Does this Sea Urchin Delay sound right to you?

[therealfindo]

I put together a delay using the MB Sea Urchin schematic (here)
But I think there's a problem.. with all the pots at 100% clockwise / full it sounds like this (excuse the "playing").

Now, am I right that it should be oscilating when FDBK is at 100%?
Do the repeats also sound dark and muddy - should they not be brighter and match the clean signal?
Should the delay be a bit longer as well?

I've run out of time today to keep messing with it, but if anyone has a suggestion of where in the circuit to look for the problem, I'd appreciate it!

Cheers.

[blackieNYC]

It's not a problem, it's a feature!  
The pt2399 is limited in its fidelity I believe.  Someone will be along in a minute to tell you what you could change in the caps and resistors on pins 11-16 (there is LP filtering going on in there). Likewise, the thing is limited in delay time. You can daisy chain them, but that fidelity issue is going to come back at you. The design seems to have around purpose of recreating "tape slap echo" (a pretty good attempt)and covering up weaknesses in the pt2399.   Analog bucket brigade delays also decreased fidelity with repeated echoes deteriorating, and its a nice sound.  Rooms and venues of every kind have echoes that deteriorate in fidelity if you think about it. For real doubling, like a stereo split kind of thing, a digital delay is what you want.  
It's data sheet should have cap and R values that are a little closer to full bandwidth, but in reading up on this I think just about every pedal design uses the LP filtering.  
It won't do very short delays either. A minimum of 20-30 ms.  
Sounds just like mine.  Which i love but I knew what to expect.  You'll find a very nice use for it.  They're great!

[therealfindo]

ah, ok.. in that case, great! 

[kingswayguitar]

in any case...
nice demo

[therealfindo]

hmm... after watching some more youtube videos of deep-blue type delays... I don't think mine is  right. It should be oscillating with the repeats all the way up - my demo is with rpts on max and it's hardly got any.. 

see, like this

which resistors / caps should be I looking at changing? Ideas?


Thanks.

[karbomusic]

Analog bucket brigade delays also decreased fidelity with repeated echoes deteriorating, and its a nice sound.

I was around when the only delays available to mere stomp box mortals were old school delays where each repeat degraded; you simply had no choice and I hated them... Oh the nights I dreamed of a delay where the repeats were pristine copies of the original. That day finally came where I sort of had that, oh joy. Then I realized that the degradation, was very much a worthwhile feature as it gets the repeats "out of the way" of the notes being played and pristine copies will walk all over the main notes depending on what you are playing. Of course then, I couldn't know what I didn't know. Now, unless it is a special use type repeat where the repeats are not happening at the same time as the next note I always want the repeats to degrade in quality. Who'd a thunk?

Obviously a YMMV thing but I always like to reflect on my perspective of my needs back before my needs were possible.

[blackieNYC]

Oh- I didn't address the number of repeats.  Hmm.  Maybe that 50k value is reducing them.  Can't tell you anything about the deep blue, but have a look at the modulus and see the 100k feedback pot.  And- the diodes.  Whatever it is in the circuit that gives you oscillation( and the modulus really does, very well), the two diodes are a perfect limiter, holding the oscillations at unity gain so they don't get out of control.  It will oscillate forever.  I don't see other differences, but look carefully.  Maybe max out the 50k pot out and add a 47k resistor to gnd and see if that does it. You would then need a 100k pot.  Don't skip the diodes.  I think my feedback could destroy things if not capped at that 1.2 volts
http://www.diystompboxes.com/smfforum/index.php?topic=72527.0

[therealfindo]

Thanks..

I tried it, but didn't notice any change.

what I did discover was another bug (or perhaps the cause of this one) namely, when I unplug the power I get a nasty squeal for a second or two, and then the dely doesn't work when I plug it back in.. when I removed and reinstert the regulator it works again. So I'm guess there's a problem with one of the capacitors in there, right?

[blackieNYC]

The squeal - is this thing operating around unity gain, or is it louder?
Since faulty schematics never make it to the Internet, you probably need to comb thru all your parts and soldering with a magnifying glass.
The reset problem sounds like the pt2399 lockup problem, for which there are some sophisticated prevention circuits, but this shouldn't happen if R18 is as high as 2.7k Ive been told   
I haven't experienced the lockup.  You can search that. But if the chip restarts and you hear delay, does that guarantee the chip is ok?  Anyone?  Does he have a typical lockup problem and a separate audio problem? 
Got another PT to swap in by any chance?

[tranceracer]

Could be a micro solder bridge someplace.  I've had success w/ flaky circuits by scraping the rosin out between pads.

[therealfindo]

(yeah, squeal is about unity I guess... I tried raising R18 to 4k7... still the same problem..
As far as I understood my googling.. locking chips don't work anymore , but this still works, once the regulator is taken out and put back in..
I can't see an solder bridges..

grr...

[therealfindo]

aha!! The chip-reset issue seems to have been fixed via this answer - soldering pins 3 and 4 together..
Still not oscilating, but it's fine as it is now, actually... thanks for your help.

[loylo]

Your repeats sound really muddy so you may kill  high frequencies in the feedback loop because of bad cap valuea  (or bad resistor), making a low pass filter.
A low pass filter in the feedback loop would prevent the possibility of self-oscillation.
So you may need to triple check your components values.