Very interesting (semi)deep dive into the most common analog delay concept. But still, even so many years later, my very favorite (sic) echo device is still the Roland 501 tape delay/chorus/reverb.
Another excellent and concise video, thank you! Just as an aside, that MXR delay schematic looks to be a good bit older than the carbon copy, though I'm not sure of the model number of their original bbd based unit. Late 70s though anyway. I've seen the switched filters on a few other units such as the rack mount digital delay MXR made that apparently was a Dave Gilmour favourite. Also, the only other delay I've seen using a CCD is the late 70s Boss DM-1 which was one of their large format pedals before they arrived at the compact stomp box design. Interesting stuff!
Yeah, I should probably but in a caveat that any time I'm showing a schematic I'm just showing one version of the unit and it might or might not match whatever photo I found.
Another fascinating lecture for electronics engineers and electronic music circuit enthusiasts, from Dr. Aaron Lanterman. The last time I saw a row of flip-flops like that was in an Engr 441? lab, where we were breadboarding a register. They were connected in parallel, of course. Thank you again for these lectures, Aaron. I've more to know more about this stuff for over 25 years, but life got in the way. Now that I'm retired, the children have all grown and have kids of their own to care for, I'm grateful for your courses. One day, your students will realize how very fortunate they were to have had you as a teacher and mentor, Aaron.
I have just begun to dablle around with various analog delay effects, primarily chorus and similar effects using BBD's. Funny coincidence you have just released a couple of videos on this subject! :D Thanks! A bit more specifically I'm working on making true stereo (output wise) chorus effects, similar to the Ensemble chorus of the Roland RS-505 string machine. With multiple delaylines each with seperate timings, modulations and mixing, and then with multiple feedback paths too. It is possible to create a plethora of differnt time effect with the same circuit, anything from flanging, chorus, vibrato and simple forms of reverb type effects all by adjusting the parameters and how the delay lines are wired together.
Thanks for another great lesson. Only small correction, the output opamp in CE-2 schematics is summing amplifier, mixing together wet and dry signal when effect is switched on.
Nice video! Even the statement that the delay is inversely proportional to the clock frequency is only true when the clock frequency changes slowly. You could keep track of the age of every sample in the delay. The age increases of course with one second per second, but is at the same time transported at a variable rate from the input to the output (and the input samples all begin with an age of zero). If the clock frequency changes slowly, then the sample age will be roughly proportional to the position in the delay. If the frequency jumps, it will have a delayed effect on the delay; initially the age of the samples coming out will be roughly the same anyway. But it will have an immediate effect in the form of a frequency shift of the signal coming out. The frequency shift depends of course on the ratio between the clock frequency when the sample reached the input and when it reaches the output. But this is the same of course with any delay that has a fixed length but variable speed, whether it's a bucket brigade, tape loop, or a digital delay. (Digital delays can of course be implemented with a moving output tap instead; then the clock frequency is probably fixed, but delay changes are instantaneous.)
As important as the electronic theory is, there is a psycho-acoustic/perceptual aspect that is crucial in establishing where one range of delay and delay-modulation effects exists/ends and another begins. At very short delays, the production of notches and sweeping comb filtering predominates, perceptually, presumably because the time difference between delayed and real-time signal is largely imperceptible, so we pat attention to the filtering. Additionally, where extremely short delays are attainable - say,
@@Lantertronics I have a huge chunk of it too, in a couple of binders. Bernie Hutchins deserves a medal of some sort, whether from NAMM or some similar organization connected with music technology.
Alas, I don't have time to dig more into it on this go-around (I need to move on to wah-wah and distortion effects), but later I may circle back to some of this.
Oh yeah CCDs. Push-broom cameras are kinda interesting. I wonder if there are any audio effects done with a CRT and/or camera tube? I know you could attempt to put a signal on a CRT and feed it through a spacial Fourier transform (a lens) and get an analogue spectrometer, though I don’t think that would be particularly useful or easy.
@@Lantertronics huh, that’s strangely similar to how modern DRAM works. Which I guess wouldn’t have been too difficult to make even then, just that the required discrete electronics would take up more room (and be more expensive if they were vacuum tubes) than using a CRT. Also I bet it would be pretty easy to disassemble a floppy disc, put the disc itself on an electric motor, and use it like a tape delay with 2 or 3 pickups.
Hey Professor! I'm a UMN Junior EE student. I really enjoy your videos and they have helped me in very basic circuit design. I was wondering if you have any signal and systems videos regarding convolution and Fourier series? Thanks!
@@Lantertronics Absolutely! Thank you so much Professor! I will absolutely be resourcing that playlist. It appears to be literally exactly what I was hoping for. Thanks again Professor!
Partly the same reason digital shift registers are good -- they're compact. A circular FIFO would need a massive amount of selection and routing fabric to be able to arbitrarily select different capacitors. Also, in a digital memory you're always "regenerating" a perfect 0 or 1 representation. That's trickier when it comes to analog values.
![Bucket Brigade Devices: How Analog Delay Pedals Work | Too Afraid To Ask [Xvive Echoman V21]](/img/n.gif)
Very interesting (semi)deep dive into the most common analog delay concept.
But still, even so many years later, my very favorite (sic) echo device is still the Roland 501 tape delay/chorus/reverb.
Another excellent and concise video, thank you!
Just as an aside, that MXR delay schematic looks to be a good bit older than the carbon copy, though I'm not sure of the model number of their original bbd based unit. Late 70s though anyway. I've seen the switched filters on a few other units such as the rack mount digital delay MXR made that apparently was a Dave Gilmour favourite.
