Pork Barrel - Chorus disappears when rate LED is connected

[benny_profane]

(1) Project Name: Pork Barrel 2015 rev.1
(2) General description of the problem:
The chorus works wonderfully until the rate LED (LED2) is wired up. Here, the engaged sound is pretty much the same as bypassed. Also, the rate LED remains solidly lit (no tracking with rate/depth).
(3) Steps that you have taken to try to resolve the issue:
I've rechecked parts and soldering. I removed and checked all off-board wiring and pots (to take a look under). Cleaned flux residue. Checked for unforeseen continuity with DMM.
(4) List any substitutions you used for parts/values:
Board is built to stock specs for the MN3101/3007 combo (sourced from SBE). Effect is powered by a 3PRR providing 12v.

I can provide measurements soon.

[somnif]

Hard to tell with the pots and LED taken out, but my first thought is something may have been shorting across the back of the pot's metal body.

Some of the snipped leads do look a little tall, but I can't tell for certain at that angle.

[benny_profane]

Hard to tell with the pots and LED taken out, but my first thought is something may have been shorting across the back of the pot's metal body.

Some of the snipped leads do look a little tall, but I can't tell for certain at that angle.

The pots have plastic dust cases/insulators on them so I don't think there's a short there. Thanks for the suggestion, though.

[madbean]

Backwards LED, maybe? Or, perhaps check the ground pad of the LED to make sure it is actually grounded.

[benny_profane]

Backwards LED, maybe? Or, perhaps check the ground pad of the LED to make sure it is actually grounded.

The LED is oriented correctly. I tried multiples and switched orientation just to be absolutely sure. Negative pad has continuity to ground. I just triple-checked the pot continuities and everything looks good there, too.

[zgrav]

So there is enough voltage there to light the LED, but (at least when it is in circuit) too much voltage to let the LED turn off.  Maybe the variation in voltage that makes the chorus work without the LED is happening above the turn on voltage for the LED.  Have you checked the voltage going to the LED mounting hole with no LED in the board?   If for some reason the voltage in that part of the circuit never dropped below the LED threshold (1.4v or so), maybe putting the LED in the circuit would turn on the LED but then pass all voltage above the threshold to ground, stopping the chorusing.

[benny_profane]

So there is enough voltage there to light the LED, but (at least when it is in circuit) too much voltage to let the LED turn off.  Maybe the variation in voltage that makes the chorus work without the LED is happening above the turn on voltage for the LED.  Have you checked the voltage going to the LED mounting hole with no LED in the board?   If for some reason the voltage in that part of the circuit never dropped below the LED threshold (1.4v or so), maybe putting the LED in the circuit would turn on the LED but then pass all voltage above the threshold to ground, stopping the chorusing.

That's an interesting idea. I just checked the anode voltages with and they're quite high! I was getting varying voltages between 8v and 6.5v. So it seems like you're on the right track with your idea.

I would omit the LED entirely, however, I've already finished the enclosure with the inclusion of the rate LED. So, what would a possible fix be? The rate LED circuit is basically a 100n capacitor in series with a 1k resistor. I would ideally like to use a green diffused LED.

I think this is on the right track!

[somnif]

I don't see anything wrong with the board itself, the resistor values in the area around the LED are all correct (visually, anyway) so I can't think of any weird current shunting going on that way.

Past that it leaves case installation faults. Something shorting somewhere that is either cutting the rate/depth portions of the circuit out and shunting that current to the LED, or wonking the rate/depth to the point where the rate is so fast (or depth so shallow) that its essentially "always on" and resulting in a similarly weird signal getting to the BBD.

[benny_profane]

I don't see anything wrong with the board itself, the resistor values in the area around the LED are all correct (visually, anyway) so I can't think of any weird current shunting going on that way.

Past that it leaves case installation faults. Something shorting somewhere that is either cutting the rate/depth portions of the circuit out and shunting that current to the LED, or wonking the rate/depth to the point where the rate is so fast (or depth so shallow) that its essentially "always on" and resulting in a similarly weird signal getting to the BBD.

So, I think it's something to do with the voltages across the LED at this point. Especially because this happens regardless of depth/rate and in/out of the enclosure. It will happen if an LED is just placed into the pads while it had previously been working properly. The issue will then resolve itself immediately once the LED is removed.

I thought one of those things had to be the case...I removed all of the off-board components and triple checked everything. Even though I wouldn't be surprised if it's something super obvious, I think the LED might just be driving the wet signal to ground.

[madbean]

Hmm. Have you tried a higher value resistor? Maybe 4k7 or 10k. Easy to check - slip a wire into the square pad of the LED (no need to solder it) then put the other end to a breadboard - 10k resistor, LED to ground.

[somnif]

So whats happening is there IS a short somewhere, but its got no path to ground until that LED is introduced so the circuit behaves normally. Use your DMM on continuity, and stick one probe in the LED's square pad, and start poking around till you hear a beep. If you get nothing, move to the other side of R44 and poke around. Then move to R30 and R31, R33, C19, etc.

You've got current jumping a queue somewhere in there.

[benny_profane]

Just gave the added 10k resistor a shot and it worked brilliantly. Did an A/B by taking away the ground intermittently with no appreciable change in the sound. I also rechecked continuity with the positive pad just to be sure and I didn't find anything unexpected. So I think the resistor is the solution.

I'll audition a few values then swap R44 out.

Cheers to everyone for the help here!

[madbean]

That's quite a surprise. I've used the 1k/LED setup on tons of different LFOs and it's never posed a problem. The only one I can think of off-hand was with a Univibe where the LED did result in a different depth response, but that's a completely different type of LFO.

Can't say why it's happening for you but I had a hunch. Don't know if the LED is acting as a current sink or somehow screwing with the depth overall.

[benny_profane]

That's quite a surprise. I've used the 1k/LED setup on tons of different LFOs and it's never posed a problem. The only one I can think of off-hand was with a Univibe where the LED did result in a different depth response, but that's a completely different type of LFO.

Can't say why it's happening for you but I had a hunch. Don't know if the LED is acting as a current sink or somehow screwing with the depth overall.

I'm running at 12v off the 3PRR board. Could the extra voltage (re zgrav's comment) be the issue?

[madbean]

That's quite a surprise. I've used the 1k/LED setup on tons of different LFOs and it's never posed a problem. The only one I can think of off-hand was with a Univibe where the LED did result in a different depth response, but that's a completely different type of LFO.

Can't say why it's happening for you but I had a hunch. Don't know if the LED is acting as a current sink or somehow screwing with the depth overall.

I'm running at 12v off the 3PRR board. Could the extra voltage (re zgrav's comment) be the issue?

Yup, that's it. In fact I ran into a very similar issues when I was working on modulation for my AD-900 project. When I powered the LFO board @ 12v the entire depth shallowed out. I changed the voltage divider supply to the LFO from two 220k to 470k and that fixed the issue. Anyway, the PB LFO runs fine @ 12v, but something about the combination of the LED flattens it. So, I guess the solutions is more resistance in front of the LED.