Insane FX loop

[toddvirgil]

Hey everybody -- need some more eyes on this and I've got some questions. I got a request to build a single effects bypass loop with everything but the kitchen sink when it comes to mods... here's the specs:

1. True bypass foot switch
2. Ability for effect to trail when switched off, but also ability to be completely true bypass (See below, I think this can be done with a toggleable "trails" Switch.
3. Mix Dry/Wet signal
4. Feedback momentary switch
5. foot switch to go 100% wet
6. Bass/Treble tone control

So I have a few questions:

1. The layout below handles 1-4. Can you guys double check me?
2. I read that trails can be accomplished by adding a buffer -- would a klon buffer in the signal path do it (see diagram)?
4. How would I add a footswitch (with LED) to go 100% wet?
4. Where would you add a bass/treble tone control? Any suggestions?
5. This guy is primarily a bass player.. any concerns with the circuit because of that?

Mashed this layout based on this, this, and this.

[toddvirgil]

No one brave enough to take a look? :)

[ana konda]

In theory it looks all go.
I won't say anything on that trails thing.
100% wet switch could be just another 3pdt switch with return to the input on the switch, you get the gist, just switching the pot out of the circuit.
i'd add tone controls last before output, but you'd probably want to have some make up gain afterwards, so add lpb-1 with internal trimmer.
Not sure what to answer to the last question, afterall this is blend in itself, but it could be useful to have master blend. Dunno bout that, someone else might have a opinion on this one.
Also why on earth would you need a led for a momentary action? Seems kinda stupid, just saying. :)
-AK

[toddvirgil]

Thanks for taking a look AK!

I think I'll just have to test the buffered trails option out to see if it works. Not sure how else trails would work.

That makes a lot of sense on the tone control. Any recommendations on which one? There's Miro's BMP tone stack with LPB1 and I saw someone else made one with a James stack.

On the momentary switch, totally not a necessity, but I've met very few guitar players who don't like things to light up, even if it's totally unnecessary :) I might do something fun with the artwork and LED placement on that thing.

If anyone else out there has any experience with tails-buffering, lemme know!


T

[toddvirgil]

AK - thanks again... Am I doing this wet switch right? I'm not sure what to do with lug one on the pot... is it fine as is?

[induction]

I don't think your Trails switch is going to work. As far as I can see, it's just a switchable input buffer. The Split'n'Blend circuit includes buffers for the send and the dry signal, so this extra buffer really only behaves like a switchable buffered bypass, which is cool, if that's what you want. But it won't give you tails.

Tails is the result leaving the output jack attached to the circuit output during bypass. It only works on circuits that delay the signal (delay, reverb, I can't think of any others). With any other type of effect (distortion, tremolo, compression, whatever) tails will have no effect because the circuit will stop producing output instantly when the input signal is removed. On some circuits (like phasers and flangers) tails will sometimes just add time-varying noise to the clean signal.

There is a tension in this project between delay-based effects and non-delay-based effects. The blend pot complicates things because we usually don't want the bypass signal to depend on the blend setting.

The blend control doesn't make much sense for delay-based effects, because they are already a blend of effected and dry signals and they usually manage that blending within their own circuits. With these circuits, tails is easy: just break the connection between in the input buffer and the part of the circuit that generates the delay. The buffered clean signal keeps coming out of the output, but the delay no longer has an input, so it outputs whatever signal it has left and then shuts up. 100% wet makes little sense here, because if the pedal can do that, it will be a setting on the pedal itself. If it can't do 100% wet, the loop circuit isn't going to make it happen. Including the external blend control causes problems because you have to choose where the tails signal comes from. If the tails signal comes from the blend output, then at anything less than maximum dry blend, the tails bypass signal will be quieter than the true bypass signal because it is attenuated by the blend pot. On the other hand, if the tails signal comes from the effect return, then at anything less than maximum wet blend, the tails will be louder than original effected signal.

With non-delay-based effects, on the other hand, you usually need to manage any blend externally to the pedal, tails is not possible, and 100% wet signal is possible. So your design solutions will look different than the ones you would choose for delay-based effects.

