Boss FA-1 Rehash Circuit Dilemna (Dilemma) Need Help

[Rob Strand]

Did a quick test from C10 and R12 with +27V power on, effect engaged, haven't tested much yet for resistance from that point.
The ones I did test first were C10, R12 to the +9V point on main board at pin 4 of the right side 4 pin header
The resistance there with +27V power selected was 0.000
Tested it in the 9V switch setting, and it tested between C10,R12 and +9V pin at over 6 Meg ohm.
So definitely something amiss there, could be that voltage select switch.
Will keep searching.

Need to measure with power off.   DMM's can do weird stuff with the power on.

[Rob Strand]

After pondering over this again, I reckon the thing to try would to add the 10k resistor after C10, and before the relay.  See if it affects the DC offset issue.

With the schematic as drawn I can't see how you can a constant 7,1V DC on the output.   It's possible something is glitching when you change over the power and that charges up C10 but if all is working that transient should settle down after a short time and the output at R12 should have no DC.

So much to do and so little time
Even the dishes wont do themselves.
Thanks for the ongoing support, always appreciated. Cheers

Unfortunately true.  I've got some projects on the back burner with long ago start dates.

[bluelagoon]

add the 10k resistor after C10, and before the relay

Okay, placed a 10k resistor to Ground straight after C10, with R12 still removed. Tested the Voltage at output, and it did lower it down to +5.9V which is better than previous above +7V, but still would prefer 0 volt at the output, as pretty sure that voltage would still amount to a lot of clunks and thuds at switching.
What do you think if perhaps I replace the 10k with the 100k R12, might that lower it further?

[Rob Strand]

add the 10k resistor after C10, and before the relay

Okay, placed a 10k resistor to Ground straight after C10, with R12 still removed. Tested the Voltage at output, and it did lower it down to +5.9V which is better than previous above +7V, but still would prefer 0 volt at the output, as pretty sure that voltage would still amount to a lot of clunks and thuds at switching.
What do you think if perhaps I replace the 10k with the 100k R12, might that lower it further?

Wow 5.9V,  that result is just nuts.    That means whatever is causing the 7.2V has the ability to supply a lot of current.  I = 5.9V/10k = 590uA. 

The fact it doesn't change much from 100k to 10k means:
- the impedance getting to the JFET/R12 is low.  A rough estimate is (7.2-5.9)/590uA = 2k2.
- the voltage supplying the JFET/R12 is in the order of 7.2V, probably a bit more like 7.2 + 2k2 * 7.2/100k = 7.4V.

When I look at the circuit the area which stands out to me is: Q2 drain (2N7000 LED MOSFET).   Perhaps R39, T1.

When the LEDs are on the voltage at the drain of Q2 is going to be a bit less than 7.7V.  That seems far to close the 7.4V estimate to be a fluke!

So maybe check for a short from the drain of Q2 (MOSFET) to one of the the pins of the Q3 (JFET).  Turn the power off and see if you can find a short or low resistance between the two parts.  If you find a low resistance look at few other points in the vacinity.  I suspect you will find a short somewhere.

I wouldn't recommend changing R12 to 10k.   It's just hiding the real problem a bit.   It looks like real short is somewhere on the the board.

Hopefully it is a simple short.  If the short goes through any relay contacts you need to use a different method:
- power the unit
- set the unit to 27V mode
- measure voltages on the pins of Q3 (JFET)
- try to find another point on the board which has the same voltage.

What still doesn't fit this picture is why the problem only shows up in 27V mode.  It could be an obscure short.

[duck_arse]

I keep trying to make sense of this section of board - it seems none of the parts designators match the circuit diagram. and what is Q2 that connects to the tip_OUT? is Q1 really Q3 in disguise?

also, to save our hard drives from accumulating a thousand local files called prism-revision, could you maybe add a revision number to each changed document? a date would also benefit your good self for future references.

[bluelagoon]

No worries Duck, I will get onto and update the circuit to show all those missing components, soon as I get a chance. Have a lot else on as usual which doesn't make it easy to get there so quick, but hopefully by the end of the week.
Those transistors are part of a temporary mute function added to the Coda bypass to ground the output temporarily at switching to overcome the adverse switch clicks, used 2n7002 transistors in place of a TLP222A photo relay, to get the same result. Coda circuit pretty close to what is shown here -
Cheers.

