dividing an asymmetric duty cycle pulse

[R.G.]

It struck me as I hit "post" that one could use a.j.'s trick on adjacent bits in the counter word in a PIC by  ANDing adjacent bits and storing that in the output register. If there's enough cycle ticks left in the error budget, one could make that configurable too.

Another thing that might make this usable is a form of internal pipelining. A tick of the interval clock would cause a transfer of the prepared output word to the output register on the pins, and then the time til the next clock tick used to update and prepare the next output word. It would put any excess cycles on the correct (i.e. noncritical) end of the timing period.

[parmalee]

I'm endeavoring to build a sort-of Vox Continental clone, albeit with 5/6 size keys and replacing the discrete blocks with CMOS chips, op amps, etc. wherever practicable. 

You would probably enjoy "Electronic Musical Instruments" by Richard Dorff. It's a book on electronic organs, and has a "deep dive" on how many of them work from both the system and circuit perspective.

If you're doing an organ, then monostables and diode-resistor-capacitor one shots will give you your desired duty cycle, approximately.

If you are into saving yourself some circuitry, a single CD4024 gives you seven stages of binary flipflops in one package. A top octave generator and 12 CD4024s generate all the musical notes. Or slice it another way: a single PIC can be crystal controlled to be amazingly close to standard tuning for a single note in the top octave. One PIC plus one 4024 gives you eight octaves of dividers, all the octaves for one note. Twelve identical circuits gives the entire note generation for an entire organ, with only two chips per note.

As for a 4PST, you can use either a CD4066 for that, or you could use JFETs as switches. The CMOS is probably cheaper.

Dorff's books are fantastic--he also wrote one exclusively on Schober organs.  I did not know that he invented the Thyratone; not surprised though.

I think I'll stick with twelve tunable oscillators (from a couple of hex inverters), the dividers, and diode keying.  I don't have any scrap organs or TOGs readily available and the (single chip) ones online are either sixty bucks or rather suspect (Chinese vendors on Ebay), and I like to experiment with alternate temperaments on occasion.

Any suggestions as to the simplest, lowest parts count method for achieving something akin to the "percussion" setting on the bass manual/pedals of the Farfisa Compact Deluxe and Compact Duo?  The envelope is essentially zero attack, fairly rapid decay, infinite sustain (so long as the key or pedal is depressed), and no release.  And it's gotta be polyphonic--so a basic EG/faux VCA for each note (18, for my setup).

I've come up with a method similar to that used in the PAIA Strings n Things: keyswitch to voltage buss triggers simple decay envelope and gates pulse wave tone (roughly 4 diodes, 2 caps, 2 resistors) AND also gates same tone at lower amplitude (just sustain--2 diodes and maybe 4 resistors).  For 18 notes, that's over one hundred diodes.  I feel like there is another way of doign this that I am overlooking.

Attack-sustain-release, just decay, or just sustain seem easily attainable via a single chain, but decay-sustain is eluding me.

[parmalee]

Sorry, I don't communicate with humans much and often omit relevant details.  Here's the Paia synth alluded to above:
(courtesy of stefanv.com)

[ElectricDruid]

Are you *sure* the bass pedal percussion is polyphonic? It's often the case that organ percussion circuits are single-triggered. All the outputs are fed to a single VCA, and the envelope is retriggered when a new key is pressed, and often not retriggered if another note is held. This is a behaviour of the percussion circuit on the Hammond tone wheel organs, and was widely copied later. My 1970s Yamaha organ worked like this.

It would make sense to only have one VCA for bass pedals on a typical home organ too, since you'd usually only play pedals with one foot, the other being occupied on the swell pedal.

Aside from that, I think your proposed design is very efficient and you'll struggle to do it in many less parts.

Tom

[anotherjim]

The Farfisa I have uses diode transmission gates with an RC (quite long time) for pedal sustain. There is a daisy chain of change-over pedal contacts that forces monophonic low note priority. The schematic can be found here...
http://elektrotanya.com/farfisa_series_250s.pdf/download.html
Here's the detail.

In this, the source pitches are much higher. A single divider chip provides the low pitches, but also a crude staircase saw wave for the rather useless (unless it had a synth filter!) bass guitar voice. This means that it cannot be polyphonic  - only one divider chip!
So, if we remove the divider and use the correct octave source pitches, bypass the change over switching, maybe it could be polyphonic.

[parmalee]

Are you *sure* the bass pedal percussion is polyphonic? It's often the case that organ percussion circuits are single-triggered. All the outputs are fed to a single VCA, and the envelope is retriggered when a new key is pressed, and often not retriggered if another note is held. This is a behaviour of the percussion circuit on the Hammond tone wheel organs, and was widely copied later. My 1970s Yamaha organ worked like this.

It would make sense to only have one VCA for bass pedals on a typical home organ too, since you'd usually only play pedals with one foot, the other being occupied on the swell pedal.

Yeah, it's definitely polyphonic on all in the Farfisa Compact series.  It is essentially the 16' tab amplified and run through a couple of passive lowpass filters--plus another segment of a board devoted to the long/short percussion effect.  The Compact tones are not diode or transistor keyed, they're pretty much switched direct from tone generators/dividers to subsequent filtering, preamp, etc.  Consequently, they tend to make a lot of noise even when you're not playing any notes!

