Joranalogue DELAY 1 - Split-Phase Bucket Brigade Delay Line

[Powerdwarf]

there are long bbd delays in pedal world, lots of. how eurorack is different, idk.

Like what? Seriously, I want to know. I have delay pedals that have digital imitations of BBD delays, but I didn't know there were actually any long delay analogue pedals.

starting from 20 euro behringer vd400 to something more luxurious like “discovery”

[tdallas]

[Keltie]

there are long bbd delays in pedal world, lots of. how eurorack is different, idk.

Like what? Seriously, I want to know.

Best pedal option IMO is relatively new AMT ADG-1, runs MN3005s for proper voltage swing, designed by Hawker, who did lots of the Moogerfoogers. Sounds absolutely glorious. The base edition (can do 17ms-350mswith 1xBBD or 35ms-700ms with 2xBBDs from the knob, up to ~900ms with CV/exp) is getting restocked on May 20th, so heads up if anybody wants one. The special edition (restock TBD) has 3xBBDs for a "long mode" of 50ms-1.1s, up to 1.35s with CV/exp, as well as a compounding send/return.

The XAOC Sarajewo is great for long echos, and obviously much more controllable than any echo, but it does unfortunately have some slew when changing delay times, I think somebody else in this thread mentioned. I suspect this is due to digital control over its internal HSVCO, but it might just be a design choice--I've been meaning to bug XAOC about it, since it means that you can't get a really crisp pitch shift effect.

The DELAY 1 seems like an excellent option for shorter lines, with ingredients from the Doepfer design and the CG 1022. Assuming he's using MN3004s and not MN3204s, I will definitely try to get one, maybe two. Interested to see if its HSVCO can easily drive the Doepfers.

I understand it like this: phaser>chorus>flange>echo.

If the delay is less than an audio cycle, phaser. If it’s inside the haas limit at about 40 ms, chorus. If it’s on the haas borderline but not yet a slap back, flange. Longer than that, so that you have two or more audible events, delay

Just wanted to respectfully chime in on this. Flange times are actually shorter than chorus times: flange is typically quoted as around 0.5 to 15ms, with chorus starting when the comb filter effect goes out of spectrum and changes into perceived pitch modulation and ending around the haas limit of ~40 (as you say). Also, while you're right to think of phasers as a type of very short delay, it's done using the group delay of a series of tuned all-pass filters, which means that the notches aren't linked to the overtone series--it's a different frequency response that you don't get by simply shortening a flanger's delay time. (Shortening a flanger to ranges below ~0.5ms, which requires pretty specialized setups to do in analog, produces a very crystalline, hollow-sounding comb filter effect that to my ear hits somewhat different than the typical flanger range, but it's still not a phaser...)

Yep, 100% on chorus/ flange. I kinda fat fingered/ brain farted that lol.

Interesting clarification on phase too. Thanks for that

[daphnid]

This looks SO sick. I have the 1024 Doepfer BBD and love it. It does great bitcrushing-like effects at longer delay times and does a great flange effect, but I've been wondering if I should get a 512 to pair it with as it doesn't extend quite into the range I want for flange and karplus effects.

And then this comes along. Absolutely everything I'd want in a BBD effect and some things I didn't know I wanted. The Joranalogue line keeps delivering. I picked up Orbit 3 and Enhance 2 a while ago and they've been all over my patches since. Would love to try the phase shifted stereo outs on the Delay 1 into Enhance 2.

there are long bbd delays in pedal world, lots of. how eurorack is different, idk.

Like what? Seriously, I want to know. I have delay pedals that have digital imitations of BBD delays, but I didn't know there were actually any long delay analogue pedals.

The Boss DM-2 Waza edition goes to like 2 seconds. There are a ton of analog BBD pedals out there. MXR Carbon Copy, EHX Memory Man, I personally really like the Malekko Dark Ekko, and have the DM-2w hooked up to my ALM SBG, they both sound excellent with synth.

[wombat]

Would owning this delay make having the Mungo D0 redundant?

