a commercial one, but I do know people that have done it.

You could do it with any SE circuit. Just make two mic preamps, one for each side of the signal.

if you include an output transformer, it accounts for the CMR alone.

what your really describing is a normal push pull circuit, like a a Langevin AM 16 or a Pultec MB-1.
You have a common ground, but a stereo amp would need a common also, so it's really the same deal.

I see no benifit to splitting the signal into two parts and then recombining, as the crossover distortion would probably be pretty bleek.
But who knows, weirder things have been done.
cj

the 'later stage' i was referring to would be a whole other circuit or piece of gear. Most likely a recorder or console input. Basically I'm talking about dual unbalanced, matched mic preamps, that each amplify one side of a balanced mic output. One pre has pin2 input, the other pre has pin3 as an input, one outputs pin2 again, one outputs pin3, signal amplified, all noise (depending on how equal the ccts are) is totally common. Then let the input of the recorder or console handle the rejection of the common tube cct noise.

just something i had started thinking about during work today.

billy

What you desribe (two separate amps for the + and - inputs) is theoretically possible, but no real use in practice. You can't have 60dB of gain and leave the CMRR to the following stage - it would saturate with any significant common-mode signal. The other problem is that any small difference in the linearity of the two amps shows up as signal at the output, and non-linearity distortion of that type is particularly nasty on the ear.

got you.... i guess that's what i was trying to say... would it be too much common mode noise for a later cmr stage to handle, and would the difference between sides be "particularly nasty" thanks... i suppose something of the aikido variety might be a better idea..... thinking of trying the aikido cct with ll1538 in and ll1517 out, but i have other stuff i should finish before i act like i am good at finishing things.

thanks for hearing out my theory 'thing' :grin:

billy

[quote author="enthalpystudios"]the 'later stage' i was referring to would be a whole other circuit or piece of gear. Most likely a recorder or console input. Basically I'm talking about dual unbalanced, matched mic preamps, that each amplify one side of a balanced mic output. One pre has pin2 input, the other pre has pin3 as an input, one outputs pin2 again, one outputs pin3, signal amplified, all noise (depending on how equal the ccts are) is totally common. Then let the input of the recorder or console handle the rejection of the common tube cct noise.

just something i had started thinking about during work today.

billy[/quote]
I realize you meant dual unbalanced, but let's do the thread title some justice and mingle this discussion with balanced circuitry as well :twisted:

Initially I had the impression that tubes were used to introduce pleasant distortion. But I learned that in 'good gear' tubes are more considered to pass along the audio natural, just amplified. So used to avoid eventual nastyness of solid-state (clumsily stated, but you'll understand).

So while I initally thought that balanced operation of tubes wouldn't be a good thing when you're after 'bottle-sound' (since the pleasant second order harmonics are reduced/cancelled), we now see that balanced tube circuits will have less distortion vs an related but unbalanced topology.
Just thinking out loud - now back to your actual topic...

[quote author="enthalpystudios"]got you.... i guess that's what i was trying to say... would it be too much common mode noise for a later cmr stage to handle, and would the difference between sides be "particularly nasty" thanks... i suppose something of the aikido variety might be a better idea..... thinking of trying the aikido cct with ll1538 in and ll1517 out, but i have other stuff i should finish before i act like i am good at finishing things.

thanks for hearing out my theory 'thing' :grin:

billy[/quote]

Well, let's take a look at it a little deeper here at what the balanced amplifier designs are trying to accomplish. In the MB-1 and the EQP-1a, the two halves of the balanced signal* are transformer-split into a dual-triode, run in (as usual), class A, functioning as a voltage amp. i.e. as two separate amps operating on each half of the signal. The power amp here is two halves of dual-triode, class A, cathode-biased, push-pull amp. Since they're operating in class A, there is no crossover distortion (although that's not to say there is no difference between the two halves).

So, what's the value of all of this? Obviously we're not deeply concerned with the efficiency of the amp, as we're dealing with a preamp here. The other significant improvement of a push-pull design is cancellation of power-supply ripple through the output stage. While to some degree you can save a few $$ on power-supply filtration using this method, the real value is that it continues to cancel out this noise as the power-supply filter caps degrade - thus giving improved performance throughout the service cycle.

So, what would this give us by carrying it out through the entire signal chain? I don't know what use it would have, but I can see plenty of problems. In this case you'd be using the entire signal chain as the two halves of a phase splitter. You're certainly not going to get a better S/N or THD out of such a design. Depending on how you choose to recombine the signals, it could be either not so bad or horrible - depending on how differently each half of your monstrously complex phase-splitter operates.

So, yeah, I'd have to agree with you - over-the-top for no discernable benefit.

*note: actually, the signal is processed unbalanced through the eq section, but is sent as a balanced signal to the amp section.