Help researching OTA phaser sweep freeze/stop - Rick Holt's Causality IV

[lomus]

Good morning from Spain and nice to meet you all! Jorge here.

It's been a long time since I've been lurking in here. I think the question I'm about to ask is the perfect excuse to start taking part of the forum's discussions, so here I am, eager to get to know you and try to help with anything I can, but I fear that I'll be the one asking for help most of the times... I'll try to give it back somehow, though.

So I've been reading related threads and testing this build of Rick Holt's Causality IV phaser (AMAZING and widely known in the forum) for a couple weeks now and I'm trying to figure out how could I add a manual switch of some kind to freeze the sweep wherever/whenever it is by the time I activate such switch - I'm not sure I'm making myselft clear enough... The search query didn't throw many results about this matter, so it's also possible that I'm looking in a wrong direction

I thought: maybe it might be possible to hold steady the LFO output that's fed to the OTAs (?)... What do you guys think? Do you know if there's any info I can use to try and come up with a solution? Of course I'm not in a hurry, so I have plenty of time to dedicate to this, hoping that I can develop my own phaser, someday, in the future.

I think something like this is achieved on EHX Bad Stone (most recent version), which has a "Manual mode" switch and a pot to select the notches position, not sure how it's done though

Thanks so much in advance, really. I hope it's not something that has already been discussed and I wish this thread could serve as a handbook to do this kind of mod to any OTA based phaser.

Salud!

[GibsonGM]

Hello Jorge. Welcome!!  And please do join in where you can, it is very nice to meet new people and hear their thoughts about DIY stomps, no matter what their level of experience!    

Can you post a link to the schematic so that we can have a common picture to refer to?  I'm sure someone with a lot of experience with phasers can help you out!

[anotherjim]

The simplest would be a normally closed momentary switch in series with the speed control pot. When the switch opens it will freeze the sweep to the voltage held on the 10uF capacitor. The disadvantage is it will drift slowly over time due to leakage currents but this can be minimised by using a low-leakage capacitor (dare I suggest tantalum?) and a CMOS op-amp for the LFO but the suggested TLxxx chip won't be too bad.
Anyway, it's easy to try the concept just by arranging to make/break the speed pot connection (either end will do).

A more involved solution might be to move the 220k resistor connection from the 10uF to the buffer amp output to the depth pot. Now when the speed is disconnected the 10uF is more like a sample and hold since the path back via the 220k is no longer at the cap but it still needs a low leakage capacitor.

[ElectricDruid]

+1 what Jim said.

I was just writing a post that said exactly the same thing, but he beat me to it!

Stick a CMOS switch in the LFO, turn it into something like a sample-and-hold stage. Normal sample-and-hold rules apply to prevent voltage droop (good caps, high input impedance for the buffer).

[anotherjim]

^But Tom beat me to the edit!

[lomus]

Thanks Mike for your warm welcome! And thanks so much to Jim and Tom for taking time to think about this and give me such valuable input. I will look into it and keep you updated with the results - or, more probably, I will come back to keep asking 

Your replies made me think about using some kind of digipot (fed by the supply voltage, for example) of which resistance value (therefore, output voltage) would be stated by the S/H circuit, to feed the LM13700s with the LFO output value when a hypothetical footswitch is pressed down - so it's not necessary to rely on the mentioned 10uF capacitor holding the charge. Does this make any sense? Might that work for holding that voltage indefinitely? That was my original idea: to hold that sweep(t) spot for as long as I would like...

EDIT: BTW, I feel like a little kid on Christmas Day when I see the people I've read for so long in this forum, and that I really admire and respect, are helping and saluting me here... haha Take care!!

[anotherjim]

The higher value of the capacitor, the problem of loss of stored charge becomes worse. Also, capacitor types that do have really low leakage just don't get made large enough. They are in <1uF values.
The 10u capacitor in the LFO is sized to suit the slow speed range required. However, there is nothing to stop a second, lower-value capacitor from being added on to act as the hold capacitor and have little effect on the speed range when connected. The hold cap fits on the +input of the LFO buffer opamp after the "hold" switch.

This is an old textbook arrangement. The switch doesn't have to be a JFET, but it shows how the capacitor is simply left connected to the +input of an opamp voltage follower.

