BOSS CE-3 Signal Allowance HEPL ME!)

[midiez1997]

Hi guys! My name is Dima, and I am their Belarus. So sorry for my incorrect English)
I am the owner japanese BOSS CE-3. And the problem is that when you turn on the effect, the output A: signal from the output sags on the volume.
The output B: signal from the output -such a problem does not have, with him everything is in order.
The pedal is completely original, the only thing I changed electrolytic capacitors there to eliminate the background.
I think to check the oscillograms and compare with the factory ones, but I heard that they often have this drawback. And can i play around with the denomination of the elements and eliminate this drawback?
Thank you all in advance!)

[midiez1997]

up

[ElectricDruid]

Welcome!

There's a similar discussion to your problem over on this thread:

https://www.diystompboxes.com/smfforum/index.php?topic=118991.0

Mark Hammer gives a good explanation of why a good reliable fix isn't simple, and why a simple fix isn't going to be that good. Without wanting to belittle anyone, the other posters are just experimenting, and their advice to "remove this" or "desolder that" can be ignored. It'd be *nice* if just lifting R18 solved the problem, but it's not that simple.

Tom

[Rob Strand]

I'm assuming the problem is only in Mode I ?

Lifting R18 isn't too bad.   There's plenty of chorus and flanger pedals out there which don't bother trying to correct for the level.

I suspect there's a little more going on here to make the level of A and B sound different.   That would be due to the fact A Out = Delay + Effect and  B Out = Delay - Effect.

For the overall level:
If you change R18 from 120k to 270k  the change in volume will be less pronounced.  If you want even less drop use 470k.

For the balance between A and B:
If you are still unhappy with A and B not being balance, add a 470K resistor across R29.  If B is still too loud try 270k.

[midiez1997]

I'm assuming the problem is only in Mode I ?

Lifting R18 isn't too bad.   There's plenty of chorus and flanger pedals out there which don't bother trying to correct for the level.

I suspect there's a little more going on here to make the level of A and B sound different.   That would be due to the fact A Out = Delay + Effect and  B Out = Delay - Effect.

For the overall level:
If you change R18 from 120k to 270k  the change in volume will be less pronounced.  If you want even less drop use 470k.
oooh thanks!!
For the balance between A and B:
If you are still unhappy with A and B not being balance, add a 470K resistor across R29.  If B is still too loud try 270k.

I want to make a signal from the output A louder. So that it is the same as the input signal.
That's all)
P.S. no matter which mode. in mode II (vibrato) the sound is also quieter...

[Rob Strand]

I want to make a signal from the output A louder. So that it is the same as the input signal.
That's all)
P.S. no matter which mode. in mode II (vibrato) the sound is also quieter...

That's a little strange.  If could be a fault.  By doing a mod you could be hiding a problem.

If output A has the correct level when the output is *clean* (that is Mode I + Effect bypassed) then there is less chance of a significant fault.  In that case, to increase the level of A try putting 470k across R36.   You could go as low as 270k.  However, if you need to go down to 270k or lower I would start thinking the pedal has a fault.  A possible fault would be something wrong with Q4.

If output A does not have the correct level when the output is *clean* (that is Mode I + Effect bypassed) then you may have a faulty opamp (IC1, pin 1, 2, 3), a faulty socket "OUTPUT A", or a faulty cap C7, or a bad solder joint in that area of the circuit.
--------------------------------------
[EDIT:   Don't do these mods yet.   I'm going to check exactly what JFET switches are on in each mode.]

OK I got the same conclusion that I wrote above.  So there's something odd about output A.

[Rob Strand]

I went through the schematic and found the schematic has a bug.

Boss CE-3
----------------------------------------------------------------------
From real unit
https://www.youtube.com/watch?v=0AJfndh4Ip4

Switch:
Mode I is counter-clockwise
Mode II is clockwise

Mode II:         
    Effect on    A = Vibrato,  B = Clean
    Bypass       A = Clean,    B = Clean
----------------------------------------------------------------------
From manual

Effect on:
    Mode I        Mode II
    A = D + E     A = E
    B = D - E     B = D

Bypassed:
    Mode I        Mode II
    A=D           A=D (Not documented)
    B=D           B=D (Not documented)
----------------------------------------------------------------------
Circuit

Bugs on schematic:
    The Mode I and Mode II *labels* on the schematic are wrong.
    Mode II is when mode switch is closed; not Mode I as shown on the schematic.


