Literature about "Bypass Capacitors" - FET and BJT

[KarenColumbo]

I'd like to have a good read concerning bypass capacitors - JFET source and BJT emitters. There's a lot of different variations - parallel to resistors, with a resistance in series etc. Haven't found a conclusive explanations with some "tricks and hints".

Could you gentlemen point me in the right direction?

[KarenColumbo]

If I don't look too closely, there's some variations that look like presence knob circuitry. But not where presence circuitry is known to exist.

[antonis]

I'm not able to draw schematic for the moment, Andreas, but there are some simple principles for your query..

1. Total Emitter (Source) bypass: Cap is completely shunting Emitter (Source) resistor..
    Advandage:    Maximun gain (R_C/r_e - AC grounded Emitter..) or (g_m X  R_D - AC grounded Source)
    Shortcomings: BJTs Low Input Impedance (h_FE X r_e) / FETs don't mind - Zero AC feedback

2a. Partial Emitter (Source) bypass: Cap in series with resistor shunting Emitter (Source) resistor..
     Advandages: "Some" AC feedback - BJTs Higher Input Impedance / FETS also don't mind..
     Shortcoming: Lower gain (Series resistor added on Emitter intrinsic resistor / FET 1/g_m for gain calculation..)

2b. Partial Emitter (Source) bypass: Cap shunting the higher value of the (two) Emitter (Source) splitted resistors..
     (e.g. a 1k Emitter resistor is splitted in 820R & 180R resistors in series & cap is set in series with 180R shunting 820R..)
     Considered equivalent with 2a..

Value of bypass cap should be calculated high enough so its impedance be considered negligible (less than 10%, say) of respective bypassed resistor value for the lower frequency of interest..
(if not, its impedance should be considered in parallel with bypassed resistor for actual AC Emitter (Source) resistance calculation..)

[iainpunk]

Eyyy

For BJT's i recommend big capacitance (1uf or moer) coulpled from ground to a trim pot's middle leg. The trim pot being the emitter resistance. Anust the trim pot to taste.

For a JFET i personally use a set of diodes and/or led's, this seems to give a very even frequency response and a good, high gain. I also use this with most valves, especially in my "modded to within an inch of its life" valvecaster, since triodes are just fancy, magical, better JFET's. 

Ill try to find some literature about this specific topic in one of my school books and post pic's if i can find this

[Rob Strand]

The different forms can be made to perform the same, so to some degree the choice is arbitrary.
This holds for any circuit:  emitter circuits, source circuits and treble boost and pre-emphasis/de-emphasis networks.

Updated:

[KarenColumbo]

Thank you gentlemen, it's becoming a bit clearer now.

[antonis]

Pending schematic post:
(sorry for the quality - exported from Tina-TI in low resolution form..)



P.S.
About literature, you can search for something like "Transistor CE Amp with Emitter degeneration" or so.. 

[KarenColumbo]

Wow, thanks for the schematics. Will be fun to try out (in a Spitfire V3, by the way).

[mac]

IMHO, the main difference between fig 3 and 5 is how fast you go from min to max gain (if Rd were a pot).

mac

[Rob Strand]

About literature, you can search for something like "Transistor CE Amp with Emitter degeneration" or so

Good stuff.  It's nice to have all the info in one place like that.

When I post stuff I often try to post web links to examples.  I can't tell you how many hours I've burned-up trying to find "common" configurations  because a large percentage of the info out there is incomplete, wrong or too academic.

[Damn, I noticed my pic had some bugs.  Now fixed.]

[antonis]

[Damn, I noticed my pic had some bugs.  Now fixed.]

I was ready to point it up...