Mix Control for LA Chorus

[Ben Lyman]

Trying the Little Angel on the bread today, I want a mix control so I did this.

How about it? any foreseeable problems with removing the 10K from pin 14 and going straight to the mix? I like this, it gives the ability to drown out the dry signal if wanted.

Other changes are reducing the opamp gain to unity with a 750K feedback resistor, engaging the effect had a dramatic boost before.
Startup delay mod carried over from the "Low Cost Chorus" and seems to work fine plus its a common BJT instead of a fancy FET. I'll give the FET network a try later and see if there's a difference.

[antonis]

Other changes are reducing the opamp gain to unity with a 750K feedback resistor,

Is it so..?? 

Also, how come 5V on start up delay BJT circuit..??
(78L06 output is 6V)

Also, what's the purpose of diode between Base & +5V..??
(if it's for cap discharging it will not work and also is risky for 78L06 OUT when power switched off..)

[Ben Lyman]

Yes, it really is. Maybe not "technically" or "mathematically" but the bypassed volume now equals the effect volume. With a 1M the volume boost was way too much. Also, I'm using a TL082 so I don't know if that could have something to do with it. Anyway, it sounds great now.

[antonis]

Previous post edited..

[Ben Lyman]

got it, thanks Antonis. I am using 7805, that schematic is Jack Orman's modified LA, so +5 for me all around.
The diode, and everything else on pin6, is exactly like anchovie's "Low Cost PT2399 Chorus" so I left it as is. Do you think its safer to use Jack Orman's design instead? I just don't want to use a FET if I don't have to. I could also experiment pulling the diode out and see if it still works, I'll try that later.

[anotherjim]

The output of an opamp is a low impedance. Pin14 is such an output. That said, the opamp in the PT2399 is a CMOS construct and may have a relatively high output impedance compared to more general-purpose opamps.
Anyway, the way you have the mix drawn, when...
The mix pot is turned fully down, pin14 is shorted to ground via the cap. There should be some resistance in pin14 output if only 1k.
The mix pot is full-on, the clean signal is shorted to pin14 (but you do have a limiting resistor). This is probably why you needed more than x1 gain at the input.

The discharge diode will work since there is a path through the PT2399. It can damage the regulator if the regulator input is shorted to ground, then there is a back path to discharge into the regulator -  they are not protected against over-current damage in that case. Then again, the same worry exists with charge stored in the "6v" supply capacitor.
The usual protection is to fit a reversed diode between regulator in and out to carry the back discharge current around it.
But, you can cure two ills by fitting a series diode after the regulator output. It will block any back discharge AND drop some voltage bringing it closer to the proper 5v required.

[Ben Lyman]

Off to the mountains for a few days, I'll check back in later for some more experiments. Chorus sounds great, only 3 knobs but a very wide range of sounds available. Thanks guys, have a great weekend!

[Ben Lyman]

I've decided that I don't like any of the various anti lock up add ons. The reason is that I don't like the noise that comes with any resistor less than 2k7 on pin 6. I know this means "too slow for chorus" delay times but I don't care, the noise is unbearable. I'm going to build it with my mix control plus Jim's advised 1K resistor, and 2k7 on pin 6 delay time. So, unless somebody has an idea of how I might quiet down the white noise and ticking time bomb sounds, but still achieve faster delay time, this will do for me

[anotherjim]

When you did the mix control...
It already had a mix control, but fixed 50:50.
The 0.1u output coupling caps can be before the 10k resistors.
The 10k resistors join together to the 100k output pull-down resistor.
The 10k resistors could be replaced by a 20k lin taper pot.
The delay output 0.1u coupling cap to the mix can be smaller. This will reduce any warbling of the fundamental tones and more effect on harmonics -  if you want more of a shimmer.

Putting the LFO in the same chip as the input buffer is always going to invite ticking noise. Good layout and supply filtering can help - breadboard may not be the best place to demo it. TL072 may not be the best choice -  do you have a TL062?

Driving the PT2399 from more than 5v supply is not going to help if its already being pushed hard by a super low pin6 resistance.

[Ben Lyman]

Awesome Jim, thanks! I'm using a TL082 and I also had tried a 55xx something or other. I like my mix control better right now. When I tried a B50k pot in place of the 2 fixed resistors there was too much change in output volume as I turned it. I don't know why I didn't think to try a B20k pot instead. I'll do that later and see what happens.

edit: perfect mix control with a B20k pot and a couple of 1k resistors on each side with the wiper going to output. I feel stupid for not trying that before, thanks Jim. there is a barely noticeable volume drop with the mix in the middle, and just the slightest increase in pure wet or pure dry signal. I was experimenting with the B50k pot before, which exaggerated the volume change.

[anotherjim]

Strictly speaking, the 1k "protection" resistors are no longer needed there, since the pot now always puts some resistance between the outputs. Only one 1k is needed in the final output to protect against a fault from outside.

