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

New Frac Items...
MUFF WIGGLER Forum Index -> Fractional Rack Modules  
Author New Frac Items...
J3RK
Well, I've never really been able to keep my mouth shut as far as new things I'm working on go. hihi So, here's a continuation of the last few posts in the "Is this thing on?" thread. I've just finished a full set of new panels for a full set of new modules. These are basically very similar to other circuits that I've done, but adapted for ease of building in 3U, as well as adding new Frac exclusive functionality on some. The panels will accept either Johnson banana jacks or the 3.5mm LJE0352-4R jacks.

PCBs, panels, kits, bundles and complete modules will be available through Synthcube.

The PCB work is well on its way, so I'm guessing we'll start with a few modules, and work up from there.

This image will probably knock the thread a bit out of proportion, but it won't have enough detail otherwise.

noisefor
now that's some action!
rezzn8r
The Ultraflop looks fantastic hyper
What's the Indication module do?
fluxmonkey
schweet!
reignbear
dude! so good! any idea on how deep these will be on average?
slow_riot
yes! long live frac !
Cat-A-Tonic
w00t Blacet Blacet Blacet w00t
Cat-A-Tonic
Level controls on Logic modules...
Can we assume these are analogue logic?
I mean like Max/Min, Peak/Trough...
Moog$FooL$
Cat-A-Tonic wrote:
w00t Blacet Blacet Blacet w00t


yeah..... this!!
werock
Some great looking stuff there. Get your extra racks ordered everyone!
synthcube
we're not worthy

very very excited about this development
J3RK
reignbear wrote:
dude! so good! any idea on how deep these will be on average?


Thanks!

The PCBs are all parallel mounted, so depths (in Frac terms) are not all that big. Some of the more complex modules like the VCOs will have 2 PCBs. I typically use 3/4" aluminum hex standoffs for my PCB mounting, so depths are in increments of 3/4"+component height.
J3RK
I just noticed that I added the wrong version of the MultiState filter, that was in transition between two version. (so the one in the image is missing some jacks, and has others that won't be in the final) I'll update the image shortly.

Fixed.
J3RK
rezzn8r wrote:
The Ultraflop looks fantastic hyper
What's the Indication module do?


Thanks!

The Indication module is very simple. It's just four bipolar LED drivers. It's basically just for monitoring rate and phase of the input signal. If you have the Blacet Mixer Processor it will be similar to how those LEDs are set up.

I'm working on a 20 segment VU meter that works in unipolar or bipolar modes, has the trailing peak dots, etc. but that's a while off. I had a design ready for this, but then we decided to go with a different chip.
J3RK
Cat-A-Tonic wrote:
Level controls on Logic modules...
Can we assume these are analogue logic?
I mean like Max/Min, Peak/Trough...


I assume you mean the MixOR. Yes, this is an analog OR circuit with level controls for the inputs. It's a bit like the OR function on the Buchla 281. The XOR circuit is just a diode-based XOR, so no controls there. I just used bigger LEDs since there was more panel space available. This one has the logic level unipolar output, and also a differential amplifier output to shift the signal for bipolar audio use (as a pseudo ring-mod.)
J3RK
I've taken a modular-modular approach with these. I've created PCBs that can be used in multiple contexts within the system, mixed/matched/stacked/etc. This will allow for some pretty flexible configurations.

For example the VCO core used in the VCOs is the same PCB across the range. (even for the built in LFO for the FadeX) The wave-shapers for sine, saw, dual PWM comparators, etc. are on a second board. That way, if the application only needs triangle and square, the shaper board isn't necessary. (again like the FadeX which is the crossfader board stacked with the VCO core board) The wave-shaper board could also be used in other contexts like a dual variable comparator module because of the two PWM comparators. The sawtooth shaper can also make some interesting waveshapes when used with other waveforms or phases of waveforms.

There are VCAs anywhere they can go, and I try to make them usable individually as well as in the module context. This way there will always be extra available VCAs in the system. For example the FoldOMix is two Lockhart folder stages preceded by a VCA for each, and then a dry signal mix. The VCAs are used to control the folding, but can be switched out of the main signal path to be used individually. The FadeX can be used as a single VCA when patched that way, the Quad Quadrant can as well when used with a unipolar CV, the Output module has two VCAs that can be used individually or as voltage control for the channels, the Low Pass Gate, etc. etc. So, in theory, one could "Always have enough VCAs" w00t

I'll be posting more as I wrap up the remaining PCB designs.
recompas
Exciting! The stereo pulse VCO looks like it would be right up my alley. Nice to see more Frac offerings!
J3RK
Well, this is moving faster than I thought. We've decided on the first four modules/PCB batches/panels. I've ordered a test panel, and so these will be available sometime in the next few weeks.

2U - Stereo Pulse VCO
Triangle Core VCO that tracks around 8 octaves well. It has linear FM input, reset sync, VCO/LFO modes, and bipolar LED for rate and phase indication.
It has Triangle, Sawtooth, and two PWM outputs. The reason for two is that they have independent pulse-width controls (CV and manual each). When combined with a crossfader or splicer this can make a stepped pulse. When the two pulse waves are sent to the stereo output mixer, it create a very spacious stereo effect. It can also make some interesting offset clocks when in LFO mode.

2U - FadeX Plus
This is a voltage controlled linear crossfader/panner/VCA. It also has a built in V/Oct VCO (the same core as the SPVCO above. This VCO can be used to control the fading/panning, used as a stand-alone VCO, etc. If the triangle and square are routed to the fader inputs, this becomes a voltage controlled waveform VCO much like the Buchla 258.

2U - Stereo Output
This is a Stereo Output Mixer. It has level controls and panning controls for each channel, and can be used as a little submixer, output stage, etc. It's designed to use the OPA2604 op amp, and set up to drive low or high impedance loads. This will easily drive headphones, and includes a jack for this. There are two AC coupled VCAs that can be used individually, or as voltage control for the inputs of the mixer. When used with the FadeX, and/or SPVCO, a lot of stereo effect, auto-panning, etc. are possible.

1U - Opto-Gate
This is my take on the standard resonant low pass gate. The topology is similar, but not quite the same as the Buchla. I typically use VTL5C1 vactrols which are faster, and have a wider dynamic range than the more common 5C3. This allows for audio rate modulation. Then, in order to get more of a ring/trail one can simply turn up the offset very slightly. The extra dynamic range of the 5C1 will allow this without leaking much if any signal. There are two CV inputs, one with attenuator for doing FM, since these will respond well to it. Otherwise, pretty much what you'd expect.

I'll post more details as they become available.
werock
Thanks for the extra info. Good to see the same multi function per module approach that Blacet uses being used here, as well as some fairly unique designs by the look of it.
solitaryzen
Interesting stuff. Will quarter inch outputs be provided on the stereo output module?
rezzn8r
J3RK wrote:
I've ordered a test panel, and so these will be available sometime in the next few weeks.


hyper MY ASS IS BLEEDING we're not worthy

In for a Stereo Output, and 1 (possibly 2) Opto-Gates.
This is all very exciting!
SlayerBadger!
chris_g
Wow! I might have to jump to Frac just for these modules! This is fantastic! Guinness ftw! It's peanut butter jelly time! This is fun!
Hermetech Mastering
Very cool to see!
Isaiah
J3RK
This is the first time I've been attracted to Frac.
I really don't want to build or buy Eurorack, but I like the potential for shallow cases.
These could form the basis of a very gig-friendly banana system to complement my planned 4U CGS/Serge stuff.

Do the PCBs (with the exception of the Dual VCO and Ultraflop) have standardised dimensions and mounting hole placements?
Could I, for example, mount two Opto Gates behind a 2U panel?
What about stacking a FadeX and an Opto Gate?
Thanks in advance!
thresholdpeople
These look fantastic.

I'm interested in making a banana version. Are there any normals that will require the panels to be modified for a switch?
J3RK
solitaryzen wrote:
Interesting stuff. Will quarter inch outputs be provided on the stereo output module?


Yes, the Left, Right, and Phones outs are 1/4". That's not fully reflected in the image above. (actually I just forgot to make the L/R ones the right size... d'oh! hihi ) It's fixed now in the actual panel file.
J3RK
Isaiah wrote:
J3RK
This is the first time I've been attracted to Frac.
I really don't want to build or buy Eurorack, but I like the potential for shallow cases.
These could form the basis of a very gig-friendly banana system to complement my planned 4U CGS/Serge stuff.

Do the PCBs (with the exception of the Dual VCO and Ultraflop) have standardised dimensions and mounting hole placements?
Could I, for example, mount two Opto Gates behind a 2U panel?
What about stacking a FadeX and an Opto Gate?
Thanks in advance!


Yes, the PCBs are all X" x 3". (1x3, 2x3, 3x3) The mounting holes are the same scheme as my 4U boards, and CGS. For a 3" board, they are 2.7" center to center. (M3 sized pad/hole)

Most of the boards are designed to stack in some way. For example the Stereo Pulse VCO has the "VCO Core" PCB stacked with a "Wave Shapers and Comparators" board. The same "VCO Core" board is stacked with the FadeX board for the "FadeX Plus" module.

