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

[Build] Dannysound - Cali Oscillator (258 clone)
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author [Build] Dannysound - Cali Oscillator (258 clone)
Monobass




The Cali oscillator is a new and improved version of the classic Buchla 258 waveshaping linear FM oscillator with an additional pulsewidth modulated output.

Buy here - https://www.thonk.co.uk/product-category/manufacturer/dannysound/





The Cali Oscillator is Dannysound’s new and improved version of the legendary Buchla Model 258 VCO. The oscillator features the classic Sine to Square and Sine to Saw wave shaping with dedicated CV control.

The Cali Oscillator differs from previous versions of the 258 in that it has a 3-way position switch rather than 2-way. This new middle position features a bonus waveform that will shape from a pulse into a kind of triangular trapezoid shape. The shaping control is more unpredictable in this middle position – it can be thought of as an extra bonus waveform!

There are dedicated inputs for linear and logarithmic FM and the Cali also adds a Pulse output with it’s own width control and PWM CV input. The behaviour of the pulse output can be affected by the position of the waveform switch and wave shape control.

The module features Coarse and Fine frequency controls and an LFO mode, LFO range is accessed via an LED pushbutton that also acts as the LFO speed indicator – the button will light up when the waveform moves through its positive cycle.

Features:

    Sine to Saw/Square Waveform Output with CV over wave shaping
    Pulse output with voltage controlled PWM
    Linear FM with attenuator
    Logarithmic FM with attenuator
    Bonus ‘trapezoid’ output
    Coarse & Fine controls with LFO mode
    Genuine Rogan USA knobs (Hex keys included)
    All Through-hole build – no SMD components!
    Width: 12HP
    Depth: 42mm


Build Documents and schematics here - https://www.thonk.co.uk/documents/dannysound/

Buy here - https://www.thonk.co.uk/product-category/manufacturer/dannysound/

euromorcego
thumbs up

saw these at superbooth. totally looking forward to get more info!

Also: there were also drum modules based on the Radio Music ... any news on them?

I see the page is up: https://www.thonk.co.uk/shop/dannysound-cali-osc-kit/

Will there also be a pcb/panel option?

Also, the build guide seems to have 53M, maybe there is a possibility to shrink the file a bit?
Monobass
euromorcego wrote:

Also, the build guide seems to have 53M, maybe there is a possibility to shrink the file a bit?


just dropped it to 4.5mb.

We'll have panel/pcb in a few weeks. Will start a new thread on the drums soon, those aren't Dannysound.
CliffordMilk
These look and sound great. I'm thinking of building a few. How well do they track?
dannysound
The EN129 tracks very well.

The Cali oscillator has a slight dip in pitch around the middle frequencies but it's not too bad.

I've been using these oscillators in synthesisers for about 5 years now (They started life in these blue monosynths that I sold). The strange thing is, even though the Cali oscillator doesn't track as well as the EN129 I do find myself turning to the Cali for the musical lines. There's something about the timbre of the Cali that I really like, especially setting the waveshape to 30% ish, sounds more acoustic somehow.
Monobass
Build Documents and schematics here - https://www.thonk.co.uk/documents/dannysound/
jimi23
Monobass wrote:

We'll have panel/pcb in a few weeks. Will start a new thread on the drums soon, those aren't Dannysound.


Any chance of it being PCB/Panel/Rare parts?
Specifically SSM2220, CA3140, CA3080. Most people who bulk buy components would generally have everything else on hand, having a quick look at the schematic
Monobass
SSM2220 - the recommendation is to actually just use a pair of matched transistors instead, which we sell on Thonk or you can match yourself.

CA3140EZ - Easily available from Farnell, Mouser, TME, Digikey , RS, Future etc

CA3080 - We have a good supply of newly manufactured ones at Thonk which isn't going to dry up
igowen
finished building and calibrating one of these last night and it sounds wonderful. one question on the "bonus" wave shape (toggle switch in the middle position) -- the range on mine is pretty limited. it starts out as a sort of truncated sine, and then around 9 o'clock it goes into a sort of unstable glitchy wave shape and then rapidly changes to a triangle wave for the rest of the range of the pot. is this the expected functionality/range? the description in the build guide doesn't give one much to go on and i wanted to make sure i didn't make a mistake somewhere in the build.
rinta
igowen wrote:
finished building and calibrating one of these last night and it sounds wonderful. one question on the "bonus" wave shape (toggle switch in the middle position) -- the range on mine is pretty limited. it starts out as a sort of truncated sine, and then around 9 o'clock it goes into a sort of unstable glitchy wave shape and then rapidly changes to a triangle wave for the rest of the range of the pot. is this the expected functionality/range? the description in the build guide doesn't give one much to go on and i wanted to make sure i didn't make a mistake somewhere in the build.


Similar behavior here with the switch in the middle position.

