Trying different transistors in a circuit

[kapsel]

Hi all,

I built a Harmonic Percolator using the "George Giblet schematic" (don't know if its ok to post it, i guess you know which one i am talking about). I socketed the transistors, used 2 silicon ones for now while I wait for some germanium. The question came to me while building this, but it is more general:

Apart from transistor type (NPN, PNP) and pinout, can i just try out different transistors or should i also change resistor values on the circuit?
I mean, some resistors, caps etc in a circuit are there to regulate the "proper" amount of voltage that a transistor needs to operate (right?).
When using a different transistor, should the rest of the circuit change as well?

Sorry if I am not expressing this well, any help or guidance is appreciated.
G.

[antonis]

Apart from transistor type (NPN, PNP) and pinout, can i just try out different transistors or should i also change resistor values on the circuit?

Apart from type and pinout the main "difference" is beta (β or hFE)
(if we don't change "category", like fast switching/general purpose/amp/power)

If we have a circuit that ensures operation independable from their individual gain (like biasing from a stiff voltage divider, compesation for different Vbe drop, e.t.c...) we don't need to change anything..

At least, at a first glance..

[tonyharker]

Also Germanium Transistors need different biasing to Silicon ones, due to the different Vbe - approx 0.2v against 0.7 for silicon. So the base resistors may need changing.

[rutabaga bob]

Do a search for the Photonic Juergulator, as regards this circuit...

[kapsel]

I see, thanks a lot for the replies!

So, in theory, if a circuit is designed for a medium gain (100-200) hfe transistor, can I use a 50 hfe and adapt it somehow or is it doomed to sound very different  from the original, even if it works?

I used some low leakage, ~50 hfe Soviet ge transistors instead of the 2n404 and while it sounds ok, it doesn't sound so extreme as the various clone and original demos I have seen online. Could I do something to make it heavier/fuzzier with these transistors?

Thanks again!

[antonis]

So, in theory, if a circuit is designed for a medium gain (100-200) hfe transistor, can I use a 50 hfe and adapt it somehow or is it doomed to sound very different  from the original, even if it works?

Using lower than suggested hfe trannys MAY lead to some circuit "upsettings"..

You'll lower transistor's input impedance (hfe X Remitter or hfe X re in case of grounded emitter..)
This may be significant, depending on the total imput impedance, leading to heavier input loading..

If bias circuit is more or less dependable on hfe (which shouldn't be, but..) you may alter Q point..
(you'll raise Vc in case of an amplifier or lower Ve in case of emitter follower leading in assymetrical clipping..)

[LightSoundGeometry]

my experience is has to be in range of original design or it will not work right. if you have to bias out a bjt then you might as well breadboard and work the circuit out from the ground up; and in my opinion this is the way to go, nothing really touches a tonebender or a muff anyways in my humble opinion.

[PRR]

> if a circuit is designed for a medium gain (100-200) hfe transistor, can I use a 50 hfe

Historically, hFE cost money.

If the original designer calculated that he "had" to spend the money for higher hFE, why do you think you can do better?

Or if my Honda sedan was designed around a 134 cubic inch engine, why would I try to run it with a 50 CI engine? Some fast math suggests a total redesign with 29 very teeny cylinders at very high RPM makes the same "Power", but needs major re-gearing, and may have shorter life and lower economy. (Not to mention that even 27-cylinder engines are prone to shake themselves to bits; anybody remember the Constellation, world's finest tri-motor?)

hFE is not "everything". You can work around low-low-hFE parts. I recall hFE nearer 20. Use more and expect less (or many more to meet a good spec.) High hFE leaves more wiggle-room. Though many historic designs go nuts with high hFE just like low hFE.

[LightSoundGeometry]

> if a circuit is designed for a medium gain (100-200) hfe transistor, can I use a 50 hfe

Historically, hFE cost money.

If the original designer calculated that he "had" to spend the money for higher hFE, why do you think you can do better?

Or if my Honda sedan was designed around a 134 cubic inch engine, why would I try to run it with a 50 CI engine? Some fast math suggests a total redesign with 29 very teeny cylinders at very high RPM makes the same "Power", but needs major re-gearing, and may have shorter life and lower economy. (Not to mention that even 27-cylinder engines are prone to shake themselves to bits; anybody remember the Constellation, world's finest tri-motor?)

hFE is not "everything". You can work around low-low-hFE parts. I recall hFE nearer 20. Use more and expect less (or many more to meet a good spec.) High hFE leaves more wiggle-room. Though many historic designs go nuts with high hFE just like low hFE.

turn them into Darlington Pairs! one of my best rangemaster builds was a low Hfe Darlington pair following Small Bears super easy walk through . The transistors were pulled from an old silvertone MIJ 10 transistor am/fm radio...one of the pair had an Hfe around 20 lol.