Reverb with more than two PT2399

[samhay]

I was working on a reverb with 3 PT2399s. The idea was to use the topology in the patent that Merlin posted and to be able to independently vary the 3 delay times - an x,y and z if you will. It was a bit of a monster, and it was a long way from working how I wanted it. About the same time, Merlin introduced his 2 chip reverbs and the smaller Belton bricks came out. I threw the towel in and moved on.
As an alternative, if a belton brick-based reverb is not enough, then you could either add a PT2399 chip and/or another belton brick (perhaps 1 long + 1 short). It gets a little expensive, but they keep the parts counts a lot more managable.

[slacker]

Interesting, I like how they do the modulation.

[artifus]

Interesting, I like how they do the modulation.

is that just a 40106 style square wave? *edit* ah - modulating one of the pt filters? so the filters cap shapes the wave a little? guessing here...

[samhay]

Interesting, I like how they do the modulation.

is that just a 40106 style square wave? *edit* ah - modulating one of the pt filters? so the filters cap shapes the wave a little? guessing here...

It is strangely implemented, but the LFO (allegedly) wobbles pin 6 at ~1 Hz - you can get a full copy of the patent here: http://www.google.com/patents/US20090003614 (see pp 4/5 and 5/6).

[merlinb]

is that just a 40106 style square wave? *edit* ah - modulating one of the pt filters? so the filters cap shapes the wave a little? guessing here...

Yes, it's a simple Schmitt oscillator, and the square wave is integrated a bit to round it off (making it into the classic 'shark fin' waveform), which then modulates the current sucked out of pin 6 (the delay pin).

[artifus]

i'd missed that pin 6 connection... thanks guys.

[seedlings]

Thank you for the Belton link.  I may end up spending the $18 to order that unit if my attempts don't work out.

CHAD

[deadastronaut]

Ive often wondered about using 3 or 4 to get a long delay by doing something i did years ago with a few delay pedals wired up
so that the delay is "accumulative" ..... E.g .  Each pedals delay time is added to the next etc. This gives a really long loop like delay.
and would be quite clean too. Ill have to try and remember how i had them wired up....cant remember off the top of my head, but it was pretty cool.

ok,this is what i meant... i got my friend to draw this up as he has several delay pedals (but is not an artist ) ...i remember getting this to work too, as long as the delays are set 'wet' it works and you get

a really long delay...each delays time is added to the next...very cool.

now if this could be applied to 3/4 pt2399's...for long delay/reverb...  4 pt2399's at 500ms...= 2 sec delay..


comprend'e?..

[seedlings]

You can see the Belton circuit here:
https://patentimages.storage.googleapis.com/US8204240B2/US08204240-20120619-D00007.png

Basically it is two delay lines in parallel, both feeding into a third one which is modulated a bit. They all have feedback too, of course.

Would it be illegal in any way to put two PT2399 circuits in parallel, feeding a third PT2399 circuit ... to borrow this patented belton topology, and sell millions of units and become rich and famous?

CHAD

[artifus]

i guess it would just be another way of generating animation , pre delay and early reflections. remember what we are faking here...

*edit* if i had that many pt's to play with i'd be experimenting with stereo fx.

[Mark Hammer]

In principle, one could use any number of these (now cheap) chips to produce a thick reverberant wash.  However, several considerations would need to be factored in:

1) Layout, layout, layout:  The possibilities for heterodyning increase as the number of different clocks (and their rates) accumulate.

2) More musical reverbs come from avoiding anything that sounds like standing waves, which means that you want to avoid mltiple delays that are multiples of each other.

3) High-frequency energy is eaten up by imperfectly reflective surfaces, such that later reflections are more treble-deficient than early ones.  Filtering needs to be used to produce/mimic that.

[artifus]

3) High-frequency energy is eaten up by imperfectly reflective surfaces, such that later reflections are more treble-deficient than early ones.  Filtering needs to be used to produce/mimic that.

that's kinda taken care of by the pt's inherent charms/limitations with regard to filtering from what i've read so far? only just recently got a couple of pt's to play with, looking forward to having the chance to breadboard.

*also* http://blog.buahdua.com/?p=67

[Mark Hammer]

3) High-frequency energy is eaten up by imperfectly reflective surfaces, such that later reflections are more treble-deficient than early ones.  Filtering needs to be used to produce/mimic that.

that's kinda taken care of by the pt's inherent charms/limitations with regard to filtering from what i've read so far? only just recently got a couple of pt's to play with, looking forward to having the chance to breadboard.

Don't confuse fixed lowpass filtering with progressive lowpass filtering.  The idea is that each subsequent "reflection", whether produced by a secondary/tertiary delay chip or via feedback from/to any of them, should have less treble than the reflection that preceded it.  No matter how dull the delay sound started out to begin with, it needs to get duller as it dies out.

If you have an analog or digital delay (such as PT2399-based) and yu have a way of adding some lowpass filtering to the recirculation/feedback/repeats path, so that each iteration is duller than the previous, you'll find it makes a heckuva difference in the naturalness of the resulting sound.

[artifus]

i thought that's what i said? 

[Mark Hammer]

I musta missed it.  Still not seeing it, but as long as we're on the same page, no harm done.

[artifus]

*re-evaluates ontological insecurities*

[artifus]