JFETs off when gate low.

           Q11/C31    Q12/C9   
Bypass     lo         hi
Effect     hi         lo

JFET conditions: (corrected Mode I & Mode II schematic labels)
Q2 and Q5 have the same control signals. 
     Always off in Mode II.
     Only on in Effect mode + Mode I.
Q4 Only on in Effect mode (Mode I or Mode II)
Q3 always on in Mode I, or on when Mode II + Bypass mode


JFET state according to Mode Switch and Foot Switch:
(Table shows corrected Mode I & Mode II schematic labels)

State      Sw       C31(Q11) C9(Q12)    Q2    Q3    Q4    Q5    Result
Bypass     Mode I   lo       hi         off   on    off   off   A=D, B=D
Effect     Mode I   hi       lo         on    on    on    on    A=D+E, B=D-E, gain reduced (A and B)
Bypass     Mode II  lo       hi         off   on    off   off   A=D, B=D
Effect     Mode II  hi       lo         off   off   on    off   A=E, B=D


Conclusions:
- The gain is only reduced (Q2 on and R18 in circuit) in Mode I when the effect is on.
- The Mode I/Mode II switch is shown with the wrong labels on the schematic.
----------------------------------------------------------------------

[midiez1997]

guys, thank you for supporting me with advice)
So, I poked a soldering iron, and here are the results:
dropped out of the scheme Q4 and make jumper source-drain. anyway the sound sags on the output a:(
replaced C7-It seems nothing has changed...

[midiez1997]

HERE! I think I won it) I dropped out, listened, and again soldered R18. Then I decided for prophylaxis to dissolve all the sockets and wires leading to the board .... and voila! the drawdown disappeared)
But there was one oddity, about which I probably forgot ...
when in output A is turned on in the chorus mode (position 2), pressing the pedal on button the sound disappears for half a second and reappears.
So it should not be?

[midiez1997]

Pancake. in the vibrato mode, there is no sound at all (and if there is, then it is loaded ...
On the board, it feels good, I washed it with alcohol ...

[Rob Strand]

HERE! I think I won it) I dropped out, listened, and again soldered R18. Then I decided for prophylaxis to dissolve all the sockets and wires leading to the board .... and voila! the drawdown disappeared)

OK.  Great.

But there was one oddity, about which I probably forgot ...
when in output A is turned on in the chorus mode (position 2), pressing the pedal on button the sound disappears for half a second and reappears.
So it should not be?

That is not normal.    I suspect there might be a fault in the circuit.
Do you have a multimeter?

in the vibrato mode, there is no sound at all (and if there is, then it is loaded ...

That's not good.  No vibrato means Q4 is not working.  Maybe you reversed Q4?, check the orientation of
Q4 using this diagram,
https://postimg.cc/K1GPkqJv

If the orientation is OK, check the solder on Q4 (not connecting or solder bridge).

[midiez1997]

I have a multimeter.
Just in case I tried to replace it with Q2.
problem still exists

[midiez1997]

dropped out P18. The signal became a little louder than the wedge!))
But it seems that a load has started to appear, or have I already had hallucinations?))
I think I need to put 470k, so the chorus will sound better!)
Or leave it as it is?

[midiez1997]

By the way, if I replaced my native electrolytic capacitors with ... Capxon, G-luxon, maybe because of this the vibrato mode does not immediately turn on?)
In general, it makes sense to change the native electrolytes in the old lotions? I heard that they dry up. and in this pedal I clearly heard pulsations at high volume.

[Rob Strand]

dropped out P18. The signal became a little louder than the wedge!))
But it seems that a load has started to appear, or have I already had hallucinations?))
I think I need to put 470k, so the chorus will sound better!)
Or leave it as it is?