[lisa]

The Boss DM-2 Waza edition goes to like 2 seconds.

800ms.

[esmooov]

I think minimum delay time of 1ms here is definitely as important, if not more, than the max (for the intended uses). Most of the newer, more polished Eurorack BBDs are built on 4096-stage chips and sacrifice minimum delay time. Sarajewo, PGH Analog Delay Unit, Erica DIY BBD, Pico BBD, all are in this category and have minimum times of ~20ms. Even the 1024-stage-based modules (Erica Black ((I believe)), Heklev) usually use a MN3007 chip with a 5ms minimum or a 3207 with a minimum of 2.5ms. The Doepfer A-188-1 lets you pick the chip, so that's a wild card.

If that doesn't sound like a big difference to you, remember that if we are Karplus Strong-ing or otherwise using the delay feedback for a tone, 1ms = 1 kHz, 5ms = 200 Hz, 20ms = 50 Hz.

[Scatre]

Would’ve been an instabuy if it were stereo. Alas…

[feelingthin]

I think minimum delay time of 1ms here is definitely as important, if not more, than the max (for the intended uses). Most of the newer, more polished Eurorack BBDs are built on 4096-stage chips and sacrifice minimum delay time. Sarajewo, PGH Analog Delay Unit, Erica DIY BBD, Pico BBD, all are in this category and have minimum times of ~20ms. Even the 1024-stage-based modules (Erica Black ((I believe)), Heklev) usually use a MN3007 chip with a 5ms minimum or a 3207 with a minimum of 2.5ms. The Doepfer A-188-1 lets you pick the chip, so that's a wild card.

If that doesn't sound like a big difference to you, remember that if we are Karplus Strong-ing or otherwise using the delay feedback for a tone, 1ms = 1 kHz, 5ms = 200 Hz, 20ms = 50 Hz.

Yes, the delay needs to go into sub 1ms to reach higher notes when using Karplus Strong synthesis. The internal clock goes up to 1mhz. Use a faster clock for higher notes. The manual does mention a faster clock can be used.

[Jumbuktu]

there are long bbd delays in pedal world, lots of. how eurorack is different, idk.

Like what? Seriously, I want to know. I have delay pedals that have digital imitations of BBD delays, but I didn't know there were actually any long delay analogue pedals.

starting from 20 euro behringer vd400 to something more luxurious like “discovery”

Thanks!

[timoka]

Would owning this delay make having the Mungo D0 redundant?

the d0 is a clean dual digital delay with a range from 0.2ms to 2s !!
but maybe you meant the c0/c1 since they have a digital bbd implementation.


regarding the delay-1 stereo configuration...it sounds cool but also a bit like it could have some phase/correlation issues?

[dragulasbruder]

My CG products delay is sweating over this news.

if it gets too sweaty for you to hold onto, you can slip it into my rack.

[MossGarden]

Would’ve been an instabuy if it were stereo. Alas…

Well, you could always pair it with an Enhance 2...

[Paranormal Patroler]

Would owning this delay make having the Mungo D0 redundant?

Nothing can possibly ever make the d0 redundant. That being said, after having used the d0 I stopped asking how slow a delay can go, and started asking how fast it can go. The sounds you can get from fast delays are a sound designer's dream; also, the fact that the Delay 1 can be clocked externally is just fantastic and opens it up for slow timing as well. I tried it and it sounds fine. Can't wait to get it on my case!

[euxine]

I tried this at Superbooth and I think it's a really really clever implementation. Loads of great sounds intuitively available.

I was a little surprised that the feedback behaviour is pretty tame, but that's a nitpick when considering the nice balanced implementation.

[beechemist]

The XAOC Sarajewo is great for long echos, and obviously much more controllable than any echo, but it does unfortunately have some slew when changing delay times, I think somebody else in this thread mentioned. I suspect this is due to digital control over its internal HSVCO, but it might just be a design choice--I've been meaning to bug XAOC about it, since it means that you can't get a really crisp pitch shift effect.

