Trying to fix a Boss BF2

[HoodTube]

Hi folks, I'm currently trying to fix a broken Boss BF2. The problem is that I'm getting no effect sound, just dry signal whether it's on or off.

I got some advice on another forum, and did an audio probe test (though my knowledge is limited and I'm very much a learner at this stage). Anyway, when I was "probing" here's what I found:
I'm getting audio output from one pin on each of the chips except MN3102. None of the pins on that pass the audio through, but a couple of them mute the whole signal, even the hum. So could it be that one that's causing the trouble? And if so, can I replace that with a CD3102, which I have one of lying around?

Also, I get no audio signal from IR9022, but I do get what I think is the clock signal from one of the pins, and it changes speed when turning the rate knob. Is that a good sign? The sound itself doesn't come through that, but does the pulse indicate that that one still works?

Any help or advice on what I should do next would be very much appreciated. Cheers!

[Mark Hammer]

There won't be any audible audio output from the 3102, even if brand new from the factory, simply because it is not an audio chip.  If you can hear audio output at 50khz, you're a better man than I.

The fact that you get clean signal in both effect and bypass mode indicates there is nothing wrong with the supply of power to all the chips.

The BF-2, like a great many commercial modulation and delay pedals, uses a single FET to switch in and block the delay/wet signal.  When there is no wet signal introduced, the "effect" disappears.  Since you have not made any mention of being unable to switch from effect to bypass or vice versa, that would suggest the issue is most likelt relegated to the FET switching.

If you can get a probe in there, check for a change in DC voltage (most likely in the range of only a couple of volts) at the collector of Q10 (where R56 and R61 contact each other).  If you get a corresponding change (DC voltage goes up when the LED goes on), then the flip-flop is clearly OK, and the problem lies most likely in the FET itself.  You can confirm this by temporarily soldering in a wire link between the output of C20 and R29.  This will mimic Q4 being in an "on" state.

[miketbass]

Its a pretty basic suggestion, but I fixed a BF2 yesterday by biasing the BBD. Picked it up non working for $20 and didn't even plug it in. Threw it on the scope to calibrate and when done it worked just fine. Ive had this happen a few times now where someone starts screwing with trim pots, nudges the bias out of range, and sells it off when its "not working".

My suggestion is to probe pin 7 of the MN3207. This is the output pin and if you are getting signal out the chip is fine. You should be able to freeze the LFO sweep by rolling all of the knobs CCW and then listening on pin 7 while turning the manual knob. You should hear a pitch bend.

Check the LFO for voltage swing and if you have a way to measure frequency check TP1 for clock frequency. These steps should narrow it down quickly - but if you dont know the history of this pedal start by slowly turning the bias pot.

[Mark Hammer]

Good call.  Pin 7 or 8.  That, and the suggestion I made earlier will tell you whether the non-effect is because the effect is not being produced, or not being allowed to get through.

[miketbass]

Good call.  Pin 7 or 8.  That, and the suggestion I made earlier will tell you whether the non-effect is because the effect is not being produced, or not being allowed to get through.

Always good to see you on here Mark!  I have truly regressed to the lazy mans approach when trouble shooting ICs.... I put my scope probe at pin 1 and go around until I have established input, clock, output etc. When I was observing the MN3207 I made it to pin 7 and saw output and stopped there!  So yes, pin 7 OR 8 will show output according to pin out....

[Mark Hammer]

Oh, it's not lazy.  I just didn't want the OP (whose query makes it sound like they are somewhat new to all this) to accidentally probe pin 8 and wonder why they are reading the same thing as pin 7. 

But your note raises one of the basic guiding principles of troubleshooting effects: break the circuit down into functional segments.  That is, is this aspect of the pedal working?  Is that aspect working?  Over time, one gets familiar enough to be able to look at a schematic and think "this is where a particular function in the pedal would show whether it is working properly or not".  Until one reaches that degree of familiarity, it remains a whole pile of parts all over a board, any of which could be "the problem".

[HoodTube]

Hello folks. Cheers for your advice and suggestions. Mark, thanks for letting me know about the audio/MN3102 thing, that's the type of stuff I'm trying to get my head around at the moment, which chips and bits do what.

And Mike, well, I turned the BBD bias just a bit, lo and behold, it's a flanger again! I can't believe the amount of head-scratching and messing about for weeks, and that's all it was.

It doesn't sound 100% "right" like before though, but I've replaced the Depth pot with one I had lying around, a 100k. (That was initially the reason for the repair, but it stopped working midway through that). Is the wrong pot value the likely reason for that, or does the BBD need more adjustment? Or both?

Or should I just quit before I break it altogether?

Thanks again folks.

