MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

[INTEREST CHECK/POLL] VC Clock/audio divider
MUFF WIGGLER Forum Index -> Music Tech DIY  

I'm interested in...
a PCB with controllers
22%
 22%  [ 11 ]
a PCB with a panel
34%
 34%  [ 17 ]
a full kit
44%
 44%  [ 22 ]
Total Votes : 50

Author [INTEREST CHECK/POLL] VC Clock/audio divider
rpocc
Hi to all.
I have developed a simple, but reliable and affordable hybrid VC clock divider based on cheap MCU, opamp and a comparator.
Possible uses: division of clock signals (especially 24ppqn), Hi-hat rhythm control, APC-alike sounds, unusual sub-oscillator, bit-crushing distortion.

KIT AVAILABLE!



Prototype:


>>>Demonstration of breadboarded prototype<<<

The circuit is tested with various wave sources, PCB and panel designs are made, now I'm thinking about making a DIY kits or assembled modules from it. Before I start the production and so on, I would like to understand how demanded this module can be to anybody besides me.
The unit should be affordable like simple Synthrotek modules.

There is a mistake in the poll. Of course, second option means PCB+MCU+panel and there is no 4th option: assembled unit. You can comment here in case of such interest.

By the way, I have successfully prototyped a module that uses MCU only for control and CMOS chips for real signal processing. It was way more complex and had no real advantages over single-MCU solution.

Specs:
Module width: 4hp
Module Depth: 59mm
Net weight: 70g
Module dimensions: 129x20x80mm
+12 power: 15–26 mA
-12 power: 1.5 mA

* Dual voltage controlled clock divider.
* Works from 12v and 15v
* Can work in audio range (Jitter becomes audible at ~5 kHz)
* Visual feedback (LED)
* Output pulse length is equal to length of the input clock pulses.
* Protection from under/overvoltage
* Reconfigurable with jumpers or extension module (coming soon)

2 modes, each have 10/16 Available division factors:
1: /1, /2, /3, /4, /6, /8, /12, /16, /24, /48, (6 extra to be discussed)
2: /1.../16
10 first factors are available with knob and 6 extra factors are available with external CV.

Input signal: any square wave crossing +1v point.
Output: Pulses 0...+5v
Max timing jitter: 20 us
Input CV: 0...+6.4v

Each of two parts contain:
-----------
Jacks:
CV In
Clock In
Clock Out

Pots:
Division

LEDs:
Divided clock
----------

Total parts:
1 PCB, 2 MCU, 3 analog ICs, 1 voltage regulator, about 30—40 passive components, 2 pots, 6 jacks.

What do you think about it? The complexifications (like reset input, additional work modes) are possible, but they will of course rise the price and make the module wider.

Anyone interested in PCB/kit/module?
euromorcego
sounds great! I would certainly be interested, in particular if it stays affordable like the Synthrotek kits. Either pcb or kit. Best for me is a parts kit, where some crucial components (MCU) are included but common parts (resistors, capacitors and such) must be supplied (also saves the hassle from counting and bagging all those common parts).

What is the CV in good for? There is a clock in, division pot, and then the CV in? The input will probably be normalled, so one can have two clock divisions from one clock?

I hope the pcb will not be too large, mcu, 3 analog ICs, 40 passive components sounds a lot for 4hp. Most cases these days are not very deep anymore. Will it fit in a 6cm deep case?

Maybe you can also consider an XL version with a bit wider panel. I'd love to have two clock input with adjustable division (just like you have) and then some additional logic outs (AND, OR, XOR, ...). Then the module can provide some interesting patterns on its own.
falafelbiels
Yeah I would love to have this. 4hp please?
cretaceousear
Verra interessing effect ..
I could be interested in a PCB but I was planning on getting Michael Barton's. http://www.bartonmusicalcircuits.com/vcclock/
His can be built a few ways, but I guess yours is simpler in not using a PIC.
Any other important differences ?
terrafractyl
yep I'm also in the market for a VC clock divider.
I will now also be tossing up between this one and the Barton.

but yeah Maybe 2 channels will be enough for me.
Thanks!

smile
Stides
This seems great to me! It's motherfucking bacon yo
rpocc
euromorcego wrote:
sounds great! I would certainly be interested, in particular if it stays affordable like the Synthrotek kits. Either pcb or kit. Best for me is a parts kit, where some crucial components (MCU) are included but common parts (resistors, capacitors and such) must be supplied (also saves the hassle from counting and bagging all those common parts).

