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

[Govmnt_Lacky]

Will a NE5517AN work in place of the LM13700? It is supposed to be a direct replacement.

+1. I ran it through DatasheetArchive and the ONLY comparable cross referenced part for the LM13600 or LM13700 is the NE5517AN. Can anyone confirm a build with this chip?

[maarten]

I got 5517's when I ordered either LM 13600 or 13700 (forgot which one) and it worked fine for me.
Maarten

[frequencycentral]

Here is my proposed implementation of the features for the Causality 6 PCB posted on page 13 of this thread:



It's a big schematic, with a lot going on, so here's a direct link which is bigger:

http://dl.dropbox.com/u/967492/Causality%206%20Phaser%20Rev%202.JPG

There are a few features which *I think* will work out well, but I wanted to discuss them before diving in.

• Regen odd/even switch replaced by a 47k pot. The idea is that odd stages regen path will be available CCW, even stages regen path will be available CW, no regen will be available with the pot at 12 o'clock. I know this implementation works because I tried similar while breadboarding my Sonic Death Ray phaser.

• Phaser/Filter implementation on the 3rd and 4th OTA stages. C1 and C2  will be mounted on the filter/phaser DPDT

• Wet Only switch replaced by a Dry Cut pot

So.......the 1st and 2nd OTA stages can be switched out. And the 3rd and 4th stages can be switched to filter mode. Which means the unit can function as ONLY a lowpass 12dB/Oct filter if desired, with no phasing. Cool! Now, Mark Hammer said this about phaser/filter mods a few pages back:

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.

That's great!

Now, as the 1st and 2nd OTA (phasing) stages can be switched out, and the 3rd and 4th stages can be switched to filter mode, and the unit can function as ONLY a lowpass 12dB/Oct filter........here comes the BIG question........

• In 'filter only' mode there are still 2 fixed phase shift stages in the 'wet' signal path. As each fixed phase shift stage shift the phase by 90^o, the two fixed phase shift stages will shift the phase by 180^o - the equivalent of an inverter (Yes?). Now, if you take a lowpass filter and mix it with an inverted dry signal in the correct amount, you get a highpass filter, courtesy of phase cancellation. So I am making the assumtion that the two fixed phase shift stages are equivalent to inverting the signal with respect to the dry signal.........so.........using the Dry Cut pot, you will be able to dial in either lowpass (by cutting the dry signal), or highpass (by mixing in the dry signal in the correct amount).

I'm making a major assumption here, that the two fixed phase shift stages are equivalent to inverting the signal with respect to the dry signal. I would really appreciate some feedback on this!

[Mark Hammer]

The impact of the 2 fixed stages will depend on the RC value.  Remember that each stage contributes up to 90 degrees of shift, but contributes less than that for frequencies below some inflection point.

So, if we have a fixed stage with a .047uf cap to the non-inverting pin, and a 10k resistor to ground, the full 90-degree-per-stage shift doesn't really kick in until around 340hz.  If you wanted that 90 degrees to apply across the entire range, you'd probably want to either increase the cap to maybe .22-.47uf, or increase the resistor to ground to 100k or so.

[tubelectron]

Hi All & Rick "frequencycentral",

Sorry to be so late to contribute for review of the Causality 4 MkII phaser - custom-made by frequencycentral, with full UFO appointments :



In front of my vintage phaser army (Morley PFA, MXR ph90 & Ph100, EH Small Stone, italian Silversound SuperPhaser), how does sound the Causality 4 MkII phaser ?

Of course, samples would be useful, but I am not sheduled to do that yet, unfortunately.

- in normal phasing situations, the N°1 remains the Phase 90, unrivalled in smooth and lush, then the Phase 100, and the others comes after.
- in sharp phasing situations, the Small Stone and the Silversound are on one side : regular and wide sweep; the Morley and the Causality 4 MkII are on the other : dramatic extroverts.
- in odd/unusual phasing situations, the N°1 ex-aequo are the Morley PFA and the Causality 4 MkII - and let me say that rivalling the PFA is not an easy task (listen to the intro of Kool & The Gang's "Ladie's night" 1979 hit for the deep/dramatic PFA sound).