I also would like to confirm if the initial waveshape is triangle, not sine like on Suptnik Oscillator which I think is also based on the half of 258. How about yours?
igowen
rinta wrote:
igowen wrote:
finished building and calibrating one of these last night and it sounds wonderful. one question on the "bonus" wave shape (toggle switch in the middle position) -- the range on mine is pretty limited. it starts out as a sort of truncated sine, and then around 9 o'clock it goes into a sort of unstable glitchy wave shape and then rapidly changes to a triangle wave for the rest of the range of the pot. is this the expected functionality/range? the description in the build guide doesn't give one much to go on and i wanted to make sure i didn't make a mistake somewhere in the build.


Similar behavior here with the switch in the middle position.

I also would like to confirm if the initial waveshape is triangle, not sine like on Suptnik Oscillator which I think is also based on the half of 258. How about yours?


mine outputs a fairly clean sine wave with the wave shape pot at 0, but i followed the calibration steps in the build guide to achieve that. did you calibrate the wave shape on yours?
rinta
igowen wrote:
rinta wrote:
igowen wrote:
finished building and calibrating one of these last night and it sounds wonderful. one question on the "bonus" wave shape (toggle switch in the middle position) -- the range on mine is pretty limited. it starts out as a sort of truncated sine, and then around 9 o'clock it goes into a sort of unstable glitchy wave shape and then rapidly changes to a triangle wave for the rest of the range of the pot. is this the expected functionality/range? the description in the build guide doesn't give one much to go on and i wanted to make sure i didn't make a mistake somewhere in the build.


Similar behavior here with the switch in the middle position.

I also would like to confirm if the initial waveshape is triangle, not sine like on Suptnik Oscillator which I think is also based on the half of 258. How about yours?


mine outputs a fairly clean sine wave with the wave shape pot at 0, but i followed the calibration steps in the build guide to achieve that. did you calibrate the wave shape on yours?


Must have overlooked something. Will look into it again, very glad to know that it can do sine. Thanks!
rinta
rinta wrote:
igowen wrote:
rinta wrote:
igowen wrote:
finished building and calibrating one of these last night and it sounds wonderful. one question on the "bonus" wave shape (toggle switch in the middle position) -- the range on mine is pretty limited. it starts out as a sort of truncated sine, and then around 9 o'clock it goes into a sort of unstable glitchy wave shape and then rapidly changes to a triangle wave for the rest of the range of the pot. is this the expected functionality/range? the description in the build guide doesn't give one much to go on and i wanted to make sure i didn't make a mistake somewhere in the build.


Similar behavior here with the switch in the middle position.

I also would like to confirm if the initial waveshape is triangle, not sine like on Suptnik Oscillator which I think is also based on the half of 258. How about yours?


mine outputs a fairly clean sine wave with the wave shape pot at 0, but i followed the calibration steps in the build guide to achieve that. did you calibrate the wave shape on yours?


Must have overlooked something. Will look into it again, very glad to know that it can do sine. Thanks!


Turned out I had a mistake when calibrating waveshape. After following instructions in the manual again, I get a nice sine! Even the way it reacts to modulation became more beautiful with correct calibration.
schege
I'm very interested in the Cali oscillator (same for the EN129). But except the Superbooth video there are no sound examples/videos to find in the internet. Would be great if Danny or someone else could give some examples. Especially I'd like to hear how the EN129 reacts on linear FM and if the pitch of the tone is stabil when FMing. Also I'd like to know how ist the tracking of these osc. How many octaves do they track approximately. Thanks!
pugix
My Thonk full kit is on order! It will be fun to compare this with the J3RK 258J clones that I built in MOTM format. I'm happy this Dannysound design is now available for Eurorack.

http://pugix.com/synth/buchla-258j-vco/
pugix
I've been looking over the schematic. A nice feature is the LED monitors on the CV inputs, pre-attenuation. You can see what the CV on the input is doing (positive excursions only). Another nice feature on this and the TZO is power supply isolation using 9V regulators for the tuning pots.

Something interesting is the normalizing of +3V to the V/OCT input. If you patch 0V to V/OCT, the pitch will drop three octaves. A similar design is seen in the TZO, but 3.8V.

Note that the schematic values for R222 and R227 are the same as labeled on the PC board. These two have been changed in the build guide.

Something a little odd I noted was the voltage divider on the pulse output: 1K/1K, which would make the pulse be quite hot at about +/-6V. On the EN129 TZO the square and pulse outs use a 1K8/1K5 divider, which makes for about a +/-5.5V output level. I'm thinking I will use 2K/1K5 on all of these, which would reduce the levels to about +/-5V. A minor detail, I know.
dannysound
pugix
Thanks very much for the nice compliments and the breakdown of the schematics. I think I forgot to mention about the power supply isolation!

The pulse wave on both the Cali and Thru zero are +/- 5v. On the Cali it's due to the TL072 not hitting the rails as it's an opamp (required for the summing stage) rather than a comparator.
It's also driven from the main output rather than a triangle wave so it's not a typical sounding pulse wave.

Cali video should be up soon, although I concentrated on demonstrating the pitch and wave shape aspects so there's no examples of the pulse output. I might see about doing another video for that in the future.
The EN129 video will follow shortly.
pugix
dannysound wrote:
pugix
The pulse wave on both the Cali and Thru zero are +/- 5v. On the Cali it's due to the TL072 not hitting the rails as it's an opamp (required for the summing stage) rather than a comparator.