Of course, in case of kit I will include a programmed chip.

euromorcego wrote:
What is the CV in good for? There is a clock in, division pot, and then the CV in? The input will probably be normalled, so one can have two clock divisions from one clock?

In current design I plan to use switching jack socket, so either pot or CV in will be used to control the divisions.

Scenario for pot: You have some source of clocks (MIDI-CV converter with MIDI clock output, DIN SYNC module, etc) and you just need one or two particular divisions like quarter and 16th notes for your sequencer. So, you set corresponding divisions with pot and you're happy.

Scenario for CV in: You want to animate some percussion instrument with a sequencer (like in my crappy video), so you insert the jack into CV in, get a fast clock source (Master clock, LFO, etc) and go tweaking.

Scenario for both: sub-osc with fixed or controllable transpose.

euromorcego wrote:
I hope the pcb will not be too large, mcu, 3 analog ICs, 40 passive components sounds a lot for 4hp. Most cases these days are not very deep anymore. Will it fit in a 6cm deep case?

Yes, current PCB design involves a Doepfer-style perpendicular single PCB having dimensions about 10x5 cm.

euromorcego wrote:
Maybe you can also consider an XL version with a bit wider panel. I'd love to have two clock input with adjustable division (just like you have) and then some additional logic outs (AND, OR, XOR, ...). Then the module can provide some interesting patterns on its own.

Thanks for the advise. I will think about it. Right now I'm after narrowing.

cretaceousear wrote:
Verra interessing effect ..
I could be interested in a PCB but I was planning on getting Michael Barton's. http://www.bartonmusicalcircuits.com/vcclock/
His can be built a few ways, but I guess yours is simpler in not using a PIC.
Any other important differences ?

My will use DIP8 AVR.
The main difference is that my circuit is compact utility module when Barton's is more like master clock source, that is wide and complex. Lot of pots, lot of jacks, lot of functions.
What I tried to develop is some kind of 4ms RCD with manual control.
Also my circuit have two completely independable dividers with 2 separate MCUs like 4ms PEG. By he way, switched jacks may be used to clock both circuits with the same wire.
Karl_Joseph
I take a pcb with panel and chip please!
rpocc
I have to wait till I get new right-angle 4ms-type jacks from Thonk since initially I planned to use Doepfer-type, but they are very expensive.

Meanwhile there is a problem with clock reset. If you turn a rate knob manually, you may get divided clocks, that are not syncronized with bars (imagine 16th triplets, that are shifted to one straight 16th).
So, this is not a problem, when you use a sequencer switching rate at exact moments of time or if you have a reset input.

Unfortunately there is no way to place 3 PCB-mounted sockets and a pot on the half-space of the 4hp module with through-hole components. So, there are three options:

1. Leave design as it is (each half: 1 pot, in, out, CV)
2. Assume manual wiring to place four jacks (in, out, CV, reset). There will be hole on the PCB.
3. Make a 6hp-wide module with two jacks (per half) pcb-mounted and two jacks manually wired. Elby-design-style mini-PCBs for additional jacks are not practical iwith such simple module.

What do you think?

As I see, the interest is extremely low, so it is unlikely that I will sell out even 10pcs batch (even if I leave pair for myself), so don't blame me if I drop this project and turn to something more unique.
paulstone
thumbs up Interest if work also for 15v
cretaceousear
I bet once you start to sell it, and if you make other modules or PCBs, interest will pick up and you will sell more over time.
sammy123
I'd be interested in a pcb/panel/chip set.
LektroiD
Great stuff, count me in for a PCB, Panel and chip set if this goes ahead smile
euromorcego
Quote:
1. Leave design as it is (each half: 1 pot, in, out, CV)
2. Assume manual wiring to place four jacks (in, out, CV, reset). There will be hole on the PCB.

i think either is ok. I am fine with wiring jacks (as long as it is only jacks, like in the frequency central and manhattan pcb/panels). But the original design is also fine.

I forgot: are the two staged normalized so that one can either have two divisions from the same input, or maybe even better: another division of the first division (maybe both, selectable via jumper on the back).

If you are unsure about interest, you could use a better title in the thread. It now seems a bit hidden and unspecific. Something like '[INTEREST CHECK/POLL] PCB/PANEL for VC clock/audio divider' would be more easy to spot. And maybe a picture or even demo (even if preliminary, or mock-ups of the panel, the video linked above is a bit difficult to understand, what you see is Orbitals).
And i agree with cretaceousear, once you have the pcb available and there are some figures and successful builds, more people will be interested (of course, you won't sell hundreds, but more than ten should be do-able, depending on price).
rpocc
euromorcego wrote:


I forgot: are the two staged normalized so that one can either have two divisions from the same input, or maybe even better: another division of the first division (maybe both, selectable via jumper on the back).

