phase splitter issue

[SY]

If by "problem," you mean "predicts the correct answer for all pairs of real (and imaginary) loads," then yes, there's a problem.

 

[CPaul]

SY, we have proven your logic is faulty. And your only response is to say, "I bet you can't use my faulty logic to prove my logic is faulty!"

Well, I gotta admit. You got me on that one.

 

[Alfred Centauri]

Chris, something else has been bothering me about SY's approach and conclusion but I haven't been able to quite put my finger on it until just now.

In addition to the numerous problems already covered, there is an even deeper, even more fundamental problem: it is inconsistent with the principle of superposition.

 

[SY]

Well, I'll ask again- what's this mysterious pair of loads which cause my model to give incorrect results? You told me it's been shown again and again, yet I can't find it. A simple schematic of these loads will do.

 

[kenpeter]

Show me a simple schematic of any two grids that simultaneously drop to zero
when driven by a Concertina. Done, put a fork in it...

 

[SY]

I would if I understood what you're talking about.

 

[Alfred Centauri]

Well, I'll ask again- what's this mysterious pair of loads which cause my model to give incorrect results?

Any pair of loads.

 

[SY]

Oh, I see.

 

[SY]

Show me a simple schematic of any two grids that simultaneously drop to zero
when driven by a Concertina. Done, put a fork in it...

A grid load is, for most concertina applications, a high resistance (the grid leak) in parallel with the input capacitance. I've experimentally confirmed that my model gives accurate results for that.

You may be confusing what the concertina drives with how one defines Thevenin source impedances for each of the output polarities (open circuit voltage divided by short circuit current- neither of these is normal operation for any amplifier).

 

[CPaul]

And SY, I'll say it again in response to your request for mysterious loads: We have proven your logic is faulty. And your only response is to say, "I bet you can't use my faulty logic to prove my logic is faulty!"

 

[Alfred Centauri]

Oh, I see.

Unlikely!

 

[SY]

I present a model. It makes predictions. The predictions are verified experimentally. No exceptions have been proffered.

And my logic is faulty?

 

[CPaul]

Yes Alfred, I've considered Superposition. But if he ignores Thevenin, why would he pay any attention to Superposition?

 

[CPaul]

SY, it certainly is, and we've explained why. You've simply ignored it.

 

[Alfred Centauri]

Yes Alfred, I've considered Superposition. But if he ignores Thevenin, why would he pay any attention to Superposition?

Indeed!

 

[leadbelly]

SY, it certainly is, and we've explained why. You've simply ignored it.

Sorry, I've read every post in this thread and I have not seen where you have disproven his work. You are attacking his derivation of the model, but simply have offered no rebuttal of his data. He's killing you.

 

[CPaul]

Maybe we should ask.

SY, are you aware of the Superposition Theorem?

 

[CPaul]

Killing me, leadbelly? With data from bad models? Just tell me one thing: how does the (two-node) Thevenin theorem possibly, conceiveably, by any stretch of the imagination, jive with an impedance that is measured by shorting three nodes together?

 

[CPaul]

Why in the world, leadbelly, would anyone give any consideration to data obtained from bad models? Should I pay any attention to somebody who wants to refine the Ptolemaic model of planetary motion with new data?

 

[leadbelly]

Killing me, leadbelly? With data from bad models? Just tell me one thing: how does the (two-node) Thevenin theorem possibly, conceiveably, by any stretch of the imagination, jive with an impedance that is measured by shorting three nodes together?

Why in the world, leadbelly, would anyone give any consideration to data obtained from bad models? Should I pay any attention to somebody who wants to refine the Ptolemaic model of planetary motion with new data?

Experimental data is data, period. It doesn't come from a model. The model is used to explain the result. So, if you want to disprove SY's model, show how the data also fits your correct model, or instead fails to cover part of the experimental space.