My first transistor overdrive / distortion

[Banjan73]

Hi!
This is my first topic on this forum. I have been a reader for a few years, but this time I thought that I would participate with my newly built overdrive / distortion pedal.
I have been drodling around with tube electronics for quite some time, but this time I decided to try to understand a little more regarding transistors. I read myself up a little bit on the net about transistor amplifying theory, searched this forum a little bit and decided to built a common emitter "thing". I don`t really remember where I got the info to use 2n3904 transistors, but it may have been on this forum.
Anyway, I started with one transistor on my breadboard, but I didn`t get enough well-sounding distortion with that, so I decided to use two in a cascade coupling. Then I did a lot of experiments with the emitter and collector transistor values, to use one or two potmeters, and experimenting with the capacitor values (I did that mostly with math (frequenzy formula)).
After a while, I got a sound that I was pleased with, but it had a lot of noise. Well, I have built some OP-amp circuits before (e.g. ECG-instrument), and I knew that the noise would probably disappear when I did a proper installation in a metal box (not on the breadboard with all the antennas that it has).
Then it was time to design a PCB. I went for this, and not point to point soldering to minimize noise.
I installed KiCAD on my computer and created a proper schematic, and layout for the PCB. The PCB was ordered from PCBway in china (had to order minimum 5 PCBs, so then I have 4 leftovers).
During the installation, I used shielded cables on the input. The shield was layed to ground (minus pole) together with the metal chassis. I used the true bypass coupling from stinkfoot.
The pedal hasn`t been tested properly on my Marshall yet, but it will come on this tread soon. Meanwhile, I am posting some pictures of the working process.

[Banjan73]

BTW. This pedal seems almost noiseless now on my small appartment amplifier

[duck_arse]

hello Banjan73, welcome to the forum. you might contact a moderator and ask to move your thread to the lounge, or pictures, or building your own sub-boards, where many millions more people will get to see it.

first point, I think your output volume control is wired -not right-. it is more usual to have the hot signal to the top of the pot, with output from the wiper.

second point - fantastic, all one watters! only, you could maybe allow them a little more copper in your trace widths, y'know, for all that current.

third point - it's disappointing to see that metalclad resistor lying on the bench, and not built into the box. is that for your next build, with all 50W resistors?

[antonis]

Hi & Welcome also ..

Apart from Stephen's notice for "weird" Vol pot wiring, allow me to add some more..

It's not wise to bias BJTs like 2N3904 (medium gain) with 1M or so resistors..
Also, the 100/1100 divider ratio results into about 800mV (for 9V PS) on Base hence about 150mV on Emitter..
Q1 will severely misbiased at low or top end of Gain pot and will clip asymmetrically (not necessarily a bad thing..)
(try, if you like, 470k for R2, 2k for Gain pot, 180R - 220R for R5, ground Gain pot lug 3 and place a 22μF cap between pot middle lug (wiper) and GND..)

Also Q2, as it is, looks like it's Collector lie at about 8.5V or so.. (another cause for asymmetrical clipping..)
(make R6 470k, R9 1k and R8 4k7)

Then we'll talk about C1 value and those #$&% nasty things called impendances....

P.S.
2 asymmetrical clipping stages of opposite headrooms could result into overal symmetrical clipping..

[Banjan73]

Hi!
Nice to have some feedback on this one.
As I said, this is my first transistor project. I tried to read up regarding the biasing, but i didn`t catch the theory. So, the 1M / 100k I stole from somewhere (don`t remember where, hehe). This project was mostly about learning a little bit.
I will try those settings on my breadboard, though I dont think I have any 2k pot.
Regarding the copper traces, I measured, early in the project, about 0.9mA current draw on the whole circuit (without LED), so I dont think the thin copper trace is a problem here. But anyway, i still agree that they could perhaps have been a little thicker;-)
About the volume pot: I am a little bit novice, actually, when it comes to the function of a potmeter. Have to read a little bit there. But the point is to "bleed of" the signal to ground gradually, isn`t it? I hav actually noticed that the volume, as it is now, works a little bit like a "second gain control". This means that, when volume is at nearly zero, the pedal has lower gain. This isn`t what I would have expected, as it has nothing to do with controling the last transistor. Or has it? Maybe the wiring you mention wil work better. To clear up: upper lug: hot, middle lug: signal out, lower lug: ground. Is this what you mean?
And again: thx for answering here. This is very usefull for me.

[Banjan73]

Hehe.
Regarding that 8ohm 250W resistor.
I don`t know why i had it on the bench. I use it as a dummy load when doing service on my Marshall :-)

[Banjan73]

Oh!
One more comment.
I seem to remember that when I measured the waves (clipping) on the two stages, it was assymmetrical on the output:-) But I will test the sound on this one before I do some serious mods. As it seems now (except from the volume / gain thing mentioned earlier) it sound pretty nice.
But anyway, I will run the other resistor setup on my breadboard in near future ;-)

[antonis]

So, the 1M / 100k I stole from somewhere (don`t remember where, hehe). This project was mostly about learning a little bit.

