A Challenge...

[aron]

I wish you could remember it too! Most times my eyes just glaze over when I start reading a programming book ..

Darn it... his name is Tom.....rrrrrrrr!!! All my programming books (old ones) are in storage......

He wrote a number of Turbo Pascal, Turbo C,

I think his name is Tom Swan! Anyone else have his books?


http://www.amazon.com/exec/obidos/tg/detail/-/0672484226/qid=1073189705/sr=1-1/ref=sr_1_1/103-1072879-8137464?v=glance&s=books

[aron]

Realbasic looks cool! I'm gonna try the demo :mrgreen:

Realbasic is pretty cool.

I own both the windows and macintosh versions. There ARE bugs, but they are trying to fix them.

Lots of 3rd party support!

[Jay Doyle]

Not to take anything away from Jay's excellent post, but coupling caps, input caps, filters etc.... were in the simple mods for years!...
...In the simple mods, check out secrets of stacked stages.

Yes, read everything on Aron's site, especially the FAQ. Read until the ideas become second nature. Remember a bunch of us learned this themselves, without EE classes, so you can too; just remember that it requires work on your part too, no one here has the inclination or time to teach you EE.

My point was to just outline the way to start rolling your own. Boscorelli has a great little section on how to start, you think of your idea and write down what it is an what you want it to do. From that description you then create a block diagram which incorporates your ideas and creates a system of "building blocks" that will fulfill the needs of your idea. From there you build it up on the breadboard and tweak until it sounds good. This last part is the important part, most ideas will work if solidly designed, the question is will it sound good? Tweaking out bad tone and creating good is hard. Then once you have it sounding good you can go onto the building stage.

Hopefully in my post I was able to make the process look easier than it sounds at first to those a little intimidated. It is possible to do this knowing a minimal amount of EE. I did not mean to imply that what I was saying was new, just write up a little bit on the process of rolling your own. There are a lot of questions you need to answer from the beginning to end and knowing which are the important ones to answer first is the key.

Jay

[Gary]

Nice thread, Jay.  There's one thing I'd like to add, though.

After awhile of doing this type of design, you will start to develop a sense for what the basic skeleton of the circuit needs to be to get the desired sound.  Maybe a mosfet stage driving a jfet, maybe a pair of mosfets, etc.  You'll develop an ear for what each component sounds like.  Then you can use certain components for their favorable characterisitcs.

I think that was what I wanted to say.  Hope it got through.

[Jay Doyle]

After awhile of doing this type of design, you will start to develop a sense for what the basic skeleton of the circuit needs to be to get the desired sound.  Maybe a mosfet stage driving a jfet, maybe a pair of mosfets, etc.  You'll develop an ear for what each component sounds like.  Then you can use certain components for their favorable characterisitcs.

This is very true, just as doing an "opamp shootout" with a TS style circuit will give you a better idea for the differences between opamps, the more you use single stages or configurations the more intimate you get with the sound of those stages. You get a "feel" for the sound ingredients of a MOSFET stage, then a feel for it at high or low current, with a bypass cap and without, at different bias points or configurations etc.

As with any learning venture, the more you experiment the more you learn that is useful later on.

Jay

[aron]

I haven't done enough work with MOSFETs. I kind of stayed away from them since they seemed to "blow up" so easily... (Static electricity) and this is in Hawaii (humid).

[george]

I wish you could remember it too! Most times my eyes just glaze over when I start reading a programming book ..

Darn it... his name is Tom.....rrrrrrrr!!! All my programming books (old ones) are in storage......

He wrote a number of Turbo Pascal, Turbo C,

I think his name is Tom Swan! Anyone else have his books?


http://www.amazon.com/exec/obidos/tg/detail/-/0672484226/qid=1073189705/sr=1-1/ref=sr_1_1/103-1072879-8137464?v=glance&s=books

thanks Aron - I checked out Amazon, looks like Tom hasn't written a programming book for a while ...

[Leftrights]

What are (or were?) Peppermill and Black Magic?  Sights?  Pedals?  If they can give me more input into designing my own pedals I'd love to know about them.

[aron]

For now we will put a pot with one outside lug to V+ the other outside lug to ground and the wiper going to a high value bias resistor, 1 Meg ought to cut it, high enough to not load down the previous stage and not too high to induce noise. This resistor goes from the wiper of the bias trim pot to the junction of the input cap and the gate of the MOSFET.

