Avoiding unnatural clipping: "Super transistor" blueprint?

[Steben]

Electro-fysical question mark. Take a look at this:



I was thinking about the collector-emitter cut-off. Once you choose your LED big enough (larger than the cut-off voltage) I think you can play with the behaviour at that side of the clipping. this way, the actual BJT clipping never comes in reach. Datasheets are very important I guess. At the other sweep side, you have the base-emitter cut-off, which compresses the signal already (as in rangemasters).

[db]

I'm not sure I'm getting this but wouldn't the LED in both cases be reverse biased and so have no obvious effect whatsoever?

[R.G.]

I think you're right, Dave.

[MartyMart]

So would they work when "forward" biased, just switch 'em around ?

MM

[DDD]

IMHO there will be some weird effect due to the "negative feedback" from collector to emitter through the LED. As far as I remember I've tried this circuit. The result was not good.
At the same time replace the LED with pair of simple diodes like 1N4148, and add a capacitor in series with it. 10-50 uF electro is to be a good choice. The resulting clipping arrangement produces interesting waveforms.

[Steben]

The point was, actually, that the LED doesn't conduct until the voltage between collector and emitter drops below the treshold.
If you reverse the diode, it will always conduct.

But I'm getting it, the emitter should have higher voltage than the collector, which is impossible.

[Sir H C]

What you will often see is a shottkey diode from the base to the collector (some transistors have this built-in) that will conduct before the base-collector diode gets too far into forward conduction. 

[Steben]

  In fact, since the base is V(e)+Vbe, you can create the "soft shortcut" from emitter to collector indeed through the base.
From Datasheet: (Vce(off)-Vbe)>Vf(diode). Brings us in modded Big Muff stories right (in series with cap)?

[Steben]

I guess this is more rewarding:



It should do what I wanted it to do.

[DDD]

Steben,
I think it's necessary to solder not one diode in series, but couple. Otherwise capacitor will take charge during some first cycles, the diode will be closed, e.t.c.

[Steben]

Steben,
I think it's necessary to solder not one diode in series, but couple. Otherwise capacitor will take charge during some first cycles, the diode will be closed, e.t.c.

That would lead us completely to Big Muff land.

[DDD]

In opposite to BMP the above circuit won't work even decently :-( :-)