west pond
Here's a take on the pup to try to get a unvibe sound out of it. Originally this was much less successful when I used the capacitor values straight from the lovepedal vibronaut without taking into account the difference in resistance values from the vcr's used between both circuits. I have since accounted for said difference and it sounds much closer to my ears
here's a demo of some of its sounds
This came out of trying to adapt the pup and its cmos vcr's to a harmonic tremolo, i may revisit that someday, who knows, but the idea for out of phase lfos started there as well as seeing the option for it on the lovetone doppleganger. i should note there's a lot with the schematic i'd do differently today. not sure why i used a dpdt switch for the lfo selection and I'm sure i'd use the extra opamp for an led driver to get the rate on the indicator led as is my preference.
tidal saffron
That’s sound incredible
west pond
thanks, it totally comes down to personal preference, but the more bass-y wibble wobble for the univibe style setup I think works pretty well
kinda interesting to hear the super fast rate sounds with this setup as its not something a real univibe can manage
you can still get too fast for this circuit and the lfo falls apart, i assume i'd need to raise the value of R42 to set the max speed where things still work but it's been months since i messed with the sim i had using this lfo setup
west pond
well i modelled the phase vibe and the vibronaut, i keep double checking the vibronaut's schematic for some typo since that low notch is so dang low my first thought was it was my mistake
im just going off the ldr specs from the datasheet for that version, but here's the schematic for anyone curious -
i didn't model it at 15v for the audio path circuitry, i wouldn't think that'd matter here
west pond
definitely more confused now, modeled the og bjt univibe and it's only really showing 2 poles
loud plinth
That is because of those 1u caps in series with LDRs and 1u caps bootstrapping impedance of those buffer/inverter. Change those to 10u and you will get those notches back. Also to kill some of that notch depth on op amp version use feedback from stage 4 to stage 3, 100k should be OK.
@west pond
west pond
i ended doing a lot more sim'in of the original univibe and found some old post about the 10u cap suggestion, i may have written about in another channel, probably WIP? but it's long enough ago I can't remember lol.
i think that site had some best guesses for the original's ldr resistances
can't find that website, but some discussion of things here plus a really cool schematic for the more recent korg univibe - #🌐│general message
loud plinth
The things those guys at Korg do with BJT transistors is just insane. Hard to wrap my head around those voltage controlled resistors O.o
What was even more interesting was that PWM controlled phaser discussed 🙂
west pond
@sinful prism makes some really cool things. I’ve been wondering how their mosfet vca has been going
sinful prism
I'm honored to be remembered by my cursed ideas hahahahaha.
The short version is: I spent a lot of time simulating every possible variation on it, including how the internal diodes of the model behave to understand why they work better "on the wrong direction", made a blackmer cell out of a 4007 and even a python tool to analyze voltage to attenuation curves, THD to gain, harmonic compositon, etc.
My intention is to write a series of articles on it to share my discoveries, but I want to do it propperly so I'm still organizing everything and improving the tool itself.
At the moment I have something like 30 different variations of VCAs and ~25 of them are the non-ideal wacky stuff I love doing, including the dual diode bridge analog multiplier, the BlackMOS Cell, and the dual gate mosfet I found on a local store (that sadly looks like a discontinued device).
again, thanks for remembering my weird concepts o/
loud plinth
Blackmer cell is the holy grail of VCA's, direct RMS control there!
loud plinth
I'm honored to be remembered by my cursed ideas hahahahaha.
The short version is...
I simulated some switched resistor/capacitor filters way back (was planning on building a simple software controlled synth) but in the end didn't do it, thought it needed very high PWM frequencies.
sinful prism
My BlackMOS Cell is kind of an unHOLY grail hahaha, lots of distortion if you go past 25mV (kind of like OTAs, but the curves are different) and the VC looks like a function of the quadratic mosfet nature, no log-antilog shenanigans like the BJTs. I managed to test a version 3 with a constant current flowing on the cell that reduced distortion a lot, but my goal was never to make ultra-linear devices, just cool audio stuff, and I imagine these designs on a more "talkback limiter" kind of compression, for paralell drums and vocals and stuff where distortion is a desirable trait, for character.
the surprising one to me was the dual gate mosfet. It was used as a gain control device on RF stuff, but never audio (to my knowledge) and I wonder why. They were very very linear control and low distortion
sinful prism
I simulated some switched resistor/capacitor filters way back (was planning on b...
