Posted: Sat Dec 13, 2008 11:06 am Post subject:
Buchla Source of Uncertainty 266
A few years ago I realized that it would be possible to build my own Buchla 266 Source of Uncertainty. I took the schematics that were online and laid each one page out on a separate press n peel pcb layout. When I started putting it together, I exchanged a few emails with Scott Stites who pointed out some useful mods, which I will include here.
Today I will start things off by posting the first board layout and some sound samples that were created with it. I will post the remaining boards and samples to go with them as time permits. The original layouts have a few errors, so I'm going through each one to correct them before sharing with the public.
So, here goes: the first board, which is based off this schematic:
Joined: Oct 25, 2005 Posts: 381 Location: San Francisco
Posted: Sat Dec 13, 2008 4:30 pm Post subject:
Luka wrote:
quick question, does anyone know how similar is the noise section in the SOC to the cgs digi noise? by description it sounds exactly the same
I imagine that they are fairly similar. There were only a couple of digital noise chips made.
I'm not that big a fan of the digital noise chips, fwiw. A transistor or zener based noise circuit sounds better, IMO. Or even a fancier (than the noise chips) digital pseudo-random shift register approach.
-C _________________ Chris Muir
http://www.eardrill.com <– My jobby (more than a hobby, less than a job)
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 97
Posted: Sat Dec 13, 2008 4:53 pm Post subject:
Yep - the reason I never finished my SOU is the Quantized Random Voltage - the shiny that distracted me from it was the first Klee sequencer. It was based off of the QRV, used a CD4006 and replaced the resistors with pots (using four of the outputs of the CD4006). IE, the Klee is a direct descendent of the QRV and Ken Stone's gated comparator.
And (yep again), the CGS digital noise would work in place of the IC. A noise transistor would work pretty well also. Noise is used for the noise outputs, and it also modulates a slow moving triangle, which is the sample source of the fluctuating random voltage section, which is my favorite (along with the cool S&H) part of the 266.
Best digital noise circuit I ever heard is Thomas Henry's design in the 21st Century book (available at Magic Smoke). Digital noise tends to suffer from chug-chug as the pseudo-random pattern repeats. This design by TH will chug in, what, two hours? IE, you will never hear any periodic element in the noise. _________________ My Site
The CGS noise module is better, imho. For the simple reason that the clock speed can be changed, both manually and with voltage control. The MM5837 is pretty lame, but I guess it was adopted by synth mfgrs since it was simpler to implement than having to hand select noise transistors/diodes.
Luka wrote:
quick question, does anyone know how similar is the noise section in the SOC to the cgs digi noise? by description it sounds exactly the same
im looking forward to the sample and hold and stored random voltages
Posted: Sat Dec 13, 2008 6:05 pm Post subject:
Subject description: Quantized Random Voltage Samples
Here are some mp3s that demonstrate a bit of what the 266 QRV can do. It's like a really unusual sample and hold. In two of the samples I fed in a VCO sine wave, both at audio and sub-audio rates. The third features the CGS VC noise as the input voltage.
266_1_01_Sine_in_2VCOs.mp3
Description:
Sine wave VCO, audio rate fed into the QRV CV input Clocked by LFO square wave. Each output fed into separate VCO. Mix between the two with VCO at audio rate, then at 01:20 I drop the input VCO to subaudio.
Filesize:
2.5 MB
Duration:
2:11
Viewed:
3642 Time(s)
266_1_2_2VCOs.mp3
Description:
Same patch as last one. Input VCO at subaudio rate.
Filesize:
1.79 MB
Duration:
1:34
Viewed:
3271 Time(s)
266-1_NOISE_IN_1.mp3
Description:
Same patch as before, except that the input voltage is CGS noise source. I played around a bit with attenuating the noise input and varying the manual control.
I see that futurlec has the 2n1711 general purpose NPN for .80, in case anyones looking. Last edited by ericcoleridge on Sat Dec 13, 2008 9:30 pm; edited 1 time in total
This is such a great project VTL5C3, thanks for posting it here. I'm really looking forward to building the whole module! I have a few questions if anyone here could answer:
Scott Stites wrote:
And (yep again), the CGS digital noise would work in place of the IC. A noise transistor would work pretty well also. Noise is used for the noise outputs, and it also modulates a slow moving triangle, which is the sample source of the fluctuating random voltage section, which is my favorite (along with the cool S&H) part of the 266.
