Phaser Idea

[tcobretti]

I was looking at these last night:

Here is Tim E's Ghost Dance:



This is Tim E's PWM:



I woke up this morning thinking that combining the very simple 4069 wah and an LFO would make a super simple phaser.  However, I lack the smarts (and I don't currently have a 4069) to put together the pieces.  For starters, is there a way to use the 4069 as an LFO so as to make the entire circuit based on one chip?  Second, Tim's LFO replaces a 500k pot, while the Ghost Dance specifies a much smaller resistance.  Is there a way to alter the resistance of Tim's LFO?

What do you guys think?  Will this work?

[slacker]

cool idea. You can make an LFO out of 3 4069 stages, have a look at this schematic http://www.eskimo.plus.com/fxstuff/slacktrem.png for an example, the LFO is the 3 stages top left. You should be able to hook that up in the same way as the 40106 in Tim's schematic. To make the transistor sweep over lower resistances just put a resistor in parallel with the collector and ground, try values between 47k and 100k to get one that works the best.

[tcobretti]

Thanks Mucho!  When I get a schem and layout together I'll post it!

[slacker]

I forgot to mention you might get LFO ticking problems using the same chip for the audio and LFO, so it's probably best to try it on breadboard first.

[tcobretti]

Ok, thanks for the tip.  I'll check it out when I get the chip.

[Transmogrifox]

I woke up this morning thinking that combining the very simple 4069 wah and an LFO would make a super simple phaser....

What do you guys think?  Will this work?

What do you mean by "phaser"?  If you apply an LFO to the wah, then you just have an auto-wah.

If you want to get a phaser sound, then you need to put a couple of these wahs in series.  One of the wah's center frequency needs to be offset from the other.   Each stage would mix its input and output so that the bandpass filter creates a notch when it's added out of phase from the input at each stage. 

The two sweeping notches would produce a phaser-type sound.  Perhaps this is what you already had in mind?

This could be done with a single 4069, I think.   2 stages of twin-T wah, and one gain recovery stage.  The other side would use 3 inverters for the LFO.   It may be possible to make an OK LFO with 2 inverters.  I'd have to think about it a bit

[tcobretti]

Thanks Transmogrifox!  Not specifically what I had in mind, but I was thinking about how to use the remaining stages of the 4069.  I guess the only part that I don't understand is how to offset the center freq of the wah.  I'll have to do some research.

What do you think about the possibility of ticking in the LFO?  Should I drop it and use a separate chip?

[B Tremblay]

The Phozer 2.0 is very similar to the ideas being discussed here - two parallel twin-T filters tuned to different center frequencies swept by an LFO: http://runoffgroove.com/phozer.html

I'm certainly not saying there's no point in developing a CMOS version of this, though.

[slacker]

I guess the only part that I don't understand is how to offset the center freq of the wah.  I'll have to do some research.

You could make the wahs sweep different frequencies by putting a resistor or pot in series with the transistor to ground on one of the stages. This way the resistance would be changing over a different range. Or I think you could change the 0.005 and 0.0015uF caps in the feedback loop of the 4069.

What do you think about the possibility of ticking in the LFO?  Should I drop it and use a separate chip?

I'd say it's worth trying, my trem worked fine with one chip. It would be a shame not to use as many of the inverters as possible and If you end up using 2 chips then you might as well do it with opamps instead.

[The Tone God]

I actually built a real phaser using 4069s but it takes a number of inverter stages to make a single phase stage. The problem is that each inverter has only one input and output so you need extra stages to provide the necessary signals for a single phase stage. You need two stages one to drive each section of the phase network. One path of the phase network needs to be 180 out of phase which can be done using another stage. Then you need to mix the two paths which is another stage. So you are looking at four stages for a single phase stage.

Opamps are looking much better now aren't they.

Andrew

[tcobretti]

The Phozer 2.0 is very similar to the ideas being discussed here - two parallel twin-T filters tuned to different center frequencies swept by an LFO: http://runoffgroove.com/phozer.html

I'm certainly not saying there's no point in developing a CMOS version of this, though.

The Phozer was certainly part of the inspiration for this; I was reading about it at ROG last night.  My thought was to take your idea and run with it by trying to do it all on one chip and hopefully without an LDR/LED combo.  I read the article so I know you guys tried to simplify it even more and it didn't really work out.  But it seemed like such a cool idea and so the Phozer combined with Tim E's ideas got me thinking about the subject of this thread.

Dang it, Andrew!  I knew somebody out there had tried this and that there had to be a reason it hadn't been published. 

You guys rock.  Thanks for the help.

[nordine]

I woke up this morning thinking that combining the very simple 4069 wah and an LFO would make a super simple phaser...
...What do you guys think?  Will this work?

hi there tcobretti,

i did read this thread and told myself "something sounds familiar"  ...in august i implemented this concept for the circuit contest, and indeed, it works 

here's a redrawn (by user RLBJR65) of the circuit i did:

http://aronnelson.com/gallery/album107/Morula_Schematic

here you can also find some details about what mods or improvements could be done with the circuit:

http://www.diystompboxes.com/smfforum/index.php?topic=48815.msg395643#msg395643


maybe you can use it as a stepping stone for this project you have

one thing about 4069's that i can't understand ( but it happens anyways) is that i've had to sort, out of 3 or 4 chips, which one works the best for LFO's... because, one had to much ticking.. another didn't oscillate at all, etc

btw, there was a little LFO ticking, but only at fast speed and with one paying lot of attention

cheers

[slacker]

I thought of your Morula when I first saw this thread, but I forgot you used an LFO, I thought you'd used an envelope follower.

[tcobretti]

I have been thinking that I need to go back and reread the contest threads.  It seems like I find a lot of cool stuff in those.  So, no I hadn't seen the Morula, and yes it is precisely what I had in mind when I started this thread. 

I think my 4046 phaser project is officially dead.  It has too many issues to fulfill its mandate of being an extremely simple single chip phaser.

Thanks again for the help, guys.

[Transmogrifox]

I'm not completely convinced It's terribly complicated to do this.  It won't be a phaser in the truest sense, but it will indeed be more like the phozer.  The idea is just to get a couple notch filters moving at different frequencies.  I personally would use the LDR, but the LED idea used in the Morula would work fine as long as a bit of distortion is acceptable and/or desirable.

[col]

I've built a couple of auto wahs, one triggered by a 555 from a Penfold book and one I got off here triggered by an op-amp LFO. Both are "phaserish" at low speeds.

[tcobretti]

Yeah, my other thought was to use the 4069 for the phase stages and a 555 for the LFO, which would still be pretty simple and would eliminate the question of clicking.

[Transmogrifox]

Another simple LFO idea is the LFO from the EA Tremolo.  That one doesn't have the speed range that you can get from the bang-bang-integrator LFO's, but it's much smoother.  It is also pretty simple

[tcobretti]

The EA Trem LFO is one of the pretty interesting aspects of the Phozer article.  It didn't sweep the way it needed to, so they used the LFO from a Ross Phaser.  I guess it makes sense that a specific LFO will interact differently with different circuits.

[Steben]

Actually, to me it seems a faux-phaser can be made best by using the T-circuit (or more than one - discussed already) but to use it in a positive feedback loop. Or use a real midband filter and use a negative feedback loop. If you use only one negative narrow midband feedback, you could have a faux phase 45.

Simulating a Phaser can be done by sweeping notches without pitch shifts.

The Phozer is indeed a nice circuit, but not a phaser and not even a faux-phaser. It is more a faux-faux-faux-phaser, or "negative"-faux-phaser.