New OTA Phaser Design - “Causality 4”

[frequencycentral]

The go-to phaser builds for DIY’ers are:

Phase 45 – Intermediate build. 2 stages. FET matching required. No regen. Very subtle.
Phase 90 – Advanced build. 4 stages. FET matching required.
Ross Ropez – Advanced build. 4 stages. OTA design.
Small Stone – Advanced build. 4 stages. OTA design.

The design concept for “Causality 4” was to achieve a musically useful intermediate build 4 stage phaser with regen, using easily available parts, and requiring no matching of components. It has two swept stages using a dual OTA, and two fixed stages using opamps. The LFO utilises a dual opamp and is based on the MXR phasers, much simpler than the LFOs in either the Small Stone or Ross Ropez. The fixed stages came from an idea suggested by Mark Hammer in this DIYSB thread: http://www.diystompboxes.com/smfforum/index.php?topic=80111.0 Having four stages, even if two are fixed, allows for the addition of regen, which for me is a phaser essential. The Regen control goes from zero to just short of self oscillation - some great sounds available here. The Depth control is particularly interesting and useful, specifying where in the spectrum the phasing occurs.

Your critiques are welcome!

Soundclips, perf and PCB layouts to follow - it's on the breadboard ATM, I want to build it first.



[/quote]

[frequencycentral]

Here's a soundclip for ya: http://dl.dropbox.com/u/967492/Causality%204%20Phaser.mp3

Signal Path:

My below average guitar fumblings  >>>  Squier Tele Custom (bridge humbucker)  >>>  Big Muff Violet Rams Head (clone)  >>>  Causality 4 (on the breadboard)  >>>  Harley Benton GA5 (VJ clone)  >>>  Shure SM58  >>>  Wavelab.

No other processing. A montage of different settings, a bit of extreme knob twiddling going on too.

[effdub]

Wow, that's intense!! 

What I've got on the bread board now is a good bit more subtle. And when you crank the speed, you get more vibe than the "wobble" I'm hearing on yours.

Very cool. 

[frequencycentral]

Thanks CJ. It is sounding good.

What I've got on the bread board now is a good bit more subtle. And when you crank the speed, you get more vibe than the "wobble" I'm hearing on yours.

Maybe you need to look at you wet/dry mix. Maybe replace R5 and R22 with a pot for wet/dry mix. You dry may be overwhelming your wet.

[frequencycentral]

.....and here's a clean clip: http://dl.dropbox.com/u/967492/Causality%20Clean.mp3

Signal Path:

My below average guitar fumblings  >>>  Squier Tele Custom (bridge humbucker)  >>>  Magnus Modulus (a bit of chorus and delay)  >>>  Causality 4 (on the breadboard)  >>>  Harley Benton GA5 (VJ clone)  >>>  Shure SM58  >>>  Wavelab.

[effdub]

Maybe you need to look at you wet/dry mix. Maybe replace R5 and R22 with a pot for wet/dry mix. You dry may be overwhelming your wet.

Son of a bitch. 

That did the trick. Now it's waaaaay better. More phasey, especially with clean tones.

1. You are a genius, Rick.
2. Why the hell didn't I think of that?

I think I'll dial in a nice sound with the pot and set a fixed resistor.

[frequencycentral]

1. You are a genius, Rick.
2. Why the hell didn't I think of that?

Well, maybe, but I put it down to having already built a couple of fully featured phasers this year, both of which have a wet/dry blend pot, so I'm aware that there are some settings where nothing happens. But you're probably right, some days I myself think I'm a genius, then again some (other) days I'm convinced that I'm Napoleon..........


EDIT (by FC's GF): Uh-oh, he's drooling again, time for his medication.

[effdub]

Rick, why did you choose 220K for the voltage divider resistors?

I have some, but I was wondering if it could be lowered to 100K. Would that pump too much current into the OTAs or something? Or is it for the LFO mainly?

Thanks

[frequencycentral]

The LFO has it's own voltage divider, the two 220K resistors. That keeps ticking out of the audio, I tried running it from the main voltage divider on the top right of the schematic, but it seems to prefer it's own divider. The LFO starts to act differently if you change those values too.

As for the main (top right) voltage divider, no particular reason why I used 220K, I'll probably use some diferent values in my build (I have some weird, seldom used values to use up). 100k should be fine.

[effdub]

Just tried to bread board this, but minus the fixed phase stages. Something ain't right, as all I get is a nasty hum. But I can hear the LFO working, so that's a start.

I've probably gone about rerouting the the input/output buffers.

Rick, if you were going to go with just the two OTA stages, where would you make the connections between in/out buffers and the dual OTA chip?

Thanks

[frequencycentral]

Have a look at the Tonepad Ross Ropez, basically you'll be breadboarding a 2 stage version of that, but with a different LFO and omitting the regen circuit.

[effdub]

Thanks

[effdub]

Other than the regen control, I have mine setup just like the Ropez. Hmm.

Well, at least I had the right idea. 

Must be something wrong elsewhere. I've checked things over, but surely I must be missing something major.

[effdub]

Actually, there is one difference I see.

I don't have diodes D1 and D2 connected to pin 3 of the input buffer. I have the 470K to vbias, but not the diodes.

[frequencycentral]

Here's the perf layout, as yet unverified, I'll go over it with a fine toothcomb and build it this weekend. The only variance from the schematic is that I've used the two OTA's of the LM13700 in the opposite order, which just made the layout easier. I've also included a LED resistor (R21) with a nominal value of 1k.



..............and here's a reveal view of the same layout:



And here's the PCB layout and it's Pnp, It's a bit different to the perf layout in that I've tried to lie the resistors flat wherever possible, though I guess you could build a perf layout from the PCB layout - much easier to follow actually, with the colour coding of +ve, ground and vref.