Author |
Message |
zipzap
Joined: Nov 22, 2005 Posts: 559 Location: germany
Audio files: 24
|
|
Back to top
|
|
|
bugbrand
Joined: Nov 27, 2005 Posts: 846 Location: Bristol, UK
Audio files: 1
G2 patch files: 1
|
|
Back to top
|
|
|
piedwagtail
Joined: Apr 15, 2006 Posts: 297 Location: shoreditch
Audio files: 3
|
Posted: Mon Sep 04, 2006 4:57 pm Post subject:
|
|
|
There's a site somewhere( or was) where a gentleman dismissed that particular example as mere folly...
He did provide a better alternative given his critic.
But be gadds if i can refind it.
Why not make the Arp 1601 quantizer?It is the industry standard,if not the only true commercial example.Can't be too expensive given the time of manufacture.
I have the service manual pdf. somewhere in the vaults....if you need?
I've given up on hard logic and have moved to soft logic with a hacked P.Rees Little MCV midi-cv converter.
Robert |
|
Back to top
|
|
|
piedwagtail
Joined: Apr 15, 2006 Posts: 297 Location: shoreditch
Audio files: 3
|
Posted: Mon Sep 04, 2006 5:15 pm Post subject:
|
|
|
If they're blacked out they should enlarge to be readable,some sort of gif dilemma i'm unsolved on.......
Robert
Description: |
|
Filesize: |
33.9 KB |
Viewed: |
625 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
79.33 KB |
Viewed: |
714 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
62.26 KB |
Viewed: |
600 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
zipzap
Joined: Nov 22, 2005 Posts: 559 Location: germany
Audio files: 24
|
Posted: Tue Sep 05, 2006 2:50 am Post subject:
|
|
|
Thank you! I had actually thought about the method described, but if you want more than 2 octaves it gets quite unefficient.
Someone in a german electronics forum suggested using an ADC0808 8Bit converter chip. costs 80cents.
What do you think about that?
Another solution for generating a stairvoltage would be to use a Binary counter like a 74HC4040 with resistors on the outputs: Q5 R, Q4 2R, Q3 4R, Q2 8R, Q1 16R und Q0 32R.
Do you think it´s a good solution?
I must say i was never into that dac stuff. There are so many types...
I will also need a midi cv. I found this, it really fits my needs, but those DAC0831s are nowhere to be found where i get my parts.
Propably easy to substitute? _________________ http://www.myspace.com/lorolocoacousticpop
http://www.myspace.com/petrolvendor
music and transcribed jazz basslines |
|
Back to top
|
|
|
Uncle Krunkus
Moderator
Joined: Jul 11, 2005 Posts: 4761 Location: Sydney, Australia
Audio files: 52
G2 patch files: 1
|
Posted: Tue Sep 05, 2006 4:18 am Post subject:
|
|
|
Those resistors would need to be spot on to make it work properly. For example, if you go up to 32R, then every resistor needs to have a tolerance of at most 1/32*100 or 3%, just to keep the notes in the right order! To get them barely in tune, you'd need 20 times that accuracy, or .15% tolerance. (and that is for every value)
It's definately fun playing around with a home made DAC though. I've got a basic schem & stripboard layout if you want it. _________________ What makes a space ours, is what we put there, and what we do there. |
|
Back to top
|
|
|
toppobrillo
Joined: Dec 10, 2005 Posts: 766 Location: oakland, ca
G2 patch files: 1
|
Posted: Tue Sep 05, 2006 11:32 am Post subject:
|
|
|
hey i suggest looking at the oberheim mini-sequencer schematic, it does use an r-2-r ladder with a counter, but, of course, you can sub any DAC for it. an 8bit is supercheap nowadays, not like back then..