I designed a PWM phaser for 2 PWM controls where the relation between them control frequency and Q, so you can have other phasing effects like bringing the notched closer together and farther apart instead of sweeping, or both. My intention is to make the thing microcontrolled with a RP2040, I already designed the PCB and everything, I just sadly didnt had the time to build it (like lots of other projects on my desk, sadly)
I hope to get to the point where I can just take some time off and build everything or order everything built from the fabhouses directly and use the time just to test it.
this phaser uses second order stages, which is also something I've never seen on comercial designs (or even hobbyist ones)
and boy, did I discovered why?! hahahah Its a pain to make it work and tune it.
loud plinth
the surprising one to me was the dual gate mosfet. It was used as a gain control...
Probably price???
RF transistors are kinda pricy.
I always wanted to play with those tetrode MOS transistors too but never had one at hand
sinful prism
afaik they were cheap and readily available at the time.
looks like they are no longer produced, and I still found them at, like ~25 cents per device, so, same as a FET would cost?
its also not complexity, making VCAs out of them is as straightforward.
I think noone was interested in R&Ding them into audio products, maybe, since OTAs and FETs were also kinda cheap and available (we are talking probably 80s, 90s here)
sinful prism
I always wanted to play with those tetrode MOS transistors too but never had one...
if you get one, please try a mos "vari-mu" design plz.
loud plinth
If I ever get one on my hands I will 🙂
Too bad they are in those stripline packages
Maybe thats why no one used them in the times of THT consumer products?
sinful prism
hmmm, thats a good theory
it would indeed require weird placement
I didnt think about that
tough terrace
can't find that website, but some discussion of things here plus a really cool ...
the service manual is available from synthxl
hallow jay
here's a demo of some of its sounds
sounds great, very juicy
west pond
I’ve been thinking of revisiting this again with a tap lfo, maybe going all the way with full stereo, tho oddly enough a harmonic trem with 2nd order filters is pretty much a phaser anyways
hallow jay
If I ever get around to a modulation pedal I almost definitely want a 2 phase LFO for stereo stuff. Yeah the whole "harmonic tremolo" thing is very phaser-y to my ears.
Being able to adjust the phase from 0 to 180 degrees continuously... or even have the offset on its own LFO. [rubs hands together]
that basically mandates a digital LFO though
loud plinth
that basically mandates a digital LFO though
I wrote it exactly to make something like this 😄
west pond
Nice, yea being able to adjust phase from 0 to 180 degrees on the lfo would be ideal
loud plinth
Dual output or phaseshift I do not have there, but could write it in
The hardest stuff is already done
west pond
Nice, there’s lots of wants id have if I knew how to do the microcontroller thing on my own but for stereo h trem I was kinda stuck with what the taplfo 3d was capable of. It still bugs me my preferred random lfo setting is stuck on the 2nd bank of waveforms but having a 2nd lfo out with gradual phase shift is up there as well
Could be cool to have separate tempo divisions between the lfos too*
loud plinth
would have to look into tap tempo also 😄
west pond
got the audio sim working but for a reason i don't understand it doesn't like to do the ac sweep when i have actual cd4069 nmos in there instead of fudging things with resistors
my poles ain't movin 😢
loud plinth
got the audio sim working but for a reason i don't understand it doesn't like to...
Where did you get the CD4069 model from? Usually logic chip models are "behavioral" so it might not go into the analog side of it at all
Or is it just that on .ac it doesn't simulate the sweep?
On .ac it just tests a single DC operating condition
So LFO wont work in that simulation mode 😄
west pond
I’m varying the dc control voltage going to the nmos’s , similar to the ota sim in wip
For the pup you’re really only dealing with the nmos’s for the cmos inverters so i didn’t bother including the full inverter
loud plinth
Where did you get the model? 😛
west pond
I’d be curious @sinful prism ‘s spice models for the cd4069 are any different than these. I’ve also found models for the fets in the cd4007
I’d be curious @sinful prism ‘s spice models for the cd4069 are any different than these. I’ve also found models for the fets in the cd4007
sinful prism
I’d be curious <@999905977358876793> ‘s spice models for the cd4069 are any diff...
iirc I shared them with you here when you first started experimenting with the harmonic tremolo, but I can get them again if you want/need and you can compare o/
+ PHI=.75 LAMBDA=1.87M RD=20.2 RS=184.1 IS=31.2F PB=.8 MJ=.46
+ CBD=47.6P CBS=57.2P CGSO=70.2N CGDO=58.5N CGBO=96.3N)
.MODEL CD4069BP PMOS (LEVEL=1 VTO=-2.9 KP=2M GAMMA=3.97U
+ PHI=.75 LAMBDA=1.87M RD=28.2 RS=145.2 IS=31.2F PB=.8 MJ=.46
+ CBD=47.6P CBS=57.2P CGSO=70.2N CGDO=58.5N CGBO=96.3N)```there u go.