There will be VCA boards, and other function block boards that can be mixed and matched to build other things behind the panels.

The main exception is the UltraFlop. I'm using the 4U board for that turned sideways. It will just fit.
Isaiah
J3RK
That is an excellent approach!

I know most of these circuits will be available in 4U anyway, but keeping the CGS standard for mounting holes will be really helpful if a builder wants to stack one of your 4U PCBs with one of your Frac PCBs in either format.
A truly modular modular. Kudos! thumbs up

I remember you discussing a few VCFs you had been working in the 4U threads.
Which one is the Multi-State? Is the signal path DC?
tron23
Extra fine additions, need to sort my racks! w00t
J3RK
Isaiah wrote:
J3RK
That is an excellent approach!

I know most of these circuits will be available in 4U anyway, but keeping the CGS standard for mounting holes will be really helpful if a builder wants to stack one of your 4U PCBs with one of your Frac PCBs in either format.
A truly modular modular. Kudos! thumbs up

I remember you discussing a few VCFs you had been working in the 4U threads.
Which one is the Multi-State? Is the signal path DC?


The Multi-State is a pretty standard state variable filter. It's the first one that I've put together that I liked the sound of. I had a couple of others going, one with discrete OTAs, and then a 4 Pole low pass. I didn't really like either of these other ones, and will eventually revisit them later.

The Multi-State doesn't have any AC coupling caps in the signal path. I also came up with a simple "ping" circuit for it. It's a different method from the Serge VCFQ (as I didn't want to copy anything from that circuit.) Seems to work pretty well though.

Basically, it doesn't have any revolutionary features, but it works very well, and I needed a good filter in the lineup. I'll be adding an SSM(ish) 4 Pole eventually too, but that one's going to be completely redesigned.
Isaiah
Thanks for the information.
J3RK
Here is the first test module. Just need to order the actual PCBs now. The actual panels will be black.

w00t




dJ dAb
Looking good!
thresholdpeople
Lovely.

How deep is it with the bananas there?
dogoftears
wow, looks wonderful!
i will surely grab some of these when they are available.
BananaPlug
Nice! That top-center jack position is the one closest to knobs so it would be better to not put V/Oct there since V/Oct is likely to have a patch cord in it most of the time. Maybe swap V/Oct and Sync. Just being picky. Not a big deal.
tron23
dogoftears wrote:
wow, looks wonderful!
i will surely grab some of these when they are available.

+1 cannot have enough VCO's and this one is so damn fine!
steffensen
J3RK
That really looks amazing man! You have me totally psyched to get started with my Frac now. smile
chris_g
Wow that is gorgeous. Banana frac is such a great looking format It's peanut butter jelly time! nanners Rockin' Banana!
DGTom
nanners

opto gate is just what I need to finally fill the last hole in my "non-Blacet" case!!

applause
J3RK
Wow, thanks for all of the comments! w00t This is fun!

I'm just about finished with the last PCB layout for these first modules. Just checking them over more or less at this point.

@BananaPlug - Not a bad idea, and I don't mind swapping those two at all on the panel layout. I'll update it, and get it over to Synthcube.
numan7
really awesome work, J3RK !!

cheers
Isaiah
J3RK
The blue panels look beautiful!

Would it be possible to get a list of the PCBs and which circuits are on each please?
Also, what are the voltage ranges of your circuits in terms on input and outputs?

Trying to weigh up whether I'd like to build some Stroh Modular stuff in 4U or Frac.
It won't be for a while yet though...
Isaiah
Also, I'm really keen to hear some examples of stereo PWM, please! hyper
J3RK
The voltage ranges are +/-5V and 0-10V. So pretty much the same as Blacet. Gates/Triggers are 10(ish) volt pulses. Envelopes are 0-10V, all bipolar waveforms and CVs are -5 to +5.

The functions of these modules are very similar to the functions in the 4U set, but are not 1:1. Some of the PCBs in the 4U versions have more features, and some of them have more in these 3U versions.

Also, how some of the functions are achieved is a bit different from one to the other. The basic underlying blocks will be very similar though.

For example, the FadeX Plus can be used as a VC Waveform VCO. This is because it contains both the FadeX circuit and the VCO Core circuit. It tracks V/Oct the same as any of the other VCOs. Patching the triangle and square waves back into the fader inputs, results in a very Buchla 258(like) sound and function. On the other hand, the VCO can be used as a CV source for auto fading and panning.

None of the actual 4U PCBs are used in this 3U set. (with the exception of the UltraFlop4U sequencer PCB) The rest have all been redesigned for this modular/stacking approach. In the process I've both added and removed features based on the overall module set to make sure there are complementary functions between all the modules. (still putting a bit of fine tuning into this as well)

As far as SPVCO demos, I have a few online. They're pretty quick and rough and have a few pops and a bit of noise in them due to the 1394 chipset on the Mac Mini on my bench not liking my Echo interface much. very frustrating However, you can get the general idea. The sounds are coming straight out of the SPVCO, and going into the Stereo Output module. (via its onboard VCAs) I'll have these posted here in just a minute.

Edit:

A couple of clips using the two PWM outs in stereo:

http://home.comcast.net/~r3cogniz3r/UPO-StereoChipTune.mp3
http://home.comcast.net/~r3cogniz3r/UPO-StereoPWM.mp3


These aren't doing the stereo thing, but give an idea of some of the other items.

http://home.comcast.net/~r3cogniz3r/UPO-Sync.mp3
http://home.comcast.net/~r3cogniz3r/UPO-LinFM.mp3

These last ones require a wave splicer or possibly crossfader to do. They take both of the PWM outputs, and splice them. One rectified above 0V and one rectified below 0V. (the Splicer is already available from Synthcube) The result is a single waveform with independent control of the pulsewidth above 0 and below 0. (making a stepped wave)

http://home.comcast.net/~r3cogniz3r/UPO-RawMultiPWM.mp3
http://home.comcast.net/~r3cogniz3r/UPO-Arp.mp3
ginorobair
Really enjoyed hearing those demos. Thanks!
J3RK
Thanks!

I'll try to work the bugs out of my recording setup, and get some better ones posted in the near future. w00t
plord
Ultra Flop next please? That's my must-have from this set of suggestions.
J3RK
plord wrote:
Ultra Flop next please? That's my must-have from this set of suggestions.


The board is already tested for this one, so it'll be a fairly simple matter once these first four are ready. w00t
J3RK
Getting ready to order the PCBs now, and the final versions of the panels are ready to get made. The various forms these will take (kits, PCBs, modules, etc.) should start becoming available in a couple of weeks. (probably longer for modules and full kits though) As soon as I get the PCBs back, I will build up a final version of each, and post some demos here.

It's peanut butter jelly time!
J3RK
These are fairly similar to the ones above, and are from the 4U/5U version of the PCB. Exactly the same functionality minus the onboard wave splicer. (which wasn't used in these demos anyway) Also, I will make the Splicer module available in one Frac unit at some point anyway. Figured I'd post them though since the 3U version will be able to do the same things. It's really just two demos. One version with FX, one just raw output from the modules. (Stereo Pulse VCO straight into the Stereo Output Module)

http://home.comcast.net/~r3cogniz3r/Black5U-002-FX.mp3
http://home.comcast.net/~r3cogniz3r/Black5U-003-FX.mp3
http://home.comcast.net/~r3cogniz3r/Black5U-002-RAW.mp3
http://home.comcast.net/~r3cogniz3r/Black5U-003-RAW.mp3

More soon!
synthcube
J3RK wrote:
Well, this is moving faster than I thought. We've decided on the first four modules/PCB batches/panels. I've ordered a test panel, and so these will be available sometime in the next few weeks.

2U - Stereo Pulse VCO
Triangle Core VCO that tracks around 8 octaves well. It has linear FM input, reset sync, VCO/LFO modes, and bipolar LED for rate and phase indication.
It has Triangle, Sawtooth, and two PWM outputs. The reason for two is that they have independent pulse-width controls (CV and manual each). When combined with a crossfader or splicer this can make a stepped pulse. When the two pulse waves are sent to the stereo output mixer, it create a very spacious stereo effect. It can also make some interesting offset clocks when in LFO mode.

2U - FadeX Plus
This is a voltage controlled linear crossfader/panner/VCA. It also has a built in V/Oct VCO (the same core as the SPVCO above. This VCO can be used to control the fading/panning, used as a stand-alone VCO, etc. If the triangle and square are routed to the fader inputs, this becomes a voltage controlled waveform VCO much like the Buchla 258.

2U - Stereo Output
This is a Stereo Output Mixer. It has level controls and panning controls for each channel, and can be used as a little submixer, output stage, etc. It's designed to use the OPA2604 op amp, and set up to drive low or high impedance loads. This will easily drive headphones, and includes a jack for this. There are two AC coupled VCAs that can be used individually, or as voltage control for the inputs of the mixer. When used with the FadeX, and/or SPVCO, a lot of stereo effect, auto-panning, etc. are possible.

