Buddy's Breadboard and Circuit Design Notes

[BuddytheReow]

Question for you all!!

I'm currently breadboarding a Distortus Maximus (will put together a Biggus Dickus shortly ). According to Chuck, this is is an LPB1 into a LM386 into a Marshall TMB stack, and I agree. I've only breadboarded the LBP1 and the LM386 and the circuit works. Instead of adding the TMB stack I put it directly into my BMP/LPB1 stripboard tool after the output pin and get this HIGH PITCHED WHINE on top of the signal and can alter the pitch of the whine with the pot. It almost sounds like RF waves. Why does this happen? Is this a specific breadboard issue or circuit design issue? Here is where I put the tone stack and marked it in BLUE.

Edit: Turning off my overhead fluorescent lights and turning off my phone doesn't change the issue, so I don't think it's an antenna issue? I also breadboarded a simple Bazz Fuss with the BMP stack after it and it worked just fine. It seems to be circuit design issues, but am not going to worry too much. Will try tacking on a baxandall instead of the Marshall TMB and report back with some other mods...

 

[BuddytheReow]

This is more of a "Look What I Built!" post than anything else...

I built the Distortus Maximus pedal sans TMB tone stack and liked the sound of it. The LM386 chip was not specifically designed for a distortion pedal, but it is used here. I added on a Baxandall tone stack after the 386 for tone tweaking.

Then I looked at the booster stage, which is an LPB1. IMO it's ok not great for a clean booster unless you have single coil pickups. So....I decided to swap that out and put in a Black Tiger Boost. I changed the pot from a 50k to 100k for more gain (OMG so metal ). In hindsight I could have also removed the 50k Gain pot on the Distorus Maximus since it's merely dumping the signal to ground before the LM386 stage, but I got a little lazy since it was already wired up and I can just keep it open. The Black Tiger Boost almost could have been removed from this circuit entirely, but it adds a bit more "creaminess" to the overall sound.

In summary:

IN ->Black Tiger Boost -> LM386 portion of Distortus Maximus -> Baxandall Tone Stack ->Hard Clipping Red LEDs (switchable on/off) -> OUT

As I'm writing this up this really isn't a single distortion pedal, it's a pedal train! Nearly each of the stages here can be boxed up individually minus the diode clipping at the end.

I call it El Tigre Distorsionado which means "the distorted tiger".

Edit: Yes, I know I have a whole other board I could have used and yes I know you can barely see the components due to the 2394832742 jumper cables.

 

[fig]

Man, that's a wild ride for sure! Nicely done!

 

[Chuck D. Bones]

Why do breadboards oscillate? In the simplest terms, it's because there is stray coupling in places where it doesn't belong.
I'm surprised when I build a high-gain circuit on a breadboard and it doesn't oscillate.

It's important to have solid grounds everywhere. None of this single-point ground nonsense, ground everything to everything else.

Power supply bypassing is critical. That 0.1uF cap should go directly from pin 4 to pin 6 of the LM386.

Keep inputs and outputs away from each other. Move the GAIN control away from the other controls and run its ground separately back to pin 2 of the LM386. Arrange your breadboard to flow like the schematic. Keep the wires short and separate them.

Do you have a schematic of the whole thing?

 

[BuddytheReow]

I'll write something up by hand tomorrow for the finished build. I haven't even attempted to learn any CAD software for schematics. Not yet at least...

That final build did have a ton of gain and I heard the slight humming, but not the squeal when I tried to tack on a BMP tone stack right after the 386.

 

[Chuck D. Bones]

Try TinyCAD. It's free and pretty easy to use. I can help.

 

[BuddytheReow]

For you viewing pleasure of El Tigre Distosionado. This is my first time using CAD software. Was pretty simple using TinyCAD. Thanks @Chuck D. Bones !

 

[Chuck D. Bones]

Not bad for the first time out. I can see what you're doing. I noticed a few typos:

1. You need a series resistor and cap between U3A's output and the LED clippers.
2. R13 & C13 should connect to the output of U3A, not the - input.
3. C10 and C11 should connect to pin 2 of the Bass pot.
4. You don't need R6 or R8 (replace with jumpers) or R9 (omit).
5. There should be a cap in series with R7.
6. You have the + and - inputs swapped on U1B.

You can drag the value and ref des for each part so they line up better next to that part.

I ended up editing several of the parts in the libraries to make them prettier. That's a nicety and something you can do when you get more comfortable using TinyCAD.

Good job!

 

[BuddytheReow]

Good catch, Chuck. Appreciate the compliments for a CAD noob. I have made the changes here. The capacitor in the negative feedback loop in U1b I'm still leaving blank since this is virtually a part for part of the Black Tiger Boost.