You might stole it wrongly..
(1M resistor might came from Collector instead of PS..)
For a voltage divider bias configuration, resistors equivalent parallel combination [R3xR2/(R3+R2)] should be 10 times lower than Q1 h_FE times Emitter total resistor..
You want for Emitter resistor to be constant for DC but variable for AC (signal)..
That's why you have to DC ground Gain pot lug 3 and AC ground wiper (via cap)..

About the volume pot: I am a little bit novice, actually, when it comes to the function of a potmeter. Have to read a little bit there. But the point is to "bleed of" the signal to ground gradually, isn`t it? I hav actually noticed that the volume, as it is now, works a little bit like a "second gain control". This means that, when volume is at nearly zero, the pedal has lower gain. This isn`t what I would have expected, as it has nothing to do with controling the last transistor. Or has it?

For sure, it has..!! 

As it is wired, when Vol pot setting is near zero, Q3 Collector is near Ground..
Any resistance between Vol pot lug 3 & wiper is effectivelly set in parallel with R8..
(power supply is considered Ground for AC - just take my word for the moment.. )

To clear up: upper lug: hot, middle lug: signal out, lower lug: ground. Is this what you mean?

Yeapp..
But reverse your pot numbering - upper should be lug 3..!!
(unless you wish to lower Volume when pot turned CW..)

[Banjan73]

Aha! Its actuall embarrasing that I didn't see the collector in paralell thing myself. Should have😉
Thank you for clearing up. I though actually I was finished with this project, but it seems not, hehe.
I think I'll fix that potmeter connection first, and listen to the sound.
As I mentioned, I have 4 leftover PCBs and plenty of components, so I'll test first with your configuration on breadboard, and then maybe I'll transfer it to a new PCB.
What is important here, though is that the pedal will have enough gain, but i will check that with the breadboard and maybe I'll adjust gain for both transistors.
We'll see.!
P.S. Is it possible to attach soundfiles here?
P.P.S. Do anyone know of a schematic of a transistor based 5 band graphic eq with 9V supply?

[antonis]

So, you like jumping into troubled water, don't you..?? 

(omit Vin, L1 & R11 circuitry - for strictly GRAPHIC EQ, omit (short) P6 pot & R7 pot (for IN-OUT unity Gain)..





But first, read R.G.'s design analysis:http://www.geofex.com/article_folders/eqs/paramet.htm 

[Banjan73]

Hehe.
I have been doing electronics for a while (even though it may not seem so😉). I need a new project after this is finished (i will look into the issues mentioned), and this time I will make something that I actually need.
The distortion box will not be used much, since I have a modified 4-channel old Marshall (modified by Tommy Folkesson in Sweden. Check out www.folkesson.nu). I also have a EVH 5150IIIs, but the Marshall is far better..
But an eq is always needed😉

[Banjan73]

And thx again for a more easy explanation of biasing.

[Banjan73]

Haha!
Wrong link.
Its www.tommy-folkesson.nu (i hope)

[duck_arse]

and, while you are at geofex, reading, go here, and please, PLEASE, learn why we don't call pot lugs by number ......

http://www.geofex.com/Article_Folders/potsecrets/potscret.htm

[Banjan73]

Hi!
Thx for the link.
I did know, deep down, how the potmeter works, I think. I just haven't been that conscious about it. After all, I have measured some of them over the years.
Anyway, I have just soldered the volume pot as described by the doctors😉. I also, when I was at it, grounded the gain pot from the unsoldered lug. I have added a 100uF between the 5,6k pot and ground.
But when testing it on my amplifier, I can't say I heard any difference, hehe. Volume pot still works as a sort of gain control (and volume off course). I will have to doublecheck what I have just done, but now its friday and tv-series time. Its either Fauda or Le bureau😉👍

[Banjan73]

I have now modified the circuit a little bit. Schematics is coming up, but for now, here's the measurement on scope. Not fully symmetrical, but almost.

[Banjan73]

The schematics from antonis had too little gain (distortion) for my taste. So modded it as follows:
R2=470k, R3=100k, R4=12k, R5=40, gain pot 5,6k, C2=4.4uF (will be changed later. Gives a cut of around 900hz, which I know is to high), R6=470k, R7=100k, R8=12k, R9=1k.
As I said, new schematics will come.
What I hope someone can help me with is how to control volume, and only that on this thing. Would it be better to have a higher value potmeter? Should I get the output from Q2 emitter (i saw that that was somekind of a solution on a catalinabread, or something, here)?
Any thoughts about this?

[Banjan73]

Ohh.
C2 is not the original C2. It is the cap from the wiper of the gain pot to ground.

[antonis]

The schematics from antonis had too little gain (distortion) for my taste.

If I only knew which schematics you refer on..

[willienillie]

4-channel old Marshall

Super PA?