I got them from a scientific paper that compared IRL curve traced components to this and it was accurate.
it was a loooong time ago so I would need to find the paper again tho, but again, iirc, thats it.
sinful prism
got the audio sim working but for a reason i don't understand it doesn't like to...
Honestly, I've spent a lot of time tweaking anything I did with "JFETs as resistors" to behave nicely.
From input signal level (there's no attenuation there for the wav file) which cause enough distortion on the allpass filters to affect the phasing when you sum back the dry signal
to setting the correct voltage swing. (mosfets have a deadzone from 0 to ~600mV or so) and no resistance change past ~4.5V or so, iirc
what I suggest is checking a .step param VC with AC analysis for voltage VS notch frequency, and only then start sending WAVs through it
west pond
iirc I shared them with you here when you first started experimenting with the h...
That rings a bell now that you mention it, sorry my memory is worse than a goldfish
loud plinth
Thanks, added it to my user models
wise tapirBOT
@loud plinth has reached level 30. Congratulations!
loud plinth
Fuckin' A
sinful prism
That rings a bell now that you mention it, sorry my memory is worse than a goldf...
mine too, I just remembered cause I was hyped up to see what you would end up with on your experiments \o/
my brain holds onto cool stuff very dearly hahaha
sinful prism
Thanks, added it to my user models
those are not for free Sir.
you are now under a contractual obligation of doing cool stuff too 😛
loud plinth
Damn
sinful prism
(and telling us about it if possible)
loud plinth
Well I have a thing in #🔊│amplifiers Simple class D amp for pedalboard
Waiting for PCB's from JLC, parts just came today
west pond
Honestly, I've spent a lot of time tweaking anything I did with "JFETs as resist...
this was the full setup for the ac sweep
west pond
testing what 0-5v cv from the taplfo/ mcu, it's being attenuated and biased, so wasn't feeding it directly to the fets
loud plinth
If you want to simulate the lfo you need to use .step
loud plinth
I'm blind it seems 😄
sinful prism
Well I have a thing in <#841620555802607656> Simple class D amp for pedalboard
Damn, that's awesome. Congrats.
west pond
I'm blind it seems 😄
this did lead to me figuring out the issue tho
sinful prism
thats the Dylan's autobias version, right?
loud plinth
thats the Dylan's autobias version, right?
No reason not using it on a 4 stager
loud plinth
the ac coupling cap c8
yeah that wont show up in DC operating point 😄
sinful prism
yep, thats smart, also tracks any chip difference IRL
loud plinth
If you will add 100k feedback from stage 4 to stage 2
Notches will look a lot more like univibe
A lot wider and less depth
west pond
dylan graphed them out here
i think they can get pretty low resistance but then its get outside their linear range
loud plinth
~290 Ohm
But probably the question should be what is the max dark resistance of univibe LDRs
loud plinth
So if we would have 10k in parallel with our NMOS
we would just use univibe cap values x100
But NMOS can't go that low in resistance
?
sinful prism
i think they can get pretty low resistance but then its get outside their linear...
they can, but
VACTROLs have other stuff going on (like memory effect) but I think scaling RC is the right aproach here.
since he's using a microcontroller for LFO anyway, you can get fancy slewing (to shape it more vactrol-like response) with some code
west pond
I think @loud plinth worked on a custom mcu that mimicked the uniqueness of an incandescent bulb as well
loud plinth
not much uniquenes, and LDR seems to have way more to do with the effect
Just digital RC filter with 2 constants, one for signal going up, other for signal going down
very simple
west pond
My understanding was an incandescent bulb is more uneven in its dimming vs ‘brightening’
loud plinth
yea, but the time constants for a light bulb are super short
if I can believe papers on that
like 100x time shorter than the time constant of LDR
west pond
I can’t find the post on instagram where they went into more detail but this builder seemed to think the bulb had a fair bit to do with things
Could well be talking things up for marketing
loud plinth
awesome aesthethic he has
west pond
It’d sure be nice to have an original univibe to test tho
west pond
I have a few myself, but not the real thing lol
loud plinth
but it is a clone
But I'm fairly sure this op amp version can be brought incredibly close
west pond
My favorite is the dry bell vibe machine, and I used it so much I burnt out its bulb, one major downside to an incandescent 😅
I’ve since replaced said bulb but haven’t used nearly as much out of worry about doing it again at an inopportune time
west pond
Aren’t those restrictions already in place?
loud plinth
There were exemptions
and I think it expired
So this project is vital for DIY Univibes 😄
west pond
sinful prism
I can’t find the post on instagram where they went into more detail but this bui...
maybe when he says "the bulb" he is accounting for the oscillator + bulb interaction, I'd guess.
west pond
Heres a diagram I’m always struggling to find when I get back to the cmos voltage controlled resistor stuffs
west pond
maybe when he says "the bulb" he is accounting for the oscillator + bulb interac...
