Newbie needs suggestions for component order

[GibsonGM]

I started the same way...from basic LED circuits, 555 timers and the like.  I had to really step back and re-learn a lot to understand that audio signals are AC, a totally different thing. The active things like opamps and transistors run on DC, but they are amplifying AC signals that have been "raised up" (biased) so that they don't cross zero inside the device as they normally would do (DC things can't go below zero, they just cut signals off).  Two things going on at once...AC signal, and DC is used to amplify it or filter and so on. 

For clarity, AC voltages rise from zero volts to some peak, then come back down and crosses zero, continuing to a negative peak, and repeat...DC is just a straight line voltage that doesn't change.  They're both "voltage", but aren't the same. 

What happens "inside of the caps" is a mix of AC/DC...the AC rides on a DC BIAS voltage that has been added to it (voltages are perfectly happy to mix together), the component does its amplifying thing, and the output cap blocks DC (strips the DC from the AC), returning the AC in an amplified form.   Input caps block the DC operating voltage from getting into things that come before the amplifying device, and also are used to set bass content at the input.    Caps "hanging off of" devices are usually filters that send some particular range of frequencies one way or another.   So after the initial understanding of how AC goes thru a DC device, you work on how capacitors act like frequency-dependent resistors. 
Then all is clear, LOL 

I still fudge a lot of fundamentals (I don't like to study the internals of transistors and such), and I've been here for probably 15 years!  But I've learned enough by osmosis and reading to help people at times, and to build most any of the stomps you have seen in your wanderings here.   If I can do it, you certainly can too!     Judging by your woodwork, you're the kind of person that will soon eat this up and challenge many of us!

[ThePracticalPeasant]

I started the same way...from basic LED circuits, 555 timers and the like.  I had to really step back and re-learn a lot to understand that audio signals are AC, a totally different thing.

I remember when I learned how a classic dimmer switch worked. Wait, you mean to tell me that you're using silicon junctions to clip every half wave?! And, a DIAC? How does an AC diode even make sense?

The active things like opamps and transistors run on DC, but they are amplifying AC signals that have been "raised up" (biased) so that they don't cross zero inside the device as they normally would do (DC things can't go below zero, they just cut signals off).  Two things going on at once...AC signal, and DC is used to amplify it or filter and so on.

Is there an advantage with this method versus using +/- supply rails?
Big caps on the output vs the complexities of a dual power supply?

Judging by your woodwork, you're the kind of person that will soon eat this up and challenge many of us!

I do like to learn, especially the technical details, but I'm flattered, thank you.

[stallik]

Thanks Mike. A lightbulb just went on after all these years

[GibsonGM]

I come in here several times every day, and I STILL get lightbulbs, LOL!  People are so clever that even now, most of these circuits still have things that can be wrung out of them...I really get a kick out of that stuff.    Not so into the physics of transistors, however   I know how they work and how others have set them up to do what they want, good enough for me!

The advantage of using a single supply, of course usually 9V, is the availability, portability and cost of the batteries.   And if your intent is to clip a signal and make distortion, you don't need to amplify it to 12 or 18V, only to cut it off at the knees and send it down the line at 1V!!   But if you want max clean headroom or to drive a speaker, dual supplies might definitely be the way to go... 

[amptramp]

The active things like opamps and transistors run on DC, but they are amplifying AC signals that have been "raised up" (biased) so that they don't cross zero inside the device as they normally would do (DC things can't go below zero, they just cut signals off).  Two things going on at once...AC signal, and DC is used to amplify it or filter and so on.

Is there an advantage with this method versus using +/- supply rails?
Big caps on the output vs the complexities of a dual power supply?

Most op amps have a common-mode input voltage range beyond which the circuit may not work properly.  If you use ± supplies, you can pretty much be sure you can stay within the usable range with signals that go above and below zero.  More importantly, if the plus and minus supplies give you a hard zero as the plus side of the negative supply and the minus side of the positive supply, this is a hard zero point that doesn't move even if a lot of circuit points are connected to it.  If you use a single supply and you have a number of connections to the Vcc/2 supply, it is possible for different stages to feed signal back through the Vcc/2 point if it is not decoupled properly.  But the use of Vcc/2 gives you an advantage: since your output is referenced to ground, your output coupling capacitor can always have the positive connected to the circuit and the negative going to the output because the output will be referenced to ground.  If you have positive and negative supplies, the output will vary above and below zero and you need a non-polarized capacitor to connect the circuit to the output.

