Parallel Reverb and Drive

[mth5044]

I was posed the question 'how can you make a reverb not get crazy loud when engaging a boost pedal?' and thought the best approach might be to put the reverb in parallel with the boost. This wouldn't work with heavier overdrives/distortion/fuzz since the tap for the reverb is before the clipping, but I'm with a tubescreamer with LED's as clipping diodes, clean reverb'd signal wouldn't be too noticeable, especially if the screamer is used mostly boost the front end of the amp.

I used the Son of Screamer and the 1776 reverb schematics as the boost and reverb.
http://www.muzique.com/tech/scream.htm
http://1776effects.com/wp-content/uploads/2014/11/Rub-a-Dub-Deluxe-BOM.pdf

(power section not shown: VB is 4.5V, VC is 5V from a 7805. Didn't want to clutter the schematic more than it is).


The reverb module is the new belton block with the nifty decay pot. I have a few questions!

1) I believe the polarity of the mixed signals will be the same ASSUMING the output of the belton brick is the same polarity as what went in. Yah?
2) Is the adaptation of the opamp after the brick ok? In the 1776 schematic the dry and reverb signal are mixed with a 22k||20k resistors, so I was thinking the 10k resistor would be ok for setting the gain of the opamp.
3) Is the single SPST switch ok to kill the reverb?
4) Any other problems with the mixing part? Is it ok to use an inverting and a non-inverting opamp for mixing as long as both output the same polarity signal?
5) Are IC1B and IC3A redundant with the opamps in the tubescreamer?

I thought at first it might be easy enough to just take the signal from the opamps in the tubescreamer, but couldn't work out how to have the reverb accessible when the boost is bypassed. Thanks for taking time to look at the schematic and questions! I appreciate it. 

[Kipper4]

Not really much help because I haven't had a go with the new btdr3s yet.
But I can't wait to see how this works out.
I would wonder if the reverb would pop when switched back into the circuit though when sw3
Good luck mate
Rich

[mth5044]

Good question about the switch pop. Perhaps the mix control will be suitable to drain the cap? Maybe it would be better to do it like fender amps and pull the signal to ground to mute it? Not sure where though

[Kipper4]

#following this.

[PRR]

> ASSUMING the output of the belton brick is the same polarity as what went in. Yah?

Broadband delays and reverbs do not have strict polarity.

A straight delay, mixed with its source, will comb-filter because it goes in-and-out of phase as the delay time does/does-not align with the signal cycle period.

An ideal reverb does that but with an infinity of "random" delay times. Comb-filtering so dense that you can't readily find any nulls.

A low-price digital "reverb" will not have the infinitely random outputs, but if it's any good at all, it will be "sufficiently mixed-up" to not cause any great "polarity problems".

Even less important when working with musical sounds as opposed to constant test-tones. A 0.01 second pluck will be gone by the time sound gets out of a long delay or reverb, so it can't null.

[mth5044]

Interesting, thanks PRR! I had read up on comb filtering in flangers, but never really realized it happened in reverberations too. So I guess as the reverb side inverts it twice and the boost side doesn't do any flips, so that should be good.

I think I'm going to take the reverb switch and connect to ground right where it is instead of just disconnecting the signal. Seems to be pretty popular in other non truebypass reverbs.

I also considered replacing the clean section of the 1776 schematic with the tube screamer but that first opamp has a gain of two and the increased signal might mess with the tubescreamer.

[Digital Larry]

I've done lots of experiments in the DSP realm with unorthodox routing, such as reverb driven from the clean signal while clean itself goes through overdrive before mixing together with the "clean" reverb, and it's not as bizarre as you might think.  It may not be what you're after but if it's easy enough I'd at least try it.

[mth5044]

Thanks Larry! I was hoping it wouldn't be bizzare! I'm aiming for something that most people wouldn't notice, more of a player pleaser if anything. Thanks for the info

[PRR]

> I guess as the reverb side inverts it twice

This really should not matter. Any repetitive wave, if you wait, will be the other instantaneous polarity on the next half-cycle. A complex wave through a complex delay (reverb) will be "all polarities". When mixed with dry, there will be nulls, either way you phase it. The nulls will be at slightly different frequencies + or - but any "good" reverb will have so many nulls that you can't readily tell the difference.

However. On your proposed plan, there is a simple way to try it both ways, in and out of notional phase. At least for testing. You don't absolutely need the gain-of-2 in IC1A. change it to unity gain (R4=100K). Switch the reverb-chip input from IC1A to IC1B. Same level, other polarity. Is there a difference? My 2 cent bet is, you can tell the difference if you play simple sounds and switch it yourself. If someone else switches it, you will become uncertain. And when you get dressed-up and the crowd is lively, nobody can know the difference.

[josephfenwick]

I'm a pedal building novice and wondered if you had a schematic for a simple overdrive/boost type pedal which would be a good place to start.

[bluebunny]

Welcome along, Joseph.  Not exactly the right thread to be asking, but head along to the Projects section from the forum home page.  You'll find a boost "Beginner Project" and an "IC-based Overdrive".  Both worth a look.  Or just keep reading this part of the forum - there are plenty of projects passing through all the time.

[mth5044]

Just wanted to update to say it works! The new brick/1776 circuit sound great too.

[Kipper4]

I've just noticed that IC1a inputs are both connected to Vb, non inverting directly connectedto Vb, inverting through R2. Is this correct?

[mth5044]

The 1M from the Vb biases the signal for both opamps, but namely IC1B. The positive input has to connect to Vb for bias as well, other options are ground and +9V, neither sound like a good thing  The bias on the + pin is there mostly because that's what 1776 said haha.