Transistor Buffer Design

eblc1388

Joined Nov 28, 2008
1,542
Wonder what difference it would make if R10/11 were 1Ω instead? 1Ω is off the shelf.
I must have seen the circuit somewhere but it seems the logical circuit connection to make.

Just open the circuit inside LTSpice, place the mouse cursor on the resistor and right clicks. Enter 1 instead of 0.2 into the value. Do this for both resistors.

Then press the "running man" icon and the results appears before your eyes.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
Like I said, I'm gonna steal it, probably redraw it to make it slightly prettier. To me there is no higher praise.

I'm guessing it doesn't draw any current if it is biased correctly.
 

eblc1388

Joined Nov 28, 2008
1,542
Like I said, I'm gonna steal it, probably redraw it to make it slightly prettier. To me there is no higher praise.

I'm guessing it doesn't draw any current if it is biased correctly.
If you like the circuit, its yours. No copyleft issues whatsoever.

The changing of final stage emitter resistors from 0.2 to 1Ω degrades the performance of the circuit and now clipping occurs with output load of 1Ω.

One need to ensure at least 20mA collector current in the final power stage or else risk cross-over distortion.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
With only 12 volts supply rail I think you are loosing too much in the Darlington connections.

Why not try the alternative of npn/pnp and pnp/npn complementary connections for upper and lower ouput compund transistors respectively?

This will put ony two BE junctions in series with the load, rather than 4.
You suggested something similar to this, but a picture is worth a thousand words.



The arrangement is called a Sziklai pair, also sometimes referred to as a compound transistor or compound Darlington.
Interesting. I must have slept through that class, or it came out after. One difference though, R4/R5 do affect the biasing somewhat.

I'd like to find more local versions for Q3 and Q4 though.
 

Attachments

Ron H

Joined Apr 14, 2005
7,063
I have used the Sziklai pair successfully in low-current applications in place of a Darlington. There is one caveat. It is a feedback loop, and it can oscillate under some conditions of loading, transistor Ft, etc.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
Yeah, I was thinking it would make a dandy pseudo ground for a dual power supply, high current output and all.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
That is digital mode. This is an analog amp. I knew it could burn up the transistors (which I've been saying from the beginning), but that is not going to happen unless it is actually driving something.

Thing I like about the layout is just sitting there it isn't using any real juice, so it won't get hot until it is used. You start having any of the transistors dissipate more than a watt (or less) without heat sinking and it gets interesting fast.

Like most Darlingtons the first transistor is not pushing as much of a load, so italo's math is a lot off. Figuring ½ of 3VPP is 1.5, that is what the driver transistor is pushing, which works out to around 30ma. The first transistor is a fraction of that.


A comment about this...

Spice boys are having a field day on this thing. first of all it is an AC amplifier. Why it cannot push a 50 load simple You expect a 2n2907 to deliiver an Ic of 120ma. wishfull thinking.
Besides being off math wise, it is borderline abusive. The idea is to expand your knowledge with this thread, point out the real problems without name calling. Math speaks for itself, and it is possible to find a circuit interesting and acknowledge it has flaws. In this case I don't think you've worked it out, since the circuit is not conducting until it is forced off it's center state, and then only one side turns on at a time, feeding the load through the capacitor. I kid my fellow members, but they can tell the difference, this wasn't kidding. I don't know if you've noticed, but we help a lot of new people, without trying to make them feel small.
 
Last edited:
Top