PT2399 headroom. 3.5Vpp before it clips? or 2Vpp?

[nonost]

Hi. I'm messing quite a bit with PT2399 delay sounds. I've found that there's no consensus about the chip headroom. In electrosmash.com there's a great analysis. These guys did their job. About the headroom we can read:

The recommended input signal levels are not specified in the datasheet. Testing it, good input levels are around 0.5 to 1Vrms (1.4 to 2.8 Vpp), with these levels the THD is <0.3% (following the datasheet) which is good to minimize the noise in the circuit.

If the input levels get higher, clipping will happen. The chip is quite tolerant to signal levels but the THD will raise and the sound quality will degrade. With the input signal with peaks over to 1.3Vrms (3.7Vpp) the THD will go over 0.5% causing clipping and audible distortion

Here the link: https://www.electrosmash.com/pt2399-analysis

However others users put this threshold so much lower. I've found 2vpp many times. I know about the led limiter trick at pin 7. Red/green leds are the ones used there. These leds are below 2Vf or so...I don't know if I'm just not getting something or what.

Any thoughts?

Cheers!

[antonis]

For +5V supply, internal op-amps are biased at +2.5V..
Considering their outputs about 500mV lower/higher than supply rails (4.5 to 0.5V), 1.4Vrms should be marginally OK..

[merlinb]

It's not an exact figure because you also get slew limiting at higher frequencies, so it depends what figure you choose as your upper limit. More bandwidth buys less headroom. I came up with a practical figure of about 3.5Vpp up to about 1kHz in my tests.

[anotherjim]

I don't think I've ever seen anything try anything like the classic NE570 compander technique with the PT2399 as was done in the more serious  BBD delays, but the PT is a cheap chip for cheap delays and I doubt anyone is drawn to making the effort to add much complexity to it.
So I tried a cheap companding option in a digital delay...

Hiding in that over-complex mess is the Compress section (which looks like a feedback diode clipper but the series resistor means it reduces gain at peaks) and the Expand which is the reverse diode placement acting on the opamp input on the way out of the delay. At the same time, there is also treble pre-emphasis on the input side and de-emphasis on the output. Since there probably already are these opamps in a PT2399 ins and outs, it's possible to try without adding much cost.

Anyway, it sounded ok in what was an 8bit MCU project mainly done to prove I could do it.

[ElectricDruid]

PT-80 is one example that springs to mind for PT2399 with 570/571 companding.

http://www.generalguitargadgets.com/effects-projects/modulationecho/pt-80-delay/

This pedal was Scott Swartz cloning some old delay (DOD something-80? Don't remember any longer, and I can't find his site any more). He kept the filtering and the companders so the sound is the same, but replaced the BBD with the PT2399. And by all accounts it makes a nice pedal.

In general the idea makes sense. The PT2399 has much the same limitations in terms of delay time/bandwidth/noise floor as the original BBDs. Companding can't fix all of those problems, but it helps a bit!

[anotherjim]

I have an idea the 570 is just about obsolete although you can still find them. On-Semi has one version (SMD) on final.

[ElectricDruid]

I have an idea the 570 is just about obsolete although you can still find them. On-Semi has one version (SMD) on final.

No, Coolaudio (Behringer's chip arm) make them as one of their old analog clones, along with the LM13700, the SSM2044, CEM3340 and others. They're only interested in multi-thousand dollar orders though, so getting hold of them might not be straightforward.

So it might be obsolete, but it's still in production!

[Krystal]

Just stumbled upon this : https://www.reddit.com/r/diypedals/comments/p4zqtf/joyo_digital_delay_inside_circuit_analysis/
It seems the Joyo Digital Delay is a Pt2399+SA571 circuit, it might just be a PT80 in fact.

[anotherjim]

This just goes to prove, that if I can think of it, several have already done it.