1U - Opto-Gate
This is my take on the standard resonant low pass gate. The topology is similar, but not quite the same as the Buchla. I typically use VTL5C1 vactrols which are faster, and have a wider dynamic range than the more common 5C3. This allows for audio rate modulation. Then, in order to get more of a ring/trail one can simply turn up the offset very slightly. The extra dynamic range of the 5C1 will allow this without leaking much if any signal. There are two CV inputs, one with attenuator for doing FM, since these will respond well to it. Otherwise, pretty much what you'd expect.

I'll post more details as they become available.


To weigh in here---
- PCB will be ordered this week
- Frac panels have been submitted for artwork review and production

Depending on timing, the panels, pcbs, bundles and kits should be available around mid-late April. We will get these loaded on the webstore so you can sign up for automatic notifications when they are stocked. We are working with Dustin on BOMs, documentation etc.

Very excited to see these new offerings in the format!!
monstrinho
Amazing stuff! This is really the first major news in frac outside of Blacet and Paia since the Wiard 1200 series.* Looking forward to these!!

*With all respect to Bugbrand, I really consider the Bug system to be a separate thing. The fact that it's in frac format is not really relevant since the modules were never available for the general public to buy.
J3RK
While researching possibilities to do a third party NoiseRing Expander, I got a bit carried away, and came up with my own take on the whole thing. It's still possible to do an expander at some point, but since I have a finished circuit for this now, I thought I might as well make a panel for it too.

BananaPlug
Woah, huge. Is there an advantage to having this particular VCO built in instead of just using any decent VCO to drive a standalone shift register module? The 6" all-in-one might be too much for people with portable systems.
I should wait for a block diagram but meanwhile I'm wondering what "SRA" and the three *D0 /*D1 switches are. Hoping the Bit1-Bit8 outputs are wire-or capable. Nice having all the bits!

Nice to see this moving along!
J3RK
Actually, the VCO is the same VCO Core PCB used in the other VCOs. So, this could easily be done without it in a small-model option. I have no problem whipping up an externally clocked version. It could use smaller LEDs too, to get them all in.

As far as those data switches, all clock and data sources pass through a multi-input OR gate. The switches just allow you to enable/disable them. So ED (External Data) RD (Recycle Data) ND (Noise+Comparator Data).

SRA and XORA just decides whether the shift register output or the XOR output gets sent to the differential amplifier to shift the signal to bipolar audio (+/-5V) output. The Shift Register will give you pitched digital noise, while the XOR will give you ring modulation. (it's a Schottky ring circuit)

As far as the bits go, those are actually large LEDs that display the state of the full register. (not jacks) I could actually make an expander to make those all available. It would require a lot of buffers, so it wouldn't fit on the current PCB. I could add a little header for that though. I could do a little 1U module that had 8 jacks, and a small PCB with all of the buffers on it.
J3RK
Also, the reason for me sticking VCOs everywhere, is that the core is a common PCB between all of them. It can be reused, and allow for plenty of modulation in a small setup. (as they can all be switched to LFO speeds) So a small two standard frack rack system like the following would have 3 VCOs without really taking a ton of room. Not that everyone will want to build exactly this, but it's the overall concept I had in mind. Something like this would be very small and very shallow. (building one for myself actually)

Edit: Oops, there shouldn't be two MixORs in there. d'oh! The last one should be the clock divider. Also the modules would be in a different order ideally.

BananaPlug
LEDs of course d'oh!
If you can get them to a header the rest could be left to DIY. Could be handy as random triggers and summing a few for additional CV outs might be nice but this is all icing on the cake.

Quote:
Also, the reason for me sticking VCOs everywhere, is that the core is a common PCB between all of them. It can be reused, and allow for plenty of modulation in a small setup.

Neat. Will stare at that for a while.
J3RK
That would be no problem at all to add them to a header. I've got a little fine tuning to do on the design anyway, so I'll add that in. I'll also come up with an externally clocked version. The same PCB can be used between the two. All of the inputs (even the internal ones that go between the two PCBs) are buffered so there really won't need to be any changes to the PCB.
thresholdpeople
I like having it all in one.

Will it be possible to use the Comparator independently of the main functions of the modules as well? Data in, Comp CV and Comp out, right?

Is it possible to switch the position of the FM/Comp CV/Scale/Weight jacks and the switches - keeping all the jacks toward the bottom and all hands on controls at the top?

Are all the panel controls wired to MTA headers or will some be PCB mounted?
J3RK
thresholdpeople wrote:
I like having it all in one.

Will it be possible to use the Comparator independently of the main functions of the modules as well? Data in, Comp CV and Comp out, right?

Is it possible to switch the position of the FM/Comp CV/Scale/Weight jacks and the switches - keeping all the jacks toward the bottom and all hands on controls at the top?

Are all the panel controls wired to MTA headers or will some be PCB mounted?


Data in goes through a transistor comparator straight to the OR gate before the shift register. However, I do have a pad for the comparator input. I would just need to change it to allow for an SPDT switch to swap between external and the white noise source. Just need to find a spot for a jack and a switch. I thought about that a while back, and forgot to add it in.

All jacks are currently on the bottom. Only knobs, switches, and LEDs are on the top.

I just noticed that I should move all of the data select switches to one side, and the rest on the other.

w00t
thresholdpeople
Ahh those are pots, of course. d'oh!

This is all looking really great. Can't wait to start building these. Thanks J3RK!
J3RK
I just ordered test panels for this, and a couple of other items, and should have prototype PCBs for the BitBucket shortly as well. (next weekish)

I also just bought my first Frac rails since 2004! w00t It's peanut butter jelly time!
ginorobair
Keeping eyes peeled for these developments. . .

J3RK wrote:
I just ordered test panels for this, and a couple of other items, and should have prototype PCBs for the BitBucket shortly as well. (next weekish)

I also just bought my first Frac rails since 2004! w00t It's peanut butter jelly time!
J3RK
I accidentally put a white banana for the CV out on the envelope, but... hihi (will have to fix it when I get a minute) Anyway, these are the test panels, and so far things are working pretty well. I need to adjust one jack placement, and these will be ready to go. This week has been nuts, with lots of training at work, so I'll post more details soon.

w00t


tIB
Lovely to see more frac on the horizon, especially since it's bananas!
a100user
Exciting
BananaPlug
Quote:
The wave-shapers for sine, saw, dual PWM comparators, etc. are on a second board.
Will the Dual VCO (page 1) be the only one with Sine and will it be coming along soon after this first set?
J3RK
Actually, there is a sine shaper on the PCB. hihi The only reason I didn't expose it, is that it's a little tight. I went ahead and added it though now. I think it will be ok based on what I've got on the test panels. w00t
J3RK
Finished the PCB layout for the VCF last night, and am about 75% done with the BitBucket layout. w00t
BananaPlug
Looking forward to the next round of prototypes. applause
Entrainer
This is a welcome development...

It's peanut butter jelly time! nanners Rockin' Banana! It's peanut butter jelly time! nanners Rockin' Banana! It's peanut butter jelly time! nanners Rockin' Banana! It's peanut butter jelly time! nanners Rockin' Banana! It's peanut butter jelly time! nanners Rockin' Banana!
indexofmetals
banana frac is always welcome! I've mainly got Bugbrand with some spicings of Metalbox and JHaible but have room for a bit more.
ginorobair
indexofmetals wrote:
banana frac is always welcome! I've mainly got Bugbrand with some spicings of Metalbox and JHaible but have room for a bit more.


I second that motion. Rockin' Banana! Rockin' Banana! Rockin' Banana!
BananaPlug
Some Serge and Modcan here but banana frac is the most versatile. Kudos to BugBrand for planting bananas in the fertile fields of frac. There's lots of DIY and convertible minijack frac and in a pinch euro modules can be brought into play even though voltage and fit are a bit off.
nannersbananafied euro delays
fluxmonkey
if panels are made i hope they have 1/4" holes for jacks... much easier to enlarge them for bananas than shrink them for 3.5mm...
BananaPlug
Any time you want to put minis in an oversized hole you have a couple choices. If the difference is not huge a good trick is to get a bag of the tiny rubber bands used by orthodontists. Use them like o rings on the mini. They center the jack in the hole and squish down out of the way, unlike shoulder washers which add thickness.
J3RK
fluxmonkey wrote:
if panels are made i hope they have 1/4" holes for jacks... much easier to enlarge them for bananas than shrink them for 3.5mm...


Actually, the holes are .328" which is both a usable size for bananas, and this type of 3.5mm jack:

http://synthcube.com/cart/index.php?route=product/product&path=74_18&p roduct_id=152

This seemed to be a good way to get the panels to work with both jack styles.
monstrinho
J3RK wrote:
fluxmonkey wrote:
if panels are made i hope they have 1/4" holes for jacks... much easier to enlarge them for bananas than shrink them for 3.5mm...


Actually, the holes are .328" which is both a usable size for bananas, and this type of 3.5mm jack:

http://synthcube.com/cart/index.php?route=product/product&path=74_18&p roduct_id=152

This seemed to be a good way to get the panels to work with both jack styles.


I've built a couple of the other Synthcube frac kits with the holes sized this way. I tested them with both bananas and mini jacks and they work fine either way with no problems at all. SlayerBadger!
Isaiah
J3RK
Dustin, these are looking great!

