MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

How to tune delay effect unit with source?
MUFF WIGGLER Forum Index -> Modular Synth General Discussion  
Author How to tune delay effect unit with source?
Homepage Englisch
I don't mean Karplus-Strong.

Suppose you have an VCO droning an A note. You plug it into 1000ms delay with some reasonable sized tail/feedback (you won't get much since it's a static drone but nevermind). Now, if you sweep the delay knob the sound will change pitch. Are there any techniques reliable enough to get the new frequency to be in some musical interval related to dry signal? Any modules (regardless of format, even standalone units) that are more suitable or even designed for this? Have you ever tried this, or know any good videos? Thanks in advance.
Dave Peck
Keep in mind that the pitch only changes while you are increasing or decreasing the delay time. As soon as you stop moving the knob, the pitch ends up the same as the original pitch once again, just with a different delay time.

This is because the change in pitch is caused by stretching or compressing the delay output in real time, meaning only while the delay time knob is actually being moved.

If you want it to steadily produce a second pitch, you need a pitch shifter, which makes use of this idea and uses multiple delays sweeping shorter and longer, (making the pitch go above and then below the original pitch) with LFOs that are out of phase and it crossfades between the resulting signals, so you only get the pitch while it is at one end or the other end of this pitch swing.
Homepage Englisch
Dave Peck wrote:
Keep in mind that the pitch only changes while you are increasing or decreasing the delay time. As soon as you stop moving the knob, the pitch ends up the same as the original pitch once again, just with a different delay time.

This is because the change in pitch is caused by stretching or compressing the delay output in real time, meaning only while the delay time knob is actually being moved.

If you want it to steadily produce a second pitch, you need a pitch shifter, which makes use of this idea and uses multiple delays sweeping shorter and longer, (making the pitch go above and then below the original pitch) with LFOs that are out of phase and it crossfades between the resulting signals, so you only get the pitch while it is at one end or the other end of this pitch swing.


But surely the delay will pick up knob movement and keep replaying it? Or (as you mentioned pitch shifters) I need 2 or more delays in series to preserve a new interval? I was thinking, maybe applying sudden CV jumps such is a square LFO would be helpful to avoid "portamento" effect.

I know there are better machines for the task, but sounds like fun experiment.
Dave Peck
It will only keep playing the pitch swoops that happened during the knob movement if you have the regeneration set to 100%.

And you need two delays to create this type of pitch shifter, but not in series. They are in parallel, and being swept by out of phase LFOs, and their outputs are sent to a crossfader that is controlled by a third phase-locked LFO that only lets them through to the final audible output one at a time, so you only hear the low pitched end (or high pitched end) of the pitch 'swoop' which comes from each swept delay.

And, perhaps a bit counterintuitive, if the LFO sweeping the delay faster and shorter is a triangle wave, THAT can cause the pitch to instantly jump from one pitch to another. If you think about it, you can see why - because the triangle wave is straight slopes that don't change their 'rate' from one end of the wave segment to the other end, so this causes the delay to be sped up by one specific amount when the triangle is ramping up, and then the delay changes its rate and slows down to a different specific amount while the triangle is ramping down.

A SINE wave sweeping the delay will create those pitch 'swoops', but not because it is going up and then down, it is because it is changing rate and slowing down it's upward and downward movement as it crests the top or bottom of the wave, and then speeds up its upward and downward movement as it passes through the zero volt line. It is not just because the sine wave is constantly rising and falling, it's because the sine wave shape is constantly changing the RATE at which it is rising and falling.
HighLordFixer
Digitech RDS 7.6 TimeMachine has CV input
so does the Korg Stage Echos for motor speed
Lexicon LXP15 series has 5 CV foot controller inputs
that can be assigned parameters
not sure if that info helps you
https://www.muffwiggler.com/forum/viewtopic.php?t=80465&postdays=0&pos torder=asc&start=25
Stereotactixxx
In theory, I think this could possibly be done by modulating the delay with a perfect saw, if the delay had a CV response optimized for this task. Getting it to actually work on your own delay though, is a whole different matter. I tried it with my Blacet Time Machine, and it didn't work very good at all, no matter how much I experimented with log and exp curves on the modulating saw. You may have more luck with some other delay.
nangu
Looks like Bastl Thyme can do something like this. It can tune the "fine" time knob based on incoming MIDI notes.

