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

KS-Sequencer
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author KS-Sequencer
Boerge
Hi @all,

thinking about making a KS-like sequencer I looked at the schematic at
http://www.hinton-instruments.co.uk/ems/ks.gif

In the lower left corner is a clock generator (E28), which uses two caps: 1µ and 330µ in parallel. Is this a typo? The 1µ-cap would have less capacitance than the tolerance of the 330µ (even if it were 1% caps, which is unlikely). My guess is that the 330µ in fact is a 330n, so the whole think has 1,33µF.

What do you think?
papz
This is a typo.

C1 is 1nF
C2 (C1X in the original schematics) is 330pF and needs selection on test, I don't know how, never had to replace this cap so far.

Both are polystyrene caps.



Emulating the KS operation with a modern micro-controller would be a better idea than copying the original circuit that uses obsolete TTL ICs and is a serious brain teaser.
guest
looks like there is also a mistake on that first D flip-flop. the feedback should be coming from the /Q output, as shown in the schematic papz included.
Boerge
papz wrote:
This is a typo.

C1 is 1nF
C2 (C1X in the original schematics) is 330pF and needs selection on test, I don't know how, never had to replace this cap so far.


Wow, that's a quite big mistake ;-) I had my doubts about the 330µ, which seems too big for a clock generator, 1,3nF makes much more sense.

papz wrote:

Emulating the KS operation with a modern micro-controller would be a better idea than copying the original circuit that uses obsolete TTL ICs and is a serious brain teaser.


Sure. But: Microcontrollers need programming, and my C knowledge is more than a bit rusty ;-) And soldering is fun anyway, much of the ICs are sleeping in my drawers, and the shift registers could be replaced with CMOS types smile

Thank you!
Boerge
guest wrote:
looks like there is also a mistake on that first D flip-flop. the feedback should be coming from the /Q output, as shown in the schematic papz included.


Thanks for the hint. I should have a closer look to the schematics, maybe there are more showstoppers ;-)
papz
The original schematics and detailed operation are in the Synthi AKS service manual.
https://elektrotanya.com/emsa_synthi-aks_sm.pdf/download.html
ersatzplanet
Remember that the KS is probably unlike any other sequencer you have ever played with. Its operation is not particularly compatible with most other gear. It can't really be synced with anything in the traditional way almost any other sequencer can. The memory is not note oriented but event oriented - basically the clock walks it through the memory locations (256 in case) and if something is happening, it gets recorded, if not, a blank is recorded. The Gates are recorded as a series of Highs and not as a gate on and gate off event. You set the clock to run through the memory at a speed that will last for the phrase you want to record (on the Synthi's the voltage meter would sweep to show you the recording time). Not the number of notes or bars.

This opens some possibilities though. I used to set the scan rate and play notes in faster than its "sample time" and when played back, it would have a semi-random "jazz" feel to it.
papz
In fact, the KS is a sequencer for musicians. hihi
Boerge
papz wrote:
In fact, the KS is a sequencer for musicians. hihi


Oops... I'm rather a solderer than a musician, but...
It was good enough for Pink Floyd, so it should be good enough for... erm... everyone.

greetings

Boer-"On the run"-ge ;-)
ersatzplanet
papz wrote:
In fact, the KS is a sequencer for musicians. hihi


It is more of a "record and play" type of machine. No thinking of step lengths or programming rest, or any of the things that modern, non-MIDI sequencers demand. The closest thing in the modular world to the KS Keyboard Sequencer, without the keyboard of course, is the Analogue Systems RS-450 CV Recorder Sequencer which is based on the KS but adds many of the features that were missing on the KA, or expanded ones that were there. It can record four channels in "Realtime Mode" which is the closest to the original KS operation mode (except four times more channels). It has 4GB of memory vs the 256 bytes of the KS (on a SD card and larger cards can be used). Ten independent memory banks. It has a fairly high sample rate and can record audio (low bit rates though).
Unlike the KS it also can sync to external gear.