[miketbass]

The schematic shows 50k for thw depth pot. If you have installed 100k, try strapping a 50k resistor (likely 47k) across the two outer pot lugs in parallel to bring the pot value down to 50k. With pot tolerances as they are this will be "in spec" and because it is a linear pot it will not mess with the taper.

As to the biasing of the BBD, ideally you would use a signal generator and an oscilloscope to find thw optimum setting. Guessing you do not have access to the tools for that you can get close enough for rock and roll with an audio probe and tone generator app on your phone or computer. Run a search for this procedure but it will involve isolating the BBD path output and adjusting for clean headroom.

A proper value pot replacement (even "cheating" with a 47k parallel resistor) and a BBD adjustment by ear, when done proper, will make this circuit sound indistinguishable from its original state.

[miketbass]

I should add this is assuming that the clock trim has not been touched. The resonance trim can and should be readjusted after the BBD bias step to properly calibrate the resonance control. Also, my apologies for any typos in my replies as i am on mobile.

[Mark Hammer]

While it is entirely feasible to adjust the bias by ear, what I usually recommend to people is to give oneself the best opportunity to hear the adjustment.  As part of that, unsolder one end of R28, which mixes clean signal in, so that ALL you hear is delay.  If you can listen to the pedal through headphones, even better.

As for the Depth pot, 100k should be okay until you can score a suitable replacement.  That pot mixes/blends a varying voltage from the LFO and a fixed/stable DC voltage from the manual pot.  What you might lose is the ability to nudge the sweep upward or downward with a bit of Manual, but nothing worse than that.

[HoodTube]

The oscilloscope stuff seems beyond my capabilities at the moment, I think.

However, I've got another BF2 and I'm just sitting comparing them now. It seems as though on the "repaired" one, the effect depth is very slightly less, the maximum rate is faster, and the resonance a fair bit weaker.

Should I try tweaking the other trim pots a bit? i.e. Turn the reso trim pot clockwise slightly, and try and match the two pedals by ear?

And will the clock trim pot change the "too fast" speed if I turn it counter-clockwise?

Thanks!

[Mark Hammer]

The clock trim sets the minimum delay time.

[thermionix]

The schematic shows 50k for thw depth pot. If you have installed 100k, try strapping a 50k resistor (likely 47k) across the two outer pot lugs in parallel to bring the pot value down to 50k.

100k resistor.

[kdutt]

Hi
I got the same fault as the original poster. LED turns red but no effect at all, just dry signal whether it's on or off.
I swapped out the MN3207 and MN3102 for new ones, but that didn't help.
I checked the switching like Mike Hammer suggested and it seems to work ok.
I also bypassed the FET mimicing Q4 being on, but still no effect.
I tried different settings on the bias pot, but it's still not flanging.
I noticed that the 5V voltage regulator gets very warm, could that indicate that any of the ICs is broken and draws to much current?
The output of the voltage regulator 78L05 is 4,8 V so it seems ok.
I suppose I should get an oscilloscope so I can see what's going on.
Any other suggestions that could be checked with a multimeter or any circuit that is likely to fail ?

Thanks.

[Fender3D]

4,8V is too low! you should have 5V + 0.6V (diode drop to GND) =5.6V @BBD's supply pin.
Any shorted cap, perhaps?

[kdutt]

ok, I will look for a shorted cap.
Thanks for the quick reply 

[ElectricDruid]

Doesn't have to be a shorted cap - any short on the 5V rail might cause that. That could be a blob of solder or a bit of wire clipping or just some muck. A dead cap is a possibility, but it's not the only one.

Check the board visually to make sure you can't see anything causing one track to touch another around the BBD and clock chip.

HTH,
Tom

[Mark Hammer]

This is why I like to keep an old toothbrush and something that can dissolve solder flux on hand (I use methyl hydrate).  Sometimes flux on the board can make solder bridges hard to see.  It's not absolutely necessary to have a board scrupulously clean of flux, but when you're troubleshooting, being able to see bare board where bare board should be is a nice thing.

[pinkjimiphoton]

This is why I like to keep an old toothbrush and something that can dissolve solder flux on hand (I use methyl hydrate).  Sometimes flux on the board can make solder bridges hard to see.  It's not absolutely necessary to have a board scrupulously clean of flux, but when you're troubleshooting, being able to see bare board where bare board should be is a nice thing.

+1
some of the old-timers in the tube days said flux is conductive. i think they're right. some things, small matter, others, can really mess ya up.

[kdutt]

Hi
I cleaned the solder side of the board with a toothbrush and some "electronics cleaner", I also scraped off some solder flux with a fine screwdriver.
I'm afraid I put the solder flux there when I tried to improve the solder joints 
And now it works  , it must have been the solder flux that shorted something somewhere.
So now I want to thank you guys for the help, couldn't have fixed it on my own.

Thanks.