In another thread, I remarked upon the polyphony aspect.  It's real subtle, and perhaps only a few milliseconds, but when playing legato passages (and I'm no virtuoso, mind--just ordinary stuff), that tiny bit of overlap seems to make a massive sonic difference.  It's somehow more than what, say, reverb or a slight delay could even accomplish.  It kind of reminds me of the few times when I've gotten to play proper pipe organs and there's that very slight latency between key strike or pedal strike.

Of course, I'm now realizing that at best I could shave off maybe 20 diodes or so--108 to ~80.  I guess it's not really a big deal.

[parmalee]

The Farfisa I have uses diode transmission gates with an RC (quite long time) for pedal sustain. There is a daisy chain of change-over pedal contacts that forces monophonic low note priority. The schematic can be found here...
http://elektrotanya.com/farfisa_series_250s.pdf/download.html
Here's the detail.
---schematic---
In this, the source pitches are much higher. A single divider chip provides the low pitches, but also a crude staircase saw wave for the rather useless (unless it had a synth filter!) bass guitar voice. This means that it cannot be polyphonic  - only one divider chip!
So, if we remove the divider and use the correct octave source pitches, bypass the change over switching, maybe it could be polyphonic.

Wow.  That is significantly different from the manner used in the Compacts and the FASTs (I think).  I'll try to locate a sample and a legible schematic.  But I think I'm gonna breadboard that one just out of curiosity.

Incidentally, Arturia makes a decent Farfisa Compact Deluxe software--with lots of added features.  The program is 100 bucks, but you can download a free demo which you can use over and over, but only for twenty minutes a go:

https://www.arturia.com/farfisa-v/overview

[anotherjim]

Incidentally, the gate sustain due to all those 25u caps is far longer than the voices actually have. D28,29 are transmission gates for 8' and 16' controlled by the true sustain timing based on C32,C33. The bass guitar has it's own envelope Q6/7 and diode gate D30. The contact bus wire at the bottom (wire 33) is basically the "key down" gate.

I've heard good things about the free ComboF and ComboV VST organs (guess which is which!). I've used one called Combo-Sister for many years - and my Nord Electro has them too. Still doesn't feel like the real thing, none of 'em.

[parmalee]

Okay, the Compact Duo bass manual/pedalboard percussion complements the tone keyed with a monostable one-shot pulse at commencement of keying:  thus, a percussive attack (of two selectable durations) decays to a lower amplitude, infinitely sustained (so long as key is depressed) tone.  Sounds *kind of* like a double bass played without a bow.

Here's the schematic:


PA-16 in the upper left is the percussive model.  Tones enter via 7 (buss p3), and proceed to preamp/filters through 8.

Same switch--for tone going into 7--drops buss p2 to ground, enters pa-16 via 3 and triggers monostable.

4,5, and 6 are for long or short mode--6 connects to 5 for long, 6 connects to 4 for short.

One thing about this scheme is that the percussion is triggered for ALL notes held, each time a new key/pedal is struck.  Not really a concern most of the time.

I kind of like this method as it doesn't involve replicating a scheme for each and every note; however, with single-pole switching I'm not sure how this can be achieved (bilateral switch maybe?). 

Alternately, I can do something along the lines of the PAIA string synth (see a few posts up): key a variable decay envelope to impose over diode (or transistor) switched full amplitude tone AND switch on a lower amplitude tone simultaneously.  This would work, but it would require considerably more components.  Surely there is a way to trigger the monostable (on every keystroke) AND switch the tone on/through without double-pole switches?  Or am I delusional?

The full schematic can be found here: http://farfisa.org/farfisa-compact-duo-schematics-133/

[parmalee]

Incidentally, the gate sustain due to all those 25u caps is far longer than the voices actually have. D28,29 are transmission gates for 8' and 16' controlled by the true sustain timing based on C32,C33. The bass guitar has it's own envelope Q6/7 and diode gate D30. The contact bus wire at the bottom (wire 33) is basically the "key down" gate.

I've heard good things about the free ComboF and ComboV VST organs (guess which is which!). I've used one called Combo-Sister for many years - and my Nord Electro has them too. Still doesn't feel like the real thing, none of 'em.

Yeah, I've got the Martinic ComboF and ComboV and like them quite a bit, but honestly, the Arturia program kinda blows them away!

And I absolutely agree that none of these are like the real things, but the analogue modeling methods seem vastly superior to any of the systems which really upon sampling.  Not surprising considering that they essentially replicate the original circuits--and even try to incorporate the (to some--though not me!--undesirable) idiosyncracies of the originals.

I'm curious about the "bass guitar" block.  With the twin-T config it looks like it could have some promise, if properly tweaked.

[parmalee]

This ought to work, yes?



(values to be determined)

c1, c2, and r1 form the decay envelope.

r4 and r5 will simply sustain, of lower amplitude.

r2 and r3 enable gating.

