[sdiy] PT-80 delay for Synth Modular

[Tom Wiltshire]

Csaba,

Like you said, Scott Swartz's PT-80 delay is delibrately setting out  
to mimic the sound of older BBD-based delays, and that means quite a  
bit of lowpass filtering.

Either BBDs or the PT2399 are both sampling the signal and then  
reconstructing it, so you need a lowpass filter after the delay line  
to smooth the sample steps out of the signal. The highest you should  
really make the filter frequency is half the sampling frequency (the  
usual nyquist limit). How high this is depends on how long the delay  
is. Shorter delays means a higher clock frequency and wider bandwidth  
(eg higher filter cutoff and better quality). At longer delays, the  
clock is running slower and you need a lower filter cutoff to help  
smooth out some of the digital steps in the waveform. At really low  
clock frequencies, you're getting something close to Tom Bugs  
"Bugcrusher" sample-rate reducer effect (http://www.bugbrand.co.uk/).

The PT2399 datasheet has a really useful table which shows the clock  
frequencies and bandwidths. If you set the filter frequency too high  
for a given bandwidth, you'll start to hear more sampling distortion,  
but nothing terrible will occur in the world, so experiment with it.

The PT2399 IC has op-amps on it, and these are used as filters, but  
the PT-80 circuit has an additional Lowpass stage (the mess of Rs and  
Cs running up to the 2N5088 tranny). If you wanted to lose some of  
the "muffled" sound you could remove this stage entirely, I'd have  
thought. Doesn't Scott make some comment on his page about copying  
this filter from some venerable analogue pedal on his page? It's  
there for historical accuracy rather than technical necessity.

Hope this helps. I've used this chip a bit, but mainly for chorus  
(short delay equals pretty good bandwidth and quality) so I don't  
know exactly how far you can push it.

Regards,
Tom



On 30 Jun 2008, at 11:05, Csaba Zvekan wrote:

> Hi Group,
>
> I have built two PT-80 delays from General Guitar Gadgets .com  
> http://www.generalguitargadgets.com/index.php? 
> option=com_content&task=view&id=125&Itemid=26 .
> Now this is originally a stomp box for guitars and has probably a  
> different input gain. I have tested it with my DIY Modular Synth it  
> works . But I am not quiet satisfied with two things.
> 1. The Wet/Dry ratio is a little  too dry. Meaning the original  
> signal is too loud ( all though the effect level is all the way up  
> to the max).
> 2. The Sound of the delay is rather dull and has not quiet a  
> brilliant sound.
>
> My questions are : Where do I modify the circuit to have less  
> original signal and more effect signal? How can I have a more  
> cleaner delay sound ?(If this is even possible with the PT2399).
> I also have to mention that I didn't have the 2N5088 high gain NPN  
> Transistor and was helping myself with the BC550C and BC547B. They  
> sound almost the same.I have purchased some 2n5088 thinking that  
> might be my problem . Can somebody verify that?
>
> here is the schematic of the delay. http:// 
> www.generalguitargadgets.com/diagrams/pt80schem.pdf
>
> Also I have some other analog  Guitar effects like  
> Chorus,Frobnicator,Tremolo and distortion built and was wondering  
> if I should modify the input stages differently for use with high  
> level synthesizers signals. And how would that look like ?
>
> Any input is of course greatly appreciated.
>
> Thanks
>
>
> Csaba
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>