PTWobble - modulated delay for weird experimental soundscapes

[ElectricDruid]

Hi All,

I've been enjoying myself messing about with this. It's a basic PT2399 delay with a vactrol to control the delay time. The vactrol is then controlled by my StompLFO chip, which means you can modulate the delay time with one of eight waveforms , including two random waves. The analog side has a basic "repeats" control and a Wet/Dry mix which goes from 0%/100% to 100%/0%.

The nice thing about using the StompLFO is that the digital PDM waveform output is ideal for driving an LED, so you get much better control over the vactrol than you would feeding a voltage-output LFO to it.

It's pretty defiantly weird, not that that's a criticism in any way. You *can* turn the modulation depth down to zero, in which case you've got a basic simple PT2399 delay and it's pretty straightforward. But as soon as you turn that "Depth" knob, all hell breaks loose.

Here's some sounds - my Pro-One mono synth through the PTWobble, recorded into a Zoom H1.

https://electricdruid.net/wp-content/uploads/2019/08/PTWobble1.mp3
https://electricdruid.net/wp-content/uploads/2019/08/PTWobble2.mp3
https://electricdruid.net/wp-content/uploads/2019/08/PTWobble3.mp3
https://electricdruid.net/wp-content/uploads/2019/08/PTWobble4.mp3

And the schematic:




Enjoy!
Tom

[EBK]

That is actually very close to what I have been working on in PT2399 weirdness:

https://www.diystompboxes.com/smfforum/index.php?topic=122769.msg1158861#msg1158861

You should try modulating two PT2399 chips in a complementary fashion.  Good times.   

I've been using your derived design formulas, by the way.  Very nice work on those!

[rutabaga bob]

Tom - I'm liking this one!  Will there be PCBs in the near future?  Nice work!

[ElectricDruid]

Yes, there probably will be PCBs. I've got a final PCB design done. The sound samples were done with a Rev.1 PCB which needed a couple of fixes, but the tweaked layout sorts those out, so it's basically there.

If anyone has advice on getting the best out of the PT2399 (that doesn't involve adding a compander!) I'm all ears...although I'm not expecting miracles.

I was trying to deign a delay pedal a couple of years ago, and eventually gave up on the PT2399 in frustration at it's hopeless quality and short echoes and designed the DigiDelay instead. But I still had a few PT2399s hanging about and connecting the StompLFO to them just seemed too tempting.

[noisette]

Loving the sound!

If I may insert my unasked for opinion here, the pt2399 has, all lofiness notwithstanding, a
unique and compelling sound, especially in it´s nonlinearities (clipping and overlong delays).
These can take an almost physical modelling sound character.
Also simplicity and cheapness make it possible to use two or more in a circuit...
I´m using diy versions of Rob Hordijk´s Zeitgeist and the beautiful Pitch Pirate.
But "all good things come in threes"

[ElectricDruid]

Maybe you're right, Noisette. It is what it is, and I should just appreciate it for that. I sometimes find it difficult to take my "engineering" hat off and finish up always wondering how it might be possible to improve X or Y. It makes me think of 1950's sci-fi soundtracks, or early Doctor Who or something. There's good fun to be had just playing the feedback, delay, and depth knobs...you don't actually have to put any sound into it once you've got it started!

There's a really crazy sound you can get if you use the very-slowest LFO speed and a sine or triangle wave, and then play a sequence through it. The echoes go from very short tight doubling almost-chorus out to very long crunchy lo-fi mayhem with the pitch dropping and dropping and the whole sequence sounds like its slowing down. Just when you think the whole thing is about to fall to bits (it sounds like it), you get to the bottom and start going up again and everything speeds up and up and the pitch goes rising higher and higher. It's bizarre. What interested me most was the way that the changing echoes gives the very strong impression that the *tempo* is changing, although it isn't. You'd almost swear that the whole thing was speeding up and slowing down, but that's only true of the echoes.

[deadastronaut]

nice, looks cool, + 1 on using 2 pt2399's current mirrored for a longer, cleaner delay.

[ElectricDruid]

nice, looks cool, + 1 on using 2 pt2399's current mirrored for a longer, cleaner delay.

Sorry, I'm not understanding you, Rob. Please explain a bit more. Using a current mirror instead of an LDR isn't going to improve audio quality, is it? (seems unlikely). And 2 x PT2399s only give a me a longer delay if I chain them, which isn't going to get any cleaner...although I suppose I'd get similar delays at a higher rate and therefore better quality - is that what you meant? But what did you mean about the current mirror?

