New Phaser Design - “Causality 4” – Built with you in mind!

[Mark Hammer]

As always, I cannot recommend the "phase-filter" mod highly enough.  It is derived from the original Blacet Phase-filter module, that is described and emulated here: http://hammer.ampage.org/files/Device1-6.PDF

The mod consists of taking the 3300pf caps that are tied to pins 10/12 and 5/7 on the LM13700, and rerouting their other end (i.e., the end that would normally be tied to pin 8 or 9 of the stage before it) to ground.  This transforms the allpass stage to a lowpass stage.  You would, of course, really need a minimum of 4 swept stages to do this, since 2 would need to be swept allpass (producing a notch), leaving 2 to be lowpass.  Simply having 2 swept and 2 fixed stages will not make the unit eligible for this mod.

What does it do, or sound like?  When combined with clean signal, Craig Anderton appropriately describes it as adding "animation".  It really comes alive, however, when the clean signal is lifted and all you have is the combination of allpass and lowpass.  Under those conditions, what you get is a bizarre mélange of what sounds like vibrato, tremolo, and autowah at the same time.  The vibrato, arising from allpass-only, makes sense.  The LFO modulated lowpass sounding like an autowah also makes sense.  And because of the way in which filtering out the treble results in shifts in apparent loudness, it sounds a bit like tremolo.  The overall package is best described as sounding "swampy" and syrupy.  It oozes mojo if your LFO has the right properties.

Note as well that, with the resonance turned up, the filtering really stands out.  With the resonance eased back, and a slow sweep (and a rising ramp would be perfect for this), it can produce some wonderful synth-pad-like tones.  Stick a bit of chorus on it, and you're cooking.

Finally, I will note that the frequency range where notches are produced might be different than the frequency range where the lowpass filtering sounds best to your ears.  What I have done in some instances is leave a pair of 3300pf caps connected to pins 8 and 9, leave another pair of 2700pf caps tied to ground, and simply switch the pin 10/12 and pin 5/7 connection between the free end of either cap, such that in phaser mode the cap values are all 3300pf and in phasefilter mode you haveallpass stages using 3300pf and lowpass using 2700pf.

Try it.  You'll like it.  I promise.

Charlie Barth has a few clips of a Small Stone with the mod, although personally I think it doesn't quite do justice to the range of tones obtainable.    http://moosapotamus.net/THINGS/frankenstone.htm

[sundgist]

Wow. Sounds interesting.
Looks like I'll be fitting in a few mods to the extra stages board!
I've really held off boxing this up. Good job too with all these extras.
This project seems to keep growing and growing.

[doc_drop]

.....and it seems I should put together a perf tutorial!

Hey, I resemble that comment. No need to be a snob!

Now go clip some leads, show off...

[Spidermonkey]

I've just finished this phaser, and sadly it is not working. The bypass works fine, the LED is fine, the signal even gets through and gets some sort of phasing, but it is not affecting the signal of the guitar, more of just a background up and down white noise sound. I did the waxing and waning LEDs mod, and only one of them lights up. It is the one that is going to ground. This leads me to suspect that this is where something has gone wrong, but I'm not sure. I have the IC voltages here-
LM13700
1.) 1.33_ 9.) 1.33
2.) 0 ___ 10.) 0
3.) 3.65 _ 11.) 3.65
4.) 4.1 __ 12.) 3.95
5.) 5.15 _ 13.) 5.15
6.) 0 ___ 14.) 8.19
7.) 5.15 _ 15.) 5.15
8.) 4 ___ 16.) 0

TL084
1.) 4.05 _ 8.) 4
2.) 4 ___ 9.) 4
3.) 3.66 _ 10.) 3.5
4.) 8.17___11.) 0
5.) 1.8 __ 12.) 3.8
6.) 4 ___ 13.) 4
7.) 4 ___ 14.) 4

NE5532
1.) 4.23 __ 5.) 0
2.) 5.1 ___ 6.) 4.62
3.) 4.64 __ 7.) 4.63
4.) 8.3 ___ 8.) 5.03

Thanks!

[soggybag]

Here's a picture of mine. I etched a board and soldered it up the other night.

http://www.super-freq.com/?p=256

[Spidermonkey]

Bump: Any ideas?

[frequencycentral]

Bump: Any ideas?

Your voltages look confused. I hope I don't come across as rude or condescending, but are you sure you understand the IC's pinouts, and which end is which? You said on BYOC:

Sadly, I only have one TL084. But after checking the data sheet, it turns out that pin one is on the side with the cutout, not the dot on the chip, so I did have it in backwards. Looks like I'll be placing another mouser order. This leads me to think, though, the NE5532 that I used also has both a cutout and a dot, and it seems that is in backwards. Would that be fried too?

The way you have laid out the voltage info suggests you need some help with this before we proceed further. Here's the NE5532 for example:



Pin 4 should read 0v, pin 8 should read 9v.

Here's datasheets (with pinouts) for the chip used in this effect:

http://www.datasheetcatalog.org/datasheet/fairchild/NE5532.pdf
http://www.datasheetcatalog.org/datasheet/texasinstruments/tl084.pdf
http://www.datasheetcatalog.org/datasheet/nationalsemiconductor/DS007981.PDF

Hope this helps as a first step to getting your pedal debugged.

