PEAQ: A colorful JFET-based booster

[niektb]

Hey, haven't done a lot on this project lately but I'm back!

I've turned the name into Milano Drive (just because I had some green, white and red knobs floating around XD)
The Tight knob works fine now, I think the breadboard caused the issue and didn't make a proper connection or so...
I've drawn a schematic and PCB (and will order the latter probably at some point this week) and included some dip switches for some alternate clipping voices.





I'm still waiting for the PEAQ PCBs from Seeedstudio, they've been lying around for almost 4 weeks in a Asian Storage Depot  but now it's finally arrived in my country so I hope to have them here soonish

[niektb]

Whew PCBs of PEAQ are finally in! That means it's picture time! 

[niektb]

Okay, I'm having an issue with the mini-booster stage. The voltage at the gate of the upper JFET is little under 9V so the bias voltage is waaaay off. The voltage at the source of the jfet is around 8 volts. Then 1.1V is dropped across the lower JFET and a whopping 6.7V is dropped across the 470 resistor...
Anybody have a clue about what might be going wrong? The FETs were taken straight from the breadboard and were tested

[PRR]

> voltage at the gate of the upper JFET is little under 9V

This comes from a 1Meg+1Meg divider. A JFET Gate will not load that. The ideal voltage should be 4.5V. Observed voltage could be 4.0 to 3.3 depending on your meter.

Check it without the JFET Gate connected.

If OK without the JFET, check JFET pinout. There are standards and AFAIK about 9 of them. So the pins could be in any order. If you really have D or S where you should have G, the ~~1k of the S-D channel sure would pull the 1Meg+1Meg up to 8.99V.

[niektb]

> voltage at the gate of the upper JFET is little under 9V

This comes from a 1Meg+1Meg divider. A JFET Gate will not load that. The ideal voltage should be 4.5V. Observed voltage could be 4.0 to 3.3 depending on your meter.

Check it without the JFET Gate connected.

If OK without the JFET, check JFET pinout. There are standards and AFAIK about 9 of them. So the pins could be in any order. If you really have D or S where you should have G, the ~~1k of the S-D channel sure would pull the 1Meg+1Meg up to 8.99V.

You are a hero and totally right!! I used smd-to92 converters but the pinning was flipped! Will upload some sound demos soon!

[rankot]

Waiting for a demo

[niektb]

Alright, I connected it up but for some reason I have some serious white noise at the output (or at least very noticeable when running into a high-gain pedal),  much worse than f.e. my Tubescreamer gives. Anybody some pointers? The noise does not change in color when turning the Eq knobs, only the Gain knob makes the noise louder.
This makes me think that the noise bleed is prolly somewhere after the Eq but befóre the Gain. But there is nothing there lol 

[11-90-an]

have you tried lowering the value of C11?

[niektb]

have you tried lowering the value of C11?

How would that affect noise? That would affect base reponse only afaik

[rankot]

Alright, I connected it up but for some reason I have some serious white noise at the output (or at least very noticeable when running into a high-gain pedal),  much worse than f.e. my Tubescreamer gives. Anybody some pointers? The noise does not change in color when turning the Eq knobs, only the Gain knob makes the noise louder.
This makes me think that the noise bleed is prolly somewhere after the Eq but befóre the Gain. But there is nothing there lol 

Check all the connections, then check your pots. Maybe volume pot is bogus.

[niektb]

Finally came around to uploading a demo! My phone stops filming after a minute or two every time so you'll have to do without XD
YouTube compression really kills me, hope the sound comes across properly! 

[niektb]

Hey all! It's been a while... been busy with graduating, holidays and what not but I'm back now and trying to figure out the noise problem!
I think I might have a ground issue! The output jack was a little loose and the whole pedal power shuts down whenever the connection was poor! Is this normal? The input jack doesn't have any effect

The DC jack is isolated and the audio jacks are non-isolated

[rankot]

This sound great, and the demo is nice, but the image doesn't follow the sound

Nevermind this, but I have a question - why both C12 and C19 (both 220p). I assume they will works as a sort of hi freq ac divider, although C19 can be partially or totally bypassed with Gain pot. Was that the intention, or not?

[niektb]

This sound great, and the demo is nice, but the image doesn't follow the sound

Nevermind this, but I have a question - why both C12 and C19 (both 220p). I assume they will works as a sort of hi freq ac divider, although C19 can be partially or totally bypassed with Gain pot. Was that the intention, or not?

Haaa thanks! I had indeed some issues with recording so after retry after retry I gave up and took some a still photo (and a couple to show what I did on the Line 6 MultiFX)!  I should get myself a new phone

Also, be sure to look at the correct schematic  I have one floating around from the Milano Boost (or the old name: PEAQ) and one from the Milano Drive The demo is for the Milano Boost
Maybe I should the thread up and do the Milano Drive into a new one

C12 forms a Low Pass Filter with the 34k R12 to mitigate the Miller Effect (see the Fetzer Valve for a more detailed explanation). C19 is a bright cap that leaks treble when the Boost pot at less than maximum, which means that it gets relatively bright if you lower the Boost pot.
The Boost has DIP switches for selecting the bright cap, right now I've only installed the 220pF and I find it a tad mellow (even though it definitely 'vibes') with a 100k boost pot.

Here is what different C19 values do (with the boost pot set low):


Some more nerdy graphs 



I btw uploaded the schematic files for both here so you can play around with cap values and whatnot:
Milano Boost: https://drive.google.com/open?id=1kgNv0svf_v1GZUtZGU6D0a7E6Q4nVdm7
Milano Drive: https://drive.google.com/open?id=1rzBbBF8amf8zKJ-M581LPgnEz4ND56Ls

[Steben]

Cant the gate be biased with the already present 4.5v through a single resistor?

[niektb]

Cant the gate be biased with the already present 4.5v through a single resistor?

I *think* it might have to do something with the local feedback through C14, not sure though. In this regard, I followed the schematic from the Shaka Braddah III, which has a 4.5V bias point but biases the FETs separately  : https://www.diystompboxes.com/pedals/shaka3.jpg

[Steben]

http://www.geofex.com/Article_Folders/modmuamp/modmuamp.htm

Here RG suggests the same thing because of ...... (shoot me) .... "noiseless biasing" 

[niektb]

http://www.geofex.com/Article_Folders/modmuamp/modmuamp.htm

Here RG suggests the same thing because of ...... (shoot me) .... "noiseless biasing" 

Oooh that's an interesting article! Definitely something I'll explore, i see that I can improve the drive capabilities on the fly too! Many thanks! 

[niektb]

I noticed there was around 12k resistance between different parts of the ground plane. The DC Jack Wirepad was connected to the the right-most wirepad of the left audio jack, but not to the middle pad.

I have no explanation for this... Anyone? Feels like I got across a poorly manufactured PCB since I flooded the stuff with ground? But it definitely explains why the power would disconnect when the input jack wasn't connected to the enclosure

In other news, I altered the FET biasing, based on Stephens suggestion.

I feel the noise is less but I haven't measured it yet, even though it's definitely still too much. There is a difference now though, it does respond to the treble knob!