Dual LFO "Crossfade" Ideas

[cbc526]

Hello everyone,

I have been designing a circuit to "blend" a dry guitar signal and an effect in interesting ways.  As it stands right now I have two of Electric Druids TAPLFO chips modulating a dry signal and an effected signal.  The two LFOs are out of phase with each other so that depending on the position of either LFOs Depth control, the two signals can crossfade back and forth.

What I would like to do is combine both of those depth controls into a single center detent "crossfade" control using a panning type circuit.  Having read up about pan law and its shortcomings, I would like this control to allow me to adjust the depth of LFO 1 from 0%-100% over the first half of the pots rotation while the depth of LFO 2 was at a static 100%.  When the center detent is reached, both Depths would be at 100%, then the opposite of the last sentence when the pot goes through the last half of its rotation.  I included a visualization of this concept below that helped me when I was trying to come up with this idea.



Essentially since the LFOs are running from the TAPLFO, then the depth control for each LFO is acting as a simple voltage divider between 0V-5V.  So essentially, I want the first half of the pots rotation to increase the Depth CV of LFO 1 from 0V to 5V over the first half of the pots rotation and keep it at a static 5V over the second half, then have the opposite be true of LFO 2.

Also, I am not against using Digital Potentiometers for this situation, as I am already implementing digital control functions into other concepts of this design.  I just want a single knob control of this function.

Excited to see what this thread comes up with!

[EBK]

Do you just need one of these?
http://www.bourns.com/docs/ProAudioDocs/bourns_pdb182_blend-balance_guitar_pots_appnote.pdf

[cbc526]

Thats pretty much exactly the idea.  my only issue with that specific potentiometer is that its resistance offerings are limited to 250k and 500k since its a replacement for guitar potentiometers.  If I could find something similar to that in a 10k resistance, that would be pretty much perfect.

[EBK]

Search for a "mn taper" pot. 

[cbc526]

It's looking like there is really nothing available from my searchings save a few ebay sales here and there of Alps Blue Velvet potentiometers, which are both expensive and not readily available.  Everything else seems like a custom order where minimum quantities are involved.  I was only able to find one other pot on mouser with a "balance" taper like the Bourns, but the information on its was minimal at best.

Any other ideas on how this kind of think could be implemented?

[EBK]

Why won't the 250k work?

[cbc526]

Because the maximum source impedance on the analog inputs of the PIC16F684 (TAPLFO base IC) is recommended not to go above 10k due to the source current being too low to process if the impedance gets higher than that.

[EBK]

Because the maximum source impedance on the analog inputs of the PIC16F684 (TAPLFO base IC) is recommended not to go above 10k due to the source current being too low to process if the impedance gets higher than that.

Ah, ok.  You could put an op amp buffer (or another buffer of your choice) between the pot and the PIC then.

[slacker]

Here's little circuit that does what you want with a normal pot and a few opamps. https://drive.google.com/open?id=0B45V-V1ycMU9Qjl1ZDEyaTY5Unc the opamps are powered off 9 volts or whatever the main supply to the pedal is, forgot to add that to the drawing. The opamps must be have inputs and outputs that can go to the negative supply, 2 LM358s or one LM324 will do the job.

U1 takes the 0-5 volts from the pot and multiplies it by 2, so you get 0-9 volts roughly out of it with 0-50% on the pot giving 0-5 volts. U3 is an ideal diode voltage clamp that clamps the maximum voltage to 5 volts. This gives you a signal that rises from 0-5 volts over the first half of the pots rotation and is then clamped at 5 volts for the second half.
U2 inverts the signal from U1, so you get 9-0 volts out with 5 volts at 50% on the pot, U4 does the same as U3. This gives a signal that is clamped to 5 volts for the first half of the pot's rotation and falls from 5-0 volts over the second half.
In theory you don't need U3, U4 and the 1k resistors you can feed the outputs of U1 and U2 into the PICs inputs through resistors that are big enough to limit the current into the pins to less than the maximum allowed. The PIC's built in input protection diodes will then clamp the voltage to 5 volts for you, this is a dirty thing to do though and I'm not sure how well it works for analogue inputs.
If you can find some opamps that will run off 5 volts and that have inputs and outputs that can go rail to rail, you could just use U1 and U2.

[samhay]

slacker - that's a nice analogue solution.
cbc526 - have you considered tweaking the TAPLFO code to do what you want in software?

[R.G.]

A multiway "morpher" circuit with various effects being crossfaded into a mix has been a longtime favorite concept of mine. I've done the manual version with separate mixing pots a couple of times, and it makes for really interesting sounds. I found that I really liked the interesting sounds that happened while the mix was changing from one set of sounds to the other.

I liked the implementation that used a multiway cross fade that faded from one effect into another in an assortment of channels, not just two. The idea is much like a loop selector box, with a number of loops each selected by one foot switch,  but with the switching happening not instantly, but as a fade in/out. The effect being selected fades up, the previous one fades out, with the fade times being adjustable. The footswitches in this version acted "radio button" style, where selecting one channel faded the previous one out, and left all the others faded out. And once you can do it manually, you can do it with a digital circuit pressing the buttons for you (in a metaphoric sense) so that the digital channel/loop selector could run through various sequences of selected loops.

One tricky part is getting a good, silent electronic crossfade/VCA. Digital pots can do this but only if they're specially set up to not have "zipper noise". Maybe a PWM approach could be made to work. OTAs would work with control voltages, but they tend to be noisy. Special purpose VCAs work fine, but they're a bit expensive.