Input 2 will be normalized to input 1. The jumper is interesting idea, I will think about it.

euromorcego wrote:
If you are unsure about interest, you could use a better title in the thread. It now seems a bit hidden and unspecific. Something like '[INTEREST CHECK/POLL] PCB/PANEL for VC clock/audio divider' would be more easy to spot. And maybe a picture or even demo (even if preliminary, or mock-ups of the panel, the video linked above is a bit difficult to understand, what you see is Orbitals).
And i agree with cretaceousear, once you have the pcb available and there are some figures and successful builds, more people will be interested (of course, you won't sell hundreds, but more than ten should be do-able, depending on price).


Thanks you for your advises. You're right, I have provided too few information. The video is a quickshot. I'll post better video soon.

Meanwhile this is mock-ups of the module with aluminum and PCB panel. I don't know yet, which method will be cheaper.

By the way, module will work both from 12 and 15v.
ABC
a rough estimation of cost would be useful, but I've been looking for a simple vc clock divider for a while, so am definitely interested.
regenbot
I'm interested thumbs up

I think you should make a aluminium panel even though it will be more expensive.
jonesboy
I would be interested in a PCB / chip / panel.
Reality Checkpoint
jonesboy wrote:
I would be interested in a PCB / chip / panel.


Me too!

Happy with either aluminium or PCB material panel.
Raytracer
I could be interested.

Does your design use Musical or Mathematical dividing?

Does the reset input trigger instantly or on the next clock input?

Quote:
1. Leave design as it is (each half: 1 pot, in, out, CV)
2. Assume manual wiring to place four jacks (in, out, CV, reset). There will be hole on the PCB.
I would prefer option 2. I didn't find it difficult to wire the jacks on the Frequency Central ADSR.
rpocc
Raytracer wrote:
I could be interested.

Does your design use Musical or Mathematical dividing?

Does the reset input trigger instantly or on the next clock input?

Quote:
1. Leave design as it is (each half: 1 pot, in, out, CV)
2. Assume manual wiring to place four jacks (in, out, CV, reset). There will be hole on the PCB.
I would prefer option 2. I didn't find it difficult to wire the jacks on the Frequency Central ADSR.


Of course it is musical in terms that output is in phase with beats, so all divisions will correlate on 1st beat
sonicwarrior
Does the PCB have mounting holes for non Eurorack formats? I could use one of these in my 5U system.
yan6
I would like one PCB/Panel/MCU, if the price is right possibly two hyper
zoot horn rollo
Im in for a couple if reset is a feature
acilator
I'm in for PCB/panel/MCU
rpocc
sonicwarrior wrote:
Does the PCB have mounting holes for non Eurorack formats? I could use one of these in my 5U system.

I'm not very familiar with 5U standards. The board by design has very busy front corners, so I can put 3.2mm holes there, but one of them will have an offset. See the attached overlay file. (It is oversized to be more clear.
Actual size is indeed smaller: about 11x5.5 cm.


zoot horn rollo wrote:
Im in for a couple if reset is a feature

What if I leave reset inputs along with some other interesting stuff on the PCB and offer additional 2HP panel with holes for vertical Thonk jacks? I have a bad feeling about cutting the PCB.
sonicwarrior
rpocc wrote:
I'm not very familiar with 5U standards.

There are no real standards so any holes will do but I guess two screws at the power connector side would be more stable. Maybe a second one at the current silk print of IC6/C4?

Is JP1 the Eurorack power connector? What is the other one then? It looks a bit small for MOTM.
paulstone
also adding 2,54 holes in the jack section can help different format
rpocc
paulstone wrote:
also adding 2,54 holes in the jack section can help different format


What is 2.54? Do you mean a SIP3 with 2.54mm distance between holes? I don't really sure I can do this because area inside the sockets is actually full of tracks. I'll try, but board is already loaded with jumpers, sockets, etc.

sonicwarrior wrote:
rpocc wrote:
I'm not very familiar with 5U standards.

There are no real standards so any holes will do but I guess two screws at the power connector side would be more stable. Maybe a second one at the current silk print of IC6/C4?

Why the one at the bottom right corner is bad?

Quote:
Is JP1 the Eurorack power connector? What is the other one then? It looks a bit small for MOTM.

JP1 is Eurorack. Other one (J1) is for MTA-156 connector - the same as used for floppies. As far as I know, this is acceptable by MOTM users as like as large Molex "IDE HDD" connectors.
sonicwarrior
rpocc wrote:
Why the one at the bottom right corner is bad?

