DOD FX35 Octaplus

[Jaicen_solo]

I've been given one of these to re-house with true bypass and i'm looking for a schematic in the hope of working out some mods. Does anbody have a copy of the schem or any hints as to the internals?? It sounds to me like a heavily filtered flip/flop octave down, but I could be wrong.

Thanks again,
*j

[Mark Hammer]

I do believe that Colin has it at the experimentalists anonymous site.

[Jaicen_solo]

Mark, you really are awesome!
Since I've got your attention, and given your knowledge about these sort of designs (as we've discussed in the past), I wonder if I can pick your brain. Having given the schematic a quick look, it uses the same 4017 IC as the blue box, but it's configured differently, so i'm guessing that it can't be modded for 1/2 octaves??
Also, the guy i'm doing this for was after a little more 'grind' from the octave signal, as it sounds heavily filtered and is very smooth. I'm thinking this filtering is coming from the 0.033uF cap coming off U6B?? If I reduced the size of this cap would it allow a more 'square' sounding octave??
I'm sure this will all become a little clearer to me tomorrow as it's been a long day, but I just thought i'd put this out while you guys are around to help Thanks again,
*j

http://experimentalistsanonymous.com/diy/Schematics/Vibrato%20and%20Pitch%20Shift/DOD%20FX35%20Octoplus.gif

[Mark Hammer]

Well, first off, you will note the large number of '?' indications on that drawing.  I fully appreciate the diligence and generosity of whoever drew it up, but please note that it is NOT a factory schem, so tread carefully.  Of course, having both the drawing and the unit itself in hand, you are in a much better position to evaluate the accuracy of the drawing than I am.

Second, it seems like there are a number of caps serving the purpose of rolling off highs on the octave channel.  Everything between pin 1 of U6A and pin 7 of U6B adds up to form a 3-pole (18db/octave) lowpass filter.  So, not only is the .033uf cap involved, but so are the .005 and 120pf caps. 

The tone control looks like it is attempting to be something like a variation on a BMP type control but isn't drawn that way.  For instance, I would think the 10k resistor is tied to pin 7 on U6B and the .1uf cap is likely soldered to the junction of the 100k pot and 10k resistor.  This would provide a low-cut path to the tone control via what is labelled as C17, and a lowpass-filtered path to the other outside lug of the tone control via the 10k/.1 network.

My first instinct would be to cut the value of that .005uf cap down to .0022uf and chop the .1uf cap to ground on the tone control ndown to .033uf.  That would reduce the amount of fixed lowpass filtering and make the tone control allow more squarishness through when the tone control is at any point other than full treble.

[hank reynolds 3rd]

http://filters.muziq.be/files/schematics/boss_oc-2.gif

http://filters.muziq.be/files/schematics/pearl_oc-07.jpg

these maybe will help...they aren't for the dod,but they follow a similar kind of layout,and are easier to read (the pearl is a factory schem,so i'm assumin it's right) ..the dod octoplus schem looks a bit wrong to me...
would chnging (switch??) the 220ks that make up the lpf at u6 give a crude 'jump' in the filters working

[Jaicen_solo]

Thanks guys, that's pretty much the same conclusion i'd come up with. I think i'll have to go through the schem and verify it myself as well. The tonestack is reported to sound like a BMP so that's probably correct. I'm thinking i'll have a mess around with the values of resistors & caps in the LPF first, and maybe install a 'tone bypass' switch. I'll let you all know how I get on later today.

[Jaicen_solo]

Just a quick update for those following this thread.
I've got into the pedal and had a bit of a play around with a few of the values of the lpf. As always Mr Hammers' suggestions proved to be bang on. I replaced the 0.005uF cap with a 0.001uF, and the 0.1uF tonestack cap with a 0.047uF cap (.033 sounded a bit too thin). This gives  more brightness to the tone and gives a good BMP feeling range for the tone control. It still does the wooly bass tone when backed off completely which is good, but there's also an interesting scooped mid tone about 11 o'clock which wasn't there before.

Now having gone over the schem, it appears to me that the octave down is being generated by feeding a half wave rectified signal into the Set inputs of 40103 and an inverted signal to the Reset pin, producing a series of pulses on every positive peak. These restart the internal clock. This is then feeding into the second half of the flip-flop where the pulses are acting as the clock input. The second half of the 4016 then flips from high to low (on/off) at the rate of the clock input, ie the pulses from the first half of the 4016. I'm guessing this is some sort of PLL implementation??

[Mark Hammer]

Gah!  Keep your chip designations straight, my friend.   I gather you are talking about the 4013 flip-flop, right?

I guess it's not something we thing of that often, but the flip-flop that does the dividing essentially has two tasks to perform.  It has to determine when to turn "on" and produce a rising edge, but it also has to determine when its time to turn off again.  How quickly it does so essentially sets the duty cycle or pulse duration of the divided version.  If the pulse-like waveform generated by the change in state of the flip-flop is brief enough, it will behave sort of like a squarewave oscillator with a duty cycle of, say, 20%.  This will produce a different tone than one which remains "On" for longer or shorter periods.  Now I'm wondering if the value of that 1uf cap after one of the FWR subcircuits feeding the 4013 reset pin could be varied to yield different pulse durations.

