For The Love Of Gawd, What Is With This Boost Section

[Ginsly]

This is part of a larger project which was met with a resounding "dunno!" when I posted about it! So it goes. I figured I would isolate the part I'm most curious about, since I'm trying to make sense of what's going on here. Might help me going forward.

The chunk below seems to be a Ge boost added before Q1 of a two-transistor fuzz. I could be wrong, but that's what it looks like.


On the back there is also a resistor leading from collector to power:



As far as I can tell, this is how it's set up (and yes, I accidentally marked the electro's negative leg toward Ground+). The transistor is a Ge MP40A, which likely (?) has an hFE around 40, give or take.


I've looked at several boost sections/circuits, and this one is just different enough that I'm not sure what to make of it, and why it was set up this way. It doesn't help that I don't know the exact values of each component! The B-C resistor is clearly 470K, and the collector-to-power biasing resistor on the back seems to be 47K. Then there's a resistor from base to ground, which may be 220K. It's very hard to make out the resistor and electro leading from emitter to ground.

So... what kind of "boost" would this be considered? In trying to figure out the values of the emitter resistor and cap, I haven't really found a similar setup that would even give me a starting point. The LPB is an excellent, simple boost, but it differs from this in many ways - it has two resistors from the base which form a voltage divider, where this only has one to ground. The LPB has a relatively small emitter resistor to ground, where this has a resistor and an electro to ground (could be anywhere from 1-220µF).

The designer says that the electro does nothing to the tone, but is simply there to reduce the footswitch pop. But wouldn't it act like a bypass cap in the position it's placed? Without the electro, the emitter resistor would seem to noticeably reduce the gain, depending on the value (which I can't see in the pic). I'm not sure why a cap would be placed in this specific spot in order to deal with switch pop.

I can certainly provide more info about the bigger picture here, but thought I'd start with the front-of-circuit boost since that's the nut I'm tryin to crack at the moment. This all seems very silly, but I'm kinda hung up on it for reasons unknown.

Any help in understanding what's going on here would be MUCH appreciated! I'd love to land on some possible values for the emitter's resistor and cap. Hell, maybe I'm wrong about some of the other ones too!

 

[Ginsly]

This comes down to me being in the process of learning how to bias transistors in different situations. I haven’t seen this particular variation though.

Given the setup on my little post-it note above - what voltages am I shooting for on my “boost transistor”, and where do I start when choosing the Base/Ground resistor as well as the Emitter/Ground resistor and cap?

@Feral Feline and @PedalBuilder, you’re good at explaining this stuff!

 

[spi]

It's not that far from a electra distortion layout.

 

[giovanni]

the topology is pretty much a common emitter with feedback resistor (although the cap means that some frequencies have no feedback resistor) which is pretty standard. The base to ground resistor seems slightly unusual to me but also not that surprising.

 

[PedalBuilder]

the topology is pretty much a common emitter with feedback resistor (although the cap means that some frequencies have no feedback resistor) which is pretty standard. The base to ground resistor seems slightly unusual to me but also not that surprising.

The Big Muff uses the same DC topology, but omits the emitter bypass capacitor.

 

[Feral Feline]

...

@Feral Feline and @PedalBuilder, you’re good at explaining this stuff!

I'm flattered. However, I'm not able to add much to the proceedings.


I redrew the diagram as a schematic...

 

[Nostradoomus]

this reeks of tgp

 

[HamishR]

this reeks of tgp

Oh no - not at all! I haven't felt the need at all to vomit.

 

[Ginsly]

this reeks of tgp

Hmm that's funny, I'd say your judgmental comment has a lot more in common with TGP behavior than my simple inquiry. Boom roasted.

Oh no - not at all! I haven't felt the need at all to vomit.

Well thanks. Just tryin to learn here but ya know, snarkers gonna snark!

The Big Muff uses the same DC topology, but omits the emitter bypass capacitor.