I have a fuzz that I have built that uses ±10 to ±15 supplies with internal regulators that give ±5 volts internally.  This gives a bit of margin over the 9 volt DC battery supply because I am using TL074 devices and the usable common-mode input voltage range is quite restrictive even though most devices perform better than spec.  I use a 10 µF non-polarized electrolytic cap to connect the circuit to the output.

[FiveseveN]

Everything in audio uses ±15V except guitar pedals because 50 years ago something something 9V batteries.

[antonis]

I suspect the majority of the gaps in my understanding are related to this magical voodoo of mixing AC and DC in the same circuit.

It's called "supeposition theorem"..
(fancy name but banal application..)

[ThePracticalPeasant]

The advantage of using a single supply, of course usually 9V, is the availability, portability and cost of the batteries.   And if your intent is to clip a signal and make distortion, you don't need to amplify it to 12 or 18V, only to cut it off at the knees and send it down the line at 1V!!   But if you want max clean headroom or to drive a speaker, dual supplies might definitely be the way to go...

I admit that I am combining the two completely unrelated ideas of entering the realm of custom/tweaked/diy FX and self-powering an electric diddley-bow; But when I realized that this tiny overdrive circuit was bordering on powerful enough to drive a small speaker directly, I started to wonder what could be achieved. Being that I can't imagine wanting to play the thing without some amount of overdrive, I thought I might try to explore this avenue, though what ultimately evolves out of this has yet to be seen!

It's complete overkill, but the easiest way for me to get this done would be to use the drive circuit as it is and stick a TDA2050 board after it. As the ones I've got are implemented with big blocking caps for a single power rail, I have a 1A buck regulator I could stick on the 9V to get the 24V the amplifier wants. This setup doesn't give me anywhere near the power for full gain of the chip, but I would be able to drive this little speaker to its limit.

Slightly off topic but related, I generally have no issue spending good money on parts, components, materials, tools, and supplies that are not only quality but the correct/best for the task. That said, I refuse to throw money away, I detest waste, and insist on getting the most I can out of everything that crosses my path. Following this attitude, I acquired two pairs of these Sony speakers originally part of a reasonable quality CD-changing mini-system (I did make a brief effort to salvage components or otherwise fix the systems, but the amplifier and power supply proved too difficult to separate from the rest of the electronics).

Two of the boards and a 24V 6A supply (would cook these speakers at full gain) is powering the speakers for the TV in my bedroom.


The first (somewhat complete) version of the concept, it used (a hugely undersized) external supply so I rigged up the power LED as a (backwards) clip indicator; the system ran the speakers off my desktop computer until I disassembled it for the components.



I think I'm in $30?

Edit: Formatting/Typo; Reminder to self, ALT+0177;

[GibsonGM]

That's pretty cool. I think many of us have 'hacked' things like that,and come up with some pretty useful things! Repurposing is part of the fun around here.

[ThePracticalPeasant]

That's pretty cool. I think many of us have 'hacked' things like that,and come up with some pretty useful things! Repurposing is part of the fun around here.

I got that vibe pretty quick! I've digging around reading old threads and have found some really neat ideas buried in the fora.

[GibsonGM]

Even modifying tape recorders to make tape echoes, LOL 

[ThePracticalPeasant]

Best way I know to confirm what I've learned and fill in the gaps is to attempt to explain it. If someone would be so kind as to point out my mistakes? I'll end up describing bits of this using the hydraulic analogy for electronic circuits where the energy is shown as balls in the pipe, and where AC currents jiggle in place and and DC currents go around in a loop. When stacked up, it looks like two steps forward followed by one step back. This is the AC riding on the DC. And, for simplicity sake I'll assume that the input to this circuit is a plain old sine wave and we visualize the AC as though the signal jiggles in place. And here's the image again so folks don't need to go scrolling.