this is the one i am planning to build, it's very simple. the polyfusion one is simple too.. i always forget where i get what schematic??? |
|
Back to top
|
|
|
piedwagtail
Joined: Apr 15, 2006 Posts: 297 Location: shoreditch
Audio files: 3
|
|
Back to top
|
|
|
bigtex
Joined: Mar 30, 2006 Posts: 323 Location: Cupertino, California
|
Posted: Tue Sep 05, 2006 1:36 pm Post subject:
|
|
|
Some R/2R stuff came up in my thread here as well:
http://electro-music.com/forum/topic-12545.html
I'd had a notion of using a microcontroller to make a MIDI to CV converter, and making the CV outputs as cheaply as possible. Apparently Semiconwell and Vishay make some very very high precision resistor ladders in many "bit" depths for just such a use. Cheap, too. You'll still need a precision reference voltage, but the dividing will be done for you with no DAC necessary (though, admittedly, DACs are getting pretty cheap these days, and you can sample a parallel DAC for free even...)
Another idea I had once would be to have a quantizer with individual note and octave detection. So the octave output (in one volt increments) would offset the reference input for note detection. You could get a 12-tone note scale output in whatever octave you want and even process the note and octave outputs separately to, say, modulate offsets on them differently.
Maybe that didn't make a lot of sense. My brain's a little foggy right now. I'm getting overa cold....
But, anyway, my idea would allow a lot more flexibility with only a little more complexity. It would integrate very well with a sequencer, to maybe have one set of sliders for octaves and another set of sliders for notes... something like that... But since you're looking for the cheap and easy route... this is probably a project for a later date...
I really need to finalize my sequencer idea so I can make the thing... |
|
Back to top
|
|
|
toppobrillo
Joined: Dec 10, 2005 Posts: 766 Location: oakland, ca
G2 patch files: 1
|
|
Back to top
|
|
|
piedwagtail
Joined: Apr 15, 2006 Posts: 297 Location: shoreditch
Audio files: 3
|
Posted: Tue Sep 05, 2006 2:49 pm Post subject:
|
|
|
Got a modularized SEM double going to ebay soonish....
The best part, apart from the build mechanics and compact design,is the +/- modulation offset pot on the filter cv.It's a small part of the schematic which any filter can benefit from.
Not commonly implemented....
Robert |
|
Back to top
|
|
|
toppobrillo
Joined: Dec 10, 2005 Posts: 766 Location: oakland, ca
G2 patch files: 1
|
Posted: Tue Sep 05, 2006 3:12 pm Post subject:
|
|
|
yeah that bipolar attenuator, which scott so thoughtfully included in his mutant, is something that alot of the serge and buchla modules had [and apparently even the moog vcf module of mosc's] and it is a very smart thing indeed.. especially on the vcf.. |
|
Back to top
|
|
|
zipzap
Joined: Nov 22, 2005 Posts: 559 Location: germany
Audio files: 24
|
|
Back to top
|
|
|
Uncle Krunkus
Moderator
Joined: Jul 11, 2005 Posts: 4761 Location: Sydney, Australia
Audio files: 52
G2 patch files: 1
|
Posted: Wed Sep 06, 2006 3:40 pm Post subject:
|
|
|
Remember the bit I said about "just to keep the notes in the right order"
I'm assuming you made the resistor ladder yourself?
1% of 1K is 10R, but 1% of 100K is 1K. Know what I mean? _________________ What makes a space ours, is what we put there, and what we do there. |
|
Back to top
|
|
|
piedwagtail
Joined: Apr 15, 2006 Posts: 297 Location: shoreditch
Audio files: 3
|
Posted: Wed Sep 06, 2006 5:38 pm Post subject:
|
|
|
heh,heh,that techno stairs....but i was waiting for a pulverising distorted 909 kick to...er,kick in.
Robert |
|
Back to top
|
|
|
zipzap
Joined: Nov 22, 2005 Posts: 559 Location: germany
Audio files: 24
|
Posted: Thu Sep 07, 2006 2:20 am Post subject:
|
|
|
It´s fun that homemade DAC.
i´ll add the kick.
