Analog tap tempo for PT2399

[earthtonesaudio]

Breadboarded and seems to work:
SIM

Press and hold the switch to reset, ramp up, and finally hold the control voltage.  Higher CV=longer delay.  Some of the values are altered to make it look better in the sim.  More appropriate values shown in the image below.  I have them this way in the sim because it's dreadfully slow to watch the circuit operation at normal speeds.

There are plenty of other ways of doing this, sure... but there's something satisfying about using the PT2399 itself to implement the tap-tempo control.



The unlabeled IC is a 555 wired as a Schmitt trigger.

[Taylor]

Fantastically clever. Looking forward to seeing this get verified. MCUs are cool, but circuits which can be dug up in years when special parts are gone, and still work, are nice.

[The Tone God]

Not really a "tap tempo" by which I mean what people refer to as "tap tempo" of tapping the tempo in and having the delay set. More of a foot switch controlled pot. Still neat. I remember doing something similar only to control wah circuit so I didn't need to build a rocker pedal.

Andrew

[earthtonesaudio]

Here's the same thing but with the latch-up prevention mod added.  I used a MOSFET instead of a BJT for the lower on-resistance.

@Andrew, you're right, it stretches the common definition of "tap tempo" somewhat.  True integrators are rarely used in effects, I'd like to see how you implemented your wah-rocker.  For a past circuit contest (design an effect with no knobs) I had the idea to make a booster with two pushbuttons, one for up- and one for down- volume.  I never got around to building one but if I eventually tackle the concept again it will probably use something like this, which is a simpler version of the integrator used in the PT2399 circuit above.

@Taylor, that's a nice way of thinking of it.  I really just did it this way for fun.  It would probably be more practical to use a higher-quality op-amp for the integrator and free up the PT's internal op-amp for audio which has much less strict performance requirements.

[The Tone God]

@Andrew, you're right, it stretches the common definition of "tap tempo" somewhat.  True integrators are rarely used in effects, I'd like to see how you implemented your wah-rocker.  For a past circuit contest (design an effect with no knobs) I had the idea to make a booster with two pushbuttons, one for up- and one for down- volume.  I never got around to building one but if I eventually tackle the concept again it will probably use something like [ which is a simpler version of the integrator used in the PT2399 circuit above.

I would have to dig through my notebooks but I think at that time it was something like a set of controls that charged or discharged a large cap which fed a S&H circuit. It had a couple of different modes where I could just step on the switch to ramp up then it would ramp back down when let go and if there was a spot I wanted in the sweep it could latch it with another switch. Another mode was where I would hold the switch then when released it at the point I wanted and it would latch. There were controls to set the ramp up and ramp down speeds.

These days I would just use a uC but thats just me.

Andrew

[earthtonesaudio]

Not really a "tap tempo" by which I mean what people refer to as "tap tempo" of tapping the tempo in and having the delay set. More of a foot switch controlled pot. Still neat. I remember doing something similar only to control wah circuit so I didn't need to build a rocker pedal.

Andrew

Your point got me thinking.  Delay pedals are about time, not frequency.  Tempo is a measure of frequency; time is in the denominator (beats PER minute).  So it does not make sense for a delay pedal to have tap tempo, that would be like advertising the maximum delay time in Hertz.

Therefore I conclude that "tap tempo" on a delay is factually inaccurate.  "Tap time" would be better.  I doubt it will catch on though.  People still call that Stratocaster lever a "tremolo" so it's hopeless.

[Taylor]

Oh, so is the delay time not going to be necessarily the same as the time you hold the switch down? In other words, you hold it down until the delay sounds about right? Or could you trim the response so that the time you hold the switch down will actually be approximately the delay time?

[earthtonesaudio]

Oh, so is the delay time not going to be necessarily the same as the time you hold the switch down? In other words, you hold it down until the delay sounds about right? Or could you trim the response so that the time you hold the switch down will actually be approximately the delay time?

The "slope trim" is supposed to let you dial in a ratio of exactly 1:1 between the button press and the delay time.  The range of adjustment may be to large though.  Right now it's set up so that with the pot rotation fully up, the ratio is pretty close to 1:1, but as you bring the rotation down, a shorter button press will result in the full delay time.  Another resistor between the pot and ground would narrow the range of adjustment to make setup easier.

