transponse on a single note with an step sequencer...

[wackelpeter]

Hi,
recently i finished the quantizer Matthias (Fonik) adapted from one of Scott's schematics or build inspired by this... however i just build this to get an reference to debug my until now unsuccesfull attempt on the DAC-quantizer Scott posted here...
Well the Fonik quantizer worked right from start and while playing with i thought it would be a nice idea to have an sequencer where i can have a sequence playing and with a kbd or any other CV source only transpose a single or a few selected notes/step CV's...

Now i'm asking myself if there's already a design for this somewhere?

My first thought to do this would be to implement a transponse input for each step and then add this CV to the initial CV of the desired step via an simple mixer before i send those to the channel CV output... but that would be a mixer for each step and a few more IC's...
First of all, would this work or is there a smarter solution which i narrow minded simply ignored?

P.S. might want to tell, that the quantizer worked curiously a bit better with an TL072 instead of the LM393 as comparator...

[Grumble]

The way my sequencer works is: basicaly I have an analogue multiplexer with on each input a potmeter that gives a voltage between 0 and 5 volt, depending on the position of the runner.
Have this multiplexed voltage added with the "offset" and in this way you transpose the sequencer voltages. This way you only need one mixer.
After the mixer feed this voltage in a quantisizer and your'e set!

[wackelpeter]

If i understand your description correct you're adding a general offset to the steps and not to a single one or a few selected steps where the voltage of the other outputs stays the same...

All my sequencers already have a transponse input... but only for the whole CV out, not for particular steps of the sequence...

My idea was that: imagine a sequence of 4 steps/notes running, where the 2nd to 4th step should stay the same and only on the 1st the pitch should be altered... well you could do this by simply adjusting the pot of that step, but it would be hard to get the same CV or intervalls with the position of the pot, especially if you wish to get some repeating pattern... Even with your sequencer output running through a quantizer this seems to be tricky to get the right spot by turning the pot.... especially when you accidentally come into a position where the comparator of those quantizers mentioned before is at a decision point between the next step, which leads to jitter... with a keyboard you have a scaled/defined voltage and can easily access and add the desired voltage by just hitting the key and let this run as long as you hit the next key and change the pitch of that note again...

[Grumble]

ow, I misunderstood your question...
so you want some notes transposed, while others are not and in an analogue fashion...
and do you want this transposed voltage be the same for every transposed note?
what you need is a switch: https://youtu.be/6e-dKVHWrgM

[PHOBoS]

You'd actually need a (mechanical) switch for each step so you can choose which step to add a CV to. I guess a Gate sequencer would work for that
which then triggers a VC switch (like in the video) but a VCA should work as well. I don't think there will be a really easy solution and adding
something direcly to the sequencer might actually be the most simple one.

[wackelpeter]

thanks for your replies.

Well with some gated schwitched this could work... somehow...

I have one of these serge 4way switches/routers...

For a 4 note sequence i had to clock it at the same rate as the sequencer is running... connect my kbd into one of the input while the 3 others left blank or maybe need 0V because as i remember there was no real ground reference on those switches which could cause some jitter maybe...
then finally hook the output to the transponse in of my sequencer...

for 4 notes this could wor, more notes could be difficult and not to mention that maybe the timing isn't that accurate...

will try that out as soon as i have some time...

but i think that should only work with even numbers... with an 8 note sequence running the 4way switch at half the rate and feed it's out to a 2 way switch at original rate.... but when the sequence is 3 or 5 notes long that shouldn't work...

As i thought a bit more of it and i guess that was what Phobos means i have to switch the CV also in and out and not only to simply add it to the step CV as it would appear all the time on the output... should only be spot on with it's corresponding step... but wait... i think of an other resolution i should try... Ken Stones pulse dividers driving a bilateral switch came to my mind...

[Grumble]

or a counting router/switch with a reset input?

[blue hell]

Not sure how you would like the control for changing a note to work, but what could work:

You could make a sum of two sequences simply where on the 2nd sequencer most values are zero but some add a modulation amount. By controlling the loop length and clock rate of that sequencer you bring variations into the sequence from the first one.

This may not give the exact control system that you want, but it is easy at least and very modular. You might need a couple of handy logic modules to make it do what you want to, like some clock dividers which can divide by 2, 3, 4, 5 ... and probably some AND / NAND / OR etc. gates.

This is what I'm using on my soft synth often, most of my sequencers have a chain input (which is simply added to the output value) for this purpose.

[wackelpeter]

Yes Jan, this could work out somehow... I'll try something of those suggested solutions as soon as i have some time from my current work of debugging my stripboard quantizers...

At least enough dividers and logic is available here... at least 2 of each from Yusynth divider, CGS pulse divider and CGS gate sequencer which are all capable of giving out uneven divisions of a clock signal.

cheers Bastian