A few questions about the VCF:
*Does it track 1V/Oct accurately and is it temperature compensated (or can a Tempco be installed)?
*Does the it only self-oscillate with an feedback loop patched?
*What is the frequency range of the oscillation and how stable is it?
*Is the circuit on one PCB, or spread across two?

I was thinking about all the different ways one could wire these PCBs to build a complex dual VCO.
I thought it might be an interesting variation to use a VCF instead of the modulation VCO.
One could mix between LP and BP outputs of the self-oscillating or pinged VCF to get a waveform other than a sine.
I might be wrong, but I think the Ping input could be used as a sort of Sync input.
The Resonance setting would dictate the decay time for FM of the VCO by the pinged VCF.
The VCO could have one of the VC-Fade PCBs for Buchla 258-like action, or just as a VCA for AM.

Just some thoughts I had today.
What do you think? Am I overlooking anything?
J3RK
This is all on a single 3x3 PCB. (same size and shape as the VCO core, Shaper, ADSR, etc.)

It is V/Oct, with tempco. Depending on how it's trimmed it will oscillate. I've gotten varying results on the range of oscillation depending on the type of caps I use. The range was much wider (no additional patching) when using film caps (as mentioned by Dr. SnE in another thread.) When using MLCCs it doesn't oscillate as easily, or as wide of a range. However the resonance has a more bubbly/mellow sound to it. So this part is really going to be up to anyone building it.

The ping circuit is just an RC network, set up similar to a gate to trigger converter. It's not a schmitt trigger like the Serge version. I didn't want to steal from that design. This works great for doing little ping sounds and such, but I don't think it would work as a sync in oscillation. It would just pulse it, not reset it.

I personally would still use two VCOs in a complex VCO scenario, but that's definitely an interesting idea to use the VCF. Since the outputs are all in phase, I'm not sure you'd get a dramatic difference in timbre by mixing them. (I'd have to try this to be sure though.)

Anyway cool idea. I think you might get more variety having two VCOs though with the additional waveforms available. Also, even though the VCF is V/Oct with tempco, it's not going to track as well as the VCOs do. As far as the 258-like idea, the Frac version of the FadeX includes a VCO core, and does exactly that. (not to mention being used in reverse as a modulator for panning/fading.)

w00t
Isaiah
Thanks for all the information.

Hmmm... Yes, I fear you're right!
Seems that it might be more interesting on paper than useful or predictable in reality.
Just to clarify, I thought the VCF outputs would be 90 degrees out of phase, is that not the case? You could mix the LP and BP outputs to get something other than a sine, right?
KNYST
Beautiful work! Guinness ftw!
J3RK wrote:
w00t

J3RK
Thanks!

I just finished up a couple more panel layouts, and the first 7 prototype PCBs will be here next week. The circuits are already tested and used in other PCB layouts, but I thought one more test pass on the new sizes/shapes was a good idea. It looks like the panels for the first 4 modules will be a 5 week turnaround, so PCBs/Panels should be available around then, followed by kits, and then some complete modules after that. I'll post more information on those as this all happens as I'm sure Synthcube will as well.

I should have the demo system above functional in about two weeks, and can start posting some sounds and maybe a few quick videos.

w00t
Isaiah
J3RK
Sorry, more questions!
Hopefully this is an appropriate place to ask.

•Do your Frac-format PCBs have headers for power distribution between stacked PCBs, or does each require a Frac PSU connection?
E.g. The FadeX Plus: are both PCBs connected to the PSU/busboard, or just one with internal power distribution to the second PCB?

•Is there a Sine wave output available from any of the Frac-format VCOs?

•With the CGS-format UFO, does the variable wave fade from Sine to Saw/Pulse with increasing positive voltage (like the Buchla 258) or negative voltage?
Is a full +/-5V CV input required (ignoring the manual offset) to fade from Sine to Saw/Pulse completely?
J3RK
The PCBs have combination MTA-156 (MOTM/Frac) and DIL-10 (Euro) headers like on the 4U boards. So you could run wires or pins between the DIL-10 headers, then run a single MTA-156 connector for module power. This is what I typically do with stacked boards.

If it was a case like the BitBucket with a ton of switching CMOS, it might be better to give that PCB a separate run to the PSU than the VCO board. Not necessary, but it could have a small advantage I would think.

Due to the size and shape of all of these PCBs, they will also lend themselves easily to other formats. For example, if you liked the functionality of one of these Frac versions more than one of the existing 4U boards, you could stack them length-wise instead of layered, and the mounting holes would be in the same positions as the 4U counterpart.

Sine wave is available on the UltraPulse, and the Dual VCO. The panels in the pictures and layout drawings are a little different than the ones that will be available. The blue panels above are for my own test system. The drawings on the first page of the thread are now slightly out of date. The final panels will be on the Synthcube site fairly soon. To be clear, the panels are laid out the same way as in the pics overall, using the same finishes and techniques, I've just added or moved a function or two here or there, some based on suggestions, others based on my test panels.

As far as CVs go, I use two standards. I use bipolar -5 to +5 CVs for anything that's a balanced behavior like panning, crossfading, etc. For things like envelopes, sequencers, etc. I use unipolar 0 - 10 CVs. The CVs will always be a 10V range though. The CV processor module has three "attenuverting" inputs, plus a 5V reference voltage that can be patched in to shift a bipolar signal all the way above or below DC, so you can pretty much shift them to wherever you want them. Furthermore, all logic and clocked inputs have comparators and can be triggered from anything over around 1.5V.
J3RK
Few more panels done, and test PCBs arrive tomorrow. w00t hyper

Edit: Boards are in! I think I might be able to have 2-3 modules functioning tonight. w00t

reignbear
J3RK wrote:
Few more panels done, and test PCBs arrive tomorrow. w00t hyper

Edit: Boards are in! I think I might be able to have 2-3 modules functioning tonight. w00t



❤❤❤
BugBrand
Lots & lots of super lovely!
fluxmonkey
J3RK wrote:
fluxmonkey wrote:
if panels are made i hope they have 1/4" holes for jacks... much easier to enlarge them for bananas than shrink them for 3.5mm...


Actually, the holes are .328" which is both a usable size for bananas, and this type of 3.5mm jack:

http://synthcube.com/cart/index.php?route=product/product&path=74_18&p roduct_id=152

This seemed to be a good way to get the panels to work with both jack styles.


sure, but i'd rather not be locked in to having to use those, or the o-ring hack. but that's just me, and i can always make my own panels... all good.

b
Cat-A-Tonic
fluxmonkey wrote:
J3RK wrote:
fluxmonkey wrote:
if panels are made i hope they have 1/4" holes for jacks... much easier to enlarge them for bananas than shrink them for 3.5mm...


Actually, the holes are .328" which is both a usable size for bananas, and this type of 3.5mm jack:

http://synthcube.com/cart/index.php?route=product/product&path=74_18&p roduct_id=152

This seemed to be a good way to get the panels to work with both jack styles.


sure, but i'd rather not be locked in to having to use those, or the o-ring hack. but that's just me, and i can always make my own panels... all good.

b

If it makes any difference, those particular 3.5mm jacks are pretty nice compared to a lot of other ones.
werock
I've got those jacks on a Synthcube module and they seem fine to me. My only issue with them is the knurled nuts, not sure what tool to use to tighten them properly. I'd prefer hex nuts, but it's a minor quibble.
J3RK
Thanks for the kind words!

Yes, I think if those jacks had hex-nuts they'd be perfect for 3.5mm. As they are, I still like them. I'm happy to help with ideas for adaptions for anyone that wants to do something else. Just let me know.

This is fun!
J3RK
The LPG and SPVCO have now been tested, and are working perfectly. w00t

Next up I'll be testing the Stereo Out Mixer, and the FadeX Plus. The first four modules will be more or less ready to go after that.

I've also started building and testing other ones too, so I'll have the next batch ready to go shortly after the first batch starts becoming available. I haven't built a new power supply for the test system yet, so I'm using my bench supplies. This makes it difficult to record audio, but I'll hook something up, even if it spans the whole room, hihi and get something posted in the near future.

Tracks 2, 3, and 8 in the following playlist are using identical circuits on a different PCB, if you want to get an idea of how they sound. The one difference being the LPG for Frac is a 2-pole with resonance (more like the traditional LPG) and the one in these tracks is a 1-pole with no resonance. The filter in track 8 is actually the MultiState which will be available in the second round of modules.

http://soundcloud.com/drj3rk/sets/stroh-modular-black-5u

More soon!
J3RK
I'm really enjoying this iteration of the LPG. I made a few refinements to my older designs, tweaked some values, etc. It's got a really nice sound to it now. I use VTL5C1 vactrols for it like with most of my other ones. The reason being, the dynamic range is much higher, and the response is quite a bit faster. So you can modulate it up into low audio range, which isn't possible with the slower couplers. Also, due to the extra dynamic range, you can actually just adjust the cutoff until you get a nice trail without audio leaking through. It adds quite a bit of flexibility. One can still build with VTL5C3s, but then you lose quite a bit of the higher speed modulation ability. (as is typical with most optical gates/VCAs/filters...)

