How do you get low HFE transistors??

[Paul Perry (Frostwave)]

HINT : millions of 2N3055s have been made since the 1970s.  They have always been cheap and really useful.  What is their "typical" hFE?  Check out the Ic vs hFE graph on the datasheet.  What is the hFE at the Ic of a typical fuzz circuit?  What kind of device is it?  Is the modern, small-package version (MJE3055) much different?  Would it work, too?

The 2N3055 might be an unfortunate example.
Back when it first came out, it had a low gain, certainly.
I believe the modern plastic encapsulated ones are made by a completely different process & have a much higher gain.
But, in any case, it's always worth checking.
A good source of low gain transistors are the early home organs.

[brett]

+1 on the BD139.  Mine are just right for Q2 in a hot fuzz (ie hFE = 150).

[zombiwoof]

Mouser has BC 107's, but they're a buck a piece!

Al

[Dragonfly]

I guess I'll just have to mention this a few thousand more times....




2N3903







http://www.onsemi.com/pub_link/Collateral/2N3903-D.PDF


Listed hfe is 50 to 150 ...the ones I get are typically 80-120.

Oh...and theyre cheap. A whopping $.07 each ... in quantities of 10 or more, $.06 ...and in quantities of 100, $.05 ...quantities of 250, $.04

....spend $10 and never need to buy another low gain silicon. Easy to design around too.

MOUSER CATALOG LISTING FOR 2N3903

[antojado]

I guess I'll just have to mention this a few thousand more times....




2N3903







http://www.onsemi.com/pub_link/Collateral/2N3903-D.PDF


Listed hfe is 50 to 150 ...the ones I get are typically 80-120.

Oh...and theyre cheap. A whopping $.07 each ... in quantities of 10 or more, $.06 ...and in quantities of 100, $.05 ...quantities of 250, $.04

....spend $10 and never need to buy another low gain silicon. Easy to design around too.

MOUSER CATALOG LISTING FOR 2N3903

Maybe only a few hundred times 

The table in the data sheet showing the relationship of the current and voltage to the gain is interesting. Here's some noobie questions about it:
1. The current is the current at the base, correct?
2. If I read the table correctly, then it's telling me the gain is greatest at about 10mA and as the current goes down the gain is reduced. So the gain of a transistor is controlled by the input current?
3. So, what is the typical current of a "typical" fuzz face circuit?

[brett]

Hi

1. The current is the current at the base, correct?
2. If I read the table correctly, then it's telling me the gain is greatest at about 10mA and as the current goes down the gain is reduced. So the gain of a transistor is controlled by the input current?
3. So, what is the typical current of a "typical" fuzz face circuit?

1. The current is the current through the collector, hence Ic.  It is virtually the same as current through the emitter (Ic is about 99% of Ie because Ie = Ic + Ib).  If you can work out the collector voltage, and know the collector resistance, you can calculate Ic.  If you don't know the collector voltage, use half of the supply, which is often about right.

2. Yes, as Ic goes down below the optimum,  hFE falls.  The gain is controlled by the collector current.  This probably explains the very large collector resistor on Q1 in the fuzzface (33k) - it makes Ic very low.  Usually, a larger collector resistor increases gain, but in this case it is probably reducing it (I say "probably" because I've never done the measurements to confirm this).

3.  It's about 6 or 7V/33k = 0.2mA or so.  With this Ic, most transistors are operating the rising slope of hFE vs Ic.  This *might* be part of the magic of the fuzzface.   

Here's one way that the "magic" might work.  As the base voltage falls and the collector voltage approaches the supply, Ic goes to almost zero.  If the transistor has just the right fall in hFE as the Ic falls, it might "soften" the clipping of the signal, because the transistor "runs out" of hFE before "crashing" into the supply voltage.

cheers

[mac]

Here's one way that the "magic" might work.  As the base voltage falls and the collector voltage approaches the supply, Ic goes to almost zero.  If the transistor has just the right fall in hFE as the Ic falls, it might "soften" the clipping of the signal, because the transistor "runs out" of hFE before "crashing" into the supply voltage.

I'll keep this in mind.

I guess I'll just have to mention this a few thousand more times....
2N3903

Sorry, I didn't hear you... again please...

Do you have a sound clip of a FF using 3903?

mac