Thanks for your post, I found it really interesting and this aspect connects to some information I've been trying to process from the long thread on the Mungo d0 - which is that on almost all available delays there is a fixed rate of slew applied to delay time changes to avoid noise and unwanted artefacts, and that the d0 gives you access to that slew as an adjustable parameter.
Have I understood that correctly? I'm curious if anyone can expand on this, and explain what's going on when noise & artefacts are introduced when a device is trying to change delay times very rapidly.

[joranalogue]

Thanks for your post, I found it really interesting and this aspect connects to some information I've been trying to process from the long thread on the Mungo d0 - which is that on almost all available delays there is a fixed rate of slew applied to delay time changes to avoid noise and unwanted artefacts, and that the d0 gives you access to that slew as an adjustable parameter.
Have I understood that correctly? I'm curious if anyone can expand on this, and explain what's going on when noise & artefacts are introduced when a device is trying to change delay times very rapidly.

There is no such slewing action going on in D1; the bandwidth of the CV circuitry is over 20 kHz.

I tried this at Superbooth and I think it's a really really clever implementation. Loads of great sounds intuitively available.

I was a little surprised that the feedback behaviour is pretty tame, but that's a nitpick when considering the nice balanced implementation.

We'd still like to tweak some minor aspects of the design before production; increasing the maximum feedback is one of them.

[boramx]

We'd still like to tweak some minor aspects of the design…….. increasing the maximum feedback is one of them.

please do. this is very important.

[Powerdwarf]

The XAOC Sarajewo is great for long echos, and obviously much more controllable than any echo, but it does unfortunately have some slew when changing delay times, I think somebody else in this thread mentioned. I suspect this is due to digital control over its internal HSVCO, but it might just be a design choice--I've been meaning to bug XAOC about it, since it means that you can't get a really crisp pitch shift effect.

Thanks for your post, I found it really interesting and this aspect connects to some information I've been trying to process from the long thread on the Mungo d0 - which is that on almost all available delays there is a fixed rate of slew applied to delay time changes to avoid noise and unwanted artefacts, and that the d0 gives you access to that slew as an adjustable parameter.
Have I understood that correctly? I'm curious if anyone can expand on this, and explain what's going on when noise & artefacts are introduced when a device is trying to change delay times very rapidly.

d0 is digital so its pretty much free from these bbd artifacts. Regarding Sarajewo I have played with it once more yesterday and yeah theres something unique going on with that thing. its reaction to triggers and changing of delay times is weird but maybe thats just me. Though I admt that i havnt explored it as thoroughly when i had it before and now I saw some other sides of it.

Going back to Joranalogue, Im really interested how it compares to cg delay as it seems to be in same department pretty much

[cloudleft]

If I've got it correctly, looks like you can extend the delay time past 50ms using a different clock source at the HF input, at the expense of introducing clock noise. I wonder how slow it can go. . . ?

[joranalogue]

If I've got it correctly, looks like you can extend the delay time past 50ms using a different clock source at the HF input, at the expense of introducing clock noise. I wonder how slow it can go. . . ?

There is no lower limit. However, at a certain point, the signal degradation will be so much that the module seizes to sound like a delay altogether.

[cloudleft]

If I've got it correctly, looks like you can extend the delay time past 50ms using a different clock source at the HF input, at the expense of introducing clock noise. I wonder how slow it can go. . . ?

There is no lower limit. However, at a certain point, the signal degradation will be so much that the module seizes to sound like a delay altogether.

That sounds like it would be fun experimental territory. Cheers Joran

[beechemist]

The XAOC Sarajewo is great for long echos, and obviously much more controllable than any echo, but it does unfortunately have some slew when changing delay times, I think somebody else in this thread mentioned. I suspect this is due to digital control over its internal HSVCO, but it might just be a design choice--I've been meaning to bug XAOC about it, since it means that you can't get a really crisp pitch shift effect.