Most chorus pedals are equivalent to removing R18, however the level does increase.  If you replace R18 with 470k it should drop the volume a small amount to make it more balanced.

Just in case I tried to replace it with Q2.
problem still exists

Do you mean Q4?

By the way, if I replaced my native electrolytic capacitors with ... Capxon, G-luxon, maybe because of this the vibrato mode does not immediately turn on?).

Normally changing the capacitor would not cause that problem.  If you put the capacitor in the wrong way around it could cause a problem. 

In general, it makes sense to change the native electrolytes in the old lotions? I heard that they dry up. and in this pedal I clearly heard pulsations at high volume.

Hard to say.   I don't hear a lot of reports of bad capacitors in Boss pedals.  I would only replace capacitors I know are faulty.

I have a multimeter.

I suggest measuring the voltages at
- IC1 pin 7
- IC1 pin 1
- IC1 pin 2

Check the voltages in each case of:
- Mode I + Bypass
- Mode I + Effect
- Mode II + Bypass
- Mode II + Effect

The aim is to see if the DC voltage in one of these cases is different to the others.  Make a note of which case is different as this will help diagnoze the the problem.

When you measure the voltages see if any of the voltage measurements have the delay when you change between bypass and effect in Mode II.

Another thing you can check is the voltages at C31 and C9 when you change between bypass and effect in Mode II.   See if any of these voltages show a delay.

[midiez1997]

That is, r18 is needed so that both outputs have the same volume?
Yes, q4 and q2-are the same.
look, I want a multimeter(check IC1), a black puppy is attached to the body, the red one poked at the right points ... everywhere it was 5.26, without delay.
But on С31 in the bypass 0, and when turned on, 2.93 rushes pretty quickly.
C9: bypass:0.78, when I turn "on"-quickly 5.69v.

[Rob Strand]

But on С31 in the bypass 0, and when turned on, 2.93 rushes pretty quickly.
C9: bypass:0.78, when I turn "on"-quickly 5.69v.

The C9 voltage looks normal.

The C31 voltage looks bad.  I would expect the C31 voltage to go from 0 to 5.69V (maybe up to 6V).  If it only gets to 2.93V the delay signal will not pass - that is exactly the problem you are seeing.  To solve the problem we need to work out what is loading C31.

It might be worth checking  the voltage on the collector of Q11.  That should change from 0V to 5.69V (maybe up to 6V).

If the collector of Q11 is good then only the voltage on C31 is bad.  Possible causes are,
- contamination on the PCB.   try cleaning the PCB.
- faulty diodes

The only parts connecting to C31 are R61, D3, D4, D6.    Try this
- Switch the bypass switch so C31 is at 2.93V.
- Using the multimeter measure the voltage on the *anode* of D3, D4, D6. The anode is the point that connects to the JFET.   I suspect one of those voltages will the different to the others.  The JFET connecting to the diode with the different voltage might be damaged.

[midiez1997]

collector voltage Q11: 5.82v.
D3=D4=2.93v.
D6=5.25v!!!

[Rob Strand]

collector voltage Q11: 5.82v.

OK, that looks good.

D3=D4=2.93v.
D6=5.25v!!!

There's definitely something wrong here but I can quite see what the problem is.

Previously you measured C31 at 2.93V, the same voltage as D3, D4.  That could be because no current is flowing through the diode, or, the diodes are shorted.  Check the diodes D3 and D4 are not shorted.

Do you know the input impedance of your multimeter?  Normally it is 10Meg but some multimeters are 1Meg.   The reason I ask is the voltage at Q11 is 5.82V, and if you multimeter is 1Meg it will form a voltage divider with R61 and produce half the voltage.  5.82 / 2 = 2.91V, which is very close to 2.93V, so perhaps your meter is loading C31.

Some other things to check are the voltages on the drain and source of Q4, Q2, Q5.  You will need to check the voltage in effect and bypass mode.

[Mark Hammer]

I have one of these pedals.  I'll try it out tonight to see how much of a volume difference there.