Even though it may not have been intended, thanks for bringing this to my attention.  Much appreciated.

[Jaicen_solo]

Dammit! I guess I look a little less clever mixing up those chips eh??
That is however a brilliant idea! I'm firing up the soldering iron right now to see what happens! Do you think i'd need to change both of the 1uF caps since it's a set/reset pair? I think I do. I'll get back to ya!

[Jaicen_solo]

Well I just tried swapping out the caps for smaller ones, down to about 0.003 but all it seemed to do was pretty much destroy the tracking so I put it back the way it was. Curses! You got me all excited then!

EDIT: I also tried different values for each position at the same time with similar results.

[Mark Hammer]

I salute your adventurous spirit in any event.  Perhaps someone with more technical knowledge than us two can suggest something to accomplish what seems like it ought to be technically possible.  Maybe its not a matter of the caps and timing, but a matter of altering the level of the reset signal such that reset will take place at an earlier or later portion of the declining side of the input waveform.

[Jaicen_solo]

Thanks for the vote of confidence, I think it's more adventurous than knowledgeable on my part

Having had a look at the datsheet again, and referred back to the schematic, it seems like the it might be possible to alter the duty cycle by inserting a cap between pins 3 & 5. It seems to me that the clock signal is being used to set the cycle width, though it might just be doing something completely different! Either way, since the pedal isn't mine i'm not about to go cutting traces but maybe an idea to bear in mind in the future. I'm sure it would be a great addition to this pedal, especially under CV.

[J. Luja]

changing the duty cycle of the first section of the 4013 isn't going to affect the output in any meaningful way since it goes directly to the flip-flop in section 2.
every other rising edge of the wave coming out of the first half is still going to happen at the same frequency regardless of duty cycle.

[Mark Hammer]

Dang!!!   

Is there ANY way to vary the duty cycle of the divided waveform?

[toneman]

Dang!!!   

Is there ANY way to vary the duty cycle of the divided waveform?

can U say "one shot"?
a 555 or a 74LS123 would work.
remember, a FlipFlop *always* has a 50/50 duty cycle.
normally, *that's* what U want   

[J. Luja]

Dang!!!   

Is there ANY way to vary the duty cycle of the divided waveform?

there's no way to do it with the parts already on the board that I can think of.
but with a daughterboard... send the output of the flip flop to a simple lowpass to "triangle" it up a bit, send that to the noninverting input of a spare opamp, then connect the inverting input to a voltage divider between V+ and ground and you've got an adjustable duty cycle on the output of the opamp.
The frequency of the lowpass and the range of the voltage divider will need to be adjusted to taste but it should work fine

[hank reynolds 3rd]

http://www.geocities.com/tpe123/folkurban/fuzz/swsh.gif

I've always wondered if this would be of any use in an octivider...
It gives a square wave, but also sawtooth etc....
would this affect anything in a noticable way (like make it more unstable etc)
a bit off topic ,but just a thought

Sam

[J. Luja]

http://www.geocities.com/tpe123/folkurban/fuzz/swsh.gif

I've always wondered if this would be of any use in an octivider...
It gives a square wave, but also sawtooth etc....
would this affect anything in a noticable way (like make it more unstable etc)
a bit off topic ,but just a thought

Sam

yeah, that would be an interesting option to use in place of the usual lowpass filtering that follows the square wave output of most octividers, since it's already square, you can omit everything before p1/p2 etc

and it shouldn't affect the stability since it's following all the sensitive areas that extract the fundamental pitch

probably not so useful in the case of this particular pedal as it seems the divided output is used to control a fet used as a switch to invert the original signal every other cycle (the scheme is obviously wrong in this section and needs some rearranging)

[Jaicen_solo]

Yeah, I'd noticed that as well. The schem is definitely not 100% correct, but it was good enough for me to get in and do what I needed to do. There's no way i'm re-doing the schem from this board tho, the traces are tiny and virtually invisible so I can understand where the errors have crept in.

[Jaicen_solo]

Ok, just one more thing I promise
I had a brainwave earlier which made me jump out of bed and fire up the soldering iron (yes, I have no life, sex or otherwise ).
I remembered I wanted to make a 'tone bypass' switch so; I found that by shorting the 1M resistor on pins 12 &9 of the 4013 to the middle lug of the tonepot, I get a really synthy square wave tone, which gives interesting results with the tonestack used. More gain as well.
Basically. i'm looking for ways to implement this into the pedal. Should I take out a leg of the 1M resistor and run a jumper to pin 7 of U6B?? Would this be a suitable way of getting around the lpf without cutting traces? (I want to keep the tone control too). Actually, could I just short pins 1 & 7 on U6??? That way I still get the buffered signal whilst bypassing the lpf.
What do you guys think?