1. Presumably the big resistor at R0 is related to noise and/or grounding of the circuit as a whole)

2. Vr is the middle of the signal, we apply a bit of current via R1 to bias C1. The signal now jiggles the cap against the bias, which in turn creates that jiggle at the +input. This is how a passive device is powered.

3. The output is fed back to the -input via D1/D2 (two, one for each direction of the flow, it's still AC after all), which by some kind of magic of their behaviour clip the wave. By forcing more or less of the output back to the -input via the divider created between R3+R10 and R4 the amount of overdrive will change. Again, Vr biases the capacitor C2 so the jiggle is biased by the same amount into both inputs.

4. I presume Ge diodes with their lower forward voltage drop result in inputs to the OpAmp at closer to the same level. I presume various diodes have differing behaviours as they clip. Asymmetrical clipping would be if the diodes behave differently, as the current flowing in each direction would be modified differently. I personally don't know how to distinguish any of these differences, it'll be trial and error for now.

5. (Ignoring the overdrive pot) The divider at R3/R4 is a balancing act. Increasing/decreasing R4 would increase the current at the -input but decrease/increase the level going to the tone circuit respectively.

5a. C1 and C2 need to be as matched as possible;
5b. The size of R2 is related to C2 which in turn is related to the voltage VR

6. The cap at C3 is related to the tone of the drive circuit.

5-6. The output of the Op-Amp is the combination of the input and a bit of the output that has been messed with. It's fairly obvious to me that the output of the Op-Amp becomes the supply rail if these resistors aren't balanced correctly.

==

7. R4/C4 is a low pass filter

8. The wiper position of R11+R4/C5 make a variable low pass filter; When the wiper is at 0, the two inputs of the Op-Amp "have the same tone" as both inputs have the same total capacitors to ground. As the wiper is moved to 20k, C5 is only acting on -input. I presume that R6 limits the effect of C5.

9. The Op-Amp is working off the difference between R11/R5, but one input has been modified by the position of the wiper. The range of R11 and C5 matter as far as the filter is concerned, the size of R11 against R5 matters as far as the Op-amp is concerned. The result of this balancing act is some frequencies being attenuated and some amplified.

==

10. C6 takes the DC bias back out; Too small the AC effectively gets absorbed. Too big and... I have no clue, but I would assume there'd be a reason to keep the size fairly close to minimum required.

Lucky 13: I can confidently say that the variable resistor at the end acts as a volume control by allowing a range from effectively no to full signal through. Other than this, how'd I do?

Here's my working drive circuit that's not adjustable in any way:


Edit: Fixed the typos, added added image.

[antonis]

1. No..
It's there to serve as anti-pop (pull-down) item..
(C1 cap leakage might cause popping issues when effect is engaged..)

2. In theory, NO current is applied via R1..
(non-inverting input is considered infinite and we DON'T want any DC jiggling there..)

3. Nothing magic there, other than diode(s) shunting signal (of level higher than a diode forward voltage drop) across R10+R3, hence taking it out of the gain loop..
Vr doesn't bias anything here..
(it's used as AV ground point - C2 could equivalently be connected to actual GND..)

4. >I presume Ge diodes with their lower forward voltage drop result in inputs to the OpAmp at closer to the same level<
I'm afraid I can't get you..

5a. Not at all..
(their values are strongly dependent on R1 and R2 values respectively - both C1/R1 and C2/R2 form HPFs..)
5b. Yes (for C2/R2 relation) and No (for C2/Vr relation)

6. Yes..
It forms a (variable) LPF with R10+R3..
(it actually forms a HPF inside NFB loop, resulting into LPF due to inverted action - just forget it..)

7. Obviously, yes..

8. http://www.geofex.com/article_folders/tstech/tsxtech.htm][url]http://www.geofex.com/article_folders/tstech/tsxtech.htm[/url]


9. See above..

10. C6 isolates (blocks) DC going to output and forms a HPF with Volume pot (in parallel with next stage input impedance)..

[ThePracticalPeasant]

I have fundamental gaps in my knowledge and as a result I can only assume that the strange questions I phrase together when I'm only half-way understanding the content are quite hilarious. Thanks for trying to fill in some gaps. I have a couple things on the go, I'll update y'all shortly!