Hmmm... was there any issue of clarity in my statements of the application???
From my initial post.
In any event, the transistor circuit without the diodes has a 1.4V deadband in which neither of the transistors are turned on to any significant degree. I can't think of a use for such a circuit, unless you actually want the level to shift in that manner for some reason. That would be a very specialized circuit, indeed.
Sorry, I have no idea how this turned into your design. Ok, whatever. I wasn't discussing some unrelated design. I was discussing an audio op amp design on a 12V power supply that required a virtual ground.
The transistor circuit does nothing useful. The 6V virtual ground doesn't source power. It's there to provide the reference voltage for the rest of the circuit. The two transistor circuit without diodes is not useful for this, as it wobbles all over if you draw any current. You would probably be better off with just resistors and no transistors.
I don't know what you are trying to say. I have no idea why you are talking about a completely different design than what is being discussed here. I have no interest in continuing this extraneous discussion. Thanks for your comments.
But it's clear from the discussion that you think the solution "doesn't work", when it fact it is a useful gizmo. Your 'audio op amp' comment doesn't specify why you think there's some significant problem, nor does my example of a TIA have a problem with a volt or three wandering.
I can't run your simulation on this machine, and wasn't referring to it.
I assume that the downstream rails will be bypassed, so one needs a c-load stable opamp, or more parts.
If you don't care about the wandering, why would you need any circuit at all? Two resistors would be very adequate from what you've said.
Ok, so you were conversing with yourself. Sorry to intrude.
With an unbalanced load, 56KR on the positive rail and 560KR on the negative, you see this:
Danke,
Please ignore previous followup. 56R needs to be connected between the positive rail and virtual ground, and not 56 KR.
Danke,
You might fix the link while you are at it.
Here are two example situations where two resistors doesn't work, and works, respectively:
Adding the two transistors turns the =( face circuit into at least a =| face, which is a non-negligible improvement for almost negligible cost
You say that, but what voltage do you get when you add the two transistors? It's no longer 6V, that's for sure. It would be 5.3V, I believe. So it's still a frowny face. That's the point!!! The transistor circuit is not much better than the two resistors.
I have no idea why you are pursuing this. It's obvious that a 6V regulator that has 1.4V of deadband is not of much use. Somehow, you seem to want to distract from the reality of this with silly circuits.
In the simulation of the two transistor circuit without the diodes, the emitters junction follows the load voltage, until about 0.7 volts is reached. Only then does it start to pass any real current. Run the simulation. You will see what I mean. This is a crazy circuit.
Don, time to renew your ham license !
boB
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.