Auto crossfader

[stickjam]

Looking for ideas on how to approach the design of an automated crossfade effect box. The idea is to provide two line-level inputs and a mono output. The only control would be a pot for fade time--possibly another for switching threshold.

In operation, the output would pass through only input A until a signal of sufficient strength appeared at input B. Once the input B signal crosses the triggering threshold, the output would crossfade from A to B at a rate controlled by the fade time pot. Once input A diminished to a lower threshold nearer to zero, input A would then become "eligible" to trigger a crossfade the other way when it hits the triggering threshold.

Here's an attempt to graphically represent the effect: http://groupdiy.twin-x.com/displayimage.php?pos=-159 It uses two colors, blue and green, to reflect signals at one input; and red/magenta to reflect the signal at the other input.

I came up with a brute-force solution with two input buffers, diode signal detectors and Schmitt Triggers into an R/S flip-flop which drives ramp generators into VCAs which feed a summing amp. Feels like sort of a Rube Goldberg solution. I'd like to get this whole thing into a small 9-volt battery powered stompbox.

Any suggestions how you might approach such a challenge?

Thanks
--Bob

 

[Svart]

hey i was about to suggest something before i read your whole post.. i was going to suggest some kind of slow rampup oscillator or cap/resistor circuit driving a FET for the attenuation. you could change the time by subbing a pot for the resistor and enable/disable with a switch.. but i see you have thought of something similar.

good luck!

 

[gyraf]

Two simple rectifier/capacitor combinations to detect level in each channel. Diode forward voltage will be a threshold function. Feed these two voltages to a comparator to find the highest one. Run comparator's output through a pot ("timing") to a capacitor. You now have a ramping voltage depending on active channel, use this as control voltage for one channel's VCA. Send this control voltage through a x1 inverter, and you have control voltage for the other channel.

This will fade to the channel most active at any given time.

What is this for? Knowing that might make it easier to come up with suggestions?

Jakob E.

 

[stickjam]

[quote author="gyraf"]What is this for? Knowing that might make it easier to come up with suggestions?[/quote]
It's a performance effect for a Stick Bass which has a stereo passive pickup--one channel for each half of the strings. Both halves are set up as a bass instrument. The effect is half "SlowGear" or E-Bow and half like a polyphonic synth patch with slow attack and release times (where one note smears into another) The difference is that the "smear" only happens when I switch sides.

I rewired a stereo volume pedal to do the effect manually (or would that be "pedally" :wink: ), but would prefer to be able to move around and let 'tronics do the work.

I'm partial to FETs in the signal path--probably a good fit for the VCAs and elsewhere in the design as well. No objection to a hybrid digital/analog solution either. I considered that an MC14093 chip might serve to provide the threshold controls with half of the chip, as well as forming an R/S FlipFlop by cross-feeding the other two gates. I also thought it would be a useful feature to have an LED indicator to show the currently active channel.

 

[stickjam]

Here's a diagram of what I was thinking.

http://groupdiy.twin-x.com/displayimage.php?pos=-160

Does this appear that it might do what I'm after? I've gotten the preamp section sorted out. Any pointers to how I might construct the VCA and summing portion in the simplest and most effective manner?

--Bob

 

[Svart]

interesting.. I'd like to hear how that works when you get it done!

http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/schmitt.html#c3 i like this link.. should help a bit if you are planning to use a comparator instead of a dedicated IC..

 

[stickjam]

Svart, I can't wait to hear it either! :grin: I'm considering the comparator route, since thresholds are more tweakable than the CMOS solution.

I found this page (http://www.pic101.com/audiosw/) that describes the use of a JFET as an audio switcher, including adding a cap to produce a fade. I understand the function of most of the components, what's the function of the source-gate cap?

In this application, -50dB attenuation as claimed there is more than adequate for the "inactive" channel. I applied that to my original drawing and added an output buffer, and humbly submit here what may be glaring evidence that my analog design chops have totally disintegrated over too many years of disuse.

http://groupdiy.twin-x.com/displayimage.php?pos=-161

I'd like to know if I'm on the right track. Component values were omitted out of convenience and ignorance. I'll take any constructive criticism, comments and suggestions eagerly with no offense.

Thanks

--Bob

 

[bcarso]

"I rewired a stereo volume pedal to do the effect manually (or would that be "pedally" ), but would prefer to be able to move around and let 'tronics do the work."

Hmmm...another pedalphile. Better watch your ...errrr... step.