Low voltage drop capacitance multiplier

Which sort of suggests using a couple of LM1117s or something...

John

John, Does that configuration meet the I/O minimum drop requirement?

...Jim Thompson

This looks OK to me up to at least a few hundred mA... has a full quarter of a volt of headroom!

John

I don't think that solution would be acceptable, considering my *whole circuit* is currently on a PCB about 0.5 inch square :)

What's your point? Neither does Spice. Our goal was to see what can be accomplished with a few parts, and to see how well it does or doesn't perform.

Right! I "invented" this when I was just a sprout, for the Winch Control Intercommunications Subsystem (catchy title, huh) for the C-5A transport. The handheld comm/control boxes derived filtered DC from the shared party-line twisted pair, and also injected audio and supersonic control tones (to control the cargo winch) into the line (by wiggling the base here.) I did a third-order version and accidentally discovered that a passive R-C network can have voltage gain... it oscillated until I added a resistor to kill the Q. All this was done in a fairly theory-free environment.

John

In article , John Devereux wrote: [...]

BTW: This circuit is also sometimes called a psudo-inductor. When seen from the Vin side, the impedance looks inductive. It is used where power and signal both travel on the same wire.

If you need more HFE you can use a "White Darlington"

__ ______ ! V_/ ! ! ! ! ! ! !

The NPN ends up with a quite small Vce but it still makes useful gain.

I wish you would use Tech-Chat or something, the circuits are excellent but the drawing is hellacious.

I had not thought of it that way... nice.

OK.

"Ken Smith" a écrit dans le message de news:djo1qs$6cp$ snipped-for-privacy@blue.rahul.net...

Unfortunatly this becomes beta dependent for its bias point and either the NPN gets saturated or the bias point will vary with the NPN's beta and temperature.

A simple cure would be either a diode drop or a divider to bias the base.

.------- --------. | v / | | --- | | ___ | | --+-|___|--+--- ---+-----+- | \\ ^ | | --- | | ___ | | '->|---|___|--+----. | D | | | --- .-. --- --- | | --- either R or D | |R| | | '-' | | | | === === === GND GND GND (created by AACircuit v1.28 beta 10/06/04

Ah, a post SMPS filter, that's a good application.

Twenty years ago, we were all nodding our heads and agreeing that you needed a linear regulator for high-sensitivity analogue circuits. Nobody has that luxury any more, and cap multipliers fix the problem.

Although their voltage regulation and low-frequency ripple suppression aren't that great, op amps have wonderful supply rejection down there. Up in the tens of kilohertz, where op amp supply rejection is poorer and all the SMPS junk lives, the cap multiplier is like a brick wall. Not bad for ten cents!

And I _have_ to be able to use that pseudo-inductor idea for something. What, I wonder? (I've been doing device work for too long, and need a good instrument to design.)

Cheers,

Phil Hobbs

In article , John Larkin wrote: [...]

I think a lot of people have invented this circuit or the version where the capacitor ties to the emitter.

I knew someone that got a patent on "Psudosilence". It is a very catchy name but you have to speak clearly when you say it.

"Beta dependant beta shmependant"

The goal was a low drop out voltage. The circuit does this to varying degrees for various transistors. Yes it is sloppy but it does work.

If you are going to do that, why not make that diode another NPN transistor and get even more HFE out of it?