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

[Skruffyhound]

You know about where I am with this stuff so this is by no means guaranteed correct, but I downloaded the datasheets and checked all the pins, cross referenced the layout with the schematic, writing in the resistor and cap numbers from the layout to the schem and crossing them off on the component list. I didn't find any mistakes.  I didn't catch the missing jumper though

[frequencycentral]

You know about where I am with this stuff so this is by no means guaranteed correct, but I downloaded the datasheets and checked all the pins, cross referenced the layout with the schematic, writing in the resistor and cap numbers from the layout to the schem and crossing them off on the component list. I didn't find any mistakes.  I didn't catch the missing jumper though

Thanks Aston, great to have another set of eyes on it. I've just completed the PCB layout, with red traces for +ve, brown for ground and orange for vref - much easier to check the connections. BTW, the missing jumper is under C1, I'll re-upload the perf layout over the weekend.

[Skruffyhound]

Oh I saw the jumper after you mentioned it (u16 to u17) and of course you were right, I just didn't catch it on my own

[ragtime8922]

I've had a bunch of CA3080's collecting dust for the past 3 or 4 years. I'll have to try this out so I can put them to good use.

[Cliff Schecht]

I would save the 3080's for projects that specifically call for them. The LM13700 is biased differently than other OTA's and so you wouldn't be able to just stick them in a circuit anyways, there would be a bit of redesign to do first. The 13700 is a solid chip though, I've used it quite extensively with a lot of luck (it has its limitations though, i.e. it can't deal with currents above about 500uA).

[ragtime8922]

I would save the 3080's for projects that specifically call for them. The LM13700 is biased differently than other OTA's and so you wouldn't be able to just stick them in a circuit anyways, there would be a bit of redesign to do first. The 13700 is a solid chip though, I've used it quite extensively with a lot of luck (it has its limitations though, i.e. it can't deal with currents above about 500uA).

I don't think there would be too much to it to drop a couple 3080's in that circuit but since the breadboard rig is tied up right now I do have time to make an order so I may try the 13700. Never used them. Are they still readily available? I'd love to find some CA3280's somewhere (a better quality dual OTA).

BTW, Hello to all. I've been absent from here/electronics for 3 years or so. I missed everyone and everything here.

[frequencycentral]

Too much dry signal for me in them clipz...

I`d replace that 10k going to the 1st stage (drawn above that input-1k) with a link...

I've taken Ton's advice (of course) and removed that 10k resistor. Now the wet is louder relative to the dry. Also the whole circuit got louder, so I've replaced the output buffer's 47k to vref with 100k to vref to compensate. I'll alter the schematic and layouts to reflect this, and re-upload them to replace the existing (as I don't want to bog this thread down with lots of updated schematic/layouts)

The option is also there for people to add a wet/dry mix pot, or a wet only switch as Ton suggests. But I want to keep it simple right now, so just three knobs, though I will include a couple of pads on the PCB for a wet only switch which can be jumpered if not required.

I don't think there would be too much to it to drop a couple 3080's in that circuit but since the breadboard rig is tied up right now I do have time to make an order so I may try the 13700. Never used them. Are they still readily available?

LM13700 are easily available. I'm assuming this circuit should also work with LM13600 and NE5517, though I haven't got any of either to try. To build this circuit with 3080's you'd have to add a couple of transistor after each 3080 configured as a darlington pair to replicate the 13700's built-in buffers. I've got a bunch of (extreme mojo) metal can 3094's and 3080's myself, but I'm saving them for another project

[frequencycentral]

I've corrected the perf layout (and reveal) on page 2, so they reflect the Rev 1.2 schematic now on page 1.

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.

[Jaicen_solo]

Sounds quite nice, I particularly like the length and shape of the LFO sweep. I suspect that you could do with refining the mix between the clean and the phased signal. It sounds as if it's not quite  as deep as it could be, slightly more vibrato than phasing.

[frequencycentral]

Sounds quite nice, I particularly like the length and shape of the LFO sweep. I suspect that you could do with refining the mix between the clean and the phased signal. It sounds as if it's not quite  as deep as it could be, slightly more vibrato than phasing.

I've taken Ton's advice (of course) and removed that 10k resistor. Now the wet is louder relative to the dry.

[Jaicen_solo]

Yeah sorry about that, the two extra pages weren't showing up on my PC for some reason!

[Fuzz Aldryn]

Hi,

built it today, but with no success. Though I have to say, that I redraw your layout in sprint layout as I don't like diy lc layouts and all that scaling trouble that comes with it. I'll double check this afternoon but at the moment I have no hope, that it will be successfull.