Check out this video at 5:03..

BugBrand
Square wave modulation, surely?
Clocks/gates in particular - unipolar (0 to +5V) rather than +/-5V
Just dial in the modulation depth to get the required pitch jump - if you do it with clock divisions then you can get it nicely rhythmic.
Homepage Englisch
Dave Peck wrote:

And, perhaps a bit counterintuitive, if the LFO sweeping the delay faster and shorter is a triangle wave, THAT can cause the pitch to instantly jump from one pitch to another. If you think about it, you can see why - because the triangle wave is straight slopes that don't change their 'rate' from one end of the wave segment to the other end, so this causes the delay to be sped up by one specific amount when the triangle is ramping up, and then the delay changes its rate and slows down to a different specific amount while the triangle is ramping down.

A SINE wave sweeping the delay will create those pitch 'swoops', but not because it is going up and then down, it is because it is changing rate and slowing down it's upward and downward movement as it crests the top or bottom of the wave, and then speeds up its upward and downward movement as it passes through the zero volt line. It is not just because the sine wave is constantly rising and falling, it's because the sine wave shape is constantly changing the RATE at which it is rising and falling.


This...just blew my mind.
Dave Peck
Homepage Englisch wrote:
Dave Peck wrote:

And, perhaps a bit counterintuitive, if the LFO sweeping the delay faster and shorter is a triangle wave, THAT can cause the pitch to instantly jump from one pitch to another. If you think about it, you can see why - because the triangle wave is straight slopes that don't change their 'rate' from one end of the wave segment to the other end, so this causes the delay to be sped up by one specific amount when the triangle is ramping up, and then the delay changes its rate and slows down to a different specific amount while the triangle is ramping down.

A SINE wave sweeping the delay will create those pitch 'swoops', but not because it is going up and then down, it is because it is changing rate and slowing down it's upward and downward movement as it crests the top or bottom of the wave, and then speeds up its upward and downward movement as it passes through the zero volt line. It is not just because the sine wave is constantly rising and falling, it's because the sine wave shape is constantly changing the RATE at which it is rising and falling.


This...just blew my mind.


Yup, same here when I first figured this out. I had an old Digitech rack effects box and I was creating a type of chorus patch with extreme modulation depth, and when I set it to 100% wet, I noticed that the audio signal from the delay output was 'jumping' between two pitches - one slightly sharp and one slightly flat, relative to the dry signal, rather than 'sliding' from sharp to flat and back again. This was because the Digitech used a triangle wave to modulate the delay in the chorus algorithm. Other delays that used a sine wave didn't create this effect.
HighLordFixer
What Digitech rack was that you used then?
Dave Peck
It was either a DSP128, a DSP 256, or a DSP 256XL I had all of them at one point and I don't remember which it was.
HighLordFixer
hmmm
have 256xl
will crawl behind the server rack with flashlight for taking look
also have some RDS time machines
7.6 has cv input
not sure about any others
AHA
Digitech 256xl has 2x footswitch inputs
completely forgot about that
thanks
ftp://ftp.digitech.com/pub/pdfs/Discontinued/Manuals/dsp256xl.pdf
RDS is probably the safer bet for using
Synthbuilder
Take a look at this page which looks at how clocking a BBD affects pitch:

https://electricdruid.net/investigations-into-what-a-bbd-chorus-unit-r eally-does/

It's a lot more complex than just the raw LFO waveform as one also needs to think about how the modulating signal affects the delay time. For example, whether the delay time is modulated in a linear, 1/x, exponential or some other sort of relationship.

Tony
Synthbuilder
Stereotactixxx wrote:
In theory, I think this could possibly be done by modulating the delay with a perfect saw, if the delay had a CV response optimized for this task. Getting it to actually work on your own delay though, is a whole different matter.


I think this is how the first pitch shifters from Eventide worked although they were digital devices. It is glitchy at the point where the sawtooth (or the digital counter) flies back. I guess one could do it with two delay lines and two modulating sawtooths (offset by 180 degrees) and quickly crossfade between the two delayed outputs to keep the part where it glitches muted.

I think it's going to need a 1/x relationship for CV to delay time to work properly.

Tony
MUFF WIGGLER Forum Index -> Modular Synth General Discussion  
Page 1 of 1
Powered by phpBB © phpBB Group