I should have some time tonight to get a bit of audio recorded. This is fun!
Isaiah
J3RK
Keen to hear this system in action, Dustin!

I'd love to hear an example of a Stroh Modular 258/259-ish"Complex Oscillator" patch:

Ultra-Pulse VCO and FadeX Plus - Primary Oscillator.
Set up to fade from Sine to Saw or Sine to Square, like a Buchla 258.

Multi-State Filter - Modulation Oscillator.
Either self-oscillating, or pinged while on the verge of oscillation.
LP and BP outputs mixed in varying levels (to provide quadrature outputs, right?) and sent via a VCA (Modulation Bus) to any of the following destinations:
Primary Oscillator Linear FM.
Primary Oscillator Wave Shape (Sine to Saw, or Sine to Square).

Of course, this doesn't have the wave-folding possibilities of the 259 (yet).
But you could do some potentially interesting things using a VCF as the Modulation Oscillator that a more traditional design wouldn't allow:
Feed one of the Primary Oscillator's outputs back into the Modulation Oscillator's Signal input, perhaps via a VCA.
Or, send the Saw (Ramp might work too) or Square output of the Primary Oscillator into the Modulation Oscillator's Ping input, perhaps via an attenuator for weak or strong Pings (which might be possible since you said you aren't using a Schmitt trigger).

Just a thought.
Could be brilliant, or rubbish!
Mr. Enjin
MY ASS IS BLEEDING I've got way too much on my plate right now but I definitely need to get busy on a couple more racks o'Frac now Blacet thumbs up
J3RK
I got the Multi-State built last night. The pinging worked better than I expected. In fact, if I trim the resonance to just below full oscillation (which actually makes it sound better anyway,) it's a ton of fun.

I actually ran the gate output of my controller into the ping input, and the CV into the V/Oct, and just played the filter straight out to the console. It tracked at least 5 octaves, and was reminiscent of an electric piano. w00t Add a little audio rate FM through the second FM input, and you've got everything from bells/chimes to percussion. I think I like having it trimmed this way best, but will see if I can strike a good balance with full oscillation and clean trigger sound. I don't think it's that critical as the SPVCO has a nice clean sine, and the proposed complex VCO will have two of them as well. The filter is quite bubbly and clean this way, and the ping is delightful. I'll give it a whirl though. This is fun!

I recorded 5 demos last night, but they all had pops in them again from my stupid interface. Luckily there's a workaround, so I'll re-record some demos again tonight.

I have a couple of duplicate panels, so I may pop a couple of these modules up in the FS section soon. I have just enough PCBs to build maybe 3-4 modules, and still have a set for myself.

The full production runs with Synthcube will follow fairly soon too.

I'm going to attempt the BitBucket build this weekend. It's a big project with a lot of variables, and a lot of wiring, but if things work as expected, it's going to be a lot of fun. More soon, and hopefully non-clicky demos tonight. w00t
Isaiah
Looking forward to hearing the electric piano-like pinging!
That description seems quite different from the examples I've heard of the Serge VCFQ being pinged.

I'm curious about the NWADSR. Looks like it can cycle.
Could you shed some light on functionality please?

This might be slightly off topic:
Have these PCBs become your priority over 4U, or are they being developed in parallel?
BananaPlug
J3RK wrote:
I got the Multi-State built last night. The pinging worked better than I expected. In fact, if I trim the resonance to just below full oscillation (which actually makes it sound better anyway,) it's a ton of fun.

Great. I agree about the trimming - go for that edge. Looking forward to hearing it and seeing how the BitBucket goes.

With the filter very near resonance, try feeding in low audio rate square around 200 Hz, tuning the filter a few octaves higher and varying the duty cycle. With certain filters it works great.
J3RK
Isaiah wrote:
Looking forward to hearing the electric piano-like pinging!
That description seems quite different from the examples I've heard of the Serge VCFQ being pinged.

I'm curious about the NWADSR. Looks like it can cycle.
Could you shed some light on functionality please?

This might be slightly off topic:
Have these PCBs become your priority over 4U, or are they being developed in parallel?


It is a bit different than the Serge, and also a bit different than say running a slow sawtooth or pulse into the audio input also. One could do the same thing with a decent gate to trigger converter, but it's handy to just have an input for it, and the values I settled on for it, seem to complement the filter well.

The NWADSR (Nicholas Woolaston ADSR for those familiar with his work) is a simple, looping ADSR. I've tweaked a few things on it, added a conditioned end of envelope pulse, etc. etc. Other than looping and EoE Pulse it's your typical non-VC ADSR. I just felt I needed one, so I whipped it up after talking to Nicholas a bit. I'll be making the board layout, updated schematic, etc. available here and on EM as a condition of its use. I have a few small bugs to work out of it before I'm completely happy though. It has a touch of "surprise behavior" at certain settings, and I think I've found a way to solve it. So nothing completely ground-breaking, but it's a solid ADSR, and I've found that I've been missing these recently.

I'm also working on a fancier version of my cycling AD envelope that will have response/curve adjustments, among some other items. No idea when I'll be done with it though. The current version works great though, so I'm not in a huge hurry on this.

Also, Blacet has very nice envelopes available too, so there are plenty of options.

On the 4U side, all of the PCBs are complete. I've tested all of them, and they're just sitting there. I'm not great with doing private runs. I've got too much going on to get everything out in as timely a manner as people like, and I've also been a little frustrated on the performance (or lack thereof) on USPS' behalf lately. That basically means that I'm making all of them available via distributors. I just wait until they're ready, or enough people ask for something, and then organize a run for the shops. I can put the boards out as quickly as there is demand for them.

So, in short, yes, the Frac project is my main focus now, because there's actually some work to be done with it for me. Boards to lay out, a fun little system to build, etc. It also translates easily to Euro projects too for those that are so inclined. Also, in the case of the BitBucket that uses a fairly large board plus a supplemental board, it's laid out to work in 3U (horizontally) or 4U (vertically). Also, since all of the 3U boards are X"x3" they can be used in 4U anyway with two stacked length-wise. So various combinations of the 3U boards could go behind a 4U panel, and still match up with 4U mounting.
J3RK
BananaPlug wrote:
J3RK wrote:
I got the Multi-State built last night. The pinging worked better than I expected. In fact, if I trim the resonance to just below full oscillation (which actually makes it sound better anyway,) it's a ton of fun.

Great. I agree about the trimming - go for that edge. Looking forward to hearing it and seeing how the BitBucket goes.

With the filter very near resonance, try feeding in low audio rate square around 200 Hz, tuning the filter a few octaves higher and varying the duty cycle. With certain filters it works great.


Was actually doing something similar last night. Now I just need to build up another VCO so I don't have to manually adjust duty-cycle. hihi
J3RK
Dead Banana waah hihi

So...


Fought with my interface for 3 hours tonight, upgrade OS, drivers, tweaked latencies until the proverbial cows came home, got bored, and left again. I was able to get one good clip with no pops and clicks. I got three more with minimal glitches, but still a few.

At this point, I think I'm done with the interface. I really liked it before, but... angry very frustrating

Anyway, the first two are the same. One with FX and one without. It's the MultiState Filter being pinged and CVed. (with FM in the FM(y) parts provided by the SPVCO) I wasn't able to get the Rhodes(y) sound that I got last night, but will attempt another recording later.

Anyway, this is just the MultiState. I'm running the gate output of my Akai controller into the Ping input, and the CV out of the controller into the V/Oct input. Then out to the... ...stupid... AudioFire...

With FX because I liked the sound...
http://home.comcast.net/~r3cogniz3r/MultiStatePinged-FX.mp3

Without FX for demo purposes...
http://home.comcast.net/~r3cogniz3r/MultiStatePinged.mp3


Edit: I've removed the ones with the recording glitches. I've got a work-around, and will re-record sometime this weekend. I've also built up most of the rest of the test rack, so only the BitBucket remains. (need to order a couple of parts for it though.) So there will be more and better demos shortly.

w00t
synthcube
Quick update- panels and PCbs on order for the first four Stroh Modular Frac modules-- Res Optogate, Stereo Out, Stereo Pulse VCO and Fadex.... more news as items start to show up.
Isaiah
J3RK
Sorry, I forgot to reply!
The filter sounds great, thanks for posting some recordings.

Just wanted to share this link:
http://cgs.synth.net/modules/cgs120_rail.html
Should be useful for mounting your Frac PCBs behind a 4U panel.
I know the mounting holes allow the builder to do that anyway, but this affords a little extra support between the mounting rails.