Huh! Looks like it's actually quite close. Aside from the emitter resistor lowering the gain slightly without it, is there any real issue with omitting the bypass cap here? For that matter, as long as it still sounds good - any harm in removing both the emitter cap and resistor?

Looking at the Big Muff, the Base resistor seems to limit gain and increase input impedance - 100K is recommended for the BMP, and I was between that and 220K, which the boost pic seems to show. I was also going to add a 1M or 2M2 pulldown resistor at the input, which it seems may also affect input impedance. Is there a way to determine how those two components may affect impedance?

 

[cdwillis]

I don't know the math, but also your emitter to ground resistor will affect impedance as well.

 

[giovanni]

Hmm that's funny, I'd say your judgmental comment has a lot more in common with TGP behavior than my simple inquiry. Boom roasted.


Well thanks. Just tryin to learn here but ya know, snarkers gonna snark!


Huh! Looks like it's actually quite close. Aside from the emitter resistor lowering the gain slightly without it, is there any real issue with omitting the bypass cap here? For that matter, as long as it still sounds good - any harm in removing both the emitter cap and resistor?

Looking at the Big Muff, the Base resistor seems to limit gain and increase input impedance - 100K is recommended for the BMP, and I was between that and 220K, which the boost pic seems to show. I was also going to add a 1M or 2M2 pulldown resistor at the input, which it seems may also affect input impedance. Is there a way to determine how those two components may affect impedance?

A common emitter with emitter resistance has an input impedance roughly equal to beta*Re, where beta is the transistor’s HFE. This configuration also has a base resistance which is in parallel so you can do the math once you read the value of Re. The base resistance cannot increase the input impedance because any resistor in parallel with another will lead to a lower resistance overall. It’s not entirely clear to me why the base resistor is needed, you would have to do a simulation (or a lot more math) to figure that out. The configuration is self biased (thanks to Rcb) so it shouldn’t significantly affect the bias but again a simulation would be the best way to verify that. A large pulldown resistor (above 1M) will not significantly affect the input impedance either.

I must say, because of C2, the input impedance will be much lower at higher frequencies and probably dominated by the transistor’s rpi, which can be less than 1000ohm. The cutoff at which this happens depends on R2 and C2. Again, a simulation would be useful to plot the input impedance as a function of frequency.

 

[giovanni]

Hmm that's funny, I'd say your judgmental comment has a lot more in common with TGP behavior than my simple inquiry. Boom roasted.


Well thanks. Just tryin to learn here but ya know, snarkers gonna snark!


Huh! Looks like it's actually quite close. Aside from the emitter resistor lowering the gain slightly without it, is there any real issue with omitting the bypass cap here? For that matter, as long as it still sounds good - any harm in removing both the emitter cap and resistor?

Looking at the Big Muff, the Base resistor seems to limit gain and increase input impedance - 100K is recommended for the BMP, and I was between that and 220K, which the boost pic seems to show. I was also going to add a 1M or 2M2 pulldown resistor at the input, which it seems may also affect input impedance. Is there a way to determine how those two components may affect impedance?

Oh one more thing: why would you want to omit the emitter resistor and capacitor? That will give you a stage with a radically different behavior. It would probably have to be rebiased and the gain will be much smaller (especially for Ge). What do you mean when you say bypass capacitor? If you are referring to the caps connecting the stage to input and output, you definitely can’t remove those or you will end up with DC in your signal path which is bad.

 

[Ginsly]

A common emitter with emitter resistance has an input impedance roughly equal to beta*Re, where beta is the transistor’s HFE. This configuration also has a base resistance which is in parallel so you can do the math once you read the value of Re. The base resistance cannot increase the input impedance because any resistor in parallel with another will lead to a lower resistance overall. It’s not entirely clear to me why the base resistor is needed, you would have to do a simulation (or a lot more math) to figure that out. The configuration is self biased (thanks to Rcb) so it shouldn’t significantly affect the bias but again a simulation would be the best way to verify that. A large pulldown resistor (above 1M) will not significantly affect the input impedance either.

