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

[jimilee]

i know it's bad practise not to mention parts substitutions, but i've just realised i have 1.5K resistors  in place of the 1.8. could this be to blame?

That's such a mild difference, pretty sure wouldn't be able to tell. Glad things re getting better,Ive been working on mine too, but watching your progress. Rhere's a causality 6 out btw! 

[MrStab]

from what i've read, there's a version with an  even more insane number of stages from FrequencyCentral. i could just do with 2 that work! lol

300Ω is small in most contexts, but all other ideas have been thoroughly exhausted. it'd mean something like 0.044V difference i guess, so you're probably right. good luck with yours!

[bside2234]

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.

I've built the PCB version on page 3 and it works perfectly so maybe that layout can give some clues to what's wrong.

[MrStab]

thanks Bill - could come in handy. i see yours doesn't have the Shape pot - i wonder if something went wrong when that was implemented in the design.

R7 in your layout appears to be 10K whereas it's 1k in mine. i've only looked at the LFO kinda region for now.

[bside2234]

Yeah, I was thinking it might be a layout issue or something that was added since the first incarnation.
Just wanted to point out that this version worked fine for me and it might be worth looking at.

Seems like it would be easy enough to remove the Shape control and the resistor value that's different around the op-amp and see how it's working for you. If all worked well, the problem lies in the shape control itself and/or the one resistors value.

[MrStab]

corrected R7, and jumpered inverted input with opamp output which was also lacking. all that did was lower the maximum speed and, interestingly, makes LED that represents the peak (?) fade but never go off.

it's hard to compare because of information overload but hey that's 2 disparities so far!

[MrStab]

at first i ruled out this thread that i came across, but i wonder if there's some common ground here as it's still OTA-based phasers involved: http://www.diystompboxes.com/smfforum/index.php?topic=88625.msg748896#msg748896 (no, not actually common g... nevermind). that specific post i linked to describes an uneven voltage divider for a more particular voltage than the 50-50 split of the Causality, but i think i recall another post from a thread about Causality where weird, un-repeatable voltage divider issues occurred until the vbias resistors were changed (still even values but lower values, iirc). might even have been FrequencyCentral? i may have imagined that.

so i wonder if it's these resistors i need to play around with. i built myself a 2nd Phase 45 last night and need to fix the sweep on one for a friend weirdly enough (probably FETs mismatched by like 0.01), but hopefully i won't get sick to death of phasers just yet!!

[gambit07]

Hi again guys. especially Mr. Stab.

I finally got it working.. but not much like the one's with youtube demo's. But still, it works... my main problem now is the speed control pot. I don't have any rev log pots, so Im currently researching about how to change linear tapers into audio tapers.. and the regen seems to dial in the sweep quite fast.. might tinker with it too..

here are some pictures I took while making it.. modified it later to markII..since i started to rebuild MarkI first to try and pick out what I did wrong before.but this one end up working the first time I fired it up..

[gambit07]

one other thing, rick, if you're still here in this thread.. How do I get the regen to "really do its job"? It seems like, I get only a subtle change when its midway then just a "good" change when maxed; not really a great change in the effect/swoosh or phasing maybe...i dont know what to call it.

And.. to no avail, I can't get the needed pot for the speed control..been tinkering on how to change the taper of linear pots, but still seems to squish all the usable speed at the last quarter turn of the pot.

also for the rate led, I couldn't get the other LED to completely turn off like the other one (wax an wane, the other one doesn't like to take rests in between..such a hardworking LED,stays on all the time).. it seems to stay on all the time. (even after endless mixing of speed,depth, width, shape pots)

I'd really like to make this my one and only phaser as I really loved it. Such a great pedal for me! and im keeping it foreveeeerr!

Thanks Rick, hope I could hear from you soon.

[gambit07]

 
bump

anything?

[gambit07]

 
no one's visiting this already..

very sad..

[bluebunny]

Not seen a lot of Rick around these parts for a while, so you may not get an answer from him (I guess he's busy with his modular synth stuff).  I built this a while back using a board from Rick - works well - so I'd suggest re-checking your build for shorts, gaps, etc.  The underside looks neat enough, but you never know...

[MrStab]

hi again Gambit, glad you got it working! i wound up using a Phase 45 i made instead, it's grown on me but for some reason i like the LM13700 chip so i might try again some day...

your Regen issue might not be a case of taper - if i remember correctly, the Regen pot decreases the resistance as you turn it clockwise, so maybe you just need a smaller pot? worth a try at least

[gambit07]

hi again Gambit, glad you got it working! i wound up using a Phase 45 i made instead, it's grown on me but for some reason i like the LM13700 chip so i might try again some day...

your Regen issue might not be a case of taper - if i remember correctly, the Regen pot decreases the resistance as you turn it clockwise, so maybe you just need a smaller pot? worth a try at least

Thanks Mr.Stab! was really stoked when I hooked it to power and saw the rate leds do symmetrical fading.. and when I plugged in my guitar, BAM! Phasing for the entire night! haha!

yeah, I think you should give it another try (and maybe talk to it while soldering the last pcs of components.. hehe.. i did.. ).. it's an amazing design.. and this thing is quite versatile.. i could get a pseudo wah kind of funky rhythm with it.. now Im actually wishing it had tap tempo! haha.. (tap tempos are in the top spot of my DIY "learning" process right now.. but it require PIC's..which i am nowhere close to understanding them..)

anyway, I wish you luck on your builds Mr. Stab!

and oh, I'll try your pot suggestion for the regen.. hmm, do you have any idea about how to change a pot's taper? I've read RG's secret life of pots.. but it doesn't work on the causality.. i did every single combination you could do to the pot (resistor in parallel to lug 1&2, 2&3, 1&3..etc..) I just couldn't spread that speed range on the entire scale of the pot.. but still, my build is usable...

here's how it looks now:

[Mark Hammer]

If the feedback/regen signal has a gain of more than 1x, the circuit will oscillate.  So the maximum regen amount is set by how close you can get to unity without breaking into oscillation.

