Triangle Big Muff circuit - swap NPN to PNP transistors?

[clarisso11]

Hi everyone,

Some background:

I'm starting a new project, retrofitting a V1 Triangle big muff circuit into an old EHX enclosure, using a Ram's head replica PCB. Here is the circuit I want to go with:



I managed to get my hands on 4 original FS37000 PNP transistors (PNP version of the NPN FS36999), pulled from a 1970s EHX pedal.

My question:

If I want to use those transistors while staying true to the circuit above, would it be enough to change the power supply polarity and double-check the transistor pin-out, or is it necessary to change the resistor values as well, to adjust biasing? There are no polarised capacitors in the circuit.

I found a Triangle circuit with PNP transistors and very different resistor values, but I'm not sure if they differ to account for the PNP design, or if they are just oddball values:



I did see some examples of Ram's head circuits with either PNP or NPN transistors, but very similar resistor values (on this page: http://www.bigmuffpage.com/Big_Muff_...ics_part2.html), which might suggest that the resistor values don't have to be adjusted?

[Helmholtz]

While the npn circuit should work with the pnp transistors, they might have adjusted component values in the pnp circuit to compensate for different hFE between the transistor types.

[sparkies]

Hi everyone,


I did see some examples of Ram's head circuits with either PNP or NPN transistors, but very similar resistor values (on this page: http://www.bigmuffpage.com/Big_Muff_...ics_part2.html), which might suggest that the resistor values don't have to be adjusted?

Nothing has to be adjusted, but I would do version 2 because of the small design errors in version 1.



Btw C9 is suppose to be a .0039 cap but the .004 will work the same. its R5 being correct in this schematic.

[clarisso11]

While the npn circuit should work with the pnp transistors, they might have adjusted component values in the pnp circuit to compensate for different hFE between the transistor types.

I see, thanks! I'm also getting two of the NPN FS36999, so I'm thinking I could test the hFE on those, compare them to the PNP types, and use the NPN circuit values if they are similar, if that makes sense?

[clarisso11]

Nothing has to be adjusted, but I would do version 2 because of the small design errors in version 1.



Btw C9 is suppose to be a .0039 cap but the .004 will work the same. its R5 being correct in this schematic.

Thanks, I didn't know about the design errors in V1, what are they? I wanted to go for a copy of the V1 Triangle circuit.

[sparkies]

Thanks, I didn't know about the design errors in V1, what are they? I wanted to go for a copy of the V1 Triangle circuit.

R5, the tone stack. Later on, they found that the PNP transistors had a lot of gain so they turned them down a little in the circuit. So if you have hotter pickups, you will get a lot of distortion effect with the older style. Not good or bad. Just different. But R5 is the correction for the tone control that can be applied independently.

[clarisso11]

Thanks for the explanation, that's good to know!

[Mark Hammer]

I wish I could remember the specific magazine, but some 15 or so years back, Mike Matthews gave an interview to a business magazine. In it, he noted that, during the '70s, if you took any four consecutive Big Muffs off the assembly line, they would all sound different from each other. Given that the resistors were all 5% tolerance, the caps generally 10-20% tolerance, the diodes likely had the usual wide range of forward voltage they still do, and the transistors all varied in hfe, I honestly don't see much point in trying to "nail" a particular issue. It's the proverbial "nailing jelly to a tree".

Certainly, play with the component values in the tonestack (something which IS different between issues and does make for a distinctive d9ifference), and maybe the feedback caps in series with the diodes. But quite honestly, beyond that one is just monkeying around in the dark. Yes, I know certain transistors are associated with certain issues and tones. But quite frankly, most such attributions are based on comparing ONE pedal from issue X against ONE pedal from issue Y, with no attempt to identify if there is something in common across MANY instances of issue X that is audibly different from what lies in common with MANY instances of issue Y. If Mike himself tells you that there will be considerable differences within a given issue, why beat your head against a wall trying to faithfully replicate something that doesn't really exist. Remember that the gain of each clipping stage is not inherent in the transistor, but rather in the components surrounding it that push it in this or that direction.

Finally, bear in mind that getting positive-ground pedals to play nicely with negative-ground pedals using a shared supply is a royal pain in the neck. Stick with an NPN build if you can. Less trouble.

[clarisso11]

Thanks Mark, these are all good points, and something to think about!

[Mark Hammer]

I'll just add that PNP/positive-ground pedals generally all stem from a period when they all relied on batteries, and not a single one had a jack for an external power source. Just TRY and find an external power jack on any germanium Fuzz Face, Rangemaster, or whatnot.

[clarisso11]

In case anyone is interested - I ended up using the V1 67#1 schematic, but with the PNP FS37000, and a small voltage inverter circuit on veroboard. Transistor gain 250-300. Sounds great!

I'll add that this was just for the fun of trying to replicate an original board as much as possible - I'm sure it wouldn't have sounded that different with some common NPN transistors, and would have made a lot more sense (no need for voltage inverter).