I must say, because of C2, the input impedance will be much lower at higher frequencies and probably dominated by the transistor’s rpi, which can be less than 1000ohm. The cutoff at which this happens depends on R2 and C2. Again, a simulation would be useful to plot the input impedance as a function of frequency.

Thanks Giovanni! I’m planning on a 1M pulldown resistor, but it sounds like 2M2 would have less chance of affecting input impedance.

When it comes to the base resistor and emitter resistor & cap, I kinda took a guess and landed on R1 @ 220K, R4 @ 470ohm, and C2 @ 100uf. R1 looks to possibly be that value in the pic, R4 seems to be a lower (120-470ohm) value in similar setups like the Electra/Big Muff Q1, and C2 I’m completely guessing based on the size of the cap in the pic! Ha… I could be way off- a Rangemaster boost has the emitter-ground resistor & cap as shown here, and they’re 3K9 and 47uf.

I know, this is all ridiculous. I’ve never used a simulator so I’m not sure how that works- sounds like a good idea in this case, though.

Oh one more thing: why would you want to omit the emitter resistor and capacitor? That will give you a stage with a radically different behavior. It would probably have to be rebiased and the gain will be much smaller (especially for Ge). What do you mean when you say bypass capacitor? If you are referring to the caps connecting the stage to input and output, you definitely can’t remove those or you will end up with DC in your signal path which is bad.

I’m a dumdum but I do know better than to remove the input and output caps at least! Ha…

I was under the impression that a cap in parallel to a resistor from emitter to ground was a “bypass cap”- C2 in @Feral Feline’s handy redraw. In breadboarding, R4 by itself seems to lower gain, depending on value- then when adding C2 it jumps back up. Omitting both and simply connecting emitter to ground sounded roughly the same as having both in place, so I figured- do I need either one?

Clearly I’m still in the early stages of figuring out some really basic stuff! I appreciate your patience, this input helps.

 

[giovanni]

Thanks Giovanni! I’m planning on a 1M pulldown resistor, but it sounds like 2M2 would have less chance of affecting input impedance.

When it comes to the base resistor and emitter resistor & cap, I kinda took a guess and landed on R1 @ 220K, R4 @ 470ohm, and C2 @ 100uf. R1 looks to possibly be that value in the pic, R4 seems to be a lower (120-470ohm) value in similar setups like the Electra/Big Muff Q1, and C2 I’m completely guessing based on the size of the cap in the pic! Ha… I could be way off- a Rangemaster boost has the emitter-ground resistor & cap as shown here, and they’re 3K9 and 47uf.

I know, this is all ridiculous. I’ve never used a simulator so I’m not sure how that works- sounds like a good idea in this case, though.


I’m a dumdum but I do know better than to remove the input and output caps at least! Ha…

I was under the impression that a cap in parallel to a resistor from emitter to ground was a “bypass cap”- C2 in @Feral Feline’s handy redraw. In breadboarding, R4 by itself seems to lower gain, depending on value- then when adding C2 it jumps back up. Omitting both and simply connecting emitter to ground sounded roughly the same as having both in place, so I figured- do I need either one?

Clearly I’m still in the early stages of figuring out some really basic stuff! I appreciate your patience, this input helps.

Oh yeah that’s what I was saying regarding the gain being different at higher frequencies because of that capacitor. The capacitor turns the circuit into a common emitter without emitter resistor at higher frequencies (note that the cutoff could be close to double digit Hz which means that all the frequencies would see a different gain). With emitter degeneration, the gain is Rc/Re; without it, it’s just gm*Rc. gm depends on the transistor (which is why Re is useful: you can get a gain that does not depend on the transistor).

 

[giovanni]

One more thing I should add: Re is also important for the DC biasing of the transistor. So if you remove it, the stage may work but it may absorb more current or have a poor operating point (which would cause clipping). These are all things you could calculate by hand (check out Wikipedia for the various characteristic values for the common emitter) or figure out in simulation.