The noise circuit is not part of *this* pcb though, right? I didn't see a Noise output on this project. I'm guessing it's probably part of the Fluctuating Random Voltage circuit.
On the SOU module, is the noise source sent to the QRV or the SRV circuits? Or do they just use CV inputs for input signal?
Last question, what does 'quantized' mean in the context of this module? I'm assuming it's not quantized to intervals for a volt/octave VCO, right? Or is it?
You must be excited, with your reply in triplicate!
I'm not sure that I can explain what is meant by Quantization, but this module does take the input signal and chops it up, sort of like a sample and hold... but only sort of.
I was going through my 266 dossier and came across an old email that Steve Ridley had sent me back in '99. In it he analyzed the 266 and did a great job of making it easy to understand. I'll post the part relevant to the QRV section here:
"The "Quantized Random Voltages" section is based around a fairly standard gpseudo-random sequence generator, using a 4006 and an EXOR gate. The 4006 is clocked by input pulses from the top left 4mm socket. The next bit is clever. Various taps of the 4006 are taken through 4016 switches and suitable scaling resistors into a mixing op-amp - a sort of analogue/digital filtering. There are two different mixes to give diferent noise distributions. The 4016 switches are driven by comparators from a resistor ladder and a CV, so increasing the CV or turning the pot brings in more switches, mixes in more stages and changes the distribution. Each output has a 4mm sockent and an LED."
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 97
Posted: Sun Dec 14, 2008 2:05 am Post subject:
Hey VTL - did Steve happen to discuss the 266 Stored Random Voltage function? That's the one that I still have trouble getting my head around. I breadboarded the thing, but it still mystifies me, what with that diode ring and all. I'd love to see Steve's take on that!
Joined: Oct 25, 2005 Posts: 381 Location: San Francisco
Posted: Sun Dec 14, 2008 10:27 am Post subject:
Scott Stites wrote:
I breadboarded the thing, but it still mystifies me, what with that diode ring and all.
Looking at your redrawn schemo, it looks to me like the diode ring is some sort of clamp, although what it's purpose is right there is slightly mystifying.
I would love to see some deeper analysis of this circuit. _________________ Chris Muir
http://www.eardrill.com <– My jobby (more than a hobby, less than a job)
wish i had room for leds
my panel design is choc-a-block a the moment
is your stored random voltage design similar to the verbos design?
There is always room for LEDs You just have to get the small ones, and find a little crack between your panel components. _________________ " I went through quite a few trannies til I found one I liked" - Wild Zebra
The digital noise is on the fluctuating random board. The integrator (slewer) is on the sample & hold board.
Do you use the MM5837 in your Fluct Rand Voltage PCB? Just wondering if I should start hunting one down. I see that the are some 'pulls' for sale on e-bay now. $24 for 10, not too bad.
Posted: Fri Dec 19, 2008 5:37 pm Post subject:
PCB Layout #2: Stored Random Voltages
Well, here's part two of this four part series. While cleaning up this layout I found one mistake that I have to fix on the actual module. I have sound samples of it in its current state, but I want to add the missing resistor to see if it's noticably different before I post any mp3s. Having used it, it seems similar to the QRV module in how it processes voltages. Anyway, here is the press and peel layout:
[Edit] I updated the parts layout to show which resistors to change if you plan on driving the 266 with 5V pulses. The actual Buchla design seems to call for 15V pulses, which is hotter than most other synths' output levels. It will still work with 15V pulses if necessary. [/Edit]
[Edit2]Removed a redundant 47K resistor in the CV2 rectifier[/Edit2]
266_2_PCB_Parts_revised_09112010.pdf
Description:
266 SRV PCB + Parts (Edited 9/11/2010 - Removed a redundant 47K resistor in the CV2 rectifier)
What a great distraction this is becoming. I guess I have to get up to speed on etching boards now or find somebody who's already got a setup. Many thanks for sharing this.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
Forum index » DIY Hardware and Software
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