[bluelagoon]

Tested between Drain of (Q2, 2N7002) and  (Q3, J175) Source with (R12, 100k) still removed, resistance was 16.4k

Between Drain of (Q2, 2N7002) and  (Q3, J175) Drain with (R12, 100k) still removed, resistance was 18.1k

Also tested the same on a seperate power board with the R12 still in place and resistance between Drain of (Q2, 2N7002) and  (Q3, J175) Source was 122k
Between Drain of (Q2, 2N7002) and  (Q3, J175) Drain was 122k

Cant see any shorts in the layout.

[bluelagoon]

Just tried something that might have been suggested earlier. Disconnected the output C10 and R12 from the VFE transistor switched bypass output side of circuit, tested the voltage at signal end straight off C10 with none of the VFE switching at end of circuit. In both power select states I get 0 voltage at the signal output. So either something needs modifying on that VFE out put side transistor switching. Or I may just try it the way I used to, using a CMOS 40106 and the Mictester style Relay Bypass, at least they gave me silent switching no clicks no excess voltage at output.
Still would be nice to remedy the situation with whatever is the problem in the VFE bypass in this situation.

Maybe I should just stick to plain vanilla style power supplies.
But I figured the klon dude did something similar, so thought I would give it a try, just seem to be let done with some kind of glitch.
Will keep pressing on.

[Rob Strand]

Just tried something that might have been suggested earlier. Disconnected the output C10 and R12 from the VFE transistor switched bypass output side of circuit, tested the voltage at signal end straight off C10 with none of the VFE switching at end of circuit. In both power select states I get 0 voltage at the signal output. So either something needs modifying on that VFE out put side transistor switching. Or I may just try it the way I used to, using a CMOS 40106 and the Mictester style Relay Bypass, at least they gave me silent switching no clicks no excess voltage at output.
Still would be nice to remedy the situation with whatever is the problem in the VFE bypass in this situation.

Maybe I should just stick to plain vanilla style power supplies.
But I figured the klon dude did something similar, so thought I would give it a try, just seem to be let done with some kind of glitch.
Will keep pressing on.

Well that's some progress.   What I'm reading into that is the grounds and supply switching are OK and aren't causing the 7.1V voltage on the output.   You already checked the ground connections before so that gives some confidence the ground and supply wiring is OK.   If for example the ground on the switching board was at ground but the ground on the analog board got switched to -9V (-7.2V) then it could cause an apparent DC voltage - it wouldn't be good.

On paper the JFET output circuit looks like it should work.   If there was some obvious design fault we should be able to see it on the schematic.   For example the output distortion and output swing issues can be deduced from the schematic.   The thing that really stumps me is the whole JFET circuit sees at most +5V but the DC voltage on the output terminal is +7.1V.   That's what makes be think there is a problem outside of what is visible on the schematic.  Otherwise I'm waiting for Penn and Teller to pop out from behind the curtain and tell us how the trick is done  .

The ways crazy bugs like this get in are:
- Shorts on the PCB/wiring/switched.  Making connections which aren't visible on the schematic.
- PCB connection issues.   For example when pins on the schematic don't match up with the PCB and the real parts.
  Imagine the relay pinout being wrong and all connections being mixed up.  That would cause all sorts of weird
  connections on the PCB which you can't see on the schematic.
- Wiring problems between PCBs is another one.   Where pin x at one end doesn't wire to pin x and the other end
  because a cable is flipped, connections numbered differently on each board, when the boards are connected
  the pin number are flipped because the connectors are place on the wrong side of the board for mechanical reasons.
- Parts loaded incorrectly.  Wrong values.  Wrong orientations.

In short - the eyes see the schematic but the actual circuit isn't following that.

IMHO the best way to find these crazy bugs is to get the circuit in a state where you are getting the +7.1DC on the output and try to find another point in the circuit which has the same voltage.    The 7.1V has to come from somewhere.    (If there is still a ground problem then you might not find such a voltage.)

