Author |
Message |
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Thu Sep 13, 2007 10:28 pm Post subject:
Help :( 1V/Octave troubles in sequencer design ...... |
|
|
Hey all - having read tons of useful info on Scott Stite's Klee Sequencer (WANT to build ONE! ) as well as other bits of sequencers built by others, I thought I'd toss out my request for help, if I may.
(((This has turned out to be a RATHER long message - but I've tried to describe the problem that I had with a past 1V/octave sequencer design of my own, as clearly as I could.)))
I've built a couple of sequencers over the years, with varying degrees of success, including my last one. (Dec. 2005 to Jan. 2006). The only trouble that I DID have with IT, was accuracy of 1V/octave over each step.
That sequencer (Dawn of The Woody - info. including samples on the sequencer are available on my old webpage at http://www.sdiy.org/damian as well as in the News section of my new webpage) was a 4 step sequencer with 12 position rotary switches (wound with 121R 1% resistors).
Each step had TWO 12 pos. rotaries, along with a switch to choose which of the 2 were active on that step.
Movement was forward or backward, as well as selectable from 1 to 4 steps.
Gates (8 steps) were switchable on / off as well.
I remember that with assistance from others online, I'd tried a Constant Current Source; changing trimmer resistor values at the bottom of each resistor network; changing the larger resistor at the top of the network; and on and on and on.
I finally scrapped the ENTIRE panel, as the sequencer engine also had an intermittently connected problem - that I couldn't find anywhere amoungst the wiring.
With all of this talk of sequencers though and my urge to build a sequencer again (spawned by talk of Steve Roach in another thread here), I've decided that I would like to try once AGAIN, to make a 1V/Octave sequencer, using 12 position rotary switches.
I don't have any troubles with different stepping functions in sequencers (except - I'd LOVE to learn how to build the SKIP STEP function of the Moog modular sequencer )
In case THIS may help - I had a NEW idea for a possible solution to my past problem tonight, at work:
I'll set up 4 of my old rotaries, that are still wound with 121R 1% resistors.
I'll drive EACH of them from a separate resistor from a Constant Current Source - which itself is being fed from a dedicated +5V source.
THAT should help in having a separate current source, for each of the strings, should it not? Due to the with 1% differences of each of the resistors on each string, I figure I'd have to set up a slightly different extra resistor value in the network of each of the 4 lines, to match their total voltage drops to Ground. No?
I doubt this - COULD it be, that the difference in values of the 121R 1% resistors (119.88R to 122.21R) could effect the tuning THAT much?
(In my scrapped sequencer, at best, tuning per rotary, was off by almost an entire 1/2 step, across the 12 positions of the rotary switch ...... thinking back to them thar days - I'm HOPING that I wasn't doing the math for 12 resistors around the rotary, when it was really 11. I don't think the problem could be THAT simple. )
Anyhoo ...... thank you greatly to any and everyone who may be able to help with this! Ultimately - I'm looking for a schematic for a PROPER (in case I chose wrong originally) power source, for driving a series of 12 position rotary switches, to play 1/2 note steps.
Oh sorry! I should have specified this portion at the beginning of the message! I was switching from step to step, through a CD4051 (1:8 multiplexer). Could it be different resistances on EACH of the 8 steps of the 4051, that caused different current amounts - therefore throwing off my 1V/Octave tuning, on each step??
(My hopefully solution to THAT above, is sending the Constant Current Source, through individual matching resistors, TO each rotary switch.
On a side note - the addressing possibilities of 4 rows of 4 columns of rotary switches could be WILD! )
Thanks again for any possible help. |
|
Back to top
|
|
|
Adam-V
Joined: Jan 29, 2007 Posts: 300 Location: Australia
Audio files: 1
|
Posted: Thu Sep 13, 2007 11:16 pm Post subject:
|
|
|
Just to clarify, you are essentially building a bog standard analogue sequencer but rather than using pots to divide the selected output down to the desired voltage, you are using rotary switches with fixed value resistors to give you "quantized" output voltages?
If this is the case, I wouldn't have thought the sequencer would behave any differently than it would have if you used pots.
I reckon to achieve what I "think" you are trying to achieve, you would need two rotary switches per step, one to determine the base voltage for the desired octave (1V, 2V 3V etc.) and one to determine the voltage for the desired note within said octave (0.8333V). These two separate voltages would then need to be added with a precision adder (DC Mixer) to form your sequencer output.
Does this make sense? Am I on the wrong track entirely?
