Bass control on Widgy Box EQ cuts overall volume when rolling past noon?

[Rob Strand]

I assume this is for the aforementioned purpose of providing the output voltage with enough room to swing in either direction? Their input buffers are also pretty much identical save for a couple of values, which reinforces what I've read about "bookending" circuits (for lack of a better term) to sort of set them up, so to speak (but please correct me if I'm mistaken!)

All spot on.

[Andon]

Alrighty, time for some updates.

I tried both methods of adding a 1M resistor from the JFET gate back to 4.5V while increasing C13 from 22nF to 47nF / and also the suggestion of removing C13, R15, Q1, R16 and adding a 1K resistor from pin 6 of IC2 to the positive side of C14, and while both methods seemed to help they didn't completely eliminate the problem. These were both done a couple of test PCBs I had, so it took a bit to try them both and report back. A couple of observations though:

Diming the volume on the pedal almost gets rid of the problem, but then you also have to run the pedal on 10 the whole time (which may not be ideal for everyone).

Playing on the neck pickup also makes the problem almost nonexistent, I assume because the bridge pickup is higher gain so it's pushing it less (this was all done using a telecaster with stock pickups).

Would increasing the resistor value of the patching 1K or the 1M further alleviate the blatty/gated performance in either of the two solutions? Keep in mind that the issue with the blatty/gated-ness is only on lower notes/chords when strummed moderately hard or more so. I know that lower frequencies require/have more energy/voltage (unless I'm mistaken?), so I could see that being an underlying issue. That said, if I turn the bass all the way down it has a negligible difference (it also doesn't cover the entire "bass" spectrum).

On that note, the issue with the bass rolling off volume past noon also persists, and removing R7 (18K) seems to be the only fix, but I don't know why. It seemed to work on the breadboard but when PRR suggested the additional 4.7uF cap, and I'm pretty compulsive when it comes to plotting my PCBs from schematics. I would also be happy to send along my PCB CAD file for someone to idiot-check, provided they're also using KiCAD.

[Rob Strand]

Just to confirm the problem is the sound cuts out when the bass control is advanced past 12'O Clock?
And, if you pull R7 (18K) it works fine?

So things come to mind a DC issue or an oscillation issue.   I know for a fact these three bands can
oscillate on occasion.   

For the DC issue perhaps there a soldering issue on the R10 side of the bass pot.  Either the counter-clockwise terminal or R10 itself.    Another  possibility is R10 itself.

One question though, if you measure the DC voltage on pin 6 of U2 and vary the bass pot from 7 O'Clock to 10 O'Clock to 12 O 'Clock, does the DC voltage on pin 6 voltage vary?  If it does that means you have an unintentional short around the tone control, perhaps to ground.   In fact it could be near the input side along R7, R8, R9.

For the oscillation issue it might be wise to increase to C12 to 100pF or even 220pF.

I'm not fond of the large value of R14 but it might be OK.  You could drop it to 1k.

[Andon]

Just to confirm the problem is the sound cuts out when the bass control is advanced past 12'O Clock?
And, if you pull R7 (18K) it works fine?

Correct, and this was actually the issue that started this whole thread. As mentioned, it seemed like it was fixed with the addition of the 4.7uF capacitor before the EQ stage, but it's back again. This would also rule out a soldering issue on the R10 side of the bass pot since it's a persistent issue, though I will say that removing R10 drops the signal down to practically nothing.

I currently have R7 removed, so the DC voltage sits at 4.6V no matter where you turn it. My soldering iron and solder sucker are warming up now, so I'll replace R7 to get a DC voltage reading for you as well as replace C12 and R14 after re-removing R7 (just in case) and report back here in a few minutes.

[Rob Strand]

I currently have R7 removed, so the DC voltage sits at 4.6V no matter where you turn it. My soldering iron and solder sucker are warming up now, so I'll replace R7 to get a DC voltage reading for you as well as replace C12 and R14 after re-removing R7 (just in case) and report back here in a few minutes.

OK cool.   

The 220pF for C12 is a little too high but for debugging problems like this it's sometimes a good idea to be heavy handed just to see the effect.  You can always back it off later to find the balance point.

At the risk of throwing too many things into the mix, another oscillation reduction scheme is to increase R13 and leave C12.  You could raise R13 to 10k or so.  The overall effect isn't much different to raising C12 although it might be more robust at fending of oscillation with different control settings.

[Andon]

Alright, did those steps backwards so I wouldn't have to un-solder and re-solder R7.

I replaced C12 with a 220pF capacitor and dropped R14 to 1K and the blatty/gated issue remains, though it seems better (I could just be wanting to hear a difference)? Anything below 220-165Hz (A3-E3) seems to blat out right when you first strike a note/chord before clearing up. There's no gated effect whenever you palm mute lower notes/chords, but that's also because you're not playing as hard.

I replaced R7 and there is indeed a difference in voltage on pin 6 of IC2:

7 o'clock: 5.20V
10 o'clock: 6.94V
Noon and beyond: 8.70V

[Rob Strand]

7 o'clock: 5.20V
10 o'clock: 6.94V
Noon and beyond: 8.70V

For the DC issue that's what I was expecting.

When you get to noon the opamp output is sitting at the supply and there is no room for the opamp to swing positively so it clips and sounds bad.   Past 12 O'clock it blocks the signal.

Set the treble and mid pots to 12 O'clock.  With all the parts in and the bass pot set to about 11 O' clock,  take DC voltage measurements
- at each of the pot connections (for all pots)
- at the point where  R7, R8, R9 join
- the opamp output pin
- the two opamp input pins

There should be a DC voltage in one of those which exposes the problem.   I'm expecting one of the voltages will be below 4.5V.

[Andon]

Bass:
1: 6.66V
2: 4.60V
3: 1.54V

Middle:
1: 6.80V
2: 4.04V
3: 1.34V

Treble:
1: 6.93V
2: 3.94V
3: 1.22V

Volume:
1: 0.00V
2: 0.00V
3: 0.00V

Gain/Distortion:
1: 4.61V
2: 4.61V
3: 4.61V

R7/R8/R9: 0.65V

IC2:
Pin 6: 7.58V
Pin 2: 4.60V
Pin 3: 4.40V

It should be noted that the only place that R7, R8, and R9 meet on the PCB are at the two clipping diodes, but every measurement from that point including the individual resistors and C18 (the added in 4.7uF after the clipping diodes) are all at 0.65V.

EDIT: This is also with the additional 1M resistor from the JFET gate to 4.5V.

[Rob Strand]

R7/R8/R9: 0.65V

The cap you added between the diodes and the where R7/R8/R9 join is faulty, or, in around the wrong way (need cap- to diodes), or, has a short across it.

That's the problem for sure!


EDIT:
FYI, all the voltages throughout the tone control should sit at 4.5V regardless of the control settings.  (4.5V, with some small error caused by the meter loading.)

[Andon]

Just wanted to update this thread and say that the problem has been resolved, and it was indeed a short as Rob had suggested, so mad props to him.

What had happened was when I was tracing the nets on the PCB that whole section had somehow looped back into itself and I must have missed it when I was clicking through the design checks that I thought were just flags for un-netted mounting holes that I use to connect a ribbon cable to my 3pdt board.

So yeah, entirely user error, but I did at least learn a helluva lot about what I should be looking for for whenever I invariably have another biasing issue down the road.

Many, many thanks again to Rob for helping me troubleshoot this (which led me to see my own error), and to everyone else that offered advice along the way - I appreciate y'all!