[bluelagoon]

Thanks Rob for the insights, things to look for. Will keep looking into it. Should have a full schematic account of complete circuit up shortly, But not sure that will even help now beyond what has already been presented.
Cheers

[Rob Strand]

Thanks Rob for the insights, things to look for. Will keep looking into it. Should have a full schematic account of complete circuit up shortly, But not sure that will even help now beyond what has already been presented.
Cheers

Maybe so but you never know what happens.   Duck's got sharp eyes.

I suspect it's going to take some quite detailed probing and checking at your end.   It's very quick to probe 20 points with the unit in front of you and you can discard "no results" on the fly.   If you had to-and-fro 20 measurements via posts it's going to be a very slow hard slog.  It only takes one obscure build problem or human error (we all do silly mistakes) to cause a problem.  Sometimes it's hard to see these things but it's "so obvious" when we find the problem.

[bluelagoon]

Alright, Finally have the complete circuit Schematic, all on the one page.
Up to Date, all anomalies fixed, circuit just as I have it assembled.

Main issue still stands, excess voltage at the output signal at around +7V and sometimes a bit more, when switched to the +27 voltage power selection, to run with higher headroom on last four op amp stages.

You can also see now the Baxandall Treble and Bass switched in and out selectable by a momentary switch using a Coda Bypass
circuit and Coda programmed MCU.

Also there is a 4PDT Slide Switch, which allows for selecting where the Electra Fuzz circuit is placed within the Fa-1 circuit, Selectable as inserted Pre EQ section after C5, or Post EQ section after C10.
Have tried in both positions, and weren't too keen on the Pre EQ insertion of the Electra Fuzz, it sounds a whole lot better Post EQ after C10, Likely would not include this option in future builds.

One thing still puzzling is that when the Electra Fuzz is put into the circuit either pre eq or post eq it still settles the voltage at output down close to zero, where it should be. its a bit puzzling in that it gets the voltage down even when placed in the earlier part of the circuit after C5.

But have established as noted previous post, that disconnecting output signal from VFE transistor switching at C10 Negative, and placing R12 at same point to ground, as in original output of Boss FA-1, testing at this point gives a reading of around 0 Volt DC at output with effect engaged in both +9V mode and +27V mode. Which is encouraging, getting closer to a resolve.

[bluelagoon]

Further analysis

Did some more probing, and think I have found the culprit in the high + DC on output signal.

Tested at traditional FA-1 output termination after C10 in circuit where R12 goes to ground
While it was totally disconnected from any following circuit. It tested at aprox. 0V DC with
both +9V and +27V power modes selected. See attached image Exhibit 1A




Tested again at same point location with the circuit continued directly on to FX Out pin of
Relay 1 on Power Board PCB. It once again tested with excess DC voltage about +7V at signal output.
See attached image Exhibit 1B



This led me to believe I had pinpointed the likely culprit as being the temporary mute transistors attached to the Guitar Out terminal of the output jack J2. These are activated with a short pulse from the Coda Bypass MCU - BAX Out Switch which sends a short pulse grounding momentarily the output signal to eliminate excessive clicks and harsh pops from switching. They usually work well in most instances, and usually I have only ever used just the one transistor on its own which has sufficed. This time I used dual transistors to emulate something similar to the TLP222G optofet with its back to back body diodes.

Long story short, I disconnected Q2 from where it joined J2 Guitar output jack signal out, and the Voltage tested then at output
is corrected back to where it should be around 0V DC in both +9V and +27V power select modes.
So me thinks its hopefully the end of dramas here with this circuit. See attached image Exhibit 1C



Just still some questions though. Does any one consider it possible that if I substituted here these transistors Q2 and Q3 for a
an optofet TLP222G or similar, would this then still get the desired result of a momentary mute without the voltage on the output signal?

And another thing that still puzzles is why when the Electra Fuzz is switched in, it eliminates any issue with the Q2, Q3 temporary mute DC Voltage on output??

[Rob Strand]

Excellent debugging.

I had a look at the full/new schematic yesterday but couldn't see anything obvious so I was going to take a longer look today.