Well they’re still using vactrols with their digital lfo that they say they’ve modeled some of the characteristics of the original bulb into
Again, could just be marketing to speak to the incandescent bulb purists lol
sinful prism
Well they’re still using vactrols with their digital lfo that they say they’ve m...
yeah, at this point I'd consider just going all the way (as in: some other form of VR + modelling the vactrol slew on the LFO as well)
hand matching vactrols to sell pedals in numbers must be a pain
west pond
Yea who knows if they really do that
One thing I’m not sure of is if the cmos vcr can vary as much as an ldr while staying as close to linear as they’ll get
loud plinth
LDR will likely have something like 2 decades of resistance range tops, fets can do that too, no problem
loud plinth
Interesting, but to take advantage of option 4 we would have to go smaller resistors feeding the gates - so also smaller resistors for bias generator output and LFO mixer, best would be to have gate control signal buffered.
loud plinth
I think that it's overkill and we wind up with 3 traces going from opamp to inverter per phase stage
gets very crowded fast
I started to draft the schematic for my version
I think I will go full SMD
Let JLC assemble this beast for me
west pond
LDR will likely have something like 2 decades of resistance range tops, fets can...
i'd still leave the caps in the all pass filters through hole for trying different things
that 100x the og univibe value didn't work out quite as well in my sims
loud plinth
I though about resistors parallel to fets, bias setup resistor and caps to be THT
wise tapirBOT
@west pond has reached level 59. Congratulations!
loud plinth
It has a regulator onboard
west pond
nice, thats handy
loud plinth
Saves some $ 😄
west pond
this is more board layout stuff but one thing i've found on my most recent taplfo projects is it helps noise wise to keep agnd and dgnd apart
it's not end of the world levels of noise, but definitely noticeable
loud plinth
OK, I think I can lock this in and do PCB tomorrow. Added some configuration options so it can either be a normal phaser or vibe.
west pond
Are the jp’s jumpers?
loud plinth
yeah
So it can either be vibe/chorus for univibe
or feedback for a phaser
switch that is
west pond
Nice
loud plinth
I wonder do I need all 3 knobs for the LFO though
I can do different firmwares for different stuff
and phasers usually dont have depth controls
and vibes use sine LFOs
so maybe 2 knobs is enough
?
west pond
I’d keep depth
Match og univibe
Personal preference tho but I like a modulation to have either a depth or a mix control
Idk what your goal with this is if it’s just a prototype to test these ideas or what but if it’s more of a test build I’d pile on the options lol
west pond
im an arduino virgin, i'll be curious how this ends of working for you
loud plinth
I'm confident the LFO should behave OK, I already tested it and it generates nice clean waves
I'm only unsure about the Seleen-Key, op amp I specified should be able to go to mV off the rails (LFO output will be centered on 2.5V)
Should work though
west pond
Ah gotcha, in case there are any issues there’s a schematic in the taplfo’d datasheets that works great. It’s pretty close to what you have already but adds a dc offset so it works with something non rail to rail
west pond
OK, I think I can lock this in and do PCB tomorrow. Added some configuration opt...
the question about a cmos switch in #❓│questions reminded me to remind you, you want to ground the input of the unused inverter 🫡
loud plinth
the question about a cmos switch in <#887320547191582762> reminded me to remind ...
It has no VDD connected, it can float 
loud plinth
Ah gotcha, in case there are any issues there’s a schematic in the taplfo’d data...