IC201B is operating as a comparator with hysteresis. But I take your point about it not hitting the rails, especially with a 2K load. I'm happy to use your values. The precise amplitude of a pulse wave isn't all that important.
dannysound
pugix
I made the same calculations as you in the beginning, took me a while to figure out why I wasn't getting +/- 5V!
Monobass
Demo video

pugix
Monobass wrote:
Demo video



Thank you! Nice demo of the wave shapes from the main output. We could have listened more to the triangle wave seen with the switch in the middle position.

Please make another that demonstrates the linear FM and the Pulse output and PWM. Linear FM is my favorite feature of the J3RK version.
pugix
I have a question about R10, 220K. It sets the maximum sensitivity of the linear FM CV, as well as setting an initial current. Compare the schematic of Mark Verbos. Note that the other parts values in the Verbos schematic are identical to those used in the Cali Osc, except around the linear FM CV input.

http://pugix.com/synth/wp-content/uploads/2011/07/258mod.jpg

http://pugix.com/synth/buchla-258j-vco/

The other differences in the linear FM CV are that the Cali Osc puts a buffer (IC3C) between the CV input attenuator and the CA3140 and DC couples it. So there is a question as to why the Cali Osc linear FM is DC-coupled (atypical for linear FM).

Let's analyze this. In the Cali Osc, with no linear FM input, the node at pin 2 of the CA3140 and the collector of U1Q2Q1is biased at around -1.5V, given the voltage divider made by R9 and R10. Whereas in the Verbos schematic, there is no voltage divider, due to the capacitor, and we see just the 1.5M resistor to -15V.

So is the reason for the 220K value due to it being DC coupled? My only concern is that this doesn't compromise the depth of linear FM available. If I were to put a 1uf capacitor in series with a 39K resistor in place of R10, it would be exactly like what I have in my J3RK build.

Maybe you did this because of an interaction with the Tune (initial frequency) trimpot and the LFO feature. I'm curious.

Thanks,
dannysound
pugix
Hi there, I sent you a PM about this earlier as I was recovering from my sister's wedding and wasn't really too with it!
It's a very good question so I'll explain for anyone else interested.

The original buchla design used a discrete Lin fm input stage which required a cap to isolate the fm input from the dc offset of the discrete stage input. As this design (based on the verbos update) uses a CA3140 it meant Lin fm can operate down to dc so I thought that might be handy if you'd like to run v/oct or a slowly changing voltage into Lin fm. The inverting buffer is used so +ve input = increased frequency and vice versa.
The junction of R9 and R10 are summed at pin 2 (virtual ground) of the CA3140. You can lower R10 to whatever value you like to get the lin fm as you prefer. I set the value of 220k as that's what I prefer, I think the verbos schematic used 10k.

Thanks for highlighting this as it's definitely something people might want to experiment with.
pugix
dannysound wrote:
pugix
Hi there, I sent you a PM about this earlier as I was recovering from my sister's wedding and wasn't really too with it!
It's a very good question so I'll explain for anyone else interested.

The original buchla design used a discrete Lin fm input stage which required a cap to isolate the fm input from the dc offset of the discrete stage input. As this design (based on the verbos update) uses a CA3140 it meant Lin fm can operate down to dc so I thought that might be handy if you'd like to run v/oct or a slowly changing voltage into Lin fm. The inverting buffer is used so +ve input = increased frequency and vice versa.
The junction of R9 and R10 are summed at pin 2 (virtual ground) of the CA3140. You can lower R10 to whatever value you like to get the lin fm as you prefer. I set the value of 220k as that's what I prefer, I think the verbos schematic used 10k.

Thanks for highlighting this as it's definitely something people might want to experiment with.


Thanks very much for this clarification. I'm going to experiment with a lower value for R10.
pugix
I just finished assembling the Cali Osc. I will test and calibrate it tomorrow. I want to point out something about the panel assembly. The jacks are not supposed to sit flush against the PC board! You can verify this from the picture at the end of the build doc. The coarse frequency pot and the waveform switch are the two mechanical support points and establish the board-to-panel spacing to be 12mm. But the jacks are meant for 11mm spacing. This is not made clear in the doc, which tells you "Solder one pin on each of the four outer jack sockets then add nuts to keep panel in place." If you happened to solder the jacks along the bottom flush to the PC board and then tightened their nuts, the panel would flex, putting stress on the pot and switch. Don't do that! What you need to do is before soldering first snug the jack nuts, so the jacks are flush with the panel. The panel at this point is cantilevered out over the bottom, being affixed by the pot and switch (which should be soldered at this point). You will note a bit of flex. Measure 12mm spacing and solder a ground leg on each of the two outer jacks. That will hold the board at the proper spacing. You want the board and the panel to be parallel. Now you can finish soldering the rest of the jack lugs, the pots, and the lighted switch.

I'll be posting my full write-up soon.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group