This led me to believe I had pinpointed the likely culprit as being the temporary mute transistors attached to the Guitar Out terminal of the output jack J2. These are activated with a short pulse from the Coda Bypass MCU - BAX Out Switch which sends a short pulse grounding momentarily the output signal to eliminate excessive clicks and harsh pops from switching. They usually work well in most instances, and usually I have only ever used just the one transistor on its own which has sufficed. This time I used dual transistors to emulate something similar to the TLP222G optofet with its back to back body diodes.

Long story short, I disconnected Q2 from where it joined J2 Guitar output jack signal out, and the Voltage tested then at output
is corrected back to where it should be around 0V DC in both +9V and +27V power select modes.
So me thinks its hopefully the end of dramas here with this circuit. See attached image Exhibit 1C

That's great progress.

So at this point I'm still not convinced the back-to-back MOSFET circuit *injects* the +7.1V because the whole MOSFET circuity is only powered from +5V.   With the Q2 drain disconnected what DC voltage do measure on the drain of Q2?   I would expect zero volts.   If there was a MOSFET pin-out issue you might see +4V to +5V.   Circuits around the JFET are also 5V.

Just still some questions though. Does any one consider it possible that if I substituted here these transistors Q2 and Q3 for a
an optofet TLP222G or similar, would this then still get the desired result of a momentary mute without the voltage on the output signal?

I've used those back to back MOSFET connections and never seen any odd behaviour.   And yes, off-the-shelf devices like the TLP222G use the same sort of drain/source connection although the gates are not DC connected.   Even if your MOSFETS were fakes and were re-marked BJTs we would not expect to see +7.1V, only +5V or so.

So in my mind the *root* cause still isn't clear.

And another thing that still puzzles is why when the Electra Fuzz is switched in, it eliminates any issue with the Q2, Q3 temporary mute DC Voltage on output??

It puzzles me a bit too.   

So two things stand out:

- The output diodes on the Electra will clamp any DC voltage.
   If +7.1V was injected then you would by the MOSFET or JFET or other circuit the
   DC voltage would be held at one diode drop.

   You should measure the voltage across the diodes to see if it's 0V or a diode drop (ge) 0.2V to (si) 0.7V.

- When the Electra is selected the opamp inverter (IC3C) connects to the Electra input not to the JFET/MOSFET
   circuits.

When power is switched the opamp (IC3C) will have a glitch at the output.  If the mute circuits is active it will
short the output at C10 to ground.  The glitch and mute will charge-up the cap C10.   The JFET ckt could also contribute.   So in this way a DC voltage could be captured at the output.  This is a normal type of thing
at the output and R12 to the added R12, which connect to ground, should cause discharge/charge C10 a short time later so the output sits at 0V.  So all being normal we would not expect the +7.1V at the output.

What we are seeing is a persistent +7.1V.

At this point I keep thinking IC3C is oscillating.  The oscillation could perhaps be induced by the Mute circuit when the power is switched.   Oscillation and JFET might be rectifying the output and charging C10.   The 7.1V comes from the opamp output.  However, with a resistor like R11 on the output of IC3C it should be pretty robust.   If you have an oscilloscope you could just probe the output of IC3C and see oscillation.

Without an oscilloscope, something worth trying would be to add say a 470pF across R28, the feedback resistor for IC3C.   That would make oscillation less likely.   Another thing would be a cap from the -Ve power pin of IC3 to ground.

So I don't think it's over yet.   Nonetheless your results have put a tighter box around the problem.  Some of the things I've suggested above would definitely help narrow things down a bit more.

[Rob Strand]

I had an idea that the problem could be to do with the back to back MOSFET switch.

When I've used back to back MOSFETs in the past there's two cases:
1) one end of the back to back MOSFETs was grounded *but* the voltage across the switch was small and AC.
2) the gate drive was floating and relative to the center point of the MOSFETs and voltage across the switch was high.
    This is more how the optical MOSFETs work.

However in your case:
3) one end of the back to back MOSFETs is grounded *but* the voltage across the switch is potentially high
   due to the high opamp swing.

