Totally clean repeats from a PT2399 based delay...can it be done?

[skiraly017]

Before I post my question, yes I did use the SEARCH function and I still could not come up with what I thought was a definitive answer. There was a link to a thread by Mark Hammer that looked helpful but I think it's long lost (I get "Page Cannot Be Found" errors).

I know that in PT2399 based delays like the Rebote it's supposed to simulate an analog delay sound by rolling off the highs a little more with each repeat. What I would like to know is if there's a way to change that. I would like no high end roll off at all, repeats as clean as the original. I saw a few threads where people have increased/decreased the amount of roll off to meet their needs but no mention of how to remove roll off completely. Did I overlook something? Any advice, direction and/or help would be greatly appreciated. Thanks.

[blanik]

as far as i know, the PT2399 will always have some noise with each repeats when the delay pot is over 50% (noise only present during the repeat) and this feature is inherent with the PT2399

for the highs getting rolled off it's made like that by design, on the Tonepad page there's also the Rebote and Rebote 2 (Rebote being (as i understand) a straignt PT2399 with no highs roll-off, the Rebote 2 has some high-cut but kind of crude and was perfectionned by Fp to give the Rebote 2.5... all this high-end roll-off might have started as a mean to get rid of the noise of the PT2399, i don't know maybe Fp would be better placed to awnser that...  :-)

R.

[Mark Hammer]

You don't want to feed an unfiltered delay signal back to the input.  Trust me.  The A/D : D/A conversion in the Princeton chips is pretty good, but not perfect.  As a result, there is error when quantization/encoding happens at input, and error at output.  When the regen signal is requantized, you get error on top of error.  That shows up primarily in the top end, which is one reason why the filtering is necessary.  It is also a reason why people complained bitterly about the tonal quality differences between analog and digital delays.

A second reason is also accumulated noise with multiple repeats.  A third reason, that I keep waving the flag for, is that multiple repeats pose an attentional distraction and increase "audio clutter".  A single lower-level repeat at short delay (slapback) can be fine with max bandwidth, I suppose.  But longer delays, especially with more than one repeat and something else going on at the same time, get real iritating real fast.

So, if you want to lift the lowpass filtering, don't make it permanent.  In the Rebote 2.5, to get a bit more treble in the recycled portion, you can sub a .0047uf cap for the .01uf one just after the 2k7 resistor by the Repeat pot.  That will change the rolloff from about 5.9khz to about 12.5khz.  Bear in mind that you have 6 poles of lowpass filtering prior to that so it won't make a huge difference.  What you might consider doing too is to re-insert the feedback signal at a different point.  If you look at the Rebote 2.5 schematic, the regen signal is routed back to the junction of the two 12k resistors and .027uf cap.  If you reroute it to the junction of the two 12k resistors and .01uf cap before the op-amp, you will essentially skip one pole of lowpass filtering in the process, preserving more high end for the regen signal.

I'm not expecting miracles, but you CAN do both those things without destroying the circuit in the process.  Personally, I would not call what comes out of that a "clean" signal, though.  Brighter, yes, but not necessarily in a good way.

[bent]

must verified that !

[aziltz]

what about boosting treble before the delay, in order to have more signal going into the feedback filtering?

[Nitefly182]

what about boosting treble before the delay, in order to have more signal going into the feedback filtering?

You would then have boosted treble on your entire signal which would potentially make your whole sound harsh.

I think the main issue is that you are trying to make the PT2399 be something its not. Youre asking it to do things it cant do. If you want crystal clear pristine repeats you are better off buying a commercial digital delay.

[robertreynisson]

A strange thing happened to my PT-80 delay from GeneralGuitarGadgets. It was working fine an than the out of nowhere it started making only one short delay. I changed the Compander SA571 and now the delays are back, but each one distorts or clips exponentially. Any ideas?