Big Muff Pi - Low Pass Question

[dwinters]

I'm relatively new to this (both pedals and electronics) however I have built a few smaller circuits and have been learning passively for about 12 months. I've been using various sources but the one I first stumbled upon I have found very useful and a great help - which is Dano's Beavis website. Anyway Dano has a schematic on his site for a Big Muff variant and I think it's the Violet Ram? Emboldened by his excellent website, I'd thought I'd give it a go.

Dano's schematic: http://beavisaudio.com/schematics/Electro-Harmonix-Big-Muff-Pi-Schematic.htm
GGG's schematic: https://generalguitargadgets.com/pdf/ggg_bmp_vram_sc.pdf

I breadboarded it using Dano's schematic and part values and although greatly relieved to hear it working. The low pass on the tone pot wasn't right, at between 47kΩ - 20kΩ between the middle Lug (Lug 2) and right Lug (Lug 3) the volume just drops dramatically. Turned all the way at 0Ω it comes back a little but is still quiet. At the other end - high pass - it sounds fine. First thing was to try a different pot, same result. I then read more about the circuit and where the drop may arise from but having tried a few different cap and resistor values around the tone stage, no luck there. I have made a few substitutions (I know...) but I don't see how they could cause these volume drop? (I'll list them below*)

Finally I noticed that most variants of the BM use a coupling cap between the Tone and Output stage. So I put in a 100nf cap and the tone pot suddenly works, full range seems to have returned. However I am worried it's papering over a crack in the circuit and that really what I am hearing is not the full tonal range? I don't have a scope so can't check but whether I'd be able to interpret it is another question - my knowledge is limited.

I'd like to understand why a coupling cap might have rescued the tone and why Dano may have been able to omit it in the variant he has used? Generally most analysis of the BM don't really mention too much about this coupling cap so I'm not able to find out more on it's powers!

I'll list the details of the readings I took as well as those subs as it could be they play a role in this I hadn't understood. I'll use Dano's circuit when referencing the part numbers

*Subs:
C2, C5, C9 - 470pf Caps swapped for 1nf
C10 3.3nf swapped for 4.8nf

Readings:

Q1
Q1 C	4v
Q1 B	0.6v
Q1 E	32mv
Q2
Q2 C	4v
Q2 B	0.6v
Q2 E	33mv
Q3
Q3 C	4v
Q3 B	0.6v
Q3 E	31mv
Q4
Q4 C	1.8v
Q4 B	2.1v
Q4 E	1.7v
D1
D1 Anode	518mv

D1 Cath

0.63v

D2
D2 Anode	0.63v

D2 Cath

522mv

I also measured the voltages at the Coupling Cap I added, as well as when it's absent - the wire between it (VR2 > Q4). I'm not sure if these are helpful...

With Coupling Cap: High pass -
Cap Anode: 1.1v
Cathode: 1.9v

With Coupling Cap: Low pass -
Cap Anode: 3.5v
Cathode: 1.9v

No Cap, turned towards High Pass: 1.3v
No Cap, turned towards Low Pass: 1.3v


Any insight would be greatly appreciated, regarding why a tone would 'drop out' and even if it's just to confirm the variant of the BM this is most similar to as I'm not sure? After this my next job is to try to convert it into a layout using Greeny's build guide.

[antonis]

Hi & Welcome...

I'd like to understand why a coupling cap might have rescued the tone and why Dano may have been able to omit it in the variant he has used?

Simply because it settled down DC difference between Q3 Collector and Q4 Base..

P.S.
Better fix should be another cap placement right after Q3 Collector to obtain a crackling-free Tone pot..
(as it is, there is DC flowing from Q3 Collector -> R19 -> Tone pot -> R18 to GND..)

edit: If Q4 voltage measurements were taken AFTER 100nF placement, you have a issue there..
If not, take new measurements with 100nF cap..
You should get something like below (give or take):
Q4 C : 4V
Q4 B : 1.8V
Q4 E : 1.1V

[dwinters]

That's great thanks Antonis, I'll give that a try tomorrow and report back!

[garcho]

The omission is probably a mistake, it happens all the time. The coupling cap stops preceding stage DC from affecting the transistor bias, which is key to a transistor acting as an amp instead of a switch. It's not a glamorous "mod cap", so no glamorous misinformation on the internet
Also, forget about the oscilloscope, it's a distortion pedal! If it sounds good, it is good.

[antonis]

The coupling cap stops preceding stage DC from affecting the transistor bias,

@ dwinters: By KVL implementation, you can find Q4 Base DC voltage..
(and verify that it's different from the (9XR21) / (R20 + R21) which should be in case of coupling cap between Tone pot wiper and Q2 Base existence..)

[Mark Hammer]

Always a good idea to follow the signal path through a circuit, and confirm that any unwanted DC is removed by a suitably placed cap.  In the case of the linked-to schematic, DC that might reach Q4's base IS blocked by C10....for that path.  But DC from the Q3 stage can still sneak through via R19.

There are several potential loci for inserting a "master DC blocking cap" that will all work by acting on common paths.  That could be at the "entrance" to the tonestack, or its exit.  Alternatively, a suitable cap in series with R19 would block DC along that route, and potentially serve to shape low end in a desired fashion.

[dwinters]

Thank you all for your replies, I added a 100nf cap between Q3's collector and R19 and that sounds good. I settled on using a 100nf after testing some differing Cap values.

Thanks @antonis Yes I need to revisit Kirchhoff's Law but the I have sometimes struggled translating theory to practical however seeing it applied to a circuit I'm working on helps.

Also @Mark yes, following the signal seems like a good idea for debugging.

Now it's onto the Veroboard!

[antonis]

Now it's onto the Veroboard!

And know is the crucial time for decisive mods, like Tonestack LPF/HPF items values, feedback diode pairs type and series caps values, gain resistors (R8 & R13) values, transistors (low, medium or high h_FE or any combo) etc..

P.S.
The vast majority of forum members should suggest for a single mod per time..