Blend control with 100% dry + 100% wet

[ihattwick]

So I've been researching blend controls, and it seems like most of them don't allow you to have 100% dry and 100% wet levels at the same time. Most circuits I've seen use a single pot with wet and dry connected to opposite lugs, meaning that as soon as you dial in wet signal, you dial out dry signal. Is there any single pot approach that will allow you to have both wet and dry at full volume, e.g. to blend a delay pedal in with a dry signal. I feel like I've seen people talk about dual-gang pots that have resistance patterns such that there is 0% resistance until you get to the center of the pot. Is this just a fantasy?

[Auke Haarsma]

You could use a splitter. Send one signal to the fx ('wet') and one signal straight to the blender part. But, have two volume controls before the blender. One for the wet signal, one for the dry signal. Use a fixed resistor for the blender (instead of the normal 'blend'pot).

[ihattwick]

Yeah- but I'd like to have only one pot, rather than two. I've looked all over for a panpot that would allow me to do this but haven't found one.

[Dragonfly]

Yeah- but I'd like to have only one pot, rather than two. I've looked all over for a panpot that would allow me to do this but haven't found one.

stereo pot or even a dual gang

[ihattwick]

Do you have any idea where I could find a stereo pot that has zero resistance on both gangs when the knob is at 50%?

[grapefruit]

I think the circuit in the box in this document from geofex will do what you want. I normally use a similar blend circuit whenever I need a blend control. The beauty is that you can have 0% of one signal and 100% of the other signal.

http://www.geofex.com/Article_Folders/panner.pdf

Cheers,
Stew.

[Sweetalk]

Hi!, for blending dry and wet signal I use the Bass Paralooper from moosapotamus and change some capacitors to fit it to guitar. It's really great blender. I also changed the Blend pot to a 500K linear one, helps a lot into the blending.

http://www.moosapotamus.net/THINGS/paraloop.htm

[Brett Sinclair]

If you're looking for balance pots, this looks like a candidate:
http://cgi.ebay.com/Dual-25K-Blend-Balance-Audio-Center-Detent-Pot-4745_W0QQitemZ130183616140QQihZ003QQcategoryZ47076QQcmdZViewItem

I'm not entirely sure, but methinks these are balance pots aswell:
http://www.music-town.de/musik-instrumente/MEC-M-85252-tandem-Potentiometer.html
http://www.banzaieffects.com/MEC-M-85252-Stereo-Pot-25k-pr-13678.html

[ihattwick]

If you're looking for balance pots, this looks like a candidate:
http://cgi.ebay.com/Dual-25K-Blend-Balance-Audio-Center-Detent-Pot-4745_W0QQitemZ130183616140QQihZ003QQcategoryZ47076QQcmdZViewItem

Cool! That looks perfect- thanks! I can't tell you how long i looked for these. Hopefully they'll have them with different values- the title says 25k but the auction text says 250k.

[Processaurus]

The verbzilla digitally has a blend that has 100% dry the entire travel of the reverb blend pot, until the last little bit, when it kills the dry and it is 100% wet.   dunno how you would do that in analog with a reasonable amount of parts.  You could do it where the dry is always in at 100%, and there is a switch to kill it if you want 100% wet.

[R.G.]

The reason that balance setups dial one signal in and one out is to maintain a constant level for the mixed signal, at least in theory.

It's not clear to me what you're trying to accomplish. Let me ruminate on it for a minute.

If you have a wet and dry signal that are the same size - and that is not always the case unless you force it to be - then a hypothetical no-loss blending pot would have 100% wet at one end, 100% dry at the other end, and 100% +100% or twice the signal in the middle. Is that what you want?

Is the business about 0% attenuation on one side that you want the taper to be such that you want, for instance, 100% dry for the first half of travel, then in the middle to jump to 100% dry +100% wet, then to jump to 100% wet for the rest of the travel? If so, we call those pots "switches".

Or is it that you want the pot to start at 100% dry and 0% wet, and increase from 0% wet to 100% wet as you turn to the middle, ending up at 100% dry +100% wet in the very middle, and then the signal remains 100% wet and dry drops off with further rotation from 100% to 0% while wet stays 100%?

Or is it that you just don't want the signal loss? If that's the case, it's much more manageable to just put a gain-of-two buffer after the pot.