Are you still working on a VC-Delay?
I was thinking recently how great that would be.
Two (or more!) MN3XXX (MN3009 for a short delay?) in series with an output tapped from each BBD.
1V/Oct VCO driving the BBDs. 1V/Oct 4-Pole VCF for, well, filtering
VCA for VC-feedback.
Karplus-Strong! Rockin' Banana!
J3RK
The fully VC PT delay is ready to get prototypes made, but I haven't started on the BBD. I've got a friend who's familiar with them (BBDs) more-so than myself, so I'm holding off until he has time to help with the project. I do plan to revisit the 4-pole filter sometime in the near future. I wasn't very happy with the last iteration. It sounded good within a certain range of settings, but I'd like it to sound good at all settings. I may go the SSM2164 route this time. This will eventually happen though. I'd actually like to come up with a switched cap filter too, but that's also a way out.
monstrinho
[quote=" I'd actually like to come up with a switched cap filter too, but that's also a way out.[/quote]

Just before I opened your post, I was reading this article. Synchronicity is weird sometimes! Anyway, it looks like there's probably a lot of overlap between the logic circuits you've already done and what you would need for a SCF (or the clocking section of a BBD for that matter), so you may be a lot closer to finishing it than you imagine. nanners
J3RK
Thanks for the link! I'll check it out.
J3RK
The first prototype for the BitBucket is almost complete. Just have a bunch of wiring to do. Dead Banana I'll report back after I get it wired and hooked up. w00t
Moog$FooL$
groovy Mr J3RK....... glad u r doing this. thumbs up
BananaPlug
J3RK wrote:
The first prototype for the BitBucket is almost complete... w00t
applause
J3RK
The BitBucket is working about maybe 70%. There are a few very minor circuit tweaks that need to happen, and the zener-diode based white noise source doesn't seem to want to work. So I'll play with it a bit more, and then do a redesign for a more standard open-collector transistor type if needed. The main board, VCA, and VCO core are all working great otherwise.

The build takes a long time, and there is a lot of wiring (and inter-board wiring even.)

I'm starting to think it might be better if I just did a single board with all of the logic, comparators, shift register, LED controllers, etc. and left the clocking to an external VCO. I think I could get the rest onto a single PCB, which would not only reduce the PCB space and wiring, but also the depth of the module.

I'm going to play with the overall design a bit, and see what makes the most sense.

I'll post a bit more about this as I work with it.

w00t
rezzn8r
Great news. Thanks for the update. Really exited that you are making this happen
applause
thresholdpeople
I'm planning on putting in an order for some vactrols for another project, but I'd like to grab a few VTL5C1s while I'm at it. How many does each OptoGate require?
J3RK
thresholdpeople wrote:
I'm planning on putting in an order for some vactrols for another project, but I'd like to grab a few VTL5C1s while I'm at it. How many does each OptoGate require?


Each one requires 2 vactrols. w00t
thresholdpeople
Ah awesome thanks!

Are there other component changes that need to happen by swapping VTL5C3s? I'd be curious to play around with response times, and maybe have headers on the pcbs to be able to swap vactrols til I decide on one.
J3RK
You could swap in 5C3s. You may need to tweak the input resistors in the CV circuit a bit, and the pot ranges might shift slightly.

The cool thing about the 5C1s is that with the extra dynamic range, you can actually open them up a bit (allowing for longer trailing decay) without audio just pouring through. So as you turn the offset knob, the trail will get longer and longer (when it's struck/plucked/whatever...) until it gets high enough for audio to continuously pass. So it's kind of adjustable in that way. Plus they respond a lot faster to modulation, so you can get up into audio ranges with them. One last thing is that I've found the 5C1s to be the most consistent couplers that I've worked with as far as response. I occasionally see one that's a little different, but they've been pretty uniform overall.

5C3s would be next choice, will respond a little slower, and not quite have the same dynamic range. They will still work as expected, maybe with a few small tweaks as I mentioned above.

Any other models, and things will start getting a squishy, and have less control. I tried some 5C9s, and wasn't too happy with them. The decay was super long (to the point of taking forever to close off) and not very useful. No reason not to experiment if you have alternatives on hand though. w00t
J3RK
Quick update on the first four modules.

I just received the PCBs, and will be sending them along to Synthcube shortly. I'm building up a few, and taking some pictures, notes, etc. to include in the build docs. These will be done hopefully over this next week. So they should be ready pretty soon.
thresholdpeople
J3RK wrote:

The cool thing about the 5C1s is that with the extra dynamic range, you can actually open them up a bit (allowing for longer trailing decay) without audio just pouring through. So as you turn the offset knob, the trail will get longer and longer (when it's struck/plucked/whatever...) until it gets high enough for audio to continuously pass. So it's kind of adjustable in that way.


Oh! That sounds really awesome.
ginorobair
Looking forward to seeing shots of them.

J3RK wrote:
Quick update on the first four modules.

I just received the PCBs, and will be sending them along to Synthcube shortly. I'm building up a few, and taking some pictures, notes, etc. to include in the build docs. These will be done hopefully over this next week. So they should be ready pretty soon.
cupwise
think i'll be interested in the stereo out (i have TWO 'stereo out' modules already, the one in the paia 9700, and the blue lantern stereoscopic vca in euro, and neither of them allow you to actually pan each input separately!) and the opto gate!
BananaPlug
Could you say a bit more about your experiences with that Bit Bucket prototype?
J3RK
BananaPlug wrote:
Could you say a bit more about your experiences with that Bit Bucket prototype?


I'm in a bit of a loop right now with it actually. On one had, I like the BIG approach, which kind of has a 266 thing going for it. IO for EVERYTHING, so each component part can be used individually (like the VC Slew instead of just a slew knob for internal use, or separate noise outputs, etc.)

Then I start thinking that maybe just a good core module for the digital functions and a companion module. But then I start thinking, well then you've basically got a Turing Machine and its expanders, or a NoiseRing, and then it may as well just be a NoiseRing expander again.

Then I start thinking more micro-scale, (just providing a set of smaller modules that can basically turn into all of this in a more customizable fashion.)

I've been around this loop a few times already.

The functionality is pretty much there minus a few adjustments, circuit tweaks, replacements (of things like the noise source, scaling VCA for the CV output, etc.)

It takes a clock in, it shifts bits, those bits can be routed, recycled, provided by external sources, turned into CVs, with those CVs being scaled by voltage, etc. However, it's a bit rough around the edges right now, and the build (with separate PCBs for a few items) is a total pain to build. This is something that having a few separate modules would alleviate heavily. But that approach doesn't leave quite the impression that a big, wide, lit up, cycling monster has. seriously, i just don't get it

I've even had thoughts of making it bigger woah and including more logic and control. The thought of assembling something like this though isn't very appealing. Dead Banana

Anyway, I think I just need to work on refining, maybe do a bit of condensing to get it all on one PCB, think about whether some features are really that important, etc. For example, an onboard scaling VCA for the CVs isn't entirely necessary. Any decent DC coupled linear VCA would provide the same functionality, and reduce the complexity of this module. Another example would be should I really include the full V/Oct VCO as the clock source. On one hand, it gives people an extra VCO in one's system, provides an excellent internal source, alleviates the use of system VCOs, and doesn't kind of lock one into a mindset of only using this as a logic and random voltage source, but use it as an audio source too.

There is a ton of tradeoff involved with this particular module, and it all came about from thinking about a NoiseRing expander. lol My projects don't usually start this way. Usually, it just hits me, I know what I want to do, and then I do it. I don't usually make tradeoffs, involve myself in debates with myself nuts go through a building up and paring down process, etc.

A bit more thought is definitely in order. I don't want to recreate things that exist exactly, but then some of those things (like the 1200 series NR aren't available anymore either.) I also don't really want to just add on to something that isn't available anymore either. That kind of where the "go big" thoughts came from. Anyway... hihi
rezzn8r
I already have a Noise Ring, and love it, so the idea of a straight up expander module appeals to me, but when you started to describe the Bit Bucket I started to get really excited. I love the idea of a mutant 266ish module available in Frac.
I don't mind too much, if it doesn't have onboard scaling of CV, or a full-featured VCO to act as a clock, as I have that functionality covered, but I do start to shy away when you start to speak of the micro-scale approach with a set of simpler modules.
More onboard logic would be cool, but if it gets too big you run the risk loosing buyers due to available rack space.
Good luck! I really hope this project comes to fruition.
SlayerBadger!
J3RK
rezzn8r wrote:
I already have a Noise Ring, and love it, so the idea of a straight up expander module appeals to me, but when you started to describe the Bit Bucket I started to get really excited. I love the idea of a mutant 266ish module available in Frac.
I don't mind too much, if it doesn't have onboard scaling of CV, or a full-featured VCO to act as a clock, as I have that functionality covered, but I do start to shy away when you start to speak of the micro-scale approach with a set of simpler modules.
More onboard logic would be cool, but if it gets too big you run the risk loosing buyers due to available rack space.
Good luck! I really hope this project comes to fruition.
SlayerBadger!


Well, it definitely will, as it's something that I want as well hihi It's just a question of revision, and how long it will take. I do have several ideas for making the PCB a bit more efficient, possibly even to the point of getting all current features onto one board. Just need some "Spare Time TM" to work on it some more. I'm about to wrap up the project that needs to be finished before moving back this sort of thing, so I should have more updates soon.
J3RK
One obvious way around the size thing, would be do do all of the logic (which is really 2/3 to 3/4 the space on a single CPLD. Interfacing and buffering would be simple, and then just do the VCO, VCA, and Integrator in the remaining space. Could probably interface with a higher precision DAC too that way, rather than use an R/2R network SIP. I've got the dev tools for this, but this would be my first project using them. eek! Luckily Xilinx ISE allows you to lay the logic out like a schematic, so I don't really need to do much in the way of code (my weakest point). I also happen to know an expert with programmable logic to help if things go... ...awry... Mr. Green

Actually, just typing this kind of made some things click. That really would solve every issue I've got with the large version BitBucket.