Just a thought - how do i translate a digital audio signal that way?
I guess it would sound terrible
Maybe a discrete 12bit AD/DA and some memory. Worst sampler in the world. _________________ http://www.myspace.com/lorolocoacousticpop
http://www.myspace.com/petrolvendor
music and transcribed jazz basslines |
|
Back to top
|
|
|
bigtex
Joined: Mar 30, 2006 Posts: 323 Location: Cupertino, California
|
Posted: Thu Sep 07, 2006 2:29 am Post subject:
|
|
|
zipzap wrote: | It´s fun that homemade DAC.
i´ll add the kick.
Just a thought - how do i translate a digital audio signal that way?
I guess it would sound terrible
Maybe a discrete 12bit AD/DA and some memory. Worst sampler in the world. |
For extra sweet tone you'll need to use magnetic core memory with that hand-made AD/DA.
http://en.wikipedia.org/wiki/Magnetic_core_memory |
|
Back to top
|
|
|
zipzap
Joined: Nov 22, 2005 Posts: 559 Location: germany
Audio files: 24
|
Posted: Thu Sep 07, 2006 2:43 am Post subject:
|
|
|
Are you serious?
Know what - say you make some of the ladder resistors adjustable and clock with a fast vco - this could be a cool digital waveform processor.
With much more steps than using a sequencer.
I´ll check that out. _________________ http://www.myspace.com/lorolocoacousticpop
http://www.myspace.com/petrolvendor
music and transcribed jazz basslines |
|
Back to top
|
|
|
bigtex
Joined: Mar 30, 2006 Posts: 323 Location: Cupertino, California
|
Posted: Thu Sep 07, 2006 12:15 pm Post subject:
|
|
|
Oh, no... sorry. I should have put in a or something. I don't think anybody has made magnetic core memory since the 1950s. But maybe you'd get a cool BBD-like sound out of it and make some kind of crazy analog delay. It would be monstrous, though, and consume a LOT of power! |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Sep 08, 2006 3:54 am Post subject:
|
|
|
Hey ZipZap,
If you're using the CD4024 in your ultimate design, make sure you allow enough settling time for it to get to count. Being a ripple counter, all the bits aren't going to settle simultaneously. It's hard to find a CMOS synchronous counter that counts high enough for this app, though. I chained two CD4516s and got good results, though. Or, you could set up a latch to make sure everything is locked in place for the conversion.
http://mypeoplepc.com/members/scottnoanh/birthofasynth/id15.html
I started this design for the Stites/Pontius sequencer. It's got far too many octaves though, and it could be reduced in parts, just never got around to it. Particularly in the comparator section - I had duals, but I found that there was enough internal leakage in the comparators to effect the channel the other side of the comparator was driving. My bandaid was just to use a single comparator for each channel.
I think 2.5 to 3 octaves would be ideal. 5 octaves just makes the decision points too narrow - 5 octaves of quantization puts a lot of steps on one little pot when tuning a sequencer. It would make the decision points more 'crisp' as well.
Cheers,
Scott |
|
Back to top
|
|
|
bigtex
Joined: Mar 30, 2006 Posts: 323 Location: Cupertino, California
|
Posted: Fri Sep 08, 2006 11:06 am Post subject:
|
|
|
Scott Stites wrote: | I think 2.5 to 3 octaves would be ideal. 5 octaves just makes the decision points too narrow - 5 octaves of quantization puts a lot of steps on one little pot when tuning a sequencer. It would make the decision points more 'crisp' as well. |
Well, why not have one pot for notes (with 12 steps) and one pot for octaves (with 10 or 12 steps) on each step, and just sum the voltage outputs? That way you could get all the precision you could possibly want and still have the steps be easy to find along the pot's travel. It would basically be a matter of building two sequencers to run in parallel, and much more flexible. |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Sep 08, 2006 7:21 pm Post subject:
|
|
|
Because my lust for panel space prohibits it
I've seen sequencers that do just that, though the quantization is implemented by using switches rather than pots and an R2R ladder or a DAC.