[Michael Weidenauer]

The 3rd case R.G. metioned is the thing I'm looking for a long time. It would be really neat in a 2-pickup guitar (or bass) to replace the pickup-switch with this blend control and have no need for individual volume pots.
It would be a stereo pot with zero resistance on the first half of the travel then rising to high resistance (1M would be cool), on the other half of the pot it should be the other way around.
Maybe it woud be possibe with pots with a center tap, but they are also hart to find.
Maybe we could mod a normal stereo-pot by putting some conducting paint on one half of the travel, but I don't know if such a paint exists (should be very conductive and mechanically resistant).
Is there any silver or copper paint that could do that job?

Michael

[Mark Hammer]

To add to that, the capability of panning from fully wet to fully dry will depend on the nature of the mixing node and how signals are introduced to the mixing node.  In a great many wet+dry modulation effects, the mixing node is the point where two resistors meet at the input to an op-amp.  A 50/50 blend often (though not always) demands equal-value resistors.  Increasing the proportion of the mix that one source provides requires its series resistance at the mixing node to be lower than the other signal.

The problem is that it can never be zero ohms or else stuff just starts working very strangely, if at all.  So, say we had a mixing node consisting of two 47k resistors.  You could certainly replace those two 47k resistors with a 50k linear pot and 22k hanging off each outside lug.  While that would certainly permit you to have more wet than dry or more dry than wet, it would never nail you wet-only or dry-only.  It would also tend to create an unintended problem.  The output of the stage (usually an inverting stage) will depend on the ratio of the input resistance to the feedback resistance.  If the stage had a 47k feedback resistor and 47k input resistors, it would be unity gain.  Decreasing one of those input resistances to 22k (with the pot fully rotated in one direction) would yield a gain of just over 2 for one of the signals, and a corresponding change in output level and need for a level control to maintain effect/bypass balance.  Myself, I like to use a combination of variable wet-level, and toggled dry-lift, simply because it avoids the problem noted above.

The other approach to wet/dry balance which CAN achieve 100% wet or 100% dry is to attenuate the signal.  The most straightforward way is to use two volume pots in a reverse-wired, dual-ganged arrangement.  That's a lotta wire, thogh, and you can't always get the dual-ganged pot value you need when you want it.  Which brings us to this:

This is the arrangement used by the MXR Blue Box and is taken from the Tonepad layout/schematic (thanks to Francisco for the drawing).  The mixing node for the fuzz and suboctave signals (our functional equivalent to dry and wet in this example) is where R18 and R19 meet at C10.  The actual resistance for each path consists of R16+R18 for the fuzz side, and R17+R19 for the suboctave side.  Yu will note, however, that the R16/18 and R17/19 paths are each interrupted by a junction with R20.  R20, in turn, goes to ground via its wiper.  In actuality, R16 and one half of R20 form a voltage divider, and R17 and the other half/leg of R20 form another voltage divider.  When the wiper moves away from the midpoint (its a linear pot), that voltage divider provides more attenuation on one side than on the other.  The critical aspect is that while the R16+18 total and R17+19 total NEVER change, the amount of signal reaching the mixing node does.  In this way, you can go from 100% fuzz (full attenuation of suboctave) to 100% suboctave (full attenuation of fuzz).  You might note that this is often the sort of arrangement that is used for Balance controls on stereos.

Can you do this for a chorus/phaser/flanger/delay?  Probably.  Lets go back to our 47k+47k mixing resistor arrangement.  Lets say, we break up the 47k to 15k+33k in series.  At the junction of those two fixed resistors, we tie the outside lug of a 100k linear pot.  Just like the Blue Box, the wiper of that pot goes to ground.  When the pot is fully rotated in either direction, one of those junctions will be tied directly to ground, while the other will have 100k between the junction and ground, fully attenuating one signal and providing some modest attenuation of the other.  Voila!  100% wet and 100% dry.

Are we home free?  Nope.  At the midpoint, there will be what is essentially a voltage divider comprised of the 15k fixed resistor and 50k of the pot, for both signal halves, which WILL lower signal amplitude.  In some instances, you can compensate by simply increasing the value of the feedback resistor of the mixing stage, but that often doesn't work with a great many commercial chorus and flangers.  They use some preemphasis and deemphasis as part of their noise-reduction scheme and that requires adjustment of not only the feedback resistor but several other components to maintain the complementarity of that process.  It CAN be done, though.  You just need to do the math and buy the parts.

[R.G.]

If there's a real need for that kind of thing, I could do some futzing around with control voltages for a dual OTA to get it accurately. What's needed is two control voltage, one which  ramps up to 100% over half a pot rotation, and one that holds at 100% until the pot gets over 50%.