[MrStab]

hi,

i don't wanna spam by making a new thread for a curiosity that's not on my project list just yet, so apologies for digging up a 2 year-old thread, but i've been eyeing this up for a few weeks now and i have some abstract questions before i go to bed (at 7am on a Tuesday. that's how i roll. lol).

at first i thought (and someone else asked, i'm sure) using a latch could make it more like a pseudo-tap tempo (or "tap time"!), but then i realised the resistance would just keep ramping indefinitely until you hit the switch again. so what about an electronically-latching momentary set-up with a "time-out" feature so it would unlatch the switch if a 2nd tap wasn't received before resistance hit an unusable limit, and maybe somehow the new resistance could only be released to the PT2399 after the 2nd tap? binary/decade counter CMOS comes to mind as a possibility (or just a simpler flip-flop system). as i say, it's all abstract and i haven't thought it through that much. seriously wish i had a breadboard. i know it just takes 2 ICs if you go the uC route, but where's the fun in that?!

anyways, just food for thought. cheers!

[earthtonesaudio]

Yeah, as drawn this is what you would hear when you press the footswitch:

1. delay time quickly goes to minimum value
2. delay time gradually rises until you release the switch

...somewhat different from conventional tap-tempo behavior.

You could use two integrators, one for storing the current value, and the other for accumulating a new value, and toggle between them with a latching switch setup.

Alternatively you could use a sample-hold and store the most recent valid value while you're acquiring a new value.

[MrStab]

much food for thought, thanks Alex. i'll need to do some reading on integrators and/or sample & hold ICs - i've seen bits & pieces about the place and it was only a matter of time before i'd have some reason to find out more about em.

[earthtonesaudio]

You should!  Those two building blocks are very useful in a lot of situations.

[jotajota0003]

Breadboarded and seems to work:
SIM

Press and hold the switch to reset, ramp up, and finally hold the control voltage.  Higher CV=longer delay.  Some of the values are altered to make it look better in the sim.  More appropriate values shown in the image below.  I have them this way in the sim because it's dreadfully slow to watch the circuit operation at normal speeds.

There are plenty of other ways of doing this, sure... but there's something satisfying about using the PT2399 itself to implement the tap-tempo control.



The unlabeled IC is a 555 wired as a Schmitt trigger.

http://www.diystompboxes.com/smfforum/index.php?topic=108251.msg986069#msg986069

[jotajota0003]

http://www.diystompboxes.com/smfforum/index.php?topic=108251.msg986069#msg986069

[jotajota0003]

[elkulon]

Well, well...

At the end of it all... does this project work or it remains only as schemes and simulations???

[earthtonesaudio]

Wow, a blast from the past!

elkulon:
I built this, minus the output buffer.  I scoped the output to verify that the ramp-hold circuit worked (it did).  Assuming the theory of operation of the PT2399 is correct, this circuit should work.

(notice we have combination of "assume" and "in theory", which should be taken to mean "keep expectations low")

The buffer is required, in theory, because pin 6 of the PT2399 is speculated to be the inverting input of an internal op-amp, with a feedback resistor connected, making the pin a low-impedance node.  Without this buffer it's expected that the circuit would behave non-linearly, or maybe not at all.

This particular buffer is suggested because it can pull enough current (up to 10mA maybe) at voltages that may go to 0.  One could substitute a regular op-amp buffer if the op-amp is powered from a bipolar supply, or by lifting the ground rail of the PT2399 to some above-zero voltage.  I thought the higher parts count of the buffer as drawn was a good choice given the higher cost and complexity of these alternatives.

[jotajota0003]

well, I designed and assembled this your idea, actually I swept the entire internet looking for a tap that worked. But this his idea after some modifications, it worked! I am using this on my delay pedal.

[earthtonesaudio]

@jotajota003, awesome!  Did you integrate the rest of the PT eventually?  It'd be cool to see a video if you get around to it. 

[jotajota0003]

I will make a video to send you