Having said all this, many find a "manual" voltage control pot that switches in place of the LFO a good way of getting a presettable and reliable way of using a phasor or flanger for fixed filtering if that is the aim.

[lomus]

The idea I posted earlier, as we say in Spanish, "it's like killing flies with cannonballs". Of course, this:

Having said all this, many find a "manual" voltage control pot that switches in place of the LFO a good way of getting a presettable and reliable way of using a phasor or flanger for fixed filtering if that is the aim.

is the easiest and most practical approach and that's what I will probably do in short term.

But I'd really like to explore the other option we were talking about, to have the ability to stop and resume the sweep wherever I want while playing in order to make it smooth, without having to lean towards the pedal and manipulate the pot, just for my own enjoyment and maybe, if it's succesful, share it with you all.

I'll keep you posted

[ElectricDruid]

But I'd really like to explore the other option we were talking about, to have the ability to stop and resume the sweep wherever I want while playing in order to make it smooth, without having to lean towards the pedal and manipulate the pot, just for my own enjoyment and maybe, if it's succesful, share it with you all.

The simple way to do this would be just to add a toggle switch between the pot and the cap. A "manual" Sample-and-hold switch, if you like. Flip the switch to stop the LFO, and switch it back to start it again. No, it won't be perfect (it will probably droop over time) but it'd be very easy to do and will give you an idea of how useful the effect is and whether it's worth pursuing further.

[lomus]

Yep! I thought about that and tried it while testing. It was just like you said, decaying over time, so I put a DPDT to switch from stock 10uF to 33uF and 470uF (yeah, I know, I'm a little "exagerado"), and as you could expect, 470u is pretty damn slow, very cool effect tho and gives you the ability to taste it properly.

But just out of curiosity and because I think it's never a bad thing to do some research and get to know how stuff works, I wanted to fully stop the sweep, as we were talking about and I just remembered this EHX Bad Stone schematic I saw on the web (it says on its watermark it's from Jack Orman's site, but I didn't find it on there):



On the lower left of the schem: there's the switch, next to an op-amp, which (I think) is the LM324, but I don't quite understand what the switch is doing there... Any thoughts? Seems like a feasible solution for this topic, and it has the pot Tom mentioned in an earlier reply

[anotherjim]

The manual switch in the Badstone selects between a control voltage from the manual pot wiper or the LFO sweep waveform. It's a different (and insanely complicated) LFO scheme but the principle is identical.
The EXH Electric Mistress flanger does it differently, the switch fixes the LFO sweep and you then use the LFO depth control to set the manual stopped effect to taste.
I don't think either of the EHX methods synchronises the LFO sweep so it pauses when switched to manual and continues from the same place when the sweep is switched back on. Really, only the sample-hold using the same timing capacitor as the LFO can work smoothly like this. Even if the cap voltage drifts, that will be the point the LFO sweep re-starts from.

Don't get the idea that a larger cap will hold better. The higher the capacitance it has the greater the leakage can be. The best you can do to keep it simple is to find a very low leakage capacitor type in place of the existing one.

Keeping the existing cap and adding a smaller hold cap is probably going to be the best solution.

It isn't only the capacitor that can leak away the voltage. The opamp +input must not take or supply any current to affect the cap voltage. A CMOS type of opamp has the best chance of doing the job. Also, the layout on the board around this must be clean. Any surface is a leakage path. The very best thing is to bend out the +input pin of the opamp and solder the cap leg and wire to the switch directly so it's no longer on a surface - air is the best insulator.

However, with this, the existing 10uF cap is still going to drift off so it will restart from a different place when switched back. With another switch, the output of the opamp can be connected back to the 10uf to maintain it at the voltage held by the hold cap.

[ElectricDruid]

If you want to avoid the droop, you could add the pot between the voltage rails like on the Badstone, and then just use a SPDT toggle to switch between the LFO output and the fixed pot output.

No, it won't be smooth, since it'll jump to the pot position from wherever it is in the LFO sweep,and similarly going back to the LFO, but it'll give you a fixed position with no droop and let you experiment with different fixed phaser positions. You might need to limit the pot's range a bit by adding series resistors at the top and bottom of the pot - the LFO won't go all the way to the rails, so if the pot is to cover a similar range, it'll need cutting down a bit.