My bad, thought this belongs to the jack footprint.

rpocc wrote:
Other one (J1) is for MTA-156 connector

MTA156 means 0.156" = 3.96 mm hole to hole grid. This looks like 0.1" = 2.54 mm.
paulstone
rpocc rigth
i mention for SIP3 with 2.54mm (or MTA connections) , try if you can otherwise no problem , just is needed to solder the wire , personally don't like solder wire in the pcb , but i can adapt myself hihi
rpocc
Thanks to MOTM-users, I see about connectors. Seems more and more like we design a PCB together. smile
New version is attached. Sorry, additional output connectors are too much for this module.
rpocc
I just made a new video that is way more clear than previous one.

Hope this will help you to understand how it works. As usual, my English is horrible smile

NEW VIDEO

Production state:
Almost all necessary parts are ordered and expected to come in the beginning of October.
By the end of the week I will have three prototype PCBs to be tested with DIY milled panels. In case of succesfull mechanical and electrical tests, 50 boards and panels will be manufactured and will be available at 1st decade of October. During this time I will be able to improve the firmware.
mckenic
Would it be possible to have the functions of JP2 in a DIY expander board - a toggle switch for "Separate/Linked"?

BTW - the demo video... the drums part is BRILLIANT! Its the clearest example of the musical uses of divider Ive seen and is really cool sounding!

Im in for a kit/whatever you are offering!
rpocc
mckenic wrote:
Would it be possible to have the functions of JP2 in a DIY expander board - a toggle switch for "Separate/Linked"?


Of course! Moreover, current version of the PCB has all jumpers combined into compact 2x5 pin header.

mckenic wrote:

BTW - the demo video... the drums part is BRILLIANT! Its the clearest example of the musical uses of divider Ive seen and is really cool sounding!
Im in for a kit/whatever you are offering!

Thank you!
Brad
Looks great. I would be interested in a kit too (depending on price).
Reality Checkpoint
Option 2 or 3 for me, and I will take a panel, PCB and chip. If you want to keep it at 4 hp then go for option 2, but I will buy one whatever you decide.
Maco
Ill take pcb and pic

applause
oberkorn
voted for PCB/Panel/PIC(s) w00t
rpocc
This version of the PCB was sent to factory.

All jumpers are combined into header connector. All jacks are connected to SIP4&5 holes. So switches, custom panels, separate inputs are possible.

But you will probably don't need a reset in most situations because reset is automatically performed when division factor is changed.
spotta
Would be up for grabbing a panel, pcb & chip thumbs up
Raytracer
rpocc wrote:
But you will probably don't need a reset in most situations because reset is automatically performed when division factor is changed.

Is it possible to build the clock divider without this?
rpocc
Raytracer wrote:
rpocc wrote:
But you will probably don't need a reset in most situations because reset is automatically performed when division factor is changed.

Is it possible to build the clock divider without this?

I can build alternative firmware and include it to the order on demand, but I would like to know, why it's so important, and what exactly do you not like in this approach. Let me explain it more.

The problem:
CV measurement in microseconds scale can be done either before clock pulse or after it. This is how background ADC works. Any step sequencer most probably will change voltage right after the clock due to propagation delay in the analogue circuits. Man can tweak knob in relatively exact time with some clock pulse or beat, but the very moment of CV measure is unpredictable — it may be millisecond before or millisecond after clock pulse. This can cause unnecessary situations when you're expecting rhythm to change at exact moment and be in sync with another beat, but you just get it one pulse delayed.

The first version of the firmware has different algorithm to avoid it: counter of clocks to be skipped - let's call it cycle - was able to change only on it's boundary. Say you has set factor to /3, then change it to /4 right after first or second pulse of original clock. Controller was waiting till the end of full cycle and then changing it to next "queued" factor. It seemed to be smart idea at first, but it's kind of thing when gear takes overcontrol upon process. If you have main musical pulsation at the rate 1/12 pulses, then you set division factor to something small like 1/6 or less, then got to greater factor in the wrong moment, you will be out of that main pulsation, all pulses will be delayed to some number of pulses. The only way to return it back was to send reset (that is optional in my module). or move back to small divisions and try to compensate the delay.
Also if reset was kept high, output was disabled, but it is not important.


The current version is dedicated especially for sequencing and play/stop commands. It automatically detects if last division factor change was made before or after first pulse in the cycle and making correction if necessary. This helps to keep pulses in beat. In all other cases, reset is performed and next first pulse in the cycle will be clock pulse being received just after ADC measurement.