I was thingking how to bias the output up, could steal this method
loud plinth
No prob, flipped it back on my end ;D
Laying out SMD sucks
I went 4 layer to make this easy for myself
west pond
Another thought, for the angular wave shapes, particularly the square wave, it pays to increase the pwm low pass filtering enough to smooth over the sharp edges of the wave shapes’ corners a bit. Can be unpleasant otherwise with a tremolo. Not sure if the effect is lessened when used in a phaser but my guess is maybe a lil
loud plinth
I have my lowpass set at 150Hz
For the reconstruction filter
So there will be some rounding
west pond
Somewhat related, I was Messing with that 2 stage ota phaser in ltspice but was curious what the schematic looks like using a more modern vca
I wonder how much tinkering it would take to get something univibey out of this
west pond
Smd giveth and smd taketh away
loud plinth
THT would never fit on a board this small
That is true
But not getting free vias with each element hurts
west pond
The jam pedals harmonious monk pedal I have is 2 stacked boards with components on both faces of each pcb and is a mix of smt and tht, it’s pretty impressive. Sometimes I’ve thought about doing that sort of front and back mounted smt component thing when things get annoying during a route but havent done it just yet
loud plinth
The jam pedals harmonious monk pedal I have is 2 stacked boards with components ...
Dual board wouldn't have been stupid for this, but what is done is done 😄
That phaser you linked is interesting, but unfortunately SSM2164 is obsolete (and pricey).
This is an alternative
Input would be at junction of R1 and OTA + terminal
Resistor to that spot
west pond
There’s more modern versions of the ssm2164 in production now
ssi2164 in particular which also has fantastic datasheets with a ton of useful info and circuits
You’re right tho, they aren’t cheap compared to cmos or lm13700
Never built this for reals but had some truly weird sounds coming from the svf in ltspice
loud plinth
SVFs are cool (especially for synths)
tough terrace
Somewhat related, I was Messing with that 2 stage ota phaser in ltspice but was ...
The SSI chips are pretty neat
loud plinth
https://github.com/JordanAceto/SSI2140_12_stage_phase_shifter
Thats noice
tough terrace
I wouldn’t mind turning that project into a 12 phaser guitar pedal
loud plinth
And the fact those SSI chips have internal expo generators is also nice
V/oct control for free
well... "free"
those chips ain cheap
west pond
I think this phaser uses those ssi2140’s https://www.instagram.com/reel/DIjcf4Ixi55/?igsh=MWN5ajA0dzVrOTRrNg==
Instagram
Still wanna get one for myself at some point but they ain’t cheap. It was easily the best phaser I’ve ever played. Really unique drive circuit too
The designer is ex moog, so I’d originally thought it was the mf103 with some changes
loud plinth
This is suspiciously similar to the link @tough terrace posted
but with internal LFO and some more selections for tap
tough terrace
Anything from AMT is guaranteed to be good
loud plinth
https://github.com/JordanAceto/SSI2140_12_stage_phase_shifter
@west pond this
It's basically the same functionality
As the one you linked
west pond
Seems like pretty standard phaser controls to me. AMT isn’t the kind of company that’d clone something
loud plinth
I didn't say they cloned it, rather that both parties expanded what is in the datasheet
And that you can get the same effect by using that modular synth project 😉
loud plinth
Not giving JLC their pound of flesh, SMD assembly price was just insane, redid the board as THT and fit it into 1590BB
Was not easy
west pond
For jlc pcba I’ll keep everything smt that’s part of the basic parts inventory
Anything outside of that I’ll usually do tht, usually quite a bit cheaper in the long run. I forgot the markup per extended inventory component, something like $3? But it can add up fast
loud plinth
yeah, $3
west pond
That said some smt ic’s aren’t too bad to hand solder but I really don’t like doing 0603 components
loud plinth
and yeah, it adds up super fast
loud plinth
@west pond
Prototype built
Arduino coded
And it works 🎉
Correction
Worked before putting into enclo
Must have broke some connection 
... just pinched a wire
Last knob is wave selector
Up left sine
Up right filtered sine
Down left tri
Down left filtered tri
loud plinth
PHASEⱽᴵᴮᴱ
Root of stomp = 
I think I gotta make the FB resistor smaller, barely noticable effect on sound
west pond
Sounding pretty cool!
loud plinth
allpass caps are:
2u2 (didn't have 1u5, might try a parallel combo of 1u+470n ceramics here)
22u
47n
470n
west pond
That could explain why adding feedback isn’t having as much of an effect, those values are pretty huge. Might want to sim things to compare
west pond
I mean interstage feedback
What some phasers call resonance, right?
loud plinth
What some phasers call resonance, right?
yea
loud plinth
This is with values that I used
didn't have 10M resistors so had to use what was on hand
There is enough drive in the LFO to "bottom out" the variable resistors
and that gives that deep univibe effect IMHO
Guts
I fucked up the foot switch board 🤦♂️ and had to bodge
west pond
ah that stinks, but hey it's a prototype
as for the cap values it's a simple enough thing to compare in a sim
snow shoal
why does this have so many responses
Bcuz posting your projects helps other get ideas and come up with new approaches to pedals/building