To sum up, the Causality 4 MkII phaser is not best at smooth, liquid and lush phasing, but think of a sharp, cutting, unusual sounding phaser. A device which would have been built by EH in it's vintage years, if it hadn't some issues :

1 - When the regen control approaches the max (3/4), it saturates and the rate is somewhat "wobbled".
2 - Sometimes there is "clicks" at the rate of the LFO, like if it's wave was clipped.
3 - The spead of the speed control isn't progressive (it may only be a matter of pot tap) : all the fast speeds are at the end.
4 - there is some interaction between the speed, range and width controls : lowering the speed narrows the width and somewhat displaces the range.

My Guitar is a 1980 stock ES-335, and the amp is a '75 Twin-Reverb. The PSU is a new 9VDC EH wall-wart (the one sold with the Holy Grail Nano reverb), so nothing special.

Since I think that point 3 can be improved, the real annoying issues are 1 and 2.

Rick : do you know how to get rid of these 2 problems ?

A+!

[radio]

Hi

Obviously im not Rick, but I bought his phaser C4 in the other forum

beginning with the f...... . Whereas I totally agree with your points 1 to 4

I was rather relieved that it didn't "swoosh" like a smallstone,

because this way you almost sometimes forget that it is altering

every note until you play several times the same note. I personally

however found the P90, well ..boring after 10 minutes,but I dig the

RedWhichAnalog MoonPhaser but this one has up to 6 stages Iirc.

(My main guitar is a Benford(kind of big apple strat with a Laney LC15)

Thanks,Rick ,for this unique project!

Regards Jean-Marie

[Pigyboy]

I got the '4 working using the pnp from page 3. Thank you Rick.  I need to read thru the thread again as it seems the regen & speed pots only work at the ends of the dial. Abd it sounds like it is clipping a bit at certain settings. Hard to tell much out of a box though.
Chris

[Pakaloabob]

Thank you Rick for designing a phaser that is not overly complicated to build and sounds great! It is currently on my breadboard as I try out the MKII mods.
I noticed on your MKI prototype you added 2 switches for depth and wave shape. Are these comparable to the width and shape pots on the MKII? I like the idea of a ramp up/triangle/ramp down switch instead of a pot. I imagine it would be a SPDT center-off switch, but I am not quite sure how to wire it. Could someone please tell me how it would be implemented? What about the depth switch?

[frequencycentral]

Thank you Rick for designing a phaser that is not overly complicated to build and sounds great! It is currently on my breadboard as I try out the MKII mods.
I noticed on your MKI prototype you added 2 switches for depth and wave shape. Are these comparable to the width and shape pots on the MKII? I like the idea of a ramp up/triangle/ramp down switch instead of a pot. I imagine it would be a SPDT center-off switch, but I am not quite sure how to wire it. Could someone please tell me how it would be implemented? What about the depth switch?

Thanks! Glad you like it.

Just replace the Shape pot with an on-off-on switch. The downside is that the saws are then faster than the triangle.

For the depth switch I did something like this: http://www.diystompboxes.com/smfforum/index.php?topic=80456.msg670900#msg670900

1 - When the regen control approaches the max (3/4), it saturates and the rate is somewhat "wobbled".
2 - Sometimes there is "clicks" at the rate of the LFO, like if it's wave was clipped.

Rick : do you know how to get rid of these 2 problems ?

A+!

I'll work on it! I'm currently building a couple of Causality 6, so I'll experiment a bit.

[Pakaloabob]

I didn't realize using a switch instead of a pot for the shape would cause different speeds between the modes. I guess I'll stick with the pot.
As for the depth switch, I did see the post about changing R5  to get a wider LFO sweep but you also said that it would slow the LFO speeds and cause asymmetry. In the MKII it looks like you have chosen to vary the resistance between pins 6 and 7 of the dual opamp instead. Did you find this method of varying width also changes the speed and symmetry?