So the obvious question is: Does the customer use delay or reverb? (Bass players often don't.) If not, forget the tails. If so, you'll need to figure out exactly they want to use this thing. Do they always want to switch on a whole bunch of effects at the same time or will they switch the individual effects on and off within the loop? Will they want to switch the delay or reverb off and get tails with the other effects still on? Do they want the bypass signal to be louder than the dry portion of the effected signal so that the overall volume is constant, or do they want to have the effected signal on top of the dry signal with no change in dry volume, so that the effected signal is louder than the bypass signal?

None of the problems I mentioned mean that what the customer wants is impossible, they just mean that you need to know exactly what they want before you start. It's easier to design this sort of thing with a specific goal in mind.

If the customer uses both delay-based and non-delay-based effects, it might be best to design two separate loopers.

In any case, you may want to include a master volume control after the blend pot. Might not be necessary, but you should probably test whether there is a volume mismatch between blended and 100% wet signals.

By the way, your 100% wet switch looks fine, but the led on your feedback switch has no path to ground and will never light up. It's easy to fix, but I'll leave it as an exercise.

[toddvirgil]

Wow! Induction, thanks for your incredible response.

That totally makes sense on the buffer -- So moving it to the output is the way to go for tails. Does any old buffer do the trick? I'll probably use a Klon, but if there' a better one for tails...

So yeah, he's a bass player and he'll primarily be using a delay, reverb and trem in the loop (in his words, "yeah, I"m that guy", lol).

I'll do some thinking on the blend/master volume issue.

Thanks for the catch on the feedback switch, I think I lost a trace when I was moving things around in DIYLayout :) Yeah, taht's my excuse. Figure I'll just move that wire up a pin, eh?

Again, can't thank you enough for the help. I'll redo with updated buffer and maybe a master volume on the blend and repost!

Todd

[induction]

I'm still not sure how useful a blend control is for delay, reverb, or trem, but if that's what the customer wants, so be it.

An output buffer by itself will not give you tails any more than an input buffer will. It's a little more complicated than that. For tails, you need to arrange for both the effected and dry signals to be sent to the output jack simultaneously. The trouble is that if the dry and wet signals are linked together, they can contaminate each other. You could end up with dry bleedthrough (making 100% wet signal impossible). Worse, you could end up with a feedback loop because the output of the wet signal is feeding the input of the wet signal. The best bet is to buffer both the input and the effected output separately, then combine them and send them to lug 1 of an SPDT switch. The unbuffered input signal goes to lug 3, and the common lug (2) goes to the circuit output lug of the 3PDT. The SPDT lets you choose tails or true(-ish) bypass. The buffers keep the wet and dry signals isolated.

Something like this:


I have never built this, so it's unverified. I recommend thorough testing before delivery.

I haven't included the 100% wet switch or the feedback switch, but you can incorporate them easily.

Technically, this is not true bypass because the input jack is always connected to the buffer. That is, the bypass signal in 'True Bypass' mode is not buffered, but is connected in parallel to the buffer. This shouldn't have any audible effect because the input impedance of the Klon buffer nice and high (any op-amp buffer is fine, just don't use a bjt buffer), but technically, you can't call it true bypass.

[toddvirgil]

Wow. Induction, I can't thank you enough. I'm still trying to wrap my head around it, but that looks great. I'll report my results!

T

[toddvirgil]

Thanks again induction. I think I've got it sorted. Below is a a new layout -- I made the trails toggle 3pdt for an LED and to also switch off the input buffer to make it really true bypass -- let me know if I'm mistaken there. Also, customer had some answers to the questions you posed earlier. Do you think this will satisfy his remarks?

Do you always want to switch on the whole loop of effects at the same time or will you switch the individual effects on and off within the loop?
I would like to be able to kill all effects within the loop. I'm gonna have trem, rv-3, and possibly an Akai E-2 but not sure yet. I don't want to dance to turn off all the effects.

Will you want to switch the delay or reverb off and get tails with the other effects still on?
 This I'm not 100% sure but truthfully it doesn't really matter to me at least.

Do you want the bypass signal to be louder than the dry portion of the effected signal so that the overall volume is constant, or do you want to have the effected signal on top of the dry signal with no change in dry volume, so that the effected signal is louder than the bypass signal?
Honestly never thought about that. I really need the the effected signal to be louder than the bypass signal. I'm pretty sure the trem, rv-3, and akai all have some form of a buffer in them. Mainly what I'd want to do is just control how much of the effected signal is being blended into the dry signal.