Tremolo-matic X from Stompboxology

[anotherjim]

One thing with an opamp rail-splitter concerns the resistive divider output bypass capacitor.
The usual single capacitor to the ground means two things...
1: On power-up, the reference voltage rises behind the rising supply voltage due to the capacitor charging. When the splitter opamp has enough juice to become functional, the reference input may still be too close to the negative rail voltage for the opamp's input range.
2: On power down, the positive supply may collapse quickly leaving a positive voltage stored on the single capacitor only feeding the opamp +input which could be damaging.

Actually, isn't there a much simpler way? Rely only on bypass capacitance for the input to the resistive divider and don't fit any capacitors on the reference input to the rail-splitter.

[antonis]

Maybe, C5/R11 time constant (100ms) is adequate for C5 rapid discharge..

[soggybag]

I've been working on this trem and had a few questions. Here is the current schematic I've got.

1) I want to bring the output up. I feel like I can't get there adjusting R16 and R17 and will have to add a transistor.
2) I get some distortion when the input is hot. If I'm playing a Strat it things are fine with one or two pickups but with all three pickups on there is some ugly clipping around the attack.
3) The 5v Reg R26 and R27 don' t seem to be doing much in practice. These seem to be there to eliminate feed through. I don't hear any feed through. Just a little guitar static. Maybe I can remove these parts?
4) The depth pot isn't doing anything for the first 25% of rotation. I'm using a linear pot. It was worse with the audio taper pot. Maybe a reverse log pot here might help?

On points 1 and 2, I feel like the solution might be to lower the gain of input op-amp IC4_D and increase the output to compensate.

[ElectricDruid]

On point (3), what voltage are you feeding the whole circuit? The 5V regulator is connected to point B, which is the midpoint bias. Given the regulator needs a volt or a volt and a half to actually work, that'd mean the minimum supply voltage to get the regulator to work would be somewhere >12V.

On point (4), if the pot is linear, you should be able to see LFO diminishing smoothly all the way to the bottom. Do you have an o'scope so you can check? If not, and you're just listening to hear the result, it's possible that the VCA isn't that linear at low voltages, or perhaps the control input needs biasing.

[soggybag]

Thanks for the reply Tom.

3) there is a note in the original article that the regulator and related parts for "optional feedthrough trim network".

I see what you are pointing out. I may have the input to the regulator connected to 4.5v when it should be connected to 9V!

4) I have a scope! I will try this out and get back on this point later...

P.S. I ordered some AS3340 chips from you and just received them in the mail, thanks!

[soggybag]

I'm having trouble taking a photo of the scope. When the time/div is down low enough to show the wave form the camera is too fast to capture with the camera.

Measuring the voltages at pins 1 and 16 of the NE570 I see a triangle that ranges from:

• +/- 0.2v at depth 25% 
• +/- 0.4v at depth 50%
• +/- 0.8v at depth 75%
• +/- 1.0v at depth 100%

[ElectricDruid]

Measuring the voltages at pins 1 and 16 of the NE570 I see a triangle that ranges from:

• +/- 0.2v at depth 25% 
• +/- 0.4v at depth 50%
• +/- 0.8v at depth 75%
• +/- 1.0v at depth 100%

That seems about right. OK, it's not perfect (0.25, 0.5, 0.75) but "within 0.05V" looks close enough to be called "linear".

If nothing happens below 1/4rd on the pot, I'd look at what's being fed rather than the pot itself. I think the problem is downstream.