Actually, if one were to go that route, one wouldn't necessarily have to make the second row tunable in octaves only. Having two rows that could tune over 3 or 4 octaves would combine to 6 to 8 octaves through a fairly low offset mixer (which would be required for half step/octave mixing anyway). This would be more versatile functionally, because you wouldn't be bound to two rows of pots to create one control output that spanned anything greater than one octave - the second row could always be used to form a second sequence in half steps. Not quite as easy to tune as a row of half steps and a row of octaves, but not overly difficult, either.
In fact, I ought to backtrack on my previous statement - I forgot I had the damn thing tuning 10 octaves on a single pot. I think five octaves would be fine for tuning (I was able to tune with 10 octaves, though delicately, so five should work out just nifty). If not, I'd go for four. The ARP 1601 quantized over only two octaves. What was the range of the Oberheim sequencer's quantizer? - can't imagine it was much more than that.
If you have a sequence spanning great gobs of octaves, I'd just not worry so much about quantizing. Most of the time I find myself sequencing only very rarely with sequences greater than an octave or two. Arpeggios are, I suppose, one area where it is nice.
Quantizers are nice, but they are, by definition, restrictive, too, even if one has a nice fancy one that quantizes to different modalities.
Cheerio,
Scott |
|
Back to top
|
|
|
toppobrillo
Joined: Dec 10, 2005 Posts: 766 Location: oakland, ca
G2 patch files: 1
|
Posted: Sat Sep 09, 2006 12:18 am Post subject:
|
|
|
the way i was going to do it was have a 5-way switch remove LSBs from the DAC, 256, 128, 64, 32, 16... steps. the range is just an voltage divider, but i wouldn't ask much more than 5 octaves from it really. at 5 1/3 octaves, 64 steps would be 1/12v, so 256 would be 1/48..
hmm |
|
Back to top
|
|
|
zipzap
Joined: Nov 22, 2005 Posts: 559 Location: germany
Audio files: 24
|
Posted: Sat Sep 09, 2006 2:55 am Post subject:
|
|
|
I´ll try chaining thos 4516s. The DACs are on their way from the philipines. Thank you AD.
Do i get it right, that for smaller range you are making the steps larger than 1/12v?
The output scaling has to correct that of course. If it makes it work better that´s good.
I had thought to have the quantizer count in 1/12v steps in any case.
By attenuating the input you could set the range. Sure, most often 2-3 octaves are enough, but sometimes not.
BTW, speaking of appegiators, is it possible (rather say realistic) to have the dac give out chords instead of halftones?
Or controll individual steps? Guess that leads to lots of work. _________________ http://www.myspace.com/lorolocoacousticpop
http://www.myspace.com/petrolvendor
music and transcribed jazz basslines |
|
Back to top
|
|
|
toppobrillo
Joined: Dec 10, 2005 Posts: 766 Location: oakland, ca
G2 patch files: 1
|
Posted: Sat Sep 09, 2006 3:16 am Post subject:
|
|
|
Quote: | Do i get it right, that for smaller range you are making the steps larger than 1/12v? |
yeah for lower resolution the steps would be larger... if 5.33v/64=1/12v per step, 5.33v/32=1/6v per step...
Quote: | BTW, speaking of appegiators, is it possible (rather say realistic) to have the dac give out chords instead of halftones?
Or controll individual steps? Guess that leads to lots of work. |
you could come up with some chord-logic! you could design a ROM that, when input a specific note data in 8bit binary, then the rom puts out a corresponding number which is the same interval from the input always. yeah lots of work. you could use analog multiplication, multiply the input voltage by a specific amount... hey actually i think i have seen "multiplying DACs' whats that all about? |
|
Back to top
|
|
|
|