Also, the only other delay I've seen using a CCD is the late 70s Boss DM-1 which was one of their large format pedals before they arrived at the compact stomp box design.
Interesting stuff!
Oh that's interesting about the DM-1, I hadn't heard that!
Yeah, I should probably but in a caveat that any time I'm showing a schematic I'm just showing one version of the unit and it might or might not match whatever photo I found.
Another fascinating lecture for electronics engineers and electronic music circuit enthusiasts, from Dr. Aaron Lanterman.
The last time I saw a row of flip-flops like that was in an Engr 441? lab, where we were breadboarding a register. They were connected in parallel, of course.
Thank you again for these lectures, Aaron. I've more to know more about this stuff for over 25 years, but life got in the way. Now that I'm retired, the children have all grown and have kids of their own to care for, I'm grateful for your courses. One day, your students will realize how very fortunate they were to have had you as a teacher and mentor, Aaron.
Loving this series of videos so far :)
I have just begun to dablle around with various analog delay effects, primarily chorus and similar effects using BBD's. Funny coincidence you have just released a couple of videos on this subject! :D
Thanks!
A bit more specifically I'm working on making true stereo (output wise) chorus effects, similar to the Ensemble chorus of the Roland RS-505 string machine. With multiple delaylines each with seperate timings, modulations and mixing, and then with multiple feedback paths too.
It is possible to create a plethora of differnt time effect with the same circuit, anything from flanging, chorus, vibrato and simple forms of reverb type effects all by adjusting the parameters and how the delay lines are wired together.
Thanks for another great lesson. Only small correction, the output opamp in CE-2 schematics is summing amplifier, mixing together wet and dry signal when effect is switched on.
Awesome topic from an awesome professor - Revolutionary.
Thank you for your kind words!
Nice video!
Even the statement that the delay is inversely proportional to the clock frequency is only true when the clock frequency changes slowly.
You could keep track of the age of every sample in the delay. The age increases of course with one second per second, but is at the same time transported at a variable rate from the input to the output (and the input samples all begin with an age of zero). If the clock frequency changes slowly, then the sample age will be roughly proportional to the position in the delay.
If the frequency jumps, it will have a delayed effect on the delay; initially the age of the samples coming out will be roughly the same anyway.
But it will have an immediate effect in the form of a frequency shift of the signal coming out. The frequency shift depends of course on the ratio between the clock frequency when the sample reached the input and when it reaches the output.
But this is the same of course with any delay that has a fixed length but variable speed, whether it's a bucket brigade, tape loop, or a digital delay. (Digital delays can of course be implemented with a moving output tap instead; then the clock frequency is probably fixed, but delay changes are instantaneous.)
As important as the electronic theory is, there is a psycho-acoustic/perceptual aspect that is crucial in establishing where one range of delay and delay-modulation effects exists/ends and another begins. At very short delays, the production of notches and sweeping comb filtering predominates, perceptually, presumably because the time difference between delayed and real-time signal is largely imperceptible, so we pat attention to the filtering. Additionally, where extremely short delays are attainable - say,
Wow! Thank you for this thorough writeup!
I love Electronotes. I have the full collection (takes up two entire bookshelves).
@@Lantertronics I have a huge chunk of it too, in a couple of binders. Bernie Hutchins deserves a medal of some sort, whether from NAMM or some similar organization connected with music technology.
The carbon copy is still my favorite delay pedal because it's not "transparent".
Got some BBD chips on my shopping list now :3
Can't buy just one! :)
I think I would be neat to learn about the circuit details!
Alas, I don't have time to dig more into it on this go-around (I need to move on to wah-wah and distortion effects), but later I may circle back to some of this.
Awesome! I love your lectures. I wish we had those at our university.
Oh yeah CCDs. Push-broom cameras are kinda interesting.
I wonder if there are any audio effects done with a CRT and/or camera tube? I know you could attempt to put a signal on a CRT and feed it through a spacial Fourier transform (a lens) and get an analogue spectrometer, though I don’t think that would be particularly useful or easy.
Waaaay back in the dark ages of computers, there was a memory element made using CRTs called the Williams tube.
en.wikipedia.org/wiki/Williams_tube
@@Lantertronics huh, that’s strangely similar to how modern DRAM works. Which I guess wouldn’t have been too difficult to make even then, just that the required discrete electronics would take up more room (and be more expensive if they were vacuum tubes) than using a CRT.
Also I bet it would be pretty easy to disassemble a floppy disc, put the disc itself on an electric motor, and use it like a tape delay with 2 or 3 pickups.
Hey Professor! I'm a UMN Junior EE student. I really enjoy your videos and they have helped me in very basic circuit design. I was wondering if you have any signal and systems videos regarding convolution and Fourier series? Thanks!
Yes! Search for my ECE3084 videos; that's my Signals and Systems class.
Here is the full playlist: th-cam.com/play/PLOunECWxELQRYwsuj4BL4Hu1nvj9dxRQ6.html
Please go tell your fellow students to watch my ECE3084 videos. :)
@@Lantertronics Absolutely! Thank you so much Professor! I will absolutely be resourcing that playlist. It appears to be literally exactly what I was hoping for. Thanks again Professor!
I wonder why they pass to the next capacitive stage instead of using them like a circular FIFO buffer?
Partly the same reason digital shift registers are good -- they're compact. A circular FIFO would need a massive amount of selection and routing fabric to be able to arbitrarily select different capacitors.
Also, in a digital memory you're always "regenerating" a perfect 0 or 1 representation. That's trickier when it comes to analog values.