Introducing the Detroit Underground DU SEQ

Cwejman, Livewire, TipTop Audio, Doepfer etc... Get your euro on!

Moderators: Kent, luketeaford, Joe.

Post Reply
kpreid
Common Wiggler
Posts: 148
Joined: Wed Aug 29, 2018 10:18 am
Location: California
Contact:

Post by kpreid » Fri Aug 23, 2019 11:02 am

The Ladik U-031 does look suitable.
Again sorry for keeping asking but why is it that the du-seq double taps when the reset trig happens at the same time as a clock pulse. And why is it our diagram fixes this?
"Double-taps" is not what I see on mine. So it is possible we're solving different problems, or your input signals are different from mine.

The problem I see is that if the reset signal comes from e.g. a clock divider or other signal that follows the main clock/gate, then the first DU-SEQ step will not happen "on the beat" but rather on the next clock pulse (gate).

This is because the reset input on the DU-SEQ is what the designers called a true asynchronous reset — it goes directly to all of the logic chips and locks them into the initial state ignoring all input. Thus, the first step of DU-SEQ's sequence occurs on the first clock rising edge that occurs after the falling edge of reset. The goal of this patch, for me, is to make sure that the clock rising edge happens after the reset falls ("at the same time as reset falls" seems to work too, but that's not guaranteed) so that the first step happens on that first clock pulse that's accompanied by a (shortened) reset pulse.

The ideal way to generate these signals would be to have a sequencing patch/module that fires a reset just before the "beat" that we want it to play the first step on, but that would require something like a step sequencer with an extra output and fine enough steps so that it can insert a reset between the notes we want to hear.

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Aug 25, 2019 3:49 am

kpreid wrote:The Ladik U-031 does look suitable.
Again sorry for keeping asking but why is it that the du-seq double taps when the reset trig happens at the same time as a clock pulse. And why is it our diagram fixes this?
"Double-taps" is not what I see on mine. So it is possible we're solving different problems, or your input signals are different from mine.

The problem I see is that if the reset signal comes from e.g. a clock divider or other signal that follows the main clock/gate, then the first DU-SEQ step will not happen "on the beat" but rather on the next clock pulse (gate).

This is because the reset input on the DU-SEQ is what the designers called a true asynchronous reset — it goes directly to all of the logic chips and locks them into the initial state ignoring all input. Thus, the first step of DU-SEQ's sequence occurs on the first clock rising edge that occurs after the falling edge of reset. The goal of this patch, for me, is to make sure that the clock rising edge happens after the reset falls ("at the same time as reset falls" seems to work too, but that's not guaranteed) so that the first step happens on that first clock pulse that's accompanied by a (shortened) reset pulse.

The ideal way to generate these signals would be to have a sequencing patch/module that fires a reset just before the "beat" that we want it to play the first step on, but that would require something like a step sequencer with an extra output and fine enough steps so that it can insert a reset between the notes we want to hear.
The problem im trying to cure is the sequencer playing step one two times on reset. Thats the same issue you are trying to cure right? If it is, i might try to post some circuit schematics of a solution ive been toying with.

kpreid
Common Wiggler
Posts: 148
Joined: Wed Aug 29, 2018 10:18 am
Location: California
Contact:

Post by kpreid » Sun Aug 25, 2019 10:50 am

andersschroder wrote:The problem im trying to cure is the sequencer playing step one two times on reset. Thats the same issue you are trying to cure right? If it is, i might try to post some circuit schematics of a solution ive been toying with.
Not sure if that's the same or not. My goal with resetting the DU-SEQ has been to get it to play a sequence that's some externally controlled number of steps long regardless of how the step-count controls are set (i.e. cut off the end). In that case, the glitch I might see is that it plays the N+1'th step "on" the reset beat and the 1st step on the second clock after reset.

I would only expect to hear the first step twice if the DU-SEQ sequence has exactly the same length as the interval between resets (and dependent on signal timing characteristics).

If that's consistent with what you're seeing, then the patch we're discussing should work, but you want to make sure to introduce a little delay into the clock signal so that it definitely doesn't get to go high at all before the reset suppresses it (so that there's no no ultrashort "glitch" pulse coming out of the gate output).

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Aug 25, 2019 11:17 am

Gotcha!!!! Same issue here. I found a bunch of appropriate easy schematics. Will post a suggestion of a circuit soon. Maybe we could tag team on getting it right together?

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Mon Aug 26, 2019 9:30 am

kpreid wrote:
andersschroder wrote:The problem im trying to cure is the sequencer playing step one two times on reset. Thats the same issue you are trying to cure right? If it is, i might try to post some circuit schematics of a solution ive been toying with.
Not sure if that's the same or not. My goal with resetting the DU-SEQ has been to get it to play a sequence that's some externally controlled number of steps long regardless of how the step-count controls are set (i.e. cut off the end). In that case, the glitch I might see is that it plays the N+1'th step "on" the reset beat and the 1st step on the second clock after reset.