Like the cost os a Synthi though, this module is not cheap. Over $1200 here in the USA.
Boerge
ersatzplanet wrote:
Remember that the KS is probably unlike any other sequencer you have ever played with.


I' aware of this, but exactly this makes the thing more interesting ;-) Even more as I'm building a VCS3-like synth. And, honestly, I didn't play much with other sequencers - MFOS 16 step, Baby-Sequencer and Arturia Keystep are not much IMHO ;-)
@ papz: Thank you very much for the electrotanya-link. I know this site, but I didn't expect to fint the sequencer schematic there, as my search via goo...whatever didn't find much.
papz
Yes, the AS module is the closest thing to the KS.
In fact it's closer to the Sequencer 256 / Synthi 100 sequencer, that can record and play 3 separate layers, each one with 2 CV and 1 gate informations. 256 events dispatched on 10000 steps in real time, like the KS.
My own Sequencer 256 was modified by the former owner (quite roughly but it works) and has 4 layers, so it can control 4 independent voices together.

I don't understand why no other sequencer offers this real time feature. This shouldn't be hard to implement in a software.
It's so intuitive and musical to record something, to have it instantly played back exactly as it was recorded, with note lengths, syncopations, gaps... and to be able to vary the speed while playing.
fg
To fix my ks I used the hinton schema, I found a lot of errors in it. I corrected the ones I found and redraw the schema but maybe there are some more. ? if you find others tell me I'll update it

[img]removed
ersatzplanet
papz wrote:
I don't understand why no other sequencer offers this real time feature. This shouldn't be hard to implement in a software.
It's so intuitive and musical to record something, to have it instantly played back exactly as it was recorded, with note lengths, syncopations, gaps... and to be able to vary the speed while playing.


I don't understand this either. Basically it acts like a digital recorder. Any digital audio recorder could be made to act like this if it was DC coupled and had the voltage range. Basically all it is doing is recording the voltage levels of the pitch CV and the voltage levels of the gate. The sample rate is the "resolution" and determines how fast the memory gets filled. Basically with processors and memory that we have around today, this could easily record hours of CV and gate. I would assume that The Expert Sleeper and other units like that are doing exactly this sort of thing but shooting it to a computer based DAW instead of local hardware memory or cards.

The biggest difference is that it is not tied to any sort of clock. Its overall clock can be synch'd but the overall clock has little true reference to the recorded gates/notes, they are basically just streams that share a high frequency sampling clock. That is also an advantage in abstract music - each channel recorded can have NO timing relationship with any of the others, they can be in whatever arbitrary time signature you want. The only thing they really share ia that the first channel recorded, sets the maximum length of the allowed time the other channels can record.

This type of sequencer is best thought of as a VC Tape deck. it is like you record one track and then cut the tape, put it back in the machine and then you can record the other tracks next to it with the recording time limited to the length of the first track. The playback speed can be changed (unlike the tape analogy, the pitch won't change since it is CV and not sound).

I wonder how high the "sample rate" on the AS module is. They say is can record low quality audio. I bet a small modern processor could really do this well. A Raspberry Pi has no problem recording stereo audio at CD level same rates. I wonder why somebody hasn't made a CV recorder using one of them. A stereo audio track could be the CV and Gate track in a sequencer.
Graham Hinton
fg wrote:
To fix my ks I used the hinton schema, I found a lot of errors in it. I corrected the ones I found and redraw the schema but maybe there are some more. ? if you find others tell me I'll update it


Please take a moment to review this forum's copyright policy which you are infringing

IMPORTANT: Copyright, Intellectual Property & Misuse

You had no permission to remove my name and copyright notice from my drawing or to publish it. Nor to alter my drawing so that it makes no sense and then stick my name on it.

What part of copyright don't you understand?
ashleym
Worse than calumny?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group