Time to install the tools on my new computer... screaming goo yo
J3RK
I'm officially going the CPLD route. I've picked the IC, found a reasonable programmer, and have the tools installed now. I also have a little help in that department. I'll be able to do a lot in much less space now, while leaving the actual analog portions analog. (not to mention learning how to do this will be useful for more projects) w00t
J3RK
Just a quick note about the first four modules. I'm ALMOST done with the build docs. I apologize for being late on this. Had a bit of an emergency to take care of over the last few days. Things are leveling out again, so I'll be back to it shortly.
Milkweg
Very encouraging to read this! I just picked up a super cheap case, power and basic modules so will be looking for some kits to flesh it out.
J3RK
Hey everyone! Sorry for my absence! I think I'm pretty much back on track now. Right after the kids started school, my wife decided to start working again, so everything went to hell for a bit. It'll be a bit of a slow start, but I'll start getting the docs that I owe on the first few projects wrapped up and posted, as well as start working on the newer ones again. I'll try to check in more for questions as well. (just finished getting back to my PMs)

More soon!
J3RK
A bit of news.

I'm taking things in a bit of a different direction due to life / time constraints, but the Bit Bucket or whatever it might finally be called is something that I really feel like I should finish.

I now have my own basic comparator/shift register/noise/serial bus/DAC/LED driver core that takes the place of the Noisering. It's slightly simplified comparatively. It's all still CMOS logic. (I didn't go to programmable logic as I was able to get the parts count down.)

Now I'm just working on peripheral functions like CV output scaling under voltage control, a way to indefinitely loop the shift register for non-changing patterns, a configurable XOR chain, etc.

I don't think I will do CV quantization on-board. There are plenty of quantizers out there that will handle this just fine.

I think I've got a good set of finished and projected functions that will fit on reasonably sized board now, so it looks like I'll be able to finish this up in the near future. The hard parts are done now.

Sorry for the lack of presence here as of late. Also, I will check on the build threads for the other modules that are available at Synthcube shortly, and make sure everyone has the information that they need.

w00t
a100user
J3RK wrote:
A bit of news.

I'm taking things in a bit of a different direction due to life / time constraints, but the Bit Bucket or whatever it might finally be called is something that I really feel like I should finish.

I now have my own basic comparator/shift register/noise/serial bus/DAC/LED driver core that takes the place of the Noisering. It's slightly simplified comparatively. It's all still CMOS logic. (I didn't go to programmable logic as I was able to get the parts count down.)

Now I'm just working on peripheral functions like CV output scaling under voltage control, a way to indefinitely loop the shift register for non-changing patterns, a configurable XOR chain, etc.

I don't think I will do CV quantization on-board. There are plenty of quantizers out there that will handle this just fine.

I think I've got a good set of finished and projected functions that will fit on reasonably sized board now, so it looks like I'll be able to finish this up in the near future. The hard parts are done now.

Sorry for the lack of presence here as of late. Also, I will check on the build threads for the other modules that are available at Synthcube shortly, and make sure everyone has the information that they need.

w00t


Dustin, can you give a breakdown of uses for us please?
J3RK
a100user wrote:
J3RK wrote:
A bit of news.

I'm taking things in a bit of a different direction due to life / time constraints, but the Bit Bucket or whatever it might finally be called is something that I really feel like I should finish.

I now have my own basic comparator/shift register/noise/serial bus/DAC/LED driver core that takes the place of the Noisering. It's slightly simplified comparatively. It's all still CMOS logic. (I didn't go to programmable logic as I was able to get the parts count down.)

Now I'm just working on peripheral functions like CV output scaling under voltage control, a way to indefinitely loop the shift register for non-changing patterns, a configurable XOR chain, etc.

I don't think I will do CV quantization on-board. There are plenty of quantizers out there that will handle this just fine.

I think I've got a good set of finished and projected functions that will fit on reasonably sized board now, so it looks like I'll be able to finish this up in the near future. The hard parts are done now.

Sorry for the lack of presence here as of late. Also, I will check on the build threads for the other modules that are available at Synthcube shortly, and make sure everyone has the information that they need.

w00t


Dustin, can you give a breakdown of uses for us please?


Its basic functionality will be similar to the Noisering, Turing Machine, Zorlon Cannon, etc. (pitched digital noise, random and looping CV patterns, random gates, etc.)

There will be a VCA to scale the CV output, so you can go from nothing to a very wide range of voltages in 256 steps. (8 bit DAC)

The XOR chain will allow you to change how data enters the shift register, which will result in interesting pulse patterns, Atari-esque tones etc. which if clocked with a V/Oct VCO will scale with the VCO musically.

I started this by trying to replicate some of the functions that the proposed Noisering expansion was supposed to do. Then decided I wanted to build my own from the ground up, so this is basically a rough approximation of an expanded Noisering. (give or take a few features)

I may add a VC Slew to it as well, as I'll have an extra OTA on hand from the VCA, so you could have smooth transitions between the CVs, or filter the noise output with it, etc.

I managed to shrink the core down by about 50% after doing some research, so I can use the extra space to add all of these other features in.

I've made a PCB with just the core functionality already, so that part is now tested. Now I'll work on the expanded functions, then combine it into a single board. The LED drivers are multi-state, and there is one for each bit, so you'll be able to see the exact state of the entire shift register at any time. (Red for low, Green for High)
a100user
Thanks for the explanations. Sounds like it would fill a functional hole in my setup.

I think the slew function is also essential - probably to much to ask if that could have CV rate smile

Cheers
J3RK
a100user wrote:
Thanks for the explanations. Sounds like it would fill a functional hole in my setup.

I think the slew function is also essential - probably to much to ask if that could have CV rate smile

Cheers


Actually, the slew is voltage controlled. It's basically an OTA based integrator. (one pole filter more or less) Depending on a few component value choices you can go from no slew, to a perfect triangle, to flat within a fairly wide range. I'll need to play with it a bit to get the best range and topology. I'm not sure currently whether the slew should come before or after the VCA that limits the CV range for example. (or maybe just make them both patch points so you can do either one)

I'll be experimenting with these portions soon. w00t
Isaiah
J3RK
Sounds interesting! Will the Bit Bucket be a 6" wide module?

Any idea when the PCBs for the previously announced Frac modules will start becoming available?
I've seen there are a few on Synthcube already.
J3RK
Isaiah wrote:
J3RK
Sounds interesting! Will the Bit Bucket be a 6" wide module?

Any idea when the PCBs for the previously announced Frac modules will start becoming available?
I've seen there are a few on Synthcube already.


I've been able to cut back the PCB size quite a bit, so it may fit into a smaller module now. It's also externally clocked and fed now, so that reduced a bit more space. Adding the other functions in will bring the PCB size up a bit, but not by TOO much. I think a 3" panel will be fine, but will think about that a bit more. I may also do a controls PCB for the first time ever, so building it would be easier. I just made some panel component footprints for pots and switches. Jacks will still be flying though because I want it to work with 3.5mm or banana.

The PCBs for the first four modules are all at Synthcube, and also the panels I believe. I know they released two of the four, but not sure on the status of the other two.

I will be doing a lot less on the DIY side of things going forward. Maybe a PCB here or there, but I'm very low on time these days. I'll be putting most of my "Synth-Time TM" into building custom modules and synths when I'm able.
Bluce
applause good job!
J3RK
Thanks!

The prototype PCB is almost done being laid out. Just seeing if I can squeeze it down a little bit more.
BananaPlug
Dustin,
I know you're busy and I don't want to be a pest but I don't seem to be able to resist bumping this thread. Any news on this?
Thanks
J3RK
I've got the simplified version built, and so far it's working quite well. I need to feed it some more sources, and make sure the comparator range is good across the knob range. It's definitely working though. I was sending standard waveforms through it, and the shit register is working, LED drivers are working, etc. I'm going to feed it some noise next, and see how the threshold feels.

I had a few ideas for routing and looping bits back into the shift register, VCA for output CV range, XOR network, etc.

I was also thinking of having "Shift" and "Shift Lock" buttons, (after a suggestion from a friend) that would invert the shift register output bits.

So, progress has definitely been made. I've got a new schematic going for the fancier version. I'm working on a few projects at once right now, but this one is definitely a focus.

More in the near future.

w00t
a100user
Sounds great Dustin

I also think all synths need a shit register screaming goo yo
J3RK
a100user wrote:
Sounds great Dustin

I also think all synths need a shift register screaming goo yo


Thanks! And I agree completely!

So a speck of news. Finally decided on the CPLDs that I'll be using going forward for various projects, ordered the programmer, and just installed the dev tools. A relative of mine also laid out a nice little breakout board that takes the SMT 44 pin IC, and puts everything on headers. This is going to allow me to do quite a bit more with this project and squish it into a small space. I'll probably do two versions of the BitBucket now. One that is CMOS, simpler, and one that uses the CPLD with more features. The CPLD version will be a little trickier to build (but not too bad).

