[sdiy] simple transistor buffers?

[Mark Rivera]

Is there a particular book that has transistor information presented
like this.. practical rather than mathematical/theoretical?


On Fri, Nov 19, 2010 at 11:37 AM, David G. Dixon
<dixon at interchange.ubc.ca> wrote:
>> So i am working on a circuit where there are a bunch of outputs. to save
>> space/opamps I decided it would be nice to use transistor buffers (emitter
>> followers)
>
> If they are precision CV outputs, then opamp followers are really the only
> defensible option.  For audio circuits emanating from opamps (provided they
> are not coming from a circuit which has compromised the normally very low
> output impedance of the opamp, e.g. an ideal diode or rectifier, etc), the
> output is already buffered.
>
>> however when i switched to what i THOUGHT was a usable buffer, the output
>> gain (versus opamp followers) dropped drastically.
>>
>> let me see if i understand this properly. ( i dont think i do)
>>
>> if i want to buffer an lfo that is running between + and - 12 volts, I was
>> putting the oscillator wave through a 1k resistor into BASE. I was then
>> putting the collector to +v. What confuses me is what to do with the
>> emitter other than
>> this is where i take the output and a resistor to ground or -v should
>> connect here.
>>
>> I am not sure if i should put emitter to GROUND or to -v through say a 10k
>> resistor.
>
> The emitter follower output will only go as low as whatever is connected to
> the emitter resistor.  Hence, if you ground the emitter, then follower
> output will not go below 0V.  This is because the transistor (presuming it
> is NPN) can only source current, so the voltage at the emitter must always
> be higher than whatever is connected on the other side of the emitter
> resistor.  If you want the emitter follower to give output over the entire
> range between the rails, then you must connect the emitter resistor to the
> negative rail.  However, even then, the output will still ride 0.6V to 0.7V
> below the input, which means that you will get clipping distortion if your
> signal goes right to the negative rail.
>
> Plus, if you take the base very negative (more than about -6V) while the
> emitter is grounded, the breakdown voltage of the base-emitter diode is
> exceeded, which degrades the transistor.  If you use grounded-emitter
> followers (for positive-only signals), then you should probably put a diode
> between emitter and base (pointing toward the base) to protect the
> transistor from negative base swings.
>
> (Ain't Horowitz and Hill grand?)
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>