Ta!

[deadastronaut]

sorry tom, i wasnt clear...

i mean so the 2 pt2399's time 'is accumulated' for a longer delay time,

then use your  'wobble' where the overall 'time' would be...if that makes sense.  sorry need more coffee.

so an npn on each pt2399 pin 6 , see my dreamtime delay schemo for an example.

https://deadastronaut.wixsite.com/effects/dreamtimedelay

[anotherjim]

It might only need to wobble one PT2399, a second one after it to make the delay longer.

[Ry]

Add a feedback control and we may be getting into Rainbow Machine territory...

[ElectricDruid]

I like the Dreamtime. Dreamtime+StompLFO would be what exactly? Stomping on dreams?! DreamyStomps?

Looks like I should try it with a second delay as well then.

I messed about with using a current mirror on pin 6 when I was working on the original PT-based circuit of what became DigiDelay. I wanted tap tempo control, you see, so I had a PIC controlling the current, and measuring the clock frequency from the output pin too. It worked, but doing tap tempo when the longest delay/slowest tap rate is about 120bpm is a bit pointless...so I did the Digidelay instead. I suppose I could have tried 2 x PT2399, but then I'd have needed to control two current sinks and measure two clock frequencies too - starting to get a bit complicated. I'd also previously tried 4 x MN3205's (again, worked, but noisy as hell), so I really did everything I could to avoid going digital, but sometimes, y'just gotta!

[deadastronaut]

ahh i see.....got ya. 

the pic lark is way above my head...

[Mr. Lime]

Pretty interesting!

I have an EQD Ghost Echo clone pcb left over and I'm wondering since months what I shall do with it.
A single PT2399 chip delay never got my attention but modulating the delay time seems cool to me.
Especially if there's a Belton Brick right behind the PT2399 for really long echos.



Also replacing the StompLFO chip with the One Shot Event Generator chip got me thinking in this application.. 

How does the LDR sound differ from transistors? Is there a significant change to patrick398's project or is it almost the same? I guess an anti-latch-up is a must for short delay times.

Thread: https://www.diystompboxes.com/smfforum/index.php?topic=102813.20

Thanks once again!

[mth5044]

If anyone has advice on getting the best out of the PT2399 (that doesn't involve adding a compander!) I'm all ears...although I'm not expecting miracles.

Have you tried midwayfair's yellow or green LED hanging off pin 7, cathode to ground? See his Hamlett delay for a pretty in depth explanation, but in short it really helps keep the headroom in check. There is a limitation in that the output of the chip becomes locked in at a certain level (I think 20db?), so if you like the sound, you may need to reconsider the rest of the gain structure.

[ElectricDruid]

How does the LDR sound differ from transistors? Is there a significant change to patrick398's project or is it almost the same?

I find it hard to believe it makes any difference, since it's only controlling the internal VCO and not getting anywhere near the audio directly. I would expect the two techniques to be equivalent, although they might (ok, "probably") follow a different "Volts in -> delay out" curve so they would bend your modulation waves in different ways.

T.

[ElectricDruid]

Also replacing the StompLFO chip with the One Shot Event Generator chip got me thinking in this application.. 

Sorry, I missed this when I read your reply the first time. Good idea! This could add a whole different set of loopy!

The OneShot chip is a drop-in replacement for the StompLFO, so wherever you can stick one, you can at least try the other. The StompLFO offers Depth control and Offset too, and you don't get those niceties on the OneShot (maximum depth all the time, and the signal settles to the minimum value, not some selectable "offset") so sometimes it won't sound great without some tweaks, but it would at least be interesting.

[Nicotep]

Simple and great project !
I wonder if modding R14 and replace it by a pot be enough to get a kind of tone control in the feedback path ?

[ElectricDruid]

Simple and great project !
I wonder if modding R14 and replace it by a pot be enough to get a kind of tone control in the feedback path ?

Unfortunately it's not that simple. That filter is a Multiple FeedBack (MFB) filter, and you can't control the filter frequency with a single resistor value without also altering other aspects of the filter (the gain, the resonance).

Try sticking the values into this calculator and playing with them and see what you get though. It might be close enough for rock'n'roll?!:

http://sim.okawa-denshi.jp/en/OPtazyuLowkeisan.htm

Tom