When the voltage across the MOSFET is high and negative the MOSFETs can't turn off even if the gate is at 0V.
Think of the drain and or source being at a negative voltage but the gate is at zero volts, here the gate is positive relative to the source and so the MOSFET is on.

I still haven't worked out the complete scenario for positive DC voltage at the output but I suspect if there is high AC level or the opamp is oscillating then the MOSFET turning-on on the negative cycles charges up the cap C10 and that results in a net positive voltage at the output.

It might loosely explain why the low voltage supply isn't problematic, or less problematic.

This simulation confirms the rectifying issue:



It might not be the whole story.  I have a feeling we might need the opamp to oscillate as well.

[bluelagoon]

latest on FA-1

Tested it with a guitar into it through the amp, with Q2 taken off, and it certainly cleaned up the footswitch noise, to an acceptable level, almost silent. The effect still sounds good, plenty of headroom in the +27V mode as compared to +9V mode, makes a real noticeable difference, and the Electra Fuzz at the end with the Germanium diodes sounds great.

The 2N7002, are all surface mount, loaded on at JLCPCB, they are marked on the top with 7002, So have no reason to doubt their legitimacy.

Installed Q2 back in place.
Tested Drain of Q2, with Guitar output signal disconnected from J2 output Jack.
with only Q2 Drain connected at J2 jack
in +9v mode with effect engaged read 0.25V
in +27V mode with effect engaged read 0.2V
in both modes tested when activating the Baxandall out selector switch. This gave a short momentary pulse up to 1.8V from Coda Bypass 12F675 MCU pin 5, as its supposed to.

Don't presently have an oscilloscope, Its on my to get list, hopefully one day.

Took off Q3 and grounded Q2 Source, to have only one transistor to send momentary temp mute pulse.
Again tested Drain of Q2, with Guitar output signal disconnected from J2 output Jack.
in +9v mode with effect engaged read similar around 0.25V
in +27V mode with effect engaged read similar around  0.2V
in both modes tested when activating the Baxandall out selector switch. This gave a short momentary pulse up to only +1V from Coda Bypass 12F675 MCU pin 5, as its supposed to, less voltage pulse than the 2 transistors together.

Connected guitar out  back onto j2 to complete circuit, this then gives only +6V DC on the output in +27V mode with FX on
When Midrange is switched in, dialed on full the voltage at output increases up to +7.9 on the output. this is with just the single temp mute 2N7002 transistor Q2 installed.

Might try next that small cap across the op amp as you suggested.
Also are considering putting together another completed PCB to see if any difference,
I also feel the Power select switch could be a bit dicky, as it seems scratchy and temperamental with the guitar plugged in.
Side switches got a bit wrenched on first attempt to get PCB's into an enclosure, as the switch levers on the side switches were too long from what was required. So a new build might turn something up better.

And yes the Fuzz diodes do take DC on the output down to zero when engaged., When taken out of the Electra circuit to leave just the Electra transistor Boost the DC jumps up to about 0.9V at output.

Thanks for persevering here with me Rob.
Cheers

[bluelagoon]

Tried a 220pF across IC3C R28, Did not have a 470pF, might try later 2 parallel 220's.
with the 220pf in, Gave a reading at output with FX on and +9V mode  0.00V DC @ out.
with the Mid switched in and dialed full clockwise gave a reading of around +2.3V on output.

in +27V mode FX on read +6.3V at output.
with Midrange switched in dialed full read +8.17V DC at output.
Both those power settings above at bypass read about +0.19V

While leaving the 220pf in place I also put across pin 1 and 7 of IC2 a 100pF cap.
This gave lower readings at output
in+9V mode with fx on gave 0.00V at output
with mid switched in dialed full +2.2V at out

+27V mode , fx on 5.6V
with mid on dialed full +7.78V

So seems those caps are not really fixing the issue when tested with the volt meter, I have not yet done a test of the guitar through to amp with the caps, will do that tomorrow sometime when the neighbors wake up.

Have to say though the effect without those caps on did go into oscillation with the treble control close to full clockwise, so those caps may at least fix that issue if nothing else.

Next step, I think I'll try a photofet as the mute switch, to see if it improves.