If reset is kept high, all clocks will go to the output, but since almost any sequencer has play/stop inputs, it's more logical to use MMC-compatible MIDI/CV converter and enable/disable steps with dedicated commands.

Anyway, chip will not sending any signal between source clocks. If reset is received somewhere between source pulses, chip will of course wait for next pulse.

It might be quite difficult to understand. See the attached diagram for visual explanation.

So, current firmware is working as shown on the top diagram. What do you want to propose?
Raytracer
Thanks for the diagram rpocc. That helps clarify how your code works. The beat magnet area looks like a good way to handle things.

When using the "reset on factor change" scheme, what happens if the division is changed outside the beat magnet area?
rpocc
Raytracer wrote:
Thanks for the diagram rpocc. That helps clarify how your code works. The beat magnet area looks like a good way to handle things.

When using the "reset on factor change" scheme, what happens if the division is changed outside the beat magnet area?


In this case reset will be performed and next uncoming clock will be first in new sequence. Same case is shown right under letter "t" in last "reset" of the title.
LektroiD
Brilliant!

Count me in for a PCB, MCU & Panel (preferably silver, not black).

thumbs up
woodster
I'd also go for a PCB, MCU and Silver Ali Panel.
sicpaul
Yes, brilliant indeed

I'd be in for a pcb and MCU
Raytracer
rpocc wrote:
Raytracer wrote:
Thanks for the diagram rpocc. That helps clarify how your code works. The beat magnet area looks like a good way to handle things.

When using the "reset on factor change" scheme, what happens if the division is changed outside the beat magnet area?


In this case reset will be performed and next uncoming clock will be first in new sequence. Same case is shown right under letter "t" in last "reset" of the title.


Thanks for the details Rpocc. I'll be happy to use your default program. I can always use a sample and hold module to achieve the same behaviour as in the second diagram.

I'll likely be up for two pcb's and two chips if you do a run. I'll put them behind a custom panel.
rpocc
Production is started.
Run of 50 boards and panels is assumed to appear in the third decade of October. I also have all necessary parts. I still can't calculate final costs, so I will tell final prices in the beginning of October, but I assume the absolute top under $100 for full kits and $20–$40 for essentials kits (PCB/Chips, Panels, maybe hardware).

Here is current design of the top side (final version) and photo of assembled prototype. Commercial version will have stylish boards with white silkscreen and anodized aluminum grey panels with black graphics.

Final version is widely configurable with set of jumpers or can be expanded by external module with switches and jacks. With jumper or switch "linear" division set can be reached: /1, /2, /3... /16 to get chiptune effects and sweeps with external modulation source.

Soon I'll make PayPal-based purchase page at my website with all available options.

If some native English speaker can help me with translation of the page and manuals (proofreading) feel free to PM me. Free full kit with free shipment is guaranteed, so Eurorack user are welcomed.

As all will be ready for sell, I'll create new threads with discussion about module itself and extension for it.

Preliminary specifications:

Module width: 4hp
Module Depth: 59mm
Net weight: 70g
Module dimensions: 129x20x80mm
+12 power: 15–26 mA
-12 power: 1.5 mA
cretaceousear
Yay - in for one pcb and mcu.
I can do proofreading/good English for you if you don't get someone who would like to do job for full kit!!
mckenic
Yeah I can proof if you need... used to in a previous life!
yan6
Ohhh, I'm excited for this. Keep up the work, its looking good.
dropmotif
pcb and chip for me please.
sammy123
Looks awesome. I am still in for a panel/pcb/chip.
rpocc
OK, please no more proofreading offers. cretaceousear PMed.

mckenic wrote:
Yeah I can proof if you need... used to in a previous life!

I can contact you next time when extension will be ready for production. smile
puzo
i like it and have been watching for a pcb, i think it'll do well. Its a simple little module, but one that is incredibly useful. Be a shame not to have it run. I'm a complete noob, can bearly understand how these thing work in a setup but it could offer some crucial applications.
Sparky
I'm interested in a pcb + panel
rpocc
puzo wrote:
i like it and have been watching for a pcb, i think it'll do well. Its a simple little module, but one that is incredibly useful. Be a shame not to have it run. I'm a complete noob, can bearly understand how these thing work in a setup but it could offer some crucial applications.


The manuals, setup examples, audio examples, etc will appear as I have "real" panel and "real" PCB.
euromorcego
any news on this one? Still interested ...
rpocc
OK, now I'm accepting orders.
The new thread is started


Thanks to all participants of this thread for support.
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 3
Powered by phpBB © phpBB Group