Can the PT2399 delay CV's/ signals down to DC?

[Processaurus]

Hi, I was wondering if anyone had tried this or had an opinion about the suitability, I was thinking it would be cool to have a modular synthesizer module that was a PT2399 based digital delay that one could plug control voltages (envelopes, gates, pitch cv, etc) into, and it would spit them out after the delay period?  Kind of like midi echo.

In the app notes in the data sheet it shows the basic digital delay for audio, in which the audio is decoupled through a 4.7uF cap, but not re-biased.  Does this mean it is internally biasing the signals to 2.5v (VCC/2)?

Thanks for any input on this, or if a BBD circuit might be more suitable.

[merlinb]

Hi, I was wondering if anyone had tried this or had an opinion about the suitability, I was thinking it would be cool to have a modular synthesizer module that was a PT2399 based digital delay that one could plug control voltages (envelopes, gates, pitch cv, etc) into, and it would spit them out after the delay period?  Kind of like midi echo.

The PT2399 is quite happy working with DC signals, so yes, this would not be difficult to do. Obviously you need to scale the CV down a bit before feeding it to to the PT, DC couple, and amplify it back up again to synth levels afterwards, and you would need to think about the effect of the inverting input amplifier. But that's all back-of-an envelope stuff. You wouldn't need such elaborate low-pass filtering of the input signal as for audio applications, which makes it all quite simple.

You need to find out what range your synth CV covers. Some brands seem to use 0-10V, others -5 to +5V, others different again...

In the app notes in the data sheet it shows the basic digital delay for audio, in which the audio is decoupled through a 4.7uF cap, but not re-biased.  Does this mean it is internally biasing the signals to 2.5v (VCC/2)?

Yes the internals are biased to 2.5V. Some more info here: http://valvewizard2.webs.com/PT2399_Data_Notes.pdf

[scratch]

it has been done ... check out the thread here ...

http://electro-music.com/forum/viewtopic.php?t=13490

[earthtonesaudio]

The PT2399 is severely under-utilized for non-audio purposes. 

[Processaurus]

Thanks very much for that, Merlin.  Your notes and diagrams especially on the chip are helpful as well. It is good to know the headroom is limited to 3.5v max.  As the synth (a homemade eurorack affair) uses 5v for the control voltages, I'm thinking about the input being DC coupled to the first stage (with the appropriate static/overvoltage protection ), and with the first internal inverting amplifier, having 2:1 reduction in gain so that it also biases it to the 2.5v, and then at the end, after the delay, using an external opamp running on +/-12v to invert and amplify it back 1:2 and hopefully get the same looking CV that went in coming out.  It will be cool to try a feedback/regen control, hopefully there's a way to do it where it doesn't drift as it decays.  Might be a good idea to buffer the input so that there is a known low impedance driving the delay's input signal, to guarantee it reacts as expected with a feedback signal.

it has been done ... check out the thread here ...

http://electro-music.com/forum/viewtopic.php?t=13490

Unless I'm missing something, they seem to be talking about a voltage controlled delay, not a control voltage delay.

[frequencycentral]

Really interested in what you're doing Ben, I'm in the process of building a Euro modular myself. And of course,  voltage control of delay time is a must too, and easy to implement, though the 'drift' as you change delay times would be, err, 'interesting'.....Anyways, I'll certainly do a layout for this once you've done the donkey work. 

[scratch]

sorry indeed, a Voltage Controlled Delay ... I figured out much later that's not what you were talking about.

[Earthscum]

I like the idea a lot. Basically what is being asked about in the OP is: Taking the original CV, for example, to one filter, and controlling (again, just example) a parallel filter with a CV that has a specific time delay, rather than a phase delay. Your second filter would always lag behind by the set time delay.

[earthtonesaudio]

Do it.

[Earthscum]

What P is asking, and I am wondering as well, is can you DC couple the signal to the PT? Could you hold the input at ground (or some voltage) for an indefinite period of time?

[Processaurus]

What P is asking, and I am wondering as well, is can you DC couple the signal to the PT? Could you hold the input at ground (or some voltage) for an indefinite period of time?

Yes we can, take a look at the audio delay circuits in the princeton datasheet, as well as Merlinb's excellent diagram that shows how the internal opamps are hooked up.  There is only the small problem of headroom, it cannot process 0-5v directly, they must be scaled down and then amplified back up after the delay, or they will clip, as the delay IC does not have rail to rail operation.

[earthtonesaudio]

Could use the clock output to drive a tachometer, into a VCO, and send that through the delay line.  Adjusting the delay time would simultaneously adjust the LFO frequency, so the resulting delayed output could be described as relating to the primary LFO in terms of phase delay rather than time delay.

You may remember the "continuously variable phase" LFO thread from a while back, Ben.  This would be sort of like that.  To get the adjustable phase you would have to make the VCO gain or tachometer time constant variable.