[Rob Strand]

Took off Q3 and grounded Q2 Source, to have only one transistor to send momentary temp mute pulse.

FYI, with a single MOSFET present the MOSFET can conduct through the internal body diode.   It creates a very similar rectifying effect that you get with the two MOSFETs in place.

Installed Q2 back in place.
Tested Drain of Q2, with Guitar output signal disconnected from J2 output Jack.
with only Q2 Drain connected at J2 jack
in +9v mode with effect engaged read 0.25V
in +27V mode with effect engaged read 0.2V

Connected guitar out  back onto j2 to complete circuit, this then gives only +6V DC on the output in +27V mode with FX on
When Midrange is switched in, dialed on full the voltage at output increases up to +7.9 on the output. this is with just the single temp mute 2N7002 transistor Q2 installed.

I wonder if the issue is due to the jack J2 or the fact some signal is present on J2.   I really suspect you need a signal present to get the positive DC output voltage and the positive DC output voltage is a result of rectification by the MOSFET switches.   All the control circuits are +5V so even a total short can only create at most +5V.

If you plug the guitar into J2 and turn the guitar volume(s) down does the DC output drop?

And yes the Fuzz diodes do take DC on the output down to zero when engaged., When taken out of the Electra circuit to leave just the Electra transistor Boost the DC jumps up to about 0.9V at output.

At least there is some positive voltage present, which is consistent with the opamp case.   The lower voltage without the diodes is puzzling, possibly related to smaller output cap and higher output impedance of the electra circuit.

So seems those caps are not really fixing the issue when tested with the volt meter, I have not yet done a test of the guitar through to amp with the caps, will do that tomorrow sometime when the neighbors wake up.

Agreed.   Perhaps the signal source is the guitar signal and not oscillation.  You have to test the water on these tricky problems.

Although the fact things change here is a little suspicious:

While leaving the 220pf in place I also put across pin 1 and 7 of IC2 a 100pF cap.
This gave lower readings at output
in+9V mode with fx on gave 0.00V at output
with mid switched in dialed full +2.2V at out

Have to say though the effect without those caps on did go into oscillation with the treble control close to full clockwise, so those caps may at least fix that issue if nothing else.

I'm a big fan of using feedback caps throughout as they certainly prevent a lot of RF and oscillation issues.

IMHO it's always a good idea to put a small cap on the Baxandall tone control.   Don't forget this thread,
https://www.diystompboxes.com/smfforum/index.php?topic=130663.0

Next step, I think I'll try a photofet as the mute switch, to see if it improves.

Yep, good idea.  Even as a test it would be a good idea to try.   It should stop any rectification effects on the MOSFET.   There's still the JFET switch not working 100% ideally.

Thanks for persevering here with me Rob.

No problem.  This issue has turned out to be quite tricky from a technical perspective.   Maybe in the top 10 of technically tricky issues over the last 20 years on the forum.   Even after your good debugging work the problem isn't 100% clear.

[bluelagoon]

latest FA-1 update

If you plug the guitar into J2 and turn the guitar volume(s) down does the DC output drop?

Hi Rob, Yes the voltage does drop when controls are dialed back, whether it is the treble or midrange  or Guitar Volume, and also I suspect would be same with dialing back the effect volume as well. Did some tests with some results to confirm as much, will post down a bit further,

IMHO it's always a good idea to put a small cap on the Baxandall tone control.   Don't forget this thread,
https://www.diystompboxes.com/smfforum/index.php?topic=130663.0

Yes the 100pF cap you mentioned back there came to mind, so went right back and reread that whole thread yesterday, before you mentioned it here, Cheers.

Next step is to test it with the guitar through into the amp with the mute circuit on the output disconnected, to see just if it is any better than when its there. After that will put a photofet in its place to see if that somehow works better.
The switch clicks are not so bad that they cant be tolerated coming from the BAX out momentary Sw without a mute circuit. So may well suffice to be rid of the transistors there altogether.

I still suspect some more testing with different feedback caps and possibly even a larger resistance at R10 coming off the Treble pot middle leg will ultimately be required.