The relationship between the mix pot value and the load value (the 100k pulldown) is important. Some simple designs rely on the load from whatever it's plugged in to, but that means a high pot resistance is needed and it picks up noise too easily in the 50:50 mix position.

[Ben Lyman]

So, what do I do if I want a second output jack to make this a stereo pedal?
I've only ever made one other stereo pedal, it was a bass fuzz with a dedicated dry out jack.
Not sure what a "stereo chorus" really is either.

[anotherjim]

Basic stereo chorus is simpler than you may think. Send dry to one channel and wet to the other. This works best when the speakers aren't too far apart. When they are far apart, you'd need to pan some of the wet and dry over to the opposite side so it isn't as obvious that the channels sound different if you listen either side of the centre

A fancy "wide stereo" method is to add an inverted version of the wet signal and create 2 wet-dry mixed outputs so each has an opposite phase version of the wet signal but the dry are both in phase. A big disadvantage is that if both channels get mixed to mono, the wet signals being equal but opposite, cancel out.

The next complication is to have 2 delays but to modulate one of them with an inverted LFO. This has a "side to side" ping-pong quality. If you want a sense of rotational direction, use an LFO with quadrature output so the difference in phase is 90deg instead of 180deg.

Next, build x2 chorus circuits. Independent delay and LFO. They can share the same input buffer and voltage regulator.

[Ben Lyman]

Hey, I'm back again. I have built the pedal twice now with a couple issues.

First build, the volume on the wet mix was muddy and quieter than my breadboard experiment. Only difference was that I used a couple 10uF tantalum caps in the build, due to stock on hand, plus box caps instead of film.

Second build, much better but still not exactly like the tones from the bread.

Second build uses mono coated ceramic discs instead of box caps and only one tantalum (coming off the depth pot wiper)

Also second build, I put 7k5 instead of 10k, off the 100n at the opamp out to the pt2399, in an attempt to raise the wet volume a bit and it did do that.

Second build, last change, was replace 100n with 10n from delay out pin 14 to mix knob, slight improvement to muddiness.

overall, the second build is much better and very useable but I wonder what I can do to really get that clear and bright wet mix from the delay chip.

I have double and triple checked my bread to schematic and 99.9% sure on that.
Schematic to layout may be suspect.
Layout to build is spot on.

Here's the breadboard schematic and it sounds great.


Here's the layout I made for it. I thought I did it right, but I don't know.

[anotherjim]

Bambi is blind?

Are you able to swap PT2399's? Pushed to the minium delays/max clock as they are in this circuit, they may not all be equal in this condition.

[duck_arse]

I would rekon the 10uF before the depth control is backwards. at least on yr circuit dia and layout, we can't tell what you've done on the builds .......

[Ben Lyman]

Bambi is blind?

Are you able to swap PT2399's? Pushed to the minium delays/max clock as they are in this circuit, they may not all be equal in this condition.

I have swapped both bread and build (socketed) several times, always the same result with bread being nice and bright but the build is muddier on the wet mix

[Ben Lyman]

I would rekon the 10uF before the depth control is backwards. at least on yr circuit dia and layout, we can't tell what you've done on the builds .......

Wow! Really? I have oriented that cap that way on bread and build. It is exactly as pictured in all the LA Chorus schematics I find online. Plus towards pin 2 and minus towards the LFO. Do you think I should flip it? Most curiously, what do you see in the circuit that would suggest flipping it? Thanks!

[anotherjim]

That 10u...
Pin2 of the PT wants to be at 2.5V. If the LFO ran on 5V, it would swing above and below 2.5v so the cap wouldn't have a right or wrong polarity - but - the LFO runs on 9v so most of the time, yes, now that it's mentioned, it is mostly + from the left of the cap. This issue "might" affect the modulation shape, but probably not the sound quality.
If you wanted to be paranoid about it you could use x2 22u in series with both + ends outermost.
If you wanted to be super-paranoid, fit a reverse diode from pin2 to +5v to catch any extra voltage over 5v.

[Ben Lyman]

That 10u...
Pin2 of the PT wants to be at 2.5V. If the LFO ran on 5V, it would swing above and below 2.5v so the cap wouldn't have a right or wrong polarity - but - the LFO runs on 9v so most of the time, yes, now that it's mentioned, it is mostly + from the left of the cap. This issue "might" affect the modulation shape, but probably not the sound quality.
If you wanted to be paranoid about it you could use x2 22u in series with both + ends outermost.
If you wanted to be super-paranoid, fit a reverse diode from pin2 to +5v to catch any extra voltage over 5v.

Thanks Jim, I'll flip it on the bread later to see what happens. Kind of crazy because all the Little Angel schematics show it like this but what you say makes perfect sense to me. There is another 10u right there going to ground. Could that one have been added for some reason related to what you speak of?