 

[BuddytheReow]

Piggybacking on the recent build, I took a look at the LM386 itself and wanted to know a bit more about it and why it's so unique. It's actually a power amplifier!! Want to hear why it's so special? Let's build it!

I took a look at the 'El Sol Distortion' which is based on the EDQ Acapulco Gold and realized it's merely a LM386 into another LM386 with some slight filtering. I wanted to know what a single LM386 sounds like so I stripped it down.

You may follow along in the schematic or my breadboard pics.


Pins 1 and 8 are jumpered together. THis is the gain control of the chip and we want it maxed out for right now
Pins 2 and 4 go to ground
Pin 3 is input
Pin 5 goes to an electrolytic cap. I chose a 4.7u. This is to filter out the DC in the circuit.
Pin 6 goes to 9v

And that's it!

Well, technically yes, but when I tried it there was a lot of "gating" when I strummed real hard. So, I decided to tone that down a bit.

Change the input to the following: Input to 10k resistor to pin 3. I also put a 100k resistor from the out of the 10k to ground.


Behold the raw power of the LM386! Honestly, it's $@^$(@ loud and gnarly! A very distinct sound indeed. If you want you can add hard clipping diodes after the 4.7u cap to ground and then to output. There is a subtle difference, but its there. Edit: Make sure you put a resistor before the diodes otherwise you may fry them. I suggest 10k.

The more you know

 

[benny_profane]

Does anyone know why most pedal applications of the LM386 merely short pins 1 and 8? The datasheet indicates that there should be a 10uF cap for gain = 200.

 

[Chuck D. Bones]

Going back to the v1.2 schematic, I think you have R3 & R4 switched.

If you get any scratching when you rotate the GAIN pot, then that cap I suggested will fix it. Make the cap 1uF or larger and it won't change the tone.

Comments on your last post, in no particular order...

If you put clamp diodes on the output of the 386 (after C1, of course), then you need a resistor in series with C1 to limit the diode current. The LM386 is a power amp and will deliver an overdose of current to the diodes.

You should really have a cap in series with the 10K. The 386 inputs are biased internally and your guitar was upsetting the bias. The 10K helped, but is not a good solution by itself.

 

[Chuck D. Bones]

Does anyone know why most pedal applications of the LM386 merely short pins 1 and 8? The datasheet indicates that there should be a 10uF cap for gain = 200.

The cap is preferred, but not required. The cap helps keep the 386's output centered for maximum headroom. Not a big deal with dirt pedals because we don't need that last little bit of headroom. Three are a few different ways to vary the gain. Check how I did it on the Biggus Dickus for an example of another way.

 

[BuddytheReow]

Going back to the v1.2 schematic, I think you have R3 & R4 switched.

If you get any scratching when you rotate the GAIN pot, then that cap I suggested will fix it. Make the cap 1uF or larger and it won't change the tone.

Comments on your last post, in no particular order...

If you put clamp diodes on the output of the 386 (after C1, of course), then you need a resistor in series with C1 to limit the diode current. The LM386 is a power amp and will deliver an overdose of current to the diodes.

You should really have a cap in series with the 10K. The 386 inputs are biased internally and your guitar was upsetting the bias. The 10K helped, but is not a good solution by itself.

To keep the raw signal, what value cap would you suggest to go with the 10k resistor?

I edited my previous post to include a resistor before the diodes.

 

[BuddytheReow]

Adding v1.3 of El Tigre Distorsionado with Chuck's r3/r4 comment.

 

[benny_profane]

The cap is preferred, but not required. The cap helps keep the 386's output centered for maximum headroom. Not a big deal with dirt pedals because we don't need that last little bit of headroom. Three are a few different ways to vary the gain. Check how I did it on the Biggus Dickus for an example of another way.

That all makes sense. I didn't know if that was just there for centering of if it also provides stabilization to prevent oscillation. The implementation in your BD is interesting. I'll have to dig into that a bit more to understand what's going on.

 

[Chuck D. Bones]

The datasheet includes a schematic. Unlike the usual opamps, the LM386 contains the negative feedback components inside. Pins 1 & 8 tap into that feedback path and allow us to tweak it.

 

[Chuck D. Bones]

To keep the raw signal, what value cap would you suggest to go with the 10k resistor?

I edited my previous post to include a resistor before the diodes.

I'd use a 220nF or larger film cap.

 

[benny_profane]

So, if I'm understanding correctly, what you're doing is adding an emitter resistor/cap to GND on the PNP darlington pair on the non-inverting input. I get that the topology is a long-tailed pair, but I think I'm confused because it seems that the +input has both the collector and emitter referenced to ground?

 

[Chuck D. Bones]

I gotta run, I'll explain it later if someone else doesn't.