To all people who own a Boss BF-2 Flanger

[smoguzbenjamin]

I've fiddled with the trimpots inside, and I've made some interesting discoveries. With one you can adjust the max depth of the flange, with another you bias the ICs (don't mess with this one cos otherwise the sound will die and leave you with clean signal), and with the other one I haven't managed to figure out what it does. Anyway I like the extra flexibility that it's given me and I wanted to share that with you guys

cheers
Ben

[Mark Hammer]

Somewhere in the archives, I must have a dozen postings regarding mods for the BF-2.

The three trimpots are for adjusting the bias (the one beside the MN3102 near the iddle of the end of the board), adjusting the clock frequency (the one nearest the corner of the board by the MN3102), and setting the maximum resonance (the one closest to the edge with the connecting wires).

[Ben N]

Can they make a silk purse from a sow's ear?  Sorry, I hate that box...

[Mark Hammer]

I'd agree, Ben, that the BF-2 is not god's gift to flanging.  On the other hand, like K-cars and Chevy minivans, they are in great abundance, reasonably priced, and common enough that repair and troubleshooting will never be a problem.   Simple mods to the unit will not transform it into a "great" flanger, but they will yield a broader tonal palette, and make it more fun to own.  I like jaw-dropping effects as much as the next guy, but when you have a song to play, its not jaw-dropping you want necessarily, its goodness of fit.  More options allows for better fit to more things.

[Ben N]

when you have a song to play, its not jaw-dropping you want necessarily, its goodness of fit

I absolutely agree with you, Mark.  In fact, one of my prefered flangers is a DOD 575 (did I get the number right?), which is rather on the subtle side.  My fave is the Electric Mistress, in part because of its versatility but more so because of its musicality.  I got a BF2 mainly because I couldn't afford the board space for an EM, but I was soooo disapointed with its clangy, metallic flavor, and overall dingy sound.  To me it sounds like a flanger that has to follow a fuzz, but can do nothing good for a clean sound.  But that's just me.  And I confess, I have never tried your mods, although I am sure they do improve it.  My solution was ebay  :wink:
Ben

[Mark Hammer]

I got a BF2 mainly because I couldn't afford the board space for an EM, but I was soooo disapointed with its clangy, metallic flavor, and overall dingy sound.  To me it sounds like a flanger that has to follow a fuzz, but can do nothing good for a clean sound.

A-ha.... Well if that was your problem, note the following about it.

The 2nd op-amp into the circuit is the one that mixes the recirculated signal and incoming signal for ship^ping off to the antialiasing filters and BBD.  One of the strategies often employed in devices that use recirculation is the use a crude limiter in the stage that mixes recirculated and unrecirculated signal; the reasoning being that when you combine them you run the risk of overloading the next device in line.

The limiter is essentially a soft-clipper in the form of a back-to-back pair of diodes and a small-value series resistor.  Those who have read any of Jack Orman's various postings on what may or may not be special about the JRC chip in the Tube Screamer over the years will immediately do a double take and say "Hey, isn't that what he suggested might nail the sound in ANY op-amp?".  And you would be right.

So how could one keep this "safety valve" aspect of the design and lose the fuzz?

The recirculation/mixer stage is configured as an inverting amplifier with a 47k feedback resistance.  For the recirculated signal, there is a 39k fixed input resistor in series with a 22k trimpot.  This enables you to set the regen path for a gain of a little more than unity (47k/39k) to less than unity (47k/[39k+22k]), with further attenuation provided by the regen pot.

If the regen+input signal is not too high, then the diode limiter is not needed.  You can do that by reducing the input signal in the first place, by reducing the gain of the regen path in that stage, or by otherwise restricting the signal amplitude by bandwidth restriction.  For instance, on the Hyperflange, Anderton uses a simple bass-cut arrangement on the regen path to take out the biggest chunk of the overall regen signal amplitude and permit use of more regen signal without clipping.   This not only keeps regen-derived overload from occurring, but also revoices high-regen sounds to be less "box-ey".   On the BF-2, that's a simple matter of dropping C7 (.047uf) down to .01uf or less.  I don't know if that would be sufficient bass-cut to justify taking the diodes out of circuit, but one can always stick in a SPST toggle to enable/disable regen-limiting.

Of course the problem/challenge to be tackled is to maintain something like unity gain when the effect is bypassed.  If one were to reduce gain in the regen/mixer stage to make high-regen settings trouble-free for the BBD  that might be helpful for the  BBD bot not for the dry/wet mixing or for effect/bypass volume balance.  Remember that the BF-2 like so many other dry+wet devices uses a dry-lift switching arrangement.  That is, in bypass mode the wet signal is simply lifted out of circuit and the dry signal you hear is exactly the same signal you hear when the effect is engaged, except when in effect mode the delay signal is added on.  If the delay signal is reduced you lose the 50/50 dry/wet balance needed for maximum effect.

My feeling is that the optimum solution is to reduce the gain of the transistor input stage a bit, cut back a little on the regen bass, and up the gain on the output/mixer stage a bit.  That way you maintain the wet-dry balance needed while challenging the BBD less.  If you add a bit of gain on the last op-amp stage, you can then stick a 100k trimpot in place of the 100k output resistor, or simply leave the resistor in place.  In principle, if I've thought this out well enough, you should get less distortion, and dry/wet balance, with unity gain overall (at the whole pedal level), at the cost of a slight increase in hiss.  That hiss can be easily compensated for with a slightly bigger cap in the feedback loop of the final op-amp stage.

I'll finish this off by saying that the whole problem arose because the BF-2 has no input attenuation and no compander to keep BBD-input levels steady and reduce noise.