Cheers
Helge

Edit: I attached my layout. Sorry, no component values at the moment, but as it is similar to yours it should be clear.

[frequencycentral]

Hi Helge, quick work! But sorry you're having trouble. The layout is not verified yet though. I think it's correct, but I need to build it to confirm it's correct. What are the symptoms of your units non-functionality?

[Fuzz Aldryn]

Hi Helge, quick work! But sorry you're having trouble. The layout is not verified yet though. I think it's correct, but I need to build it to confirm it's correct. What are the symptoms of your units non-functionality?

Hi,

yeah, I'm sorry. I have to admit that my malfunction description wasn't that detailed. First to name the things that I think do work: The LFO works and the phase stages around the OTA (only a guess). So my suspicion is that something around the fixed phase stages and the mix stage of the Tl074/84 doesn't work. I only can hear some phasing sounds with the depth and speed pot up while R13 is unconnected. With R13 connected the phasing sound is barely noticable also on lower speed settings the effect seems to dissappear. I knew that the layout was still unverified, but to me it looked correct too. What a pitty!:D

Cheers
Helge

[frequencycentral]

Hi Helge, quick work! But sorry you're having trouble. The layout is not verified yet though. I think it's correct, but I need to build it to confirm it's correct. What are the symptoms of your units non-functionality?

Hi,

yeah, I'm sorry. I have to admit that my malfunction description wasn't that detailed. First to name the things that I think do work: The LFO works and the phase stages around the OTA (only a guess). So my suspicion is that something around the fixed phase stages and the mix stage of the Tl074/84 doesn't work. I only can hear some phasing sounds with the depth and speed pot up while R13 is unconnected. With R13 connected the phasing sound is barely noticable also on lower speed settings the effect seems to dissappear. I knew that the layout was still unverified, but to me it looked correct too. What a pitty!:D

Cheers
Helge

Helge, I've spotted an error in your layout. There is a missing trace under R26. If you look on my layout and PnP you'll see there is an orange trace to vref that is under R26. Probably easier to see on the PnP than on the layout itself. Which means that the second OTA stage, the final fixed stage and the output buffer on your build are not receiving a vef. Hope that helps.

[Fuzz Aldryn]

Hi Rick,

thanlks for your help. I did fix that little mistake and now I have somewhat of phasing. But now new strange bugs appeared. I guess this phaser is not supposed to be an envelopephaser, that stucks in the low sweep if you hit youir strings harder? Mine does that atm. Also it tends to distort (I'm using Mini HB with about 8kOhm) and the regeneration pot is pretty... hm, touchy, not to say useless? I gues I still have something made wrong here....  After building about 25 FXs without debugging the last year I guess now it's time to have some bad luck.

Cheers
Helge

[frequencycentral]

I just drew up my layout again, referencing the schematic and my original layout. A very intensive part by part examination. It came out the same. I can't find any issues with it, so I'm just gonna build it.

@ Helge: Your envelopephaser thing sounds very weird. I used a humbucker for my soundclips, I'm not getting any distortion. Did you use any part substitutions in your build? I did a few experiments on my breadboarded prototype tonight, the only way I could get it to @#$% up was to replace the TL084 with a LM324.

[frequencycentral]

Just finished building this onto perf. It worked first time, sounds exactly the same as the one on the breadboard. I still have a little fine tuning to do, Helge is right about the regen being touchy, maybe it need to be 10k or 22k. I also need to up the value of the resistor to ground from the depth pot. I perf'ed mine referencing the PCB layout, which is very close to the perf layout but neater and easier to follow.

[frequencycentral]

Here's a couple of photos of my verified build:




I did experience some strange behaviour though when I first turned it on (with the Depth control at minimum), which could have passed for the 'envelopephaser' effect that Helge mentioned. It turns out that the Iabc inputs of LM13700 don't like to be so close to ground (actually two diode drops above) , and phasing didn't start for me straight away. I played around on the breadboard and replaced R8, 10k on my PCB layout, with a 47k. Now that problem is solved.

I also used a 47k Log for the Regen control instead of 47k Lin, which makes sense as it's actually audio that the pot is dealing with. The most usable resonance sounds are higher up, but I'm still finding I want the whole range of the pot rather than using a lower value pot.

Finally, the regen on my build is a little further from self oscillation than on my breadboarded version, so I'm going to experiment with maybe dropping the value of R24 to 8k2.

I will update the schematics and layouts with these changes ASAP.

[doc_drop]

Hey Rick,

I've been watching this thread. I'm curious how this one compares sound-wise with your Phase Revolution? I've been playing the heck out of that guy pretty much every day, and I love it, especially in the box with a Magnus Modulus. Does this phaser bring something new to the table in your opinion?

Thanks!