Silly Idea for a Phaser...

[SpaceCowboy]

Well, so while I'm waiting for my PCB to get printed, I was thinking about getting into analog pedals... and I was thinking of doing a MXR P-90 based phaser,
with the op amp all-pass filters but with Opto-FETs such as the  H11F1 as the variable resistor to adjust the quadrature freq and using a microcontroller with an external
12 bit or more DAC to generate sine and triangle waves for the LFO.

So, perhaps someone can entertain my ideas? Is the design practical for just a silly daydream? Lol. Thanks.

[iq01221]

It's a good idea! I don't know 'bout controlled lfos, but I've already installed a trémulus lune's LFO to a 2nd gride of a subminature CK5678 tube. With the simetry mod, it gives U ramp up, rump down and some mixes of the wave form.
Just get on it! It will be a niiice mod, I bet on it!

[SpaceCowboy]

Yeah I couldn't quite get how those Wien Bridge oscillators worked in the original MXR schematic, so I'm gonna stick with something I do know lol.
Does anyone know how much resistance the H11F1 optoFET give when in linear mode? Searching on the interwebs seems to give a ballpark of 60-400 ohms.

Edit: scratch that... using regular N-FET instead. Now the question is, how the heck does the biasing work on the MXR pedal? I'm having a hard time figuring it out...

[SpaceCowboy]

Yeah I couldn't quite get how those Wien Bridge oscillators worked in the original MXR schematic, so I'm gonna stick with something I do know lol.
Does anyone know how much resistance the H11F1 optoFET give when in linear mode? Searching on the interwebs seems to give a ballpark of 60-400 ohms.

Edit: scratch that... using regular N-FET instead. Now the question is, how the heck does the biasing work on the MXR pedal? I'm having a hard time figuring it out...

Lol, FET matching sounds like a b*tch so time to do the OTA phase stage a la Ross or Casuality. Still going with the uCom LFO though... I think that's gonna give
quite the interesting sound.

[R O Tiree]

FET matching is actually easy - 5-10 minutes to set the circuit up on a breadboard, then about 20 seconds or so for each FET under test. That's the time-consuming bit.

I got some little tiny sticky labels and numbered my FETs 1-130 (because that's how many I bought at the time). I then opened a spreadsheet and made one column 1-130 and then tested each FET and entered them into the sheet. From there, it was a simple matter to sort the 2 columns by voltage and select pairs, quads and outliers (that would not be a good match).

From that point on, any time I want to build a P45, I go for a pair and if I want a P90, I select one of my quads. So, time consuming to sort them, but very quick ever since. Like anything else, preparation pays dividends.

[armdnrdy]

IMHO the best sounding phasers I have heard have been vactrol based for the variable resistor. The sound is very liquid, un-mechanical.

It probably has to due with the slow decay of the vactrols.

[SpaceCowboy]

IMHO the best sounding phasers I have heard have been vactrol based for the variable resistor. The sound is very liquid, un-mechanical.

It probably has to due with the slow decay of the vactrols.

Hmm vactrols? I see the settling time may give that particular sound quality... I'll give that a try. Is there a particular brand of vactrol that's best? Looking on Newark gives a few (expensive) ones. Well, might as well order both the vacs and OTAs just for the the breadboard then.

[armdnrdy]

The best way to go is to use dual vactrols VTL5C3/2. The response time and resistance is perfect for phasors

Saves on board space and the cost of buying two.

Look to the Mutron Phasor II for implementation ideas. (The Phaser II used a custom six LDR, 1 LED vactrol)

[SpaceCowboy]

Gotcha. Thanks.

[slacker]

Lol, FET matching sounds like a b*tch

If you're using a ucontroller just use PWM, then they're either on or off, no need for matching.