I would only expect to hear the first step twice if the DU-SEQ sequence has exactly the same length as the interval between resets (and dependent on signal timing characteristics).

If that's consistent with what you're seeing, then the patch we're discussing should work, but you want to make sure to introduce a little delay into the clock signal so that it definitely doesn't get to go high at all before the reset suppresses it (so that there's no no ultrashort "glitch" pulse coming out of the gate output).

I found two diagrams that will work as 12v euro as well.

First the gate to trigger circuit (or trigger shortener).
https://www.elby-designs.com/webtek/cgs ... igger.html
Spoke to the designer, she said it would be able to send a 0.5ms trigg if you replace the 10nF cap with a 1nF.

Second, the mixer with inverting input.
https://www.elby-designs.com/webtek/cgs ... mixer.html

I have remixed these into a new schematic. Please comment!

Image

kpreid
Common Wiggler
Posts: 148
Joined: Wed Aug 29, 2018 10:18 am
Location: California
Contact:

Post by kpreid » Mon Aug 26, 2019 9:39 am

Looks reasonable.

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Mon Sep 02, 2019 4:07 am

Ill try to build this within the next few weeks. I'll let you know how it goes.

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Sep 15, 2019 10:31 am

kpreid wrote:Looks reasonable.
it doest work!! The clock comes through fine if i disconnect the trigger circuit. But when the trigger circuit is is connect nothing is happening. Can you spot the mistake??? HEEEEELP

kpreid
Common Wiggler
Posts: 148
Joined: Wed Aug 29, 2018 10:18 am
Location: California
Contact:

Post by kpreid » Sun Sep 15, 2019 11:05 am

If nothing is happening, that suggests that reset is always high. I looked harder at the schematic you posted, and the - input of IC1A in the trigger shortener is pulled up to +V so it isn't going to do anything interesting. Comparing to the original schematic, you left out the 15k to ground (that's drawn next to IC1B).

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Sep 15, 2019 1:03 pm

kpreid wrote:If nothing is happening, that suggests that reset is always high. I looked harder at the schematic you posted, and the - input of IC1A in the trigger shortener is pulled up to +V so it isn't going to do anything interesting. Comparing to the original schematic, you left out the 15k to ground (that's drawn next to IC1B).
Thank you. Do you think you Could help draw it up for me? Would help me so much. much
Last edited by andersschroder on Sun Sep 15, 2019 1:23 pm, edited 1 time in total.

kpreid
Common Wiggler
Posts: 148
Joined: Wed Aug 29, 2018 10:18 am
Location: California
Contact:

Post by kpreid » Sun Sep 15, 2019 1:22 pm

Just delete all the parts that are the same for the two halves. The resistors are connected to different rails so they need to stay, and you'll find them that way. (It is unfortunate that the way the schematic is drawn it's easy to make this mistake.)

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Sep 15, 2019 1:44 pm

Added the 15k. Still not working. Hmmmm

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Sep 15, 2019 2:13 pm

If i disconnect the trigger into inverted mix this happes:

When i run a trigger signal into trig-in then at the trig-out its audible ver very short. So it seems like that is working.

When i run a clock signal into clock-in then the clock comes through fine. If i change the layout so i run the clock into the inverted mix in, nothing comes through.

I think its the inverted mixer part thats messed up right? Clock should work inversed. Maybe its because i replaced the 50k pot with a 47k resistor compred to the original circuit?

kpreid
Common Wiggler
Posts: 148
Joined: Wed Aug 29, 2018 10:18 am
Location: California
Contact:

Post by kpreid » Sun Sep 15, 2019 3:07 pm

You'll have to troubleshoot the circuit — probe it with an oscilloscope and see what the signals are at each stage and whether they are reasonable. All I can do here is produce a bunch of wild guesses (and I am not interested in doing so).

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Oct 27, 2019 2:58 pm

...

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Oct 27, 2019 3:12 pm

kpreid wrote:You'll have to troubleshoot the circuit — probe it with an oscilloscope and see what the signals are at each stage and whether they are reasonable. All I can do here is produce a bunch of wild guesses (and I am not interested in doing so).
Ok here are some news regarding the issue where DU-SEQ plays step one twice on RESET.

I simultaneously recorded both CLOCK and RESET from my clock source Yarns, into my DC coupled interface.

I then played BOTH into my DU-SEQ. All i needed to get it to work was shorten the RESET signal so it done playing before the next clock high signal.

See screenshot. Now i just need to get myself some kind of ultra quick trigger shortener.

stay tuned, i'm not giving up on this shit.Image

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Sun Oct 27, 2019 3:19 pm

...
See attached.
Picture file

andersschroder
Learning to Wiggle
Posts: 31
Joined: Fri Mar 15, 2019 7:45 am
Location: denmark

Post by andersschroder » Tue Nov 12, 2019 9:44 am

All RESET issues sorted using Ladik's new 090 Clock/Reset Corrector!!! Get this: http://ladik.ladik.eu/?page_id=2857

Post Reply

Return to “Eurorack Modules”