Jay,

Can you explain a little on what that 1MEG resistor does? How does it affect a previous stage when:

This is the first stage in the pedal. If so, does it matter?

If it's the 2nd of cascaded stages, do we need a pot for every stage and how does that 1M resistor affect a previous stage? What does it load down exactly? It is limiting current?

Thanks,

Aron

[Ge_Whiz]

Aron

Been playing with MOSFETs for years, and never managed to blow one up yet.

A humid environment ought to be safer - less risk of static.

[Jay Doyle]

For now we will put a pot with one outside lug to V+ the other outside lug to ground and the wiper going to a high value bias resistor, 1 Meg ought to cut it, high enough to not load down the previous stage and not too high to induce noise. This resistor goes from the wiper of the bias trim pot to the junction of the input cap and the gate of the MOSFET.

Jay,

Can you explain a little on what that 1MEG resistor does? How does it affect a previous stage when:

This is the first stage in the pedal. If so, does it matter?

If it's the 2nd of cascaded stages, do we need a pot for every stage and how does that 1M resistor affect a previous stage? What does it load down exactly? It is limiting current?

Thanks,

Aron

Aron,

That first one Meg resistor is just the biasing resistor. By previous stage I meant anything from another pedal to the raw guitar signal itself. It sets the input impedance of the stage.

No, you don't need a pot to do this you can figure it out with resistors, but I feel that using a pot allows you to dial in any bias point possible for the stage and therefore you can use it to tune the bias easier than switching out resistors. Once the effect is ready to come off the breadboard you can decide whether you want to keep it a trim or use resistors, either is just as good, though the trim allows you to adjust it. I use the same idea with the JFET but it is in fact less useful there with a small source resistor, but having it on the breadboard can't hurt I figure...

Sorry if what I wrote was confusing.

For MOSFETs here are a few things to keep in mind:

1. They have a SUPER high input impedance, literally a physical barrier of metal.
2. They are static sensitive.
3. They work like FETs but bias like BJTs meaning that they don't draw any current through their gate but their gate needs to be higher than the source by some amount.

The trim pot allows you to dial in the voltage needed in #3.

Let me know if you have any other questions.

Jay

[Phorhas]

Jay, I can just sum it all up by saying: "thank you"
I hope it's eanough...

Regards,
Dan.

                     :lol:  

[Jay Doyle]

No Problem. Stay tuned, I am in the middle of putting the information down in a paper...

With some pretty pictures  

[Phorhas]

Oh, well then please do E-Mail me when it's done...

Phorhas@Icqmail.com

Thanx once More...

[Mark Hammer]

When it comes to the "dry" parts of design - the electronic theory - I'm at a total loss.  I couldn't bias a transistor if someone had a knife at my family's throat.  What I DO have a good sense of though, is how it all fits together as a system.

For me, the challenge is not so much building something new in terms of what copyright law would acknowledge as different.  Rather, the challenge is to design things that DO something different, which, in some respects can be a whole lot easier, OR a whole lot harder.

I've made stack of distortion units, and at a certain point it's bit like eating fries somewhere different every meal for a month or locking yourself away with a mountain of porn.  It eventually all looks, feels, sounds, tastes, the same, or so minimally different that you can't justify the existence of most of it.

Once in a while, though, you get something that makes you think and play differently.  For me, THAT is the challenge.  I was playing through a Univibe-modded phaser last night, and the difference between how you play with a phaser and a Univibe is really quite remarkable.  Where the obvious moving notches  in a phaser get you to steer away from holding notes that might be suddenly obliterated by a notch, the more watery and less peaky sound of a Univibe encourages you to hang onto clean notes longer because there will be no sudden dips to erase them.  Similarly, I can see why Adrian Belew is as fond of his Foxx Tone Machine as he is.  The octave function begets a certain style of playing that doesn't occur to you normally, and the way octaves arise is something different than a mere fuzz.

So, the challenge is not to make the same thing over again with different parts and slightly different EQ-ing.  A less fizzy Fuzz Face will not change how you use the pedal, merely how much or little you need to tinker with the amp's treble control or how pleasingly the sound records.  The challenge is to think first about what would throw a curve into how you approach the effect.  For me, this often turns into performance controls, rather than core design elements, but it can translate into core design elements too.