Thanks for your post, I found it really interesting and this aspect connects to some information I've been trying to process from the long thread on the Mungo d0 - which is that on almost all available delays there is a fixed rate of slew applied to delay time changes to avoid noise and unwanted artefacts, and that the d0 gives you access to that slew as an adjustable parameter.
Have I understood that correctly? I'm curious if anyone can expand on this, and explain what's going on when noise & artefacts are introduced when a device is trying to change delay times very rapidly.

d0 is digital so its pretty much free from these bbd artifacts

I think there are separate issues here which I'm finding confusing. Users of the d0 clearly do get noise and glitches with certain slew settings on the module. This is a post by jjclark (who designed Rainmaker) from the d0 thread explaining why: it's clearly a separate issue to the one being discussed here relating to analog modules, but I don't fully understand the technical information.

Suppose you had a 1-second delay, and feed a 1KHz 10Vp-p triangle wave through it. The slope of the triangle waveform is +/- 20V/msec.
Now suppose you modulate the delay time with a signal that gives a full-scale variation (so the delay time can be modulated from 0 to 1000 msec). If there is a noise step in the modulation signal of 1/2000th (1/20 %) of the whole range there will be a 0.5msec change in the delay time. Doesn't sound like much, but on a 1KHz triangle it can result in an output change (glitch) of 10V!
Even if the noise step is 1/2,000,000 of the whole range there will be a glitch of 1/10V on the output, which would be clearly audible. In the digital world 1/2,000,000 is 3 bits of noise in a 24 bit signal. A Digital-to-analog converter can easily produce this sort of noise level.

The way to avoid this is to slew the delay time control signal so that glitches turn into frequency shifts instead of step amplitude changes. Some may find the frequency shifting caused by the slew objectionable, but the alternative is a noisy output.

I also own an OTO Bim which freaks out and produces bitcrushed noise sounds when switching between tapped tempos rapidly - I have no idea what's going on there!

[Keltie]

If I've got it correctly, looks like you can extend the delay time past 50ms using a different clock source at the HF input, at the expense of introducing clock noise. I wonder how slow it can go. . . ?

There is no lower limit. However, at a certain point, the signal degradation will be so much that the module seizes to sound like a delay altogether.

Now you’re talking…

[joranalogue]

I think there are separate issues here which I'm finding confusing. Users of the d0 clearly do get noise and glitches with certain slew settings on the module. This is a post by jjclark (who designed Rainmaker) from the d0 thread explaining why: it's clearly a separate issue to the one being discussed here relating to analog modules, but I don't fully understand the technical information.

Suppose you had a 1-second delay, and feed a 1KHz 10Vp-p triangle wave through it. The slope of the triangle waveform is +/- 20V/msec.
Now suppose you modulate the delay time with a signal that gives a full-scale variation (so the delay time can be modulated from 0 to 1000 msec). If there is a noise step in the modulation signal of 1/2000th (1/20 %) of the whole range there will be a 0.5msec change in the delay time. Doesn't sound like much, but on a 1KHz triangle it can result in an output change (glitch) of 10V!
Even if the noise step is 1/2,000,000 of the whole range there will be a glitch of 1/10V on the output, which would be clearly audible. In the digital world 1/2,000,000 is 3 bits of noise in a 24 bit signal. A Digital-to-analog converter can easily produce this sort of noise level.

The way to avoid this is to slew the delay time control signal so that glitches turn into frequency shifts instead of step amplitude changes. Some may find the frequency shifting caused by the slew objectionable, but the alternative is a noisy output.

I also own an OTO Bim which freaks out and produces bitcrushed noise sounds when switching between tapped tempos rapidly - I have no idea what's going on there!

This seems to assume a digital delay line with a fixed sample rate and variable buffer length. In BBDs, the buffer is fixed while the sample rate is variable, which avoids these glitches.

It is still true that small amounts of clock jitter causes noticeable distortion, which in most BBD modules is a major source of THD, as the waveform is essentially being 'wiggled' slightly as it passes through the delay line. We solved this by designing a fully analogue high-stability, high-frequency voltage controlled clock source.