 

[Ginsly]

Oh yeah that’s what I was saying regarding the gain being different at higher frequencies because of that capacitor. The capacitor turns the circuit into a common emitter without emitter resistor at higher frequencies (note that the cutoff could be close to double digit Hz which means that all the frequencies would see a different gain). With emitter degeneration, the gain is Rc/Re; without it, it’s just gm*Rc. gm depends on the transistor (which is why Re is useful: you can get a gain that does not depend on the transistor).

This is a bit over my head, but it's starting to make sense. I'm wondering if the values I've chosen for the emitter resistor/cap could be way off..? Not sure what it would do to the sound, but the gain being different at different frequencies doesn't sound like a good thing... 100uf seems a bit high, and the resistor might need tweaking too.

One more thing I should add: Re is also important for the DC biasing of the transistor. So if you remove it, the stage may work but it may absorb more current or have a poor operating point (which would cause clipping). These are all things you could calculate by hand (check out Wikipedia for the various characteristic values for the common emitter) or figure out in simulation.

Interesting - why is it that most emitters are connected straight to ground, then? The other two transistors in this Fuzzrite go directly to ground. Is it that the B-C resistors in those gain stages bias them enough already?

 

[GAStan]

The configuration is self biased (thanks to Rcb) so it shouldn’t significantly affect the bias but again a simulation would be the best way to verify that. A large pulldown resistor (above 1M) will not significantly affect the input impedance either.

What I'm seeing, and this is a question as much as statement because I'm a bit rusty at this, is Rcb (R2) without R1 (Rb) will pull the Base high turning the transistor to FULL ON because it does not have a DC isolating capacitor that would make it a pure feedback circuit. Rb (R1) forms a voltage divider along with Rcb (R2) and Rc (R3) to set bias. IMHO best thing to do would be replace Re with a trimmer and using a SigGen and O'scope set bias for the type of clipping (asymmetric vs symmetric) desired. Alternatively a live guitar by ear could be used to set bias if test equipment isn't readily available.

 

[mybud]

Interesting - why is it that most emitters are connected straight to ground, then? The other two transistors in this Fuzzrite go directly to ground. Is it that the B-C resistors in those gain stages bias them enough already?

The bits I understand about common emitter amps seem to imply that your reading of the Fuzzrite bias is correct.

I am more than glad to be corrected by those in the ‘actual’ know.

‘The fine print giveth and the fine print taketh away.’ Tom Waits

 

[giovanni]

This is a bit over my head, but it's starting to make sense. I'm wondering if the values I've chosen for the emitter resistor/cap could be way off..? Not sure what it would do to the sound, but the gain being different at different frequencies doesn't sound like a good thing... 100uf seems a bit high, and the resistor might need tweaking too.


Interesting - why is it that most emitters are connected straight to ground, then? The other two transistors in this Fuzzrite go directly to ground. Is it that the B-C resistors in those gain stages bias them enough already?

You can definitely correctly bias a common emitter without Re. My point is that you can’t just remove Re from an existing circuit and expect it’ll work just as well. A common emitter without Re is used for specific reasons. In a fuzz circuit we know that we want to let the quirks of the transistors take over the design. So omitting Re will give you the “features” that people expect from fuzzes: interaction with the guitar volume (because of low input impedance), temperature effects (because the operating point depends on temperature more without Re), different sounds with different transistors (because Re helps removing differences between transistors). From a circuit design perspective, you could consider a fuzz as a “broken” circuit, but those quirks are what make these circuits special. As to how we got here to begin with, my guess is a combination of poor design skills, fortuitous accidents, the need for lowest part count for price cutting and… tone.

 

[giovanni]

One more thing: it’s pretty common to have an emitter cap in order to have a different gain value for high frequencies (which again, could be and often is all frequencies!). Usually this is done to have a specific biasing scheme that doesn’t fit with the type of gain you want for your signal.