How much planning and intentionality is involved is another thing too.  Ansil, for instance, tends to prefer "design by monkey wrench", tossing in obstructions to normal functioning, in the hopes that something new and different will emerge.  At the other extreme, some of the Z-Vex effects are painstakingly planned out to be new and different.  Both are valid.

Why have we generally drifted from "true" DIY and innovativeness to "clone-only"?  Well, first, let me say that I don't think we have.  There have been as many fascinating and innovative ideas here in the last few weeks as there were in the same time period in 1998.  Hell, maybe even more.  I'm reading a book now entitled "Enabling Knowledge Creation" about how large successful organizations bring people together informally to develop concepts that no single person could ever do on their own.  Satisfyingly, this forum tends to illustrate many of the best practices outlined in the book.

Are there more "clone-only" people than in the past?  Yeah, I think so.  Partly that reflects an increasing technical competence on the part of musicians (or maybe an increasing musical interest on the part of electricians!).  Partly that reflects an increasing awareness of the value of historical pedals and especially the virtues of analog and all its quirks.  Partly that reflects a desire by current musicians to have as broad a tonal palette as possible, incorporating both the full range of the known in addition to a smattering of the unknown.  And partly it reflects the fact that a forum like this is not such a well-kept secret anymore.

Is there a virtue to attempting to accomplish the same goals in other ways?  Sure.  I was looking through a huge stack of distortion schems in bed last night (I can only take so much of the Golden Globes) and stumbled across Jay's assorted boosters that are essentially discrete op-amps.  Do we need something as simple as an op-amp made more complicated?  Sure.  Experimentation with effects ideas is good but experimentation with making the familiar behave more predictably (or unpredictably as the case may be)  is also good.

Ultimately, the challenge to pedal freaks and DIY-ers is the same as the challenge to musicians:  How much do you want to play "the classics" the way they deserve to be played, and how much of an obligation do you have to do something entirely different?

Happy Australia day!

[smoguzbenjamin]

:shock: Wow. Well after all that, my words are going to sound infantile :mrgreen: Ahem.

It's time to bodge some wierd and cool circuits together!

[Jay Doyle]

Mark,

My point wasn't to help to design yet another distortion. It was to show those who may be intimidated with starting to design with a blank piece of paper and an idea how to take that idea and turn it into an effect. The process you use to design a new distortion, if done thoroughly and thoughtfully, should be the same process as designing a new phaser, or a combination tremolo/phaser that can have LFOs synched together or seperate for each effect.

I was merely using Jake's idea of a "MOSFET into a JFET" as the basis for showing the process. And my paper will address that process more specifically.

I do understand your point though.

Jay

[swt]

Thanks a lot for an unbelievable piece of information. Of course i'd love to see that paper when it's finished!!

[aron]

Jay,

Sorry, I am still confused. OK, here's our 1MEG pull down resistor to avoid pops right?

>OK, we have a power supply and reverse polarity protection so we are ready to start with the actual circuit. We start with a high value resistor, say 1 Meg, to pull down excess voltage, into a DC blocking/AC coupling cap...


Ok, here is where I am confused...


>...with the bias arrangement of the stages. For now we will put a pot with one outside lug to V+ the other outside lug to ground and the wiper going to a high value bias resistor, 1 Meg ought to cut it, high enough to not load down the previous stage and not too high to induce noise. This resistor goes from the wiper of the bias trim pot to the junction of the input cap and the gate of the MOSFET.

So we have a 100K pot with outer lug to V+, other lug to ground (Voltage divider) then wiper to a 1M resistor which goes directly to gate right? OK, so this gives the signal a DC offset/bias.

Assuming the above is correct, what does the 1M resistor do? I assume if we adjust the 100K pot to middle on a 9V source, we get 4.5V out of the wiper. Now, what does the 1M resistor in series do when connecting to the gate?

I'm so used to seeing voltage dividers on the gates. On subsequent stages, do you need a gate resistor to ground or does this 1M suffice?

I hope you understand what I am asking.

[gez]

Aron, take a look at the EA trem at runoff.  The input stage is along the lines of what Jay is describing.

Because different MOSFETs have different thresholds a trim is used to set the bias.  The down side to this is that input impedance will vary according to whatever resistance is on the pot - it’s not consistent from circuit to circuit - so a resistor is tied from wiper to gate and the input coupled directly to the gate.  The 1M sets the input impedance (the pot being decoupled by a large value cap) and you can make it whatever value you desire.

Hope this helps.