Cheers,
Adam-V |
|
Back to top
|
|
|
Randaleem
Joined: May 17, 2007 Posts: 456 Location: Northern CA, USA
|
Posted: Fri Sep 14, 2007 9:07 pm Post subject:
Re: Help :( 1V/Octave troubles in sequencer design ...... |
|
|
Rykhaard wrote: | I've decided that I would like to try once AGAIN, to make a 1V/Octave sequencer, using 12 position rotary switches.
Thanks again for any possible help. |
Hi Rykhaard,
Have a look here for inspiration (and a citrcuit that I believe does what you're asking for).
http://www.wiseguysynth.com/larry/jlh-822/822.htm
Kind Regards,
Randal |
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Fri Sep 14, 2007 10:05 pm Post subject:
Re: Help :( 1V/Octave troubles in sequencer design ...... |
|
|
Randaleem wrote: | Rykhaard wrote: | I've decided that I would like to try once AGAIN, to make a 1V/Octave sequencer, using 12 position rotary switches.
Thanks again for any possible help. |
Hi Rykhaard,
Have a look here for inspiration (and a citrcuit that I believe does what you're asking for).
http://www.wiseguysynth.com/larry/jlh-822/822.htm
Kind Regards,
Randal |
Thank you greatly for the link, Randal! I just had a gander at it. Whilst I think it's a great idea, it's still a bit huge per rotary switch.
To try and clarify my long windedness in the 1st message, for everyone:
There're 2 main things that I'm looking for help / suggestions on, for designing my own sequencer:
1) Could anyone point me to a URL which will have a schematic for a proper current source, for a 1 volt per octave sequencer?
2) Could anyone point me to a URL (or tell me how) I can drive more than 1 step of a sequencer, from the unit in my first request?
The trouble that I had on my 1st and only sequencer that was 1V/octave, was the CMOS CD4051 multiplexer chip - that I used to switch the Current between 4 steps.
Each of those steps which I 'tuned' for the same over all resistance, had different pitch ranges around their rotary switches.
I'm trying to find a proper way to have 1 current source, to provide to each step of a sequencer. In each step, I'm using a 12 position rotary switch to select 1 of 12 notes within a single octave.
(I do hope that spells it out a little more clearly - I've been terrible for babbling on and on and on and .... for years. Sorry about that. )
Again - thanks for your help Randal!
(Edit) P.S. I have no troubles designing the rest of the sequencer myself. I have many different sequencer type ideas, that I would love to build. My only hook at the moment, is coming up with the proper electronics for providing me with 1V/octave control. (I'm not a fan of continuous controller pots, for Pitch, as you can't work with them quickly, in a live scenario - as per my past experiences.) |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Sat Sep 15, 2007 9:42 am Post subject:
|
|
|
If the CD4051 was giving you grief in your first design, perhaps a better variant of the same thing would take care of the problem?
I'm thinking one of the higher quality "DG" switches that Maxim and others put out. They handle bipolar voltages quite well, which is beside the point here, but mainly I think they have a much lower and uniform internal resistance than the stock CMOS CD50XX. I betcha Ian Fritz would know more about the subject....
Cheers,
Scott _________________ My Site |
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Sat Sep 15, 2007 4:17 pm Post subject:
|
|
|
Scott Stites wrote: | If the CD4051 was giving you grief in your first design, perhaps a better variant of the same thing would take care of the problem?
I'm thinking one of the higher quality "DG" switches that Maxim and others put out. They handle bipolar voltages quite well, which is beside the point here, but mainly I think they have a much lower and uniform internal resistance than the stock CMOS CD50XX. I betcha Ian Fritz would know more about the subject....
Cheers,
Scott |
Thank you greatly for the thoughts, Scott! I've read of other switches / multiplexers out there, and have wondered at them. If I'm not mistaken, Maxim at least, is still giving away samples. I can check into that one.
I'll email Ian and ask his advice as well. I should have thought of that in the first place.
Another angle that I've been pondering, is similar to one in one of the sequencer designs in Howard's article / thread, that I read earlier today - sending the current through 4 switches of a CD4016 / CD4066. (I'll have to measure their resistances). That could be another possible working solution.
I'm still trying to find out how others have done it. Scott - hadn't you mentioned having built a sequencer with a portion of it taken from the Arp Sequencer?
My 2 biggest info. quests are - a proper Constant Current Source schematic, as well as maintaining the same resistance on each channel.
Thanks for you suggestions.
(In a quivering voice....) I ..... MUST ...... BUILD ..... A ...... SEQUENCER! (shivers) LOL |
|
Back to top
|
|
|
|