[Pakaloabob]

I wanted to report some funny behavior I  noticed with my C4 phaser build.
I built the perfboard following the markII layout, but left out the shape and width pots. There was a noticeable ramping up on my build. I double checked all my solders and traces. Everything seemed fine. I then checked voltages and noticed that my Vref was fluctuating. I noticed that R33 was the only connection between the LFO and Vref. When I removed the resistor, Vref stabilized and the ramping disappeared.
After I did this I re-read this entire thread and saw that Rick did mention the fluctuating Vref, but I don't think there has been any further talk about it. However, there was this:

Other mods I made to my MkII:

• R5 = 220k
• C4 = 47uF (I just paralleled a 33uF with the existing 10uF)
• R33 = 47k

Does the R33 substitution stabilize Vref?

Since I am not using the shape pot, is it still possible to use the width pot effectively? Do I need to have the resistor tied to Vref?

[frequencycentral]

Damndest thing. Last weekend I built a C6 using the layout on page 13. When I fired it up I just got clean sound - no phasing. On inspection, the vref was really low, like just above 1 volt, and both 13700 were getting hot. I checked for shorts and incorrect component values. Nothing. I triple checked the layout against the schematic and previous layouts. Nothing. So I decided to start populating another PCB, checking the vref after soldering each part. Turns out that after inserting the resistors to ground from pins 8 and 9 of the 13700 with the 220k/220k voltage divider I've previously used for vref, it didn't want to play nice. So I removed the 220k/220k voltage divider and broke it out onto breadboard. It worked with a 10k/10k voltage divider. What's really got me foxed is that I only populated one 13700 and it's resistors, so it's not the fact that this is a '6' and not a '4'. And this is the fourth Causality I've built, all three of the others used either 220k/220k or 100k/100k voltage dividers for vref without a problem. I fixed my non-functional build by using a 10k/10k divider, but I just can't figure out why it didn't work with the normal values I use. Confused of UK.

Anyway, the 2 extra swept stages are really cool, adds another dimension. I didn't add the regen parity switch to this build, but I'll try it on the PCB I just half populated. I'm also gonna cludge in a 78L05 for with a 1N4148 vref instead of the resistor voltage divider. This will sort out the problem of the vref fluctuating reported in the above post too.

[tubelectron]

Hi Rick,

Would it be related with my C4MkII issue :

Quote from: tubelectron on November 03, 2010, 03:51:29 AM

1 - When the regen control approaches the max (3/4), it saturates and the rate is somewhat "wobbled".
2 - Sometimes there is "clicks" at the rate of the LFO, like if it's wave was clipped.

Rick : do you know how to get rid of these 2 problems ?

A+!


I'll work on it! I'm currently building a couple of Causality 6, so I'll experiment a bit

Or have you find a solution to it ?

Thanks !

[askwho69]

Happy New Year

[frequencycentral]

I've got a second C6 board half populated. I'm gonna just populate the audio section and hook it up to a breadboarded LFO. I'm convinced that the LFO part can be improved to smooth out the irregularities. I'm also gonna try the lowpass/highpass filter mods mentioned above. Happy new year!

[askwho69]

really digging this project!

[Xide88]

Which D1 and D2 i should use? 1n914??

[frequencycentral]

Which D1 and D2 i should use? 1n914??

Yup, that'd work. I use 1n4148. I haven't ironed out the bugs in the C6 design yet, won't get to it for another month maybe, so until then I'd recommend people stick to C4.

[Xide88]

Thanks a lot Rick!!!

But i have another question(

Where are connections for Width lug 3 and Regen lug 1?

It seems in the schematics that lugs 1 & 2 for Width and lugs 1 & 2 for  Regen are soldered together 

Or i make a mistake and there are simple n/c to these lugs?? 

[frequencycentral]

Thanks a lot Rick!!!

But i have another question(

Where are connections for Width lug 3 and Regen lug 1?

It seems in the schematics that lugs 1 & 2 for Width and lugs 1 & 2 for  Regen are soldered together  

Or i make a mistake and there are simple n/c to these lugs??  

Either. Solder the unused lugs to lug 2, or n/c. It's all the same.