Muff's Modules & More

fonik

below you see pictures of my finished divider module. i posted this over at e-m.com as well, but i have a couple of questions and wanted to be sure to get the opinions of you patch experienced folks.

for the divisions i used inverters. why? this way all divisions are "ON" when reset pulse is received.
i never have had a divider before, and for me it has been just natural to have it this way. i looked about it as it would be a sequencer: i.e. the Klee2 sequencer reloads on the pulse and is at the "one" of the pattern when reload (reset) pulse is received. i would think a divider should operate the same way, when it gets controlled by another divider or sequencer or whatever. am i wrong here? woult it be better to have all divisions "OFF" when reset is received? then the reset pulse would have to be the last step in a sequence!?
any opinion would be appreciated...

Cat-A-Tonic · Posted: Sun Feb 08, 2009 6:39 pm Post subject:

Quoted from Ken Stone (on his CGS Pulse Divider):

neandrewthal · Posted: Sun Feb 08, 2009 8:48 pm Post subject:

plord · Posted: Sun Feb 08, 2009 11:31 pm Post subject:

Regarding the inverters, I think no matter what you do some people will complain, because not every clocked OR clock-generating module is going to behave the same. I get all kinds of different behavior from clocked and clockable stuff around here. The CGS outputs all go high on reset, as noted, which may or may not be what you wanted to have happen if, for example, you're using the /4 to step through slow sequencer key changes on a /8 or /6 pattern confused

On the other hand, as the Analogue Soutions people are fond of pointing out, resetting to one on step 16 won't always be the right thing either, which is why they reset to step 16 on the Oberkorn.

Anyway, sometimes I lose the 1, and the only way to get it back is a hard reset on the whole system. Drag :(

plord · Posted: Sun Feb 08, 2009 11:33 pm Post subject:

Oh, and: I'm totally in for 2 or 3 of these regardless of their reset behavior. A skinny little Frac divider just dedicated to making the long divisions? I'm all over it.

consumed

A Dingleberry Monstrosity · Posted: Mon Feb 09, 2009 1:26 am Post subject:

looks fine to me, but thats only after a quick glance. At first I was rather confused about having the 1/ there, but if you you wanted to tap the stock signal then that would save you from having to use a mult.

throw in some "and" gates to get everything else

for example "1 and 2" = 3
"1 and 4" = 5
....

etc etc SlayerBadger!

fonik

wow. thanks for all your professional opinions and thoughts about this. great input. w00t

will have to think about all this for a while...

i wanted to keep things simple and not suffer from feature creep. so no odd divisions yet. however, i will think about it. and it seems this divider could be a nice addition to existing designs - that is what i get from some of your statements.

Leisure Cove · Posted: Mon Feb 09, 2009 10:18 am Post subject:

fonik · Posted: Mon Feb 09, 2009 11:29 am Post subject:

otherunicorn · Posted: Tue Mar 03, 2009 6:49 am Post subject:

The reason for resetting to ON is simple - You want to reset to step 1, not step 0. This is music, not maths.

As for resetting to step 16 instead of step 1, there is reason for that too - if you want to advance to the first step with the first pulse. It all depends on how you play.
I have long thought that a sequencer should have an extra 1/2 stage (not 1/2 step!!) - an initial "0" stage with no pots or outputs. A sequence would be "initiated" to this stage, so that it advanced to step 1 on the first clock pulse. It would then be out of the loop, with "reset" taking the sequence back to step 1 with each successive cycle.

fonik · Posted: Tue Mar 03, 2009 7:06 am Post subject:

otherunicorn · Posted: Tue Mar 03, 2009 6:04 pm Post subject:

There is another difference I forgot to mention - binary counters advance on the falling edge of a clock signal. Synth modules are triggered by a rising edge (trig or gate) signal.

fonik · Posted: Wed Mar 04, 2009 4:43 am Post subject:

zthee · Posted: Wed Mar 04, 2009 4:53 am Post subject:

The 4024 goes high and remains high for the entire duration of that division? I.e. the /8 output goes high on the 8th step and low after 8 steps?

Wouldn't it be good to add a trig output to the circuit? So you can have a short pulse sent at every division as well?

fonik · Posted: Wed Mar 04, 2009 5:37 pm Post subject:

zthee · Posted: Thu Mar 05, 2009 1:56 am Post subject:

fonik · Posted: Thu Mar 05, 2009 7:12 am Post subject:

zthee · Posted: Thu Mar 05, 2009 7:57 am Post subject:

You rock man! we're not worthy

I'll def. take 2 of the quantizer board when it done! w00t

Kwote · Posted: Fri Mar 06, 2009 2:32 pm Post subject:

I missed the mention of the quantizer but I’d definitely be interested in that board so please keep us posted.

fonik · Posted: Fri Mar 06, 2009 5:54 pm Post subject:

consumed · Posted: Fri Mar 06, 2009 7:12 pm Post subject:

i definitely like this behavior even though my clock divider works like this:

i dont see a 'right' way in either design. it would be fantastic to have both behaviors in a system.

ignatzthemouse · Posted: Wed May 27, 2009 5:10 am Post subject:

oooh ive just been thinking about how to do this. Will find this post and the diagrams useful.
The panel looks pretty btw.
Cheers fonik.

wetterberg · Posted: Wed May 27, 2009 8:31 am Post subject:

just re-read this thread - I like the debate about reset to nothing or step 1, actually. From my perspective I'd actually prefer the reversed behaviour - that resetting "primes the counter", but that it's the CLOCK that determines when we get the first count.