If you look at the stage where the regen signal returns to, yo will see a 10k resistor between input and out on that op-amp (its feedback loop), a 10 resistor from the regen going to the '-' pin, and another 10k going from the input buffer to that point as well.  The 10k input resistors and 10k feedback resistor in that stage produce a gain of 1x....assuming that their 5% tolerances do not dictate otherwise.



If the regen resistor is 9.5k, and the feedback resistor in that stage is 10.5k (both at the outer edges of 5% tolerance), then the gain of that stage (for the feedback signal) is 1.1x.  That may not seem like much, but it is enough to produce oscillation/howling.

So the 10k log ot for adjusting the regen is really more of a fine-tuning control, to add a bit more resistance in order to be able to stay safely away from oscillation.  If a person wants to have a variety of regen settings, then I would suggest using a 50k pot, since a 10k pot doesn't really add enough resistance to change things appreciably.  With a total regen resistance of 10k+50k, there will be very little regen fed back to the earlier phase-shift stage.  As you reduce the pot setting and approach 10k, it will start to be sharper and sharper.

Ideally, a person probably wants to use a 9k1 fixed resistor, instead of 10k, and a 2k trimpot to set the absolute maximum regen possible before oscillation, and then use the suggested 50k pot to produce degrees of feedback less than that maximum.  Either that or use 1% resistors all round, such that the actual values correspond precisely to the values shown in the schematic.

You'll find that a lot of commercial phasers and flangers incorporate a trimpot to set maximum regen, for precisely the reasons outlined.  If a manufacturer uses 5% resistors, they have no guarantee that each and every unit will stay safely enough away from oscillation when the regen knob is maxed.

[MrStab]

that looks awesome, Gambit! looks like you used the same knobs i usually do in my builds. wish i had the patience to be more comprehensive with enclosure measurements.

any ideas on the uneven sweep, Mark? it's been a while since i tried the circuit, so i can't remember the ins & outs that well, but it definitely had me stumped. it didn't help my perseverance that the board eventually came filled with confusing kludges in an attempt to fix it.

i gave my P45 a proper enclosure yesterday, after using a butt-ugly mains transformer box for months (at shows! when placeholders become permanent...). i've been looking at Rick/Frequency Central's not-too-talked-about "Phase Revolution" (adds 2 more stages), which seems pretty easy to retrofit with a switch, but i can't remember the values of the FETs i matched and 2n5952s aren't the kinda thing you can just grab 10's of for cheap. not hijacking, just redundant phaser talk! lol

[gambit07]

Thanks Mr.Stab! yeah, it is a pain to measure pot alignments relative to the enclosure down to the very millimeter..but it's really worth the effort.. although I get excited most of the time and just pick a spot where to drill... but this time, it was an overall exemption to my habits. . everything was made with patience.. even down to the hand drawn "blue print-ish" design with a phase shift graph..haha..  and yeah, those knobs (my last pcs) look really cool.. it actually came from my mxr dist + clone and another 3 knobber donor pedal.. but, what i do not like about them is the knurled plastic inside which creates a "memory crease" of it's own.. so you can't re position it with small increments..but overall, it looked great combined with the enclosure design!

Mr. IVIark, thank you for that in-depth explanation about the regen pot.. well, i know now why I hit the oscillation zone.. im using cheap resistors..(these may have 10% tolerance, ) but I'll try that trimmer to set the max regen.. I'm sure that'l fix it. (or measure ones that are close to the exact value or maybe make some series resistors to achieve the exact values and sub them with those 10k's)

now, all im left is the speed knob.. having a modified linear taper pot to function as a log taper really spread out the slows but not the fasts.. now it's even harder to find a good slow speed..
I know the schemo calls for a reverse log.. but will it not be the same as the modified log pot? or if i used the reverse log pot, the fasts will be the ones spread out?

[MrStab]

your enclosure art could be quite educational, Gambit - just try not to stare at it when you're meant to be playing a song lol.

you could try wiring up your modified Speed pot backwards, and although it'll work in the wrong direction, if your pot truly is logarithmic then it'll be reverse-log when backwards. if that makes sense. then you can see if a reverse log pot will resolve the issue.

you've probably read up on it already, but i've found the graphs & diagrams on this thread (specifically this page) to be quite handy:
http://www.diystompboxes.com/smfforum/index.php?topic=70732.40 for working out the actual taper of modified pots.

[pinkjimiphoton]

finally built this.... hot damn, i'm lovin' it! thanks rick!! 

[Mark Hammer]

Of course, Sir James, with all of the many distortions you have at your disposal, now is the time to see which of them responds most favorably to a phaser in front.

The thing about moving notches before a clipping circuit is that they will pull the signal down below the clipping threshold for different parts of the signal at different times, resulting in differing amounts of harmonic content for notes in different parts of the spectrum.  The net effect is one of "animating" the tone.  This stands in contrast to sticking a phaser after a distortion, where everything has pretty much the same amount of harmonic content, and the sweeping notches make some notes "hide" while others stand out.