Distortion on the first repeat in Deep Blue Delay Clone

[Rethfing]

Ooooho, you're my Master I would buy you a beer if I could, Mike! I hear distortion now only when I hit really, I mean REALLY hard on the strings, unusually hard with normal playing and with Mix knob turned around half (which is a bit too much for using it actually). I guess lowering R3 and raising R5 a bit would solve the problem completely If you know the proper values then it would be awesome, but even now the delay works great, so thanks again!

[R O Tiree]

Set R3 equal to R2 (180k on the schem) and R5 equal to R4 (22k on the schem).

Are you playing a humbucker-equipped guitar though this thing?  Humbuckers typically push out about +/-1V, maybe a bit more, so a gain of 2 at the input buffer sends the total to over +/-2V which is probably what is overloading the PT2399.  The output buffer/mixer has a gain of 0.5 to reset the levels so Vin = Vout.  The designer did this so that weaker signals from single-coils get boosted going in, so noise added by the PT2399 is small(ish) compared with the signal going through it and then is reduced by half in the mixer.  It's not the best method, by any means, but it's better than a poke in the eye with a sharp stick.

[Rethfing]

I did this mods after your previous post. There is still little distortion on Mix above half and when I hit string really hard. Will lowering a bit R3 and raising R5 help?

Yes, Im using active pickups (cheap ones that come stock with Schecter Demon, good for slaughter though). This is a bit weird because my previous built on same PCB worked very good without any mods. But I did use different PT2399 (it came with PCB then, but for this build I ordered few from some other shop).

[slacker]

I would leave R3 and R5 how you have them now and instead make R8 bigger that will only lower the volume going to the delay and not affect the dry sound. If you keep cutting and then boosting the dry signal you're just adding noise.
This will make the output of the delay quieter as well so if it's then too quiet make R17 the same proportion smaller, so if you double R8 half R17. You might find making R8 bigger makes the delay sound duller, if it does then make C6 smaller to compensate.
Another thing to try would be make R11 smaller the distortion could be coming from that stage of the PT2399 not the input.

I'd also have to agree with Mike C, your pin 7 and 8 voltages look wrong, I would have thought they should be much lower, but I'm not sure if them being higher means there's anything wrong.

[Rethfing]

Well I guess I just leave it all as it is now Im happy with the sound so I see no point in messing with so many parts again + I hate desoldering parts

About these voltages - I have no idea what could be wrong.

[blue_tokai]

Are you using the same power supply as you did when you breadboarded this?
I see you have 8 volts into the regulator... I was wondering how many milliamps the power supply is delivering? The fact you are getting distortion on the first hard hit string almost sounds like a bit extra draw on the power supply then its stabilizing??!!

[Harold]

My freshly built DBD clone distorts too, especially when hitting it hard with my EP booster: "tube screamer like distortion".

I'll have a look into the proposed mods posted above, as this renders the delay a bit useless.

PS: I used the Sea Urchin schematic.

[Harold]

The gain on the input buffer is 2, at present. Try dropping R3 to 180k as well?

That will make things very quiet at the output, though, so you'll have to raise R5 to 22k to balance things out again.

Yes, this fixes my overly distorted repeats! 

Could it be that different brand PT2399 have a different "headroom"? For example, this batch I have on my bench right now don't require the digital and analog ground to be connected (pin 3 and 4), but the previous batch I had wouldnt "boot" without them connected.

[dromilious]

Lets make active this topic again !If i change the 360K and the 180K resistors i notice a different tone at the output.This is due to the low impedance of the input?
As you can see the atached schematic i replaced them with 4.7K and 18K the feedback resistor.So now the new input impedance is 4.7K?
I think the 1M pop resistor is not considered as the input impedance but the 4.7K going to the inverting pin of the op amp.



Then if the low impedance is a problem if it would be better to use a non inverting input?

[slacker]

Your input impedance is now 4.7k in parallel with 1M which which makes slightly less than 4.7k. This is  far too low, why have you chosen those values?  Yes you could make the first stage non inverting, that means that the clean signal is then inverted over all which probably isn't a problem.

[dromilious]

Tonepad delay as i can see uses 510 K resistors to increase the impedance.This could be a good solution i think.I don't want to increase them more to avoid thermal noise.Non inverting input is a nice solution but i am afraid of out of phase signals.

[slacker]

Another thing you could do is keep the low resistor values and put a FET buffer in front of it. That keeps everything in phase and avoids the thermal noise problem caused by needing large resistor values in the opamp stage.

[dromilious]

After many tests i saw that changing the resistors R2 ,R3,R5,R7 affects the tone reaction.For instance if i change the R7 resistor to 47K to have a gain of 2 ,i notice a high cut reaction when i plug before the circuit a distortion circuit .

The correct value of C4 i use is 1uf not 22nf.

[Renegadrian]

I made one that distorts as well  tried to put a booster after it and I get an overdrive...

[midwayfair]

It has nothing to do with the "brand" of PT2399. It's because the PT2399 runs on 5V, which is not a ton of headroom.

There's an actual solution I discovered a while ago, but I guess everyone around here's forgotten about it

Look at pin 7 here and the read the How It Works section:

https://docs.google.com/document/d/1P7uOYFRH6ZW5eGoquhAiMQMyUU76bod_gNKkiSlPNJM/

It is absolutely 100% impossible to overdrive the PT2399 regardless of what your dry path is doing if you do that.

Note that there may still be some harmonic distortion present in the PT2399 that's caused by something completely different -- it increases with the delay time, has to do with the chip's memory, and there is absolutely nothing you or anyone else can do about it except to use a more sophisticated digital chip like the FV-1 or SHARC or something that takes actual programming knowledge.

[Renegadrian]

ok I know that the 2399 is not the perfect chip, but other delays I built (rebote 2, 2.5, and DBD) didn't act that way, yeah they had this "low-fi" repeats but not that kind of distortion at the first repeat...I guess I have to try that led on pin 7, easy mod hoping it will take out some fizziness...
EDIT
tried a green led paralllel to the pin 7 cap, it didn't work...

[midwayfair]

ok I know that the 2399 is not the perfect chip, but other delays I built (rebote 2, 2.5, and DBD) didn't act that way, yeah they had this "low-fi" repeats but not that kind of distortion at the first repeat...I guess I have to try that led on pin 7, easy mod hoping it will take out some fizziness...
EDIT
tried a green led paralllel to the pin 7 cap, it didn't work...

Does the LED light up when you play something? If it just blinks a little when you first play a note or two, then it's not the signal level causing your distortion.

Are you sure the filtering capacitors in your delay line (all the stuff pins 13-16 up to the repeats pot) are correct? What is the delay time you're working with (measure the exact resistance and check the datasheet)?

You've been fiddling a lot with the input stage to deal with an issue in the repeats of the delay. You will get noticeable distortion on the repeats themselves (not the analog circuitry of the chip) above about 300mS unless you have the bandwidth EXTREMELY constrained.

If it's not (a) the signal level or (b) a lengthy delay time, then you've probably actually got something wrong or a bad chip (which does happen, particularly if you're paying 50c each for seconds from an ebay supplier or Tayda).

If you have other delays sitting around you can at least pop their chips in this delay to check that it's nothing to do with the chip.