Fonik, I'm quite interested in this circuit, and I figure I'd breadboard this as soon as possible, with the two remaining /64 and /128 divisions too - might as well wink

Quick questions:

1) could you explain to me the circuitry surrounding the Reset In? Is it debouncing for the button? From the way I've understood it the 4024 is heavily buffered internally already?

2)Are there simpler methods of outputting, without going through transistors?

cheers. help

fonik · Posted: Wed May 27, 2009 11:15 am Post subject:

wetterberg · Posted: Wed May 27, 2009 12:18 pm Post subject:

thanks for the help, man! I really appreciate it! Having a bit of a lightbulb-diy moment here, and loving it!

fonik · Posted: Wed May 27, 2009 1:54 pm Post subject:

wetterberg · Posted: Wed May 27, 2009 2:01 pm Post subject:

fonik · Posted: Wed May 27, 2009 2:46 pm Post subject:

wetterberg · Posted: Wed May 27, 2009 2:56 pm Post subject:

ignatzthemouse · Posted: Wed Jun 03, 2009 6:23 am Post subject:

My limited experience means I'd prefer to just build from your schematic fonik but I have a load of hcf4040be ics knocking about so I'm using one of those to build a clock divider instead.

I hadn't intended to use transistors on the outputs and wasn't sure about doing so for the inputs. It seems to cope with clock from the A190 but isn't 100% reliable so I should experiment a little with that.

I mainly intended to use it to drive a 4017 based sequencer so I suppose I could use a very stripped down version.

I'll end up just building a prototype and seeing how it behaves but if anyone has suggestions of any kind id like to hear them.

fonik · Posted: Wed Jun 03, 2009 8:00 am Post subject:

i would say go with the transistors on the input stage. would be some protection for CMOS as well.
the 4040 delivers 6.8mA on the output (+15V operation). way enough to drive LEDs and the 4017:
say you run the 4040 from +15V. following ohm's law that would give you about 14.95V on a high output. with a 6.8k resistor in series to drive a LED you would suck 2.2mA (3.2mA for 4.7K).

the 4040 has 12(!) stages: /1, /2, /4, /8, /16 ... /1024 - wow. that's for real slow motion, i think.

ignatzthemouse · Posted: Wed Jun 03, 2009 8:32 am Post subject:

Interesting. I'm going to run it in a box off a 9v battery for now. I've never used anything other than batteries to power diy stuff. Just picked up a Ken Stone synthacon filter pcb which will be my first foray into the mains.

I was thinking of making one of the outputs switch able since the 4040 divides so insanely far.

I'm sure you make a good point about an input transistor protecting the ic. I've already fried one in the last week.

The pics in another thread of your diy racks are great btw and have inspired me to replace my piss poor soldering iron with something better.

talkboxert · Posted: Sat Oct 01, 2011 10:55 am Post subject:

Good to have it with the converters, i keep inverting my pulse before it goes to my doepfer divider to keep evrything on beat...

knob_alchemist · Posted: Mon Mar 05, 2012 2:41 pm Post subject:

Interesting...I'm just in search of a simple DIY Clock divider...

ersatzplanet · Posted: Mon Mar 05, 2012 3:02 pm Post subject:

Looking at the schematic it seems pretty easy to put a header (3 pins for each inverter) that would allow bypassing the inverters so the user could choose "high on one" action or not. Even be able to choose different settings for each divisions output. Then they could make an expansion board with switches to allow front panel selection. Only if you haven't laid out the board yet. People who wanted to make it without the expense of the headers could just bridge the connections with bus wire.

fonik · Posted: Mon Mar 05, 2012 3:30 pm Post subject:

everything is possible, of course! alas, then it is not that simple anymore hihi

james, i have a SMT layout in the drawer, already. the only switch is a switch to select the division of the 5th output socket, though.

maybe i should/could do a TH DIY layout with all bells and whistles?

ersatzplanet · Posted: Mon Mar 05, 2012 3:39 pm Post subject:

emdot_ambient · Posted: Mon Mar 05, 2012 4:25 pm Post subject:

Some say that less is more...but I say added features is more. Mr. Green

I always want more.