I'll be able to fit a VCO on the CPLD version, have a longer shift register with movable taps and routes, etc.

Another thing is that I'll make all of the files available, so people can tweak the module, or make it do more, if they're so inclined.

Anyway, more on all of this soon.
revtor
Dustin, glad to hear you have settled on a CPLD and have the tools etc.. I wish my DIY chops were at that level! Looking forward to the BitBucket.

Anyone have thoughts on a "must have" module or two that would complement or be useful to mesh the BB into your average (VCO VCF VCA EG LFO etc etc) modular?

~Steve
rezzn8r
revtor wrote:


Anyone have thoughts on a "must have" module or two that would complement or be useful to mesh the BB into your average (VCO VCF VCA EG LFO etc etc) modular?

~Steve



I would say a Quantizer and a bias/offset module would go well with a BB.
revtor
Quantizer, yeah that sounds good indeed. Im all set with bias/offset.. I'm thinking a clock divider would be pretty awesome with this too, maybe VC.

Dustin, do you have any rough draft front panel layouts for the BitBucket that you wouldn't mind sharing?

~Steve
(A delay module still in the works?)
BananaPlug
I want to try Rungler inspired patches. One oscillator clocking, one oscillator being the input bit, bits from the buffer making CV for modulation.
revtor
I'm going to go right for sub oscillator duties (miss my MOTM 120) and also looking forward to Sid style arpeggiatior patterns.

Bring it on
J3RK
BananaPlug wrote:
I want to try Rungler inspired patches. One oscillator clocking, one oscillator being the input bit, bits from the buffer making CV for modulation.


I did a few patches like this, and it was pretty cool, especially with a bit of FM applied to the VCOs.
J3RK
revtor wrote:
Quantizer, yeah that sounds good indeed. Im all set with bias/offset.. I'm thinking a clock divider would be pretty awesome with this too, maybe VC.

Dustin, do you have any rough draft front panel layouts for the BitBucket that you wouldn't mind sharing?

~Steve
(A delay module still in the works?)


I haven't laid out any panels yet, as I haven't quite finalized the features, but things are working. I'm juggling a few projects right now, so things are moving a little slow, but they're moving.

I was thinking of expanding on the roundish layout (but with more controls and IO) that I used for the very simple version I put in my 4U panel. (See the BitShifter section...)

slow_riot
really nice looking panel. long live frac!
J3RK
Thanks!

The programmer for the programmable logic ICs arrived today. Just need to get some of the chips. We laid out a little breakout PCB that will contain the CPLD and programming header. This will piggy-back onto the main board, which will have all of the analog circuitry. I'm thinking this will make a good first project, since it uses a lot of simple logic functions, but a bunch of them strung together in a configurable way.

Will update soon!
rezzn8r
So
Fucking
Excited
hyper
revtor
x2!
rezzn8r
I really like the arrangement of the LEDs. It reminds me a little of the JAG.
SlayerBadger!
J3RK
This may be taking a slight turn. There was a recent thread about phase modulation and phase distortion oscillators. It kinda got me thinking about these.

So, I laid out a sawtooth core VCO, implemented a phase shift circuit, and am working on wavetable lookup right now. So I may put this, and all the BitBucket LFSR stuff into the same module. All of it is relatively simple stuff, and so far it's a very low parts count even with all these features. It think it would be fun to have all this digital stuff available in one module.

Anyway, most of the initial development is done. Just picking components, and then will lay out a prototype to test it all with sometime in the next week or two.

Kind of a "CZ Noise Ring City" sorta thing. hihi
revtor
Hmmm.. I'd think a dual CZ osc would be a module on its own.. Bit bucket is its own thing too..? But if there's some connection btwn the functionality that I'm missing, of course -you da man..
-steve
rezzn8r
I wonder if the panel might get out of hand, if you put it all in one module.
J3RK
I don't think it will be too crazy. The VCO portion is little more than a sawtooth core with a few basic VCO controls. Then there is a phase knob/CV. The lookup table is all internal. There would be a couple of outputs (digital waveform, one or two analog.)

The reason they're good together is that both can be clocked by the same core, all of the bit bucket functions are coming off of a single IC (CPLD) and the two digital outputs can be combined in interesting ways via logic functions using a few panel buttons or switches.

They don't necessarily HAVE to be combined, but there is a lot of duplicated circuitry between the two that could be consolidated into one device.

As far as the second VCO in the CZ config for sync, phase, etc. this could be any external VCO. (though really with the simplicity of the current saw core, it could be on the same PCB too without a problem.)

Anyway, I've got all of the component parts figured out, and am just working on the ROM and conversion portions of the VCO parts, and have just started working on the logic for the CPLD for the BitBuckety parts.

They could easily be mixed and matched or separated out in any proportions.
J3RK
I decided to keep these two items separate. hihi Mainly because I thought of a lot more that I want the PD oscillator to do, and since it will be a mostly analog implementation, it's going to take some PCB space.

Just placed an order for the CPLD breakout board, and am continuing to work on the logic.

In the meantime, I posted the following over in the DIY forum. If anyone is inclined, here's a schematic I whipped up last night that could be fun.

Pasting other thread here:

Couldn't sleep, so I whipped this one up. eek!

This one is an 8 bit register made up of CD4013 D-Type flip flops. It also has a 1V/Oct VCO for the clock (Triangle and Square outs are available for normalish VCO use.) There is an R/2R to create stepped voltages, that is then scaled through a VCA. There is a differential amplifier off of the both the serial output of the register, and it can be switched to the output of the LFSR-Loop. There is an external data comparator input. So data either comes in there, or is generated by the feedback loop. There is a bank of four switches to route the bits through the XORs. I tested quite a few configurations. Lot's of tones/patterns available there. (See PDF for better view)

J3RK
My CPLD breakout boards arrive later this week. Once I have those, I'll be able to start getting familiar with programmable logic properly. I've got some schematics drawn up in the dev environment to try out. I don't think it will take long to get the hang of it. We're getting closer hyper

In the meantime, I've been working on some buffering and scaling for the CPLD IO pins to interface with the outside world.

More on this soon!
J3RK
Also here's a Frac/Euro friendly Phase Modulation VCO I whipped up last night, while I'm battling with my Phase Distortion and ThruZero design. I order prototypes that will be here next week. If everything checks out, this will be available shortly. PCB is 4"x4" so the panel will need to be a bit bigger. It will have plenty of controls and IO though to utilize the space.

J3RK
While I wait for CPLD ICs to show up, I made one last CMOS version of the BitBucket thing. PCBs will be to me by next week, and I'd be happy to do a short run of them while working on the more fully featured version. It's really not much more or less than a NoiseRing, but for anyone that needs this functionality and can't get it for Frac, it may be useful.

It has a maximal linear feedback loop like the NoiseRing, but can switch to external data only, or just sampling white noise, or just using the ring. It has a bipolar output for doing digital noise, a serial output that can be used to feed other data circuits, or modify the data stream and feed it back into the input, (or use as random gates.)

There is an LED for each stage of the shift register.

There is also a CV output for random voltages.

It's externally clocked on this version, so clocking with a V/Oct VCO will give you tones in certain modes, or being fed certain types of data.

The extended version using the CPLD will have longer shift register, movable taps, configurable XORs, onboard VCO, more logic gates, etc. etc.

Anyway, I'll have this one tested next week, and will talk to the distributors about carrying a few of them. Assuming the prototype PCBs work fine, I could also send a few of those out in the meantime.

I'd put the extended version out a couple of months, so this may be useful to some.

J3RK
Got the Bit Station boards back today. I'll get one built up and tested after work today, and post some updates.
revtor
Would love a blue banana frack Bit Station!!! . .

keep up the good work!
J3RK
Thanks!

I tested it last night, and other than needing to tweak the reference voltage for the comparator slightly, it's working perfectly. As it is right now, it has fairly similar function to the Noise Ring, but it's a little more flexible in some ways, and a little more simplistic in others. A VCA to scale the CV output would be nice, as well as an XOR separate from the internal LFSR chain. It's been a lot of fun so far. Anyway, I'll figure out the right voltage for the comparator tonight (one resistor value change) and then it's pretty much ready to go. Still working on the more complex one, but it will be a little further out. I could create a stackable companion board for this that would contain XOR, VCA, V/Oct Clock, and maybe a simple slew or integrator. That might be kind of nice, and would be relatively easy to whip up. (I already have the functions laying around, so it would just be board layout work.)

It's nice being able to see the entire shift register contents. Maybe I'll upload a quick/tiny video of this in a bit. It's a bad phone vid, but will show the indication well enough.
J3RK
Ok, horrible video, but it shows a basic CV-Out patch, and the indication.

Moog$FooL$
nice.

thumbs up
J3RK


(testing the Bit Station and Four Pole VCF) My test builds are 5U, but the boards are very 3U compatible. I just have one more tweak to make on the filter, and both of these should be ready to make some more PCBs.

w00t

Here's the filter PCB. It's SMT, but was pretty easy to build by hand. I settled on Polypropylene capacitors. They seemed to have the best overall sound.

MUFF WIGGLER Forum Index -> Fractional Rack Modules  
Page 1 of 7
Powered by phpBB © phpBB Group