The thing that lets this work for control voltages is that you can clamp the CV in the middle with a zener or equivalent so it simply can't go higher. Could also be an inherent clampb the range of the CV generator's upper limit.

Visualize:
A 10K linear pot with 5V across the outer lugs. The wiper produces 0-5V. Add two 10K's in series across the pot. The midpoint is always 2.5V. Now with the opamps. Use R-R I/O opamps for simplicity. One opamp has its + input tied to +2.5, Rf/Ri=1, and Ri goes to the pot wiper - it inverts the wiper voltage around 2.5V. You now have two identical stages, each with Rf=Ri, and Ri tied to ground. The + input of one goes to the wiper, the + input of the other goes to the inverted-wiper-voltage. This generates the necessary control voltages, now you only need to pipe that into a VCA.

[Ben N]

Here ya go:
http://www.stewmac.com/shop/Electronics,_pickups/Potentiometers_and_push-pull_pots/4/Potentiometers_and_Push-Pull_Pots.html

It would be really neat in a 2-pickup guitar (or bass) to replace the pickup-switch with this blend control and have no need for individual volume pots.
It would be a stereo pot with zero resistance on the first half of the travel then rising to high resistance (1M would be cool), on the other half of the pot it should be the other way around.

This is something I have been thinking about, too, specifically for my LP clone. I'm used to Fenders, so a single master volume makes more sense to me than separate pickup volumes (and I don't much like the way they interact, either). What I would like to have (without losing the pickup switch) is separate tones and the blend pot in place of the two volumes ahead of the switch, and a master volume between the switch and the jack. What concerns me is loading: with 500k elements in the blend and a 500k, either pickup is looking at an effective 250k loading (not even counting the tone controls), which is too low for humbuckers, and maybe too low for my P90s.  As long as Michael brought it up, anyone have any thoughts?

Ben

[Processaurus]

a low tech solution: How about a 12 position switch that would act like your ideal mix pot?  Could be done with a grip of resistors and some opamps.

By the way this is a very reasonable want, it is a serious drawback of panning dry wet controls that the there is a volume drop in the dry as you blend.  I had that problem using the deluxe memory man and holier grail live, that the instrument would disappear in the mix when I kicked one on.  The 1st rule is that something has to be loud enough to even be in the running for sounding good.

[ihattwick]

Sorry, there have been a bunch of responses to this thread I haven't seen yet. Turns out that the stew-mac blend pots do work like I was hoping, and at the center detent there is zero resistance to the output lugs. Unfortunately I have found the other problem of this, which michael mentioned, which is that if you just connect the output lugs of both halves of a 500k blend pot, even at full volume each side will see only a 250k load (two 500k loads in parallel). Right now I am trying to use a blend pot to connect  a lipstick pickup and a contact mic, and the load the contact mic is seeing is so little I am not getting a decent signal from it.

[tommy.genes]

I'm glad that this topic got bumped. Since the last of the original posts in January, I spent some time doing overdubs in the studio. I was trying to get a good blend of clean and effected sound using the passive blend control in my Paralooper and it just wasn't happening. Finally, we took a clean signal right off the bass with a DI before the effects, and got the effected sound from the mic that was already on the amp. Not only will this give the engineer more options during the mix, but it sounded much better right there in the control room too.

I've been thinking about (since I rarely have time to do anything about) a better blend circuit that would capture that experience. I was wondering if two separate volume knobs would be better, since that would most accurately recreate the original situation, or whether a 100%/100% "balance" knob as discussed here, followed by a master volume to adjust for the effective 200% signal that would result, would be a more useful control set in a live setting.

Any thoughts?

-- T. G. --

[ihattwick]

Hmmm. . . seems the paralooper should already do what you are looking for. I guess if the blend control on it is passive, it would work better if there was an active mixer to blend the two signals- you could have the dry and wet have individual levels and pan ideally. Basically, it's just splitting your signal to two effects and using a mixer to combine them.

Probably a big part of what you liked about splitting your signal in the studio was not having the dry and wet signals run through your amp simultaneously- running into two channels of a mixing board will sound like using two amps, especially if you pan the signals at all.

[tommy.genes]

Probably a big part of what you liked about splitting your signal in the studio was not having the dry and wet signals run through your amp simultaneously- running into two channels of a mixing board will sound like using two amps, especially if you pan the signals at all.

There's probably something to that, but eventually the signal from the board goes into a single, well - stereo, amp and out through the studio monitors. My bass amp is an Eden and it really doesn't color the sound much at all if you don't want it to. I'll just have to play with an active mixer, or some variation on R.G.'s panning ideas.

-- T. G. --