[Spidermonkey]

Ahh... I have the pinouts as counting down on the right side instead of up. Thanks for the help! Here are the corrected voltages-
LM13700
1.) 1.33_ 16.) 1.33
2.) 0 ___ 15.) 0
3.) 3.65 _ 14.) 3.65
4.) 4.1 __ 13.) 3.95
5.) 5.15 _ 12.) 5.15
6.) 0 ___ 11.) 8.19
7.) 5.15 _ 10.) 5.15
8.) 4 ___ 9.) 0

TL084
1.) 4.05 _ 14.) 4
2.) 4 ___ 13) 4
3.) 3.66 _ 12.) 3.5
4.) 8.17___11.) 0
5.) 1.8 __ 10.) 3.8
6.) 4 ___ 9.) 4
7.) 4 ___ 8.) 4

NE5532
1.) 4.23 __ 8.) 0
2.) 5.1 ___ 7.) 4.62
3.) 4.64 __ 6.) 4.63
4.) 8.3 ___ 5.) 5.03

[frequencycentral]

NE5532
1.) 4.23 __ 8.) 0
2.) 5.1 ___ 7.) 4.62
3.) 4.64 __ 6.) 4.63
4.) 8.3 ___ 5.) 5.03

Pin 4 should read 0v, pin 8 should read 9v.

[Spidermonkey]

After going back and remeasuring voltages, it turns out that pin 4 of the NE5532 is in fact 0v and pin 8 is 9v. And now pin 7 is about 2v instead of 4.62v. Pin 9 of the LM13700 is also now getting 2 or 3 volts, and I'm not sure why, as i previously had it at 0. My circuit is out of the box now, so could something be shorting?
Thanks for all the help

[doc_drop]

What I do when I get to this point, and I haven't been able to solve my problem with an audio probe, is print a clean schematic and go through it point by point, connection by connection for continuity with my DMM. And I also check adjacent traces, etc., for continuity where it shouldn't be. I highlight the checked areas on the schematic as I go along to make sure I don't miss something. And I measure each resistor to make sure it is the correct value. I have almost always been able to isolate and fix the problem like that.

It sounds to me like with all the confusion about pins it would have been very easy to get something wrong there. (I am sure I would have!)

[Spidermonkey]

Hmm... all of the 27k resistors I installed have the right color bands, but they are measuring around 10k on my DMM, rather than 27k. This seems really really wide for the tolerances, so could that be my culprit?

[TELEFUNKON]

Hmm... all of the 27k resistors I installed have the right color bands, but they are measuring around 10k on my DMM, rather than 27k. This seems really really wide for the tolerances, so could that be my culprit?

My bet is that if you measure them before installing on the board, they`ll read very close to 27k!

[doc_drop]

My bet is that if you measure them before installing on the board, they`ll read very close to 27k!

True that. I should have mentioned that if there is any other componant with resistance in parallel with the resistors in the circuit, you can't measure them with a DMM without disconnecting one side. I would assume the color bands are correct, but I know I have made mistakes thinking the red band is orange, etc., so be careful...

[Spidermonkey]

Okay, that makes sense. I measured others in the bunch that I got them in, and they are all very close to 27k. I disconnected all the wires, so i can really go through the board and check everything. Wish me luck!

[in3x0rable]

I just built this pedal. However, I am experiencing a huge volume drop. Should I replace the 68k resistor with either a trimpot or something with a resistance around 100k?

[frequencycentral]

I just built this pedal. However, I am experiencing a huge volume drop. Should I replace the 68k resistor with either a trimpot or something with a resistance around 100k?

What values are you using for R10 and R12?

[jacobyjd]

Ok, Rick, I bit on this project--I've been reading through this monstrosity of a thread, and I have a couple questions now that I'm populating my board--

1. I didn't see anywhere what the diode values should be--I'm assuming an 1n4148 or similar will do the trick. Can you confirm?

2. I don't have any 27k resistors, so I subbed 22k. Do you see any immediate issues with this? I also have some 33k resistors if the 22k won't work.

3. There was some mention with trouble w/  ICs other than the specified ones. I've got 4558s and TL072s, and I have TL074s instead of the specified opamps. I guess this isn't really a question, but I'll be sure to report in if I have any trouble using unspecified ICs.

Thanks to Rick for putting this project together, and thanks to all who put their work into it--layouts, mods and whatnot I can't wait to get this thing up and running!

[frequencycentral]

Hi Josh,

The diodes can be 1n4148. 4558 or TL072 is fine for the LFO. Not sure about the TL074, should work, but socket it just in case!

Not sure about the resistor subs either. I seem to remember when I breadboarded it I used some 22k until my 27k arrived. Didn't notice much difference at the time. Go for it maybe, if you run into issues you'll know where to look!

[jacobyjd]

Hi Josh,

The diodes can be 1n4148. 4558 or TL072 is fine for the LFO. Not sure about the TL074, should work, but socket it just in case!

Not sure about the resistor subs either. I seem to remember when I breadboarded it I used some 22k until my 27k arrived. Didn't notice much difference at the time. Go for it maybe, if you run into issues you'll know where to look!

Right on--thanks for the reply. I'll ring in here with what I've got in a couple days