Can someone help me understand a few parts of the Boss Dimension C circuit?

[solderfumes]

I decided that it was finally time to tackle The Big Boy, the chorus pedal I've been curious about since I was 15 years old and reading Guitar Player magazine: the Boss DC-2.  I've got it working on my breadboard, but I hoped someone out there would be kind enough to explain how the clock circuits feeding the BBDs work.

I've been working from the schematic that's floating around out there (http://www.freeinfosociety.com/electronics/schemview.php?id=123), substituting BL3207s and BL3102s for the BBDs and clocks.  I'm using TL072s for all the op amps, and BC327s for the PNP transistors in the clocks.  Once I sorted out some small issues, it made sounds, but those sounds were awful noises.  Poking around with my scope, it appeared that the problem was that the clock wasn't producing a usable pulse to feed the BBD.  By dumb luck, I decided to probe the signal fed to pin 7 of one of the BL3102s while the other one was connected to one of its outputs, and all of a sudden the output pulse was nice and clean.  Figuring it must be the parasitic capacitance of the probe, I added small capacitances to ground from pin 7 of both BL3102s and now it works.

So, my question to you is: why??  What did this do?  How does the clock work at all?  I'll break down some of my questions into point form, referencing parts by their numbering on the above linked schematic, and focusing on a single delay line, IC9.

- I get that op amp 6b feeds an attenuated copy of the LFO signal to the non-inverting input of 6a.  I *think* that Q8 forms a current source, that charges the 100p capacitor I guess?

- Probing the output of 6a, I see that it produces a triangle wave of constant absolute slope (i.e. positive and then negative with the same amplitude) but with wavelength and amplitude changing according to the LFO signal.  I didn't understand why the op amp's output could be anything but square, as it looks like it's wired up to work like a comparator, but am I correct in guessing that, by feeding the output of 6a into the oscillating circuitry of the BL3102 and back out through some other stuff into the inverting input, there is somehow negative feedback occurring here?

- Moreover, the triangle wave coming out of 6a doesn't go very high or low.  The datasheet of the MN3102 (I can't read the full datasheet of the BL3102 as it's in Chinese) says that the input to pin 7 registers high as a signal between Vdd-1 and Vdd, and low as a signal between 0V and 1V.  The triangle wave doesn't span this range, so how is it driving pin 7?

- Lastly, what did adding the small capacitance to ground at pin 7 do?  Why does it magically work now?

Any and all help is welcomed.  I didn't find anyone doing a real circuit explanation/analysis of this in any of my Googling, so whatever you can provide is greatly appreciated.

[czapa tranzystor 2]

 Hi
Orginal schematics to contain Flip flop switch (BA634)
Better schematics with bypass to have Fromel Lush Seraph.
Maybe  need different transistors.

[solderfumes]

Oh right, I should have mentioned that I omitted all of the FET switching stuff.  What I've got on the breadboard is more or less the same as the schematic on this page: http://topopiccione.atspace.com/PJ18BossDC2.html

[czapa tranzystor 2]

 What layout to use at build,utilised layout topopiccione ?
If you can to show board foto ?

[solderfumes]

It's still on breadboard at the moment.  I'll take a picture when I get a chance, but I doubt it will be that useful.  I'm more interested in how the circuit is *supposed* to work in theory.

[Scruffie]

I don't have time for a thorough reply to your questions but...

LFO is formed by standard intergrator/schmitt at IC7, amplitude is increased via IC5, then inverted/buffered via IC6b (non-inverted version fed to IC5b so that both BBDs are swept opposingly) divided by the resistors at its output and fed in to IC6A/Q8 business which is some kind of comparator/astable thingy i'll have to look at properly later.

[solderfumes]

I don't have time for a thorough reply to your questions but...

LFO is formed by standard intergrator/schmitt at IC7, amplitude is increased via IC5, then inverted/buffered via IC6b (non-inverted version fed to IC5b so that both BBDs are swept opposingly) divided by the resistors at its output and fed in to IC6A/Q8 business which is some kind of comparator/astable thingy i'll have to look at properly later.

Cool, thanks.  Yeah that's what I had figured, which is why I'm surprised that the comparator/astable thingamabob (technical term) is putting out triangle waves.

[Scruffie]

The triangle is formed by the intergrator/schmitt LFO.

[solderfumes]

No I mean the pulses that the comparator/astable part is putting out are triangle waves, whose amplitude is modulated by the Schmitt/integrator LFO output.

[Scruffie]

Oh right, well yeah, the output modulates the clock frequency up and down and we want a triangle modulation waveform, not a square, the 3102 gives the BBD its square wave clock.

[solderfumes]

Yeah, okay.  I think I've convinced myself that the path through OX1 and OX3 of the MN3102 does in fact provide the negative feedback.  I was also wondering why the signal doesn't seem to be big enough to actually trigger the MN3102, but I'm realizing that it by definition *has* to be big enough, as the wave wouldn't change directions if it didn't manage to trigger OX1.

Given that, I'm thinking now that the reason the small capacitance I added at pin 7 helped to filter out some high frequency noise that was messing up the BBD.  Does this sound reasonable to anyone out there who understands this stuff better than me?

[Scruffie]

The MN3102 is basically an astable multivibrator & divider in one package, pins 5,6 & 7 forming the vibrator bit.

Anyway, did these noises sound like 'chirping' if so I would guess you were experiencing heterodyning from the 2 clocks, it may be that your extra capacitor altered the clock frequency to put it out of the range of the the other or just smoothed out the VCO to cure it... something like that.

[solderfumes]

No it was an awful ka-chunk ka-chunk static ka-chunk ka-chunk static noise.

[Scruffie]

Still may have been heterodyning, breadboards aren't a quiet place.

[aion]

Not sure what your plan is for getting it off the breadboard and into a box, but if you can wait another 6-8 weeks I'll have a DC-2 DIY project ready to go. Just sent off for the prototypes this past Friday.



The PCBs are double-decker: signal & switching on the top PCB, clock & LFO on the bottom (with holes for top trimmer adjustment). 125B enclosure, all 1/4w flat-mounted resistors and standard box caps. Supports either stock buffered bypass (signal goes through pre-emphasis and de-emphasis filters) or minimally-buffered bypass (signal only goes through the first op-amp stage and is split to each output).

I don't mean to drop in and advertise... just one of those "if it were me I'd wanna know" situations