*************************************************

Here's todays test results -

Tested further the FA-1 with the guitar through it into the amp. with the 2 ceramic caps placed over op amps as mentioned earlier.
One thing was that the oscillation was far more noticeable as a high pitch squeal.

Tested first in +9V Mode, FX on, All dials, Treble, Bass, Volume on full, No Mids selected.
Read +1.45 at output with noticeable high pitch oscillation.
Dialed down the treble control aprox. 1/10 inch and oscillation stops and DC at output goes down to 0.00V at output.

Switched in mid range dialed on full, with the treble just off a bit as in last test, and voltage at output went back up to +1.73V
with further noticeable oscillation.
Dialed down the mid range control to half way and voltage at out goes down to 0.00V and oscillation stops.

One drawback in this setting is no discernable difference between midrange in and midrange out, as most the noticeable kick of midrange is in the last half of the mid range dial.

Tested again with with FX on, all dials on full in same +9V select mode, no mids.
There was noticeable oscillation with +1.45 at output
This time dialed down the guitar volume to 5 and the oscillation stops and voltage at output drops to 0.00V
switched in the midrange on full, the voltage goes up again at output, and more oscillation, dialed the Guitar volume control to 4.5 and oscillation stops and voltage dropped back to 0.00V again

**********************************************

Same test in +27V mode, FX on, All dials, Treble, Bass, Volume on full, No Mids selected.
Reads 6.37V at output plus noticeable oscillation.
Dialed down the treble control aprox. 1/10 inch, lowers the output voltage to 0.00V and cuts out the oscillation.
Switched in the midrange, oscillation comes back and voltage at output runs high again at +6V
Dialed down the midrange to half way and voltage dropped to 0.00V and oscillation stopped.

Again no discernable difference with mid switched in or out at this setting

Tested with all controls back full, No mids, Guitar Volume full, and gives +6.4V at output. with oscillation
This time turned the Guitar Volume down to 4.5 and this cuts the oscillation, and sends output voltage down to 0.00V
still with some noticeable oscillation artifacts at last 1/10 inch on treble dial.
Similar as before with the mid range switched in and guitar volume turned down to 4.5

One thing to mention was, in both power mode settings, with all controls set on full and midrange switched in on full, there is noticeable oscillation and voltage up to around 6V at output. Then dialing down the Treble pot alone aprox. 1/4 inch cuts out the oscillation and drops output voltage to 0.00V, While still allowing the noticeable emphasis of the midrange control when switched in and out.

[bluelagoon]

Had a re-visit to testing with the guitar through FX into the amp after removing all mute transistors, and while it does read at 0.00V DC on output apart from spikes upward after switching which always settle quick back to 0.00V, there is no concern for the voltage on the output with the mute transistors removed. however on second thoughts and hearing again the switch noise without those mute transistors there, The BAX Out switch is far too noisy to be an acceptable thing.

The signal into the amp hasn't changed anything different without the muting transistors there. Still same oscillations and concerns as noted last post tests.
Will try next the photofet as the temp mute trigger from the BAX Out switch, If that doesn't do it, I may revert to redrawing the circuit to possibly replace the Coda Bypass MCU with a second VFE Bypass MCU circuit or a Mictester type bypass that uses the 40106 CMOS

Also still not real happy even with the VFE MCU Bypass switching with the inherent clicks and too noisy thunks on the main Footswitch.
The VFE system does work well in most instances, It may just be on account of the higher output from a boost circuit as is the FA-1, and it is also compounded all the more when it is switched to run at the higher +27V power supply.

Just wondering does anyone have any idea of any extra circuitry to add to the VFE switching to mute the circuit quieter than it is already when switching.? I have the actual code for the VFE Bypass MCU, but not to sure how to refine it, or even to view it, as it was only released recently to the public in some form other that C programming code so not sure how to open it to read it or rewrite it.
I can program the chips alright but other than that are a bit clueless on it.

I think after considering all else, that you nailed it correct Rob, saying the DC at output was an issue resulting from the opamps oscillation
So I guess main criteria now is to tame that oscillation down to a point of nothingness, then that should ideally fix the voltage on the output down to 0.00V as previous tests seem to conclude, even with the original Q2 and Q3 muting transistors back in place.