Simulation Analysis of several unique Allison-based output stages.

[magnoman]

Originally Posted by keantoken

To my knowledge the transistors in that CCS are not saturated, even though Vce is 0. I

Again I'm relatively new to the forum as well as many of the acronyms, but if CCS refers to a constant current source with one BJT having its collector tied to the base Vcb = 0 then the Vce still has some potential relative to the emitter and the beta action of the transistor will prevent saturation.

Hope this helps
-Antonio

[kenpeter]

In the case of Allison: As long as the output device presents two emitter drops of
threshold to turn on (Darlington pair or enhancement MOSFET), error amp collector
stands plenty high enough. With bootstrap, no excuse VCE<VBE ever need happen.

Still would affect output device as it approaches the rail, no bootstrap help here...
But thats same for just about every amp...

Strange you would obsess on .0000000000000000000000013% distortion, and yet
be OK with operating a small signal transistor in a region where its been cursed by
the evil eye. Damn the real world physics of it!

[keantoken]

Magnoman, welcome to the forums. Here is the CCS I'm talking about:

Simulation Analysis of several unique Allison-based output stages.

If we see the transistor as 2 diodes, then Vce=0 should roughly be where all transistor action stops. I've found that the breakover voltage of the B-C and B-E junctions aren't always the same, so this relationship is not precise.

The text you quoted contains an error I made, I meant to say Vbc. I keep mixing them up!

Your explanations are helping. You seem to be confirming my suspicion that when the B-C diode becomes forward biased, it's recovery time and time constants (not to forget natural diode action) are what degrades the net gain of the transistor, even though the B-E junction is behaving like any normal diode. But this would mean that charge is stored in the B-C junction, not the B-E junction.

So far none of this points to driving small-signal transistors at Vcb=0 as a bad idea, all other things considered.

You say that it's closest to a dynamic capacitor on top of the Vbe junction. Do you mean in series with the base? Parallel to the B-E? Parallel to the B-C?

If I seem angry or annoyed, I am not and didn't mean to come off that way. We are making progress.

- keantoken

[keantoken]

Kenpeter, I have found that the beta of the Allison transistors is unimportant, as long as they have normal transconductance (the bases are connected to a very low-impedance high-drive node). This is true if you don't use a very large resistor in series with the base for compensation.

Most of us know that transistors are more linear with more than 2V Vce. However I never heard of an abrupt decrease in speed below this. Where do you draw your conclusions from, and where can I read about this?

- keantoken

[kenpeter]

My father is dead (Harley D. Peter - Engineering Fellow Raytheon/E-Systems).
Helped design a surveillance receiver abused by Nixon to spy on Watergate.
http://watkins-johnson.terryo.org/CE...ers/RS-111.htm
Designed and wire wrapped his own 20MHz minicomputer in 1973, compatible
with DEC PDP-8 instruction set. Nobody had a 20MHz computer then, NOBODY.
I should say he finished it in 73, obviously that didn't all happen overnight.

I can't ask him specific details why he taught me that rule. All his technical
books and handwritten notes are wherever his wife decided to dump them.
Mint Radiotron Designer's Handbook 4th ed, gone who knows where...

Hundreds much rarer texts, optics originally worth tens of thousands scratched
and broken. Sometimes I get a call when something she thinks is important hits
the curb, most of the time not. Its all just garbage to her, and our relationship
is one of smoldering hostility.

He said don't do it. Therefore I don't need a logical reason for my conclusions.

[sawreyrw]

keantoken,

I can't describe BJT storage time effects much better than I did in post 608. What don't you understand about it?

Rick

[keantoken]

All I know is that it is taking time for electrons to get from here to there. I want to characterize this phenomenon in terms of a dynamic reactive network. This would be meaningful for me and others in the hobby, and help me to know what I need to do to adjust to it.

- keantoken