Basic question about IC PT2399

[marcos10067]

Hi there, i have a question about the IC PT2399. I observed a lot of circuits and tested one delay/echo circuit. Here is the situation that I found, if the delay circuit is making a sound at the output, and then you want to change the time delay, it also changes the note of the sound at the output (can goes to a higher note or a lower note, depending if the time delay raise or not) also, if it have a repeat function happens the same, why this occurs? It's because the feedback? Or its just a IC inconvenient haves that it can't be resolved?

[EBK]

When you change the time, you are really changing the clock rate that data moves through the memory array.  If you make the time shorter, samples move faster, making anything already in the memory sound higher pitch (sped up) when it comes out.  Anything that enters the memory cells after your time change will sound normal pitch (assuming you aren't still changing it).

I'm presently working on an idea that I call a P.O.D. (Polyphonic Octave Delay) that exploits this effect.

Edit: Welcome to the forum! 

[marcos10067]

When you change the time, you are really changing the clock rate that data moves through the memory array.  If you make the time shorter, samples move faster, making anything already in the memory sound higher pitch (sped up) when it comes out.  Anything that enters the memory cells after your time change will sound normal pitch (assuming you aren't still changing it).

I'm presently working on an idea that I call a P.O.D. (Polyphonic Octave Delay) that exploits this effect.

Edit: Welcome to the forum! 

Thanks for receiving me. I see, it have sense, one solution can be make a negative feedback of this notes during the lapse of time you want to change the time delay. But for this, its necessary another IC with memory. Another solution is reset the memory of the IC, but this IC doesnt have a reset input

[Mark Hammer]

Any time that you make the "sample out" rate different than the "sample in" rate, you will change the pitch.  There is only a small difference between this and playing back a tape or vinyl disc at a different speed than it it supposed to be played.  The Chase Bliss Thermae pedal uses this to change the pitch of whatever you play twice.

In the 1970's Electro-Harmonix produced the Memory Man delay.  They also produced a unit called the Hot Foot.  It was a unit that looked like a very large wah, with something that looked like a bicycle brake cable coming out of it.  When you moved the Hot Foot treadle, it rotated the cable.  If you attached the free end of the cable, you could turn a pot on a stompbox.  One of the intended uses of the Hot Foot was to "bend" the pitch of what you played, by attaching the cable to the delay-time control on the Memory Man.  Quickly increase the delay time, and the pitch would drop.  DEcrease the delay time and the pitch would go up.

[marcos10067]

Any time that you make the "sample out" rate different than the "sample in" rate, you will change the pitch.  There is only a small difference between this and playing back a tape or vinyl disc at a different speed than it it supposed to be played.  The Chase Bliss Thermae pedal uses this to change the pitch of whatever you play twice.

In the 1970's Electro-Harmonix produced the Memory Man delay.  They also produced a unit called the Hot Foot.  It was a unit that looked like a very large wah, with something that looked like a bicycle brake cable coming out of it.  When you moved the Hot Foot treadle, it rotated the cable.  If you attached the free end of the cable, you could turn a pot on a stompbox.  One of the intended uses of the Hot Foot was to "bend" the pitch of what you played, by attaching the cable to the delay-time control on the Memory Man.  Quickly increase the delay time, and the pitch would drop.  DEcrease the delay time and the pitch would go up.

Interesting thank you. So, there is a delay you know that it doesn't produce this effect?

[EBK]

If you don't want the pitch shifting, the easiest way to avoid it would be to bypass/mute the delay while you mess with the time. 

[ElectricDruid]

Interesting thank you. So, there is a delay you know that it doesn't produce this effect?

Yes. There are two types of delay, fundamentally: Variable sample rate, and fixed sample rate.

Variable sample rate delays like the PT2399 and every analog delay pedal that I've ever heard of change the delay time by speeding up or slowing down the delay clock. This causes a change in the pitch of the signal that is stored in the delay at that point. On a BBD-based pedal, this is the only option. Digital pedals can be done either way.

Fixed sample rate delays change the delay time by changing the point in the delay buffer from where the delayed signal is read. Most digital delays fall into this category, unless they're deliberately trying to copy the sound of the earlier analog delays (and many are, it's true...). With a fixed sample rate delay, you get a sort of "CD skipping" or "audio scrubbing" sound as the delay changes, depending on how well smoothed it has been. Some early digital delays simply muted the sound while the delay moved to a new position because "delay tap interpolation" was expensive in terms of calculation back then. These days, you'd expect to move smoothly to the new location without hearing anything hugely glitchy.

I looked into this in some detail while working on my own DigiDelay pedal. It uses a fixed sample rate, and changing the delay time gives no pitch change and is fairly glitch free.

HTH,
Tom

[marcos10067]

Interesting thank you. So, there is a delay you know that it doesn't produce this effect?

Yes. There are two types of delay, fundamentally: Variable sample rate, and fixed sample rate.

Variable sample rate delays like the PT2399 and every analog delay pedal that I've ever heard of change the delay time by speeding up or slowing down the delay clock. This causes a change in the pitch of the signal that is stored in the delay at that point. On a BBD-based pedal, this is the only option. Digital pedals can be done either way.

Fixed sample rate delays change the delay time by changing the point in the delay buffer from where the delayed signal is read. Most digital delays fall into this category, unless they're deliberately trying to copy the sound of the earlier analog delays (and many are, it's true...). With a fixed sample rate delay, you get a sort of "CD skipping" or "audio scrubbing" sound as the delay changes, depending on how well smoothed it has been. Some early digital delays simply muted the sound while the delay moved to a new position because "delay tap interpolation" was expensive in terms of calculation back then. These days, you'd expect to move smoothly to the new location without hearing anything hugely glitchy.

I looked into this in some detail while working on my own DigiDelay pedal. It uses a fixed sample rate, and changing the delay time gives no pitch change and is fairly glitch free.

HTH,
Tom

Hi Tom, nice information, your link at your website is broken or it doesn't exist. On the other hand, I was thinking to make a linear control of the VCO of the PT2399. I take the voltage VCO like reference, make a feedback with an integrer and make a plant with a specific dinamic, take out the error signal, when the error is 0, it desactivates a relay, so when the error isn't null, the relay opens the output. The dinamic of the plant can be made with resistors and capacitors, the diferencer and the integrer with OpAmps.

[bluebunny]

I looked into this in some detail while working on my own DigiDelay pedal.

Hi Tom, nice information, your link at your website is broken or it doesn't exist.

  Seems to be fine for me?

[ElectricDruid]

I looked into this in some detail while working on my own DigiDelay pedal.

Hi Tom, nice information, your link at your website is broken or it doesn't exist.

  Seems to be fine for me?

Sorry, the site has been moving server, so it's been a bit up and then down the last day or two. Should be ok from now though.