Dual PT2399 with tap tempo

[Marshall Arts]

Hey there,

I am almost done with the design of an arduino nano based tap tempo control board with modulation. Control of a single PT2399 delay works perfect, also modulation works quite nicely. Modulation depth is controlled via a second digital pot. I programmed a calibration routine for the whole setup, so tapped tempo and resulting delay will be very precise, disregarding the 30% tolerances that digital pots come with. With this being said, I think that the setup should be able to work for multiple PT2399 delays like the MBP Deathclaw as well. Can somebody confirm this?

Quick explanation of the schematic (excerpt) below:
T2 is fed with an oscillating signal, therefore pin 7 of the MCP41010 is fed with an oscialting resistance against ground. The digitally controlled resistance of the MCP41010 "mixes" the oscillating signal with a fixed ground signal from Pin 5/6. This "mix" is fed to the deathclaw resistors (basically a "shaking ground" :-)).
MCP41050 again controls the two BC550s in the deathclaw schematic, controlling the average delay time (which again varies with the "shaking ground" around that value).

I think that compared to other approaches (http://www.madbeanpedals.com/forum/index.php?topic=8053.0), this should work, as I have control over the MCP settings via the arduino.

[flanagan0718]

i wish i had something to offer other than my money when this is finished...

[Marshall Arts]

i wish i had something to offer other than my money when this is finished...

I wish somebody would offer to try it with a deathclaw (or any other similar approach). I am quite happy with my deep blue delay (single PT2399) and the way this thing works with that. I have buttons for tap tempo, subdivision, modulation on/off, up to three presets, rotary encoders for delay time, depth and speed. Still thinking about PCB design (as small as possible with a lot of offboard wiring or for a 1590B with pcb-mounted pots).

[daleykd]

Considering I know jack squat about this stuff (though I'm a C# developer by day), is this more accurate than the Taptation?

[Marshall Arts]

Considering I know jack squat about this stuff (though I'm a C# developer by day), is this more accurate than the Taptation?

I never built the taptation, but it works like this: The microprocessor is hooked up to the input and output of the delay. It sets the digital pot to zero resistance and measures the delay time by sending out audio beeps and measuring the audio delay it receives from the delay output. This is repeated for increasing values of resistance for the digital pot up to the maximum resistance. Tolerance of the digital pot (and the pt2399) thus does not matter, because the controller stores the setting specific to your components.

This calibration has to be done only once.

How do you like that? 

[Marshall Arts]

Considering I know jack squat about this stuff (though I'm a C# developer by day), is this more accurate than the Taptation?

I should mention, that modulation with a PT2399 will always have a (neglegtable, IMHO) effect on the perceived accuracy. But with unmodulated delays, this IS precise.

[daleykd]

I should mention, that modulation with a PT2399 will always have a (neglegtable, IMHO) effect on the perceived accuracy. But with unmodulated delays, this IS precise.

So, I can use a DBD with this with precision?  TAKE MY MONEY NOW!

[Marshall Arts]

I should mention, that modulation with a PT2399 will always have a (neglegtable, IMHO) effect on the perceived accuracy. But with unmodulated delays, this IS precise.

So, I can use a DBD with this with precision?  TAKE MY MONEY NOW!

Yes, you can. For the time being, you would have to live with THIS, though:

[daleykd]

Yes, you can. For the time being, you would have to live with THIS, though:

Want me to help with PCB layout?

[Marshall Arts]

Yes, you can. For the time being, you would have to live with THIS, though:

Want me to help with PCB layout?

I don't think it is necessary, thanks - It is all drawn up in eagle already, minor bugfixes (coding/breadboarding/back to eagle,...). I am planning to make the software open source anyway, once I have mine running (so people can tweak values like maximum delay, default modulation values for zero delay (basically a chorus or a leslie ;-)), colour settings for the multicolor LED, subdivision factors (triplets, dotted eight,...), increase/decrease values for the rotary encoders,... The code is well documented, anybody can tweak it.

Like the idea?

[samhay]

What's the minimum delay you can get with this setup - my guess is that the 'shaky ground' will add substantially to this?

[daleykd]

I don't think it is necessary, thanks - It is all drawn up in eagle already, minor bugfixes (coding/breadboarding/back to eagle,...). I am planning to make the software open source anyway, once I have mine running (so people can tweak values like maximum delay, default modulation values for zero delay (basically a chorus or a leslie ;-)), colour settings for the multicolor LED, subdivision factors (triplets, dotted eight,...), increase/decrease values for the rotary encoders,... The code is well documented, anybody can tweak it.

Like the idea?

Love it!

The thing I was going to try with a PCB layout is to match the same footprint as your arduino and sandwich them together with pin headers.  Not sure if that'd help or not.  I just HATE offboard wiring.

[Marshall Arts]

What's the minimum delay you can get with this setup - my guess is that the 'shaky ground' will add substantially to this?

I have a physical switch to switch off modulation completely. I don't know how short the delay is with a minimal setting (there is some), but it is not audible (so, way below a slapback). It adds just a bit of "space" to the tone. I will post a video to give you a better idea.

[Marshall Arts]

I don't think it is necessary, thanks - It is all drawn up in eagle already, minor bugfixes (coding/breadboarding/back to eagle,...). I am planning to make the software open source anyway, once I have mine running (so people can tweak values like maximum delay, default modulation values for zero delay (basically a chorus or a leslie ;-)), colour settings for the multicolor LED, subdivision factors (triplets, dotted eight,...), increase/decrease values for the rotary encoders,... The code is well documented, anybody can tweak it.

Like the idea?

Love it!

The thing I was going to try with a PCB layout is to match the same footprint as your arduino and sandwich them together with pin headers.  Not sure if that'd help or not.  I just HATE offboard wiring.

I will do it that way - use the arduino nano as shield. With 3 encoders, power supply, up to 5 switches, control out, RGB-LED,... there is still a lot of offboard wiring. Maybe I come back to your offer once I think I am done with the micro-bugfixes. Probably, there should be more than one layout for different use cases anyway...

[nzCdog]

Great work man, awesome to see your progress