Dave's Boost A/B Pedal

[ashcat_lt]

(This post was composed for the GuitarNutz forum, so there are some reference to things I've not discussed on this board.  I'll add some links when I'm not on the iPhone so that some of it makes more sense.  The details of the pedal itself are pretty clear anyway.)
Dave is the other guitarist from Circadian Nations.  I've built custom pedals for our bassist, Luke (1337drive) and myself (a Rat-based thing which I still need to report on) and felt I should make something for Dave.  It was his birthday recently, and I also wanted to solve an issue we've been experiencing in our live sets.

As I've mentioned previously, Dave switches back and forth between an electric guitar (ES335 clone with HBs) and an electric mandolin with as single-coil.  The guitar is a bit louder than the mando.  We had been running 2 seperate V-Amps - one for each - but my original blue tabletop model finally gave up on us.  So we had set up two custom patches on the V-Amp Pro which he is using.  They're exactly the same except the mando's has a bit higher gain setting.  Works fine, except that Dave doesn't generally stand close enough to the rack to make switching easy, and I don't have a footswitch for the thing.

So I decided to make him a simple opamp based booster.  He could set the gain rather low, and when he plugs in the mandolin turn it on and problem solved.  Then, too, on the rare occasions where he actually gets to use one of his real tube amps (he's got a couple vintage Fenders which I won't let him use with the band) he could crank the gain for some nice crunch.  All quite simple, and he'd probably been happy with that.

Of course, as you may be aware, I'm a little Nutz.  I couldn't leave things there.  Enter the Complications.

Complication #1 -  A Push/Push Gain pot.  By making this switch introduce a static resistor of the same value as the pot in series with said pot we get a "More" switch.  Turn the gain all the way up.  Still not enough?  Turn it back down and push the switch and you can get more!  The pot I got from RadioShaft is a 500K pot with only two lugs.  So the gain pot works backwards - clockwise is down.

Complication B - Adjustable minimum and maximum gain.  I don't see any point in allowing the gain pot to turn all the way down to unity, but that 0 setting makes for an easy to remember setting for the amount of gain required to compensate for the mando.  Of course, I didn't have access to his instruments to be able to find that exact ratio.  So I used a trim pot to set this minimum gain.  While I was at it, I threw in a trim pot on the ground leg of the feedback path to set max gain.  I mean, why not?  Speaking of that "ground leg"...

Complication the Third - By switching in a smaller cap in this ground leg, I can convert the thing to a treble booster.  Don't know exactly why he needs this, but it's an easy mod.  This is somewhat reactive to the gain pot in that as you turn up the gain, it gets more trebley.

At this point I was happy that I had a pretty versatile box that would work for CN and give him options for other things.  I bought all the components - including a fancy little LED indicator built into its own mount with an integral resistor and everything.  Then I talked to Dave (this was supposed to be a surprise) and he introduced...

The Final Complication - Not knowing what I had cooking, he mentioned that he'd like to have an A/B switch.  This would allow him to plug in both instruments and avoid the time and loud noises through the PA entailed in unplugging and re-plugging.  So I worked that in here, too.  I left an important part of this out of the schematic below.  Specifically, the Input B jack is an NC switched type.  The tip's switched lug is connected to the tip of the Input A jack.  When nothing is plugged into B, signal flows to Sw1a via both paths, and it works like a buffered bypass.  Plug guitar into B and mando into A, and when guitar is selected it's buffered and when the mando is selected it gets boosted.  Using the DPDT stomp switch that I had, it meant I had to jettison the fancy LED, but not a big deal.  It's easy enough to tell if it's on or not.

So a very simple opamp booster turned into this:

Now there's nothing particularly groundbreaking here, but I did design it from scratch, chose values based on 5spice models, and I'm kind of proud of it.  The bypass switching is something I've never seen before.

I wired this all up on Friday night and wanted him to be able to use it at our big Samedi Gras party the next night.  I was "done" at about midnight.  Plugged it in to test it and !  Opened it back up and found I'd transposed the wires going to Sw1b, and had shifted the connections meant for pins 6 and 7 on the opamp to pins 7 and 8.  It's tough to see those little pins poking through the perfboard!  Fixed that and it worked fine in bypass, and worked fine when the gain was all the way down.  Turning up the gain made it fart out and go silent, as did pushing the "more" switch.  Darn it!

You generally don't see Gain pots in the 500K range, and I was worried that this might be the problem.  But it worked fine (if very noisy) on the breadboard.  So, like WTF?  Tweaking the min and max trims I could get it to where I could turn gain up a couple clicks (the pot has thirty something detente) but still not acceptable.  Measuring voltage I found that with gain all the way down, the feedback path read right in the middle of the supply - where it should.  Turning up the gain caused this voltage to drop.  I measured across the gain pot and found that it was actually dropping DC voltage.  The caps to ground should stop that happening, but...

About 3AM, after the second time I'd decided that it just wasn't going to happen, and cleaned up all my crap, I looked one more time.  This whole thing is crammed into a single-gang light switch box (I'll get a photo soon, you guys are gonna love it!) and one of the lugs on the Treble switch had ended up touching the grounded lug on Input B.  This defeated those caps to ground and created a DC voltage divider in the feedback path.  As I turned up gain the bias voltage drove toward ground and made the thing just not work.  A quick and gentle bend if those lugs and it was working!  He used it Saturday night and it was perfect!  We may need to tweak that min gain trimmer, but otherwise he was surprised and touched and very happy.