blackfire question

[Ansil]

would it feaseable to make a blackfire out of mpsa18's???  i was wondering if it would self oscialte.

[WGTP]

I remember when it first came out someone used MPSA018, and indicated it worked.

[rx5]

hey man,

gonna do my blackfire on BC549C (low noise,highgain <500>) ....cause Im having a hard time getting 2N5088/89 and MPSA18..... currently have 2 549's..still  waiting for the other 3....



     

[brett]

MPSA18 = 2N5089, so there's absolutely no problem with using them.  (Anybody know why 2 different labels for what appears to be the same transistor?)
BC549Cs are a little noisier (NF=4dB) than the above, but far better than 2N3904s etc (NF=10dB and up).

[petemoore]

Answer all these questions, with extremely elaborate, instantly recognizable detail.
 If you we're to choose just one transistor to try, the MPSA18 would be a good choice, very like the 2n5089's in gain character I think [I surmize from my experiments with BF and different transistors], tone comparisons may vary but are probably  percieved as not huge.
 Anything lower gain like 5088's doesn't work the same.

[Paul Marossy]

"MPSA18 = 2N5089, so there's absolutely no problem with using them. (Anybody know why 2 different labels for what appears to be the same transistor?)"

It's probably two different manufacturers, one is probably in response to the other's product.

[rocket]

I have 2 comments about the design:

I cannot quite follow the describtion.
IMO the first transistor acts as a voltage source rather than current source, even if the output impedance might be rather high (~2kOhm) .
I can' see how an output impedance of 2+10 k should limit the base current of a transistor with an Hfe of 500+ .

I would also susggest to use "noiseless biasing" like it's done in most commercial circuits. That would save 1 transistor an cost 1 capacitor.
The circuit might get a little more complicated.
But I think it's worth to avoid the thermal noise of the biasing resistors, especially in a high gain circuit.

[rocket]

I don't want  to critisize the design - just discuss.
I am really interested how it works.

[WGTP]

The 2N5089 and MPSA 018 are slightly different, with the 018 having higher gain.  Check the Fairchild comparison chart for hFE of 500+.

Some of the the 2N5088's I've measured are above that.

[Joe Davisson]

My hypothesis:
Q1 is configured for high-impedance. The circuit works with no 100k emitter resistor on Q1, but adding it raises the gain a touch. Lower it to 10k (like a "typical" buffer), and it overloads Q2 big time. So there is a balance factor between the amount of gain and harshness. The 100k resistor achieves the balance.

The voltage on Q1's emitter will always follow the input, with the emitter current remaining constant. Transistors will always draw current from the input, and in this case Q2 can only get as much as Q1 will provide. This lets you control Q2's operating speed with Q1's emitter resistor.

A single transistor gain stage has low impedance. With one stage (like a Rangemaster) there isn't enough current from the guitar pickups to overload it. But in a multiple gain-stage configuration there is a low->low impedance match (not a exact match, but enough) which causes a high amount of current to transfer between stages along with the voltage. This slams the outputs of the transistors to the power supply rails, creating square-waves.

So the Blackfire works by repairing the transistor's impedance problems, and slowing down the transistor's response to input voltages. And the impedance between stages is kept low->high, preventing one stage from loading down the other.

Hopefully this makes more sense. If some engineers can help clarify anything please jump in. I have abolutely no test equipment besides a DMM so it's very hard for me to verify anything. This is the latest schematic:
http://www.joefus.com/pedals/images/blackfire.gif

-Joe

[rocket]

I see.

btw - the noiseless biasing would of course save 1 resistor no transistor.

[Jered]

Hi Rocket, can you give a link to a schematic of the noiseless biasing your refering to?  
 Thanks,  Jered