[anotherjim]

Looks good.
D1 anode must go to ground to block the tone. Tone wave must not suffer overshoot causing it to dip negative -  maybe resistor needed in tone feed (examples always seem to have this).
Mix output must be ground referenced so D2 can turn on. Maybe R7 can be a common mix resistor to ground with output from D2 cathode? 

[amptramp]

Are you *sure* the bass pedal percussion is polyphonic? It's often the case that organ percussion circuits are single-triggered. All the outputs are fed to a single VCA, and the envelope is retriggered when a new key is pressed, and often not retriggered if another note is held. This is a behaviour of the percussion circuit on the Hammond tone wheel organs, and was widely copied later. My 1970s Yamaha organ worked like this.

It would make sense to only have one VCA for bass pedals on a typical home organ too, since you'd usually only play pedals with one foot, the other being occupied on the swell pedal.

Aside from that, I think your proposed design is very efficient and you'll struggle to do it in many less parts.

Tom

Add to that one consideration: if you do manage to hit two adjacent pedal notes simultaneously and you are at high volume, you will be remembered as the person at the epicentre when the earthquake struck.

[parmalee]

Hugh Banton--of Van der Graaf Generator fame and designer of digital pipe organ systems--briefly used a setup in which he incorporated a Farfisa Professional organ into the wiring of the lower manual of a Hammond C3, separated all of the outputs, and essentially made it so that various effects--fuzz, dividers, tape echo, Schaller Rotosound, etc.--could be easily re-routed.  He also designed some sort of system which incorporated 32' stops, along with appropriate amplification and cabinets to handle the demands.

To be honest, I'm not sure how much actual use--in studio or on stage--this system got, or even how much of it simply remained in the theoretical stages, but...  interesting idea, all the same.  In VDGG, both Banton and David Jackson (saxes and flutes--he often played a baritone and tenor simultaneously) used octave dividers to great extent.  Unfortunately, most of the live recordings from the seventies aren't of sufficient quality to really capture the lows, but some of the BBC sessions sound great.  Organ, saxes, drums, and dividers--who needs guitars?

Anyways, here's some of the info  (full story starts here):

[anotherjim]

I was going to mention HB's giant RTR speakers and I think this is the near mythical HB1 organ described in that report. I went to see VdGG all those years ago in Manchester, and the organ broke down, never to be heard that night . Luckily HB also played bass guitar, so they got quite a bit done, but I never heard the full VdGG experience up close and personal. Funnily enough, a few years ago I saw them in Manchester again...  and no sound from the organ. Luckily, it was only that something hadn't been switched on in his backline. They did Gog and Magog that night and it was downright scary!

[parmalee]

I'd kill to have seen VdGG back in the day, though they're still good.  Just not the same.

I have about 15 boots from 71-72 and 74-76, some of which are of fairly decent quality.  No soundboards, expert perhaps one Italian show from '75.  I've devoted an awful lot of time and energy to each and every version of Theme One I've got.  Hugh's pedal playing has always been fairly restrained, but he always seems to hit the right note--the last few minutes of "Childlike Faith...", in particular, features some extraordinary, whilst minimal, pedal work.

[ElectricDruid]

One thing about this scheme is that the percussion is triggered for ALL notes held, each time a new key/pedal is struck.  Not really a concern most of the time.

That's the "single triggered" thing I mentioned earlier. Pretty common for percussion on organs of all types, it seems.

Tom

[parmalee]

Looks good.
D1 anode must go to ground to block the tone. Tone wave must not suffer overshoot causing it to dip negative -  maybe resistor needed in tone feed (examples always seem to have this).
Mix output must be ground referenced so D2 can turn on. Maybe R7 can be a common mix resistor to ground with output from D2 cathode?

Could you clarify what you mean in the underlined portion?  Actually, I think you already did state it clearly, but for some reason it's eluding me.

I'm finding that I can omit R4 and R5, and simply make R3 sufficiently larger than R2 to achieve the lower amplitude sustain tone.  However, the overall drop--meaning, the peak amplitude of the beginning decay portion, as well-- is greater than anticipated and I'm not entirely sure as to why...  Maybe I'm just really tired and my tiny studio space is struggling to retain heat--it's about 0 F/ -18 C outside.  I'm using 1n914 diodes, so I do not think them the culprit.

[parmalee]

That's the "single triggered" thing I mentioned earlier. Pretty common for percussion on organs of all types, it seems.

Tom

That type of triggering I don't find all that troublesome.  I've encountered some schemes for which percussion is only re-triggered upon full release/opening of all key switches.  I can see how that might work for a dedicated bass manual, or pedals--albeit somewhat a challenge for legato passages--but on the manuals, it just seems odd.

Incidentally, there's a truly staggering number of patents registered between the 1950's and the early 1980's on this very subject.  Some entail an absurd number of components--30, 40 and then some--for every single key switch.  One design incorporates a number of switches within the envelope shaper to approximate a variety of instruments from harpsichord to piano to organ to percussive organ, and so forth--and all this just for envelope, the subsequent filtering is not accounted for within said patent.  Kind of like that one RMI beast--not the one that Tony Banks used (which was mostly piano sounds, IIRC), but the other one.  The Rocksichord, I believe?