One post, One component - An experiment

[phector2004]

Not really... just a BJT and a FET, and an IC for an LFO probably...

The mosfet is being used as a clipper

[Steben]

Not really... just a BJT and a FET, and an IC for an LFO probably...

The mosfet is being used as a clipper

yeah ok, but where is that FET going for?
And mosfet is being used as nothing at the moment.

I would've liked a diode-wired 2N7000 from collector to base. (body diode acts as clipper then) A mosfet bias.   But the bias is there already.

[Quackzed]

seems like some discussion of what is needed, to make this thing into whatever it is to become, should be  carefully considered...
like resolving the components that are floating without 'doing' anything yet...
like that 100k pot off of the diode from the jfet gate... what can we do there...
shall we feed that inductor/cap junction to a gain stage, what is a good topology there to take advanage of that indunctor...
i know, its a sort of 'tame' perspective, as opposed to just letting this thing explde with possibilities, but it seems so far that generally we've kept this thing from being absurd with some restraint and sence. the bigger it gets the harder it will be to do so.
so i'd just like to reccomend to those with some ideas ,to tie up the little loose ends before diving into a big adding frenzy?
i'm surprized that this thing is as sensible as it is with so many different people contributing...
then again, i dont want to stifle any creative juices. rather i'd hate to see this thing become a mess, as its on its way to being an interesting device.

[phector2004]

I agree... Aside from the "standard" components (reverse polarity diode, power cap), a lot of the other stuff doesn't have a specific purpose. I have no clue why the MOSFET was added, nor the inductor. Anyone wanna cue me in to what they're after?
Richon's got the right idea: state what you're getting at with your component additions.

E.g. I added the MAX1044 to run JFET gain stages at 18V, but it can also give +/- 9V for the opamp.

Is the BJT/FET pair supposed to be a SRPP? I think R.G.'s site has suggestions for that kind of stuff.

[vendettav]

I agreed all the way. i just took a look at it again and, really, that diode with a pot or the inducter or mosfet...

anybody explain please??

[deadastronaut]

30.000volt AC psu...

seriously though, an inductor has already halted my possibility of  breading this ....and the mosfet!......and....

[.Mike]

OK, so, my hunch is that the inductor serves two purposes: 1. They look cool. 2. Squiggly lines are fun to draw.

Now, I think I have figured out the third. I think that the inventor(s) of this circuit have read R.G.'s Dirty Tricks.

I am absolutely convinced that the inductor is a ruse to trick potential cloners.



Or not... whatever.

Mike

[flintstoned]

My bad, I missed the suggestion to steer the tl072 into an lfo. I threw in the mosfet for some ocd clipping, was gonna suggest the asymmetrical mosfet on my next page vote but do to strong protest,  I'll accept the coalition's plea for removal.

[phector2004]

Technically, we can still use 1/2 of the TL072 as a gain stage or a buffer, but then we'd need another TL chip to keep the circuit from ticking.

Took the MOSFET off, as requested. No idea what's going on at the inductor part.
I also added a small todo list, to give this SOME structure



Oh and my contribution... are 1N4148's good for charge pump diodes? I usually use 1N5818s, but you're all more likely to have these in stock

EDIT: Ok, this is a bit of a "hand-still-on-the-chess-piece" move, but I'm switching to bipolar supply... 10uf electrolytic!

[earthtonesaudio]

Hey, since there has been some desire expressed for the reasoning behind the suggested additions, I'll add mine.

The 1N4148 on the JFET gate is intended to constrain the JFET for switching duties only, not for amplification.  When the JFET is on, the BJT is biased by the trimpot.  When the JFET is off, the collector sees a very high impedance and it's not likely to be biased correctly. 

What I envisioned was perhaps taking the output for the first transistor stage from the TOP of the JFET, rather than directly from the collector.  This way when the JFET is off the output is essentially just the trimpot value to V+. 

This could be used in conjunction with a second, identical setup of JFET+diode+trimpot to the collector, for multiple bias/preset levels.

Two different JFET switches gives you 4 possible presets for the BJT bias.  (off/off, off/on, on/off, on/on)
You could switch these automatically with a 2-bit counter, and perhaps send each output to a different circuit for parallel processing.


The addition of the pot off the JFET spoils my plans but it could still be worked into something useful.  For instance you could put a square wave LFO at one end of the pot and ground (or the negative rail if the charge pump ends up making this a bipolar supply) for a sort of "depth" control.  It would be nonlinear but maybe that doesn't really matter.

[phector2004]

 

Sounds like something interesting, but I'm completely lost...

If the JFET is off, wouldn't the output just be the bias DC? i.e. 0V after hitting a decoupling cap?

What would this be doing to the signal? What would it sound like?

I'm actually thinking of undoing my last addition, using 2 caps instead to get bipolar power...

[earthtonesaudio]

Here's what I had in mind.  Maybe it will help explain, maybe not...



Say you make EQ1 a high pass filter and EQ 2 a low pass filter.  By clocking the 4024, with a debounced pushbutton or a square wave, the mixer output will cycle through:
Q0 Q1 MixerOut
0   0   off (effect bypass maybe?)
1   0   low pass
0   1   high pass
1   1   notch or peak (depending on if the mixer is adding or subtracting the two signals)


But that's just one person's deranged ramblings. 

[Steben]

holy mother jezus!!!!


yeah, that sort of stuff definitely isn't possible laissez-faire component per component!

[vendettav]

yeah but still that's a nice idea... tho i can see myself nt being able to breadboard this in the end...

[phector2004]

Is that a non-latching SPST  4-mode splitter?

I've been looking for something that can do this for a while now  

No wait... tap-tempo EQ cycler?

[ayayay!]

OK, so, my hunch is that the inductor serves two purposes: 1. They look cool. 2. Squiggly lines are fun to draw.

Now, I think I have figured out the third. I think that the inventor(s) of this circuit have read R.G.'s Dirty Tricks.

I am absolutely convinced that the inductor is a ruse to trick potential cloners.

Haha.  While I do admire an inductor symbol, I was wanting to do more than just that.  C'mon guys, you can't think of anything that has a series inductor like this?  I'm so disappointed...  

[phector2004]

Wah?

You know, you can add another component to hint at what it's doing 

[phector2004]

Any new ideas?

[Hides-His-Eyes]

I like where earthstone went with it; we need an LFO, an inverting op-amp stage at the very least for that. (why go all the way when you can slide between?) You can draw those on as 'blocks', there's no point us assembling such standard subsets one component at a time.

[CynicalMan]

How about number 2 here:


We can always adjust the DC bias and LFO amplitude later to go with the FET.

I think we need to do a bit of cleanup here. There are components here that probably aren't going to amount to much. So I think people need to use rule 1 to its full extent:
1. You can only place, modify, or remove one component per page of this thread.
In particular, I don't think the 1044 or inductor are going anywhere (unless ayayay! tells us what he's talking about  ).

Here's what I had in mind.  Maybe it will help explain, maybe not...

Say you make EQ1 a high pass filter and EQ 2 a low pass filter.  By clocking the 4024, with a debounced pushbutton or a square wave, the mixer output will cycle through:
Q0 Q1 MixerOut
0   0   off (effect bypass maybe?)
1   0   low pass
0   1   high pass
1   1   notch or peak (depending on if the mixer is adding or subtracting the two signals)


But that's just one person's deranged ramblings.  

That's wacky! Whether or not this thread ends up making it, I think you should look in to that.

Anyway, my contribution this time is removing the 1N4147 diode. We can still use the FET as the switch, but now we can use it in linear mode for a tremolo-ish thing.