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

Using Silent Way with an AC-coupled audio interface
MUFF WIGGLER Forum Index -> Expert Sleepers Goto page 1, 2, 3, 4, 5  Next [all]
Author Using Silent Way with an AC-coupled audio interface
os
I just made this:


What you see there is a quarter inch TRS jack, with the wiring somewhat modified. At the other end of the cable is a 3.5mm TRS jack, but that's only because I bought off-the-peg TRS-TRS leads - it could just as well be a 3.5mm TS jack.

The modifications are:
- ring connection lifted and soldered to shield (so R & S at the other end are both S)
- 2x 1N4148 diodes, anodes soldered to plug's T & R, cathodes soldered together
- T lead soldered to diode cathodes
- 1 micro farad capacitor soldered across the lead's T & S

The other essential ingredient is a new plug-in that I've written which takes the usual Silent Way outputs and encodes them on an AC signal.

Using this and my Metric Halo 2882 interface, which is not DC coupled, I managed to calibrate an ASys RS-95 VCO over about 6 octaves, and it seems to play back rock steady.

Of course you only get positive voltages out of this arrangement, so if you need bipolar CVs you can just apply a DC shift.

The magic plug-in isn't ready for release yet but that shouldn't take long. Prospective beta testers are invited to drop me a line (you'll need to solder you're own leads though!).
bsmith
YESSSSSSS!!!!!!! This is huge news!
goiks
brilliant! this is exciting, i have a 2882 also. your continued silent way efforts make me want to give you $$!
os
The software is ready for beta testing, if anyone's interested.
JP
What does it take to beta test this for you? I might be able to put some time into it this weekend, but I'm assuming I'd need to make one of the plugs you have above.
computer controlled
Excellent! Sent you a PM.
os
JP wrote:
What does it take to beta test this for you? I might be able to put some time into it this weekend, but I'm assuming I'd need to make one of the plugs you have above.

You need
- a cable as above
- an audio interface with balanced outputs
- a PC or Mac to run Silent Way on
- the magic software (email me for the download links)

NB if you have a MOTU or other DC-coupled interface, you can run Silent Way without the magic software, so that wouldn't be such a useful test of this new technique.
jenamu6
If this works it will be a huge breakthrough.


You might get a Nobel price for Modular Hero for this.
os
I think I replied to all the PMs. If I missed any, I apologise - drop me an email.
bsmith
off to fry's to grab Stuff - any recommendations on that capacitor? 1uF good?

*edit*
Wow this is looking good! Ended up needing to go with a .47 uF capacitor due to selection at store, am getting close to 8 octaves which play solidly. Also tried a 10 uF capacitor and the pitch would still calibrate, but when playing it kind does a kind of glide thing into each new note. Don't really know what I'm doing or how this works, so using the 'throw at wall and see what sticks' method. Am having a little trouble with gate so far, need to boost it with a .bam to get it usable with my envelator (no love from the kraftzwerg yet in the gate department) - will work with more over the weekend. Clearly this approach works!!

As with all the expert sleepers tools & toys - great stuff, Os!
You're going to save a bunch of us money for more modules... It's peanut butter jelly time!
Klipspringer
Yes I'm just about to go and get the parts now. I'll use my Edirol FA-66 for the test.

I have a question: I've never used Silent Way before. Do I add the acencoder plugin to the same audio channel as the voicecontroller and I then just calibrate using the voicecontroller?

If this works I'm buying Silent Way today!
bsmith
yeah, insert the encoder after the voice controller on the same channel. I use a stereo aux track, so left is pitch and right is gate is how that defaults, and then obviously audio in to that same track so it can calibrate.
sgnelson
So, I'm trying to get in on this action as well, and I'd rather not rip apart most of my cables for this yet, so does anyone have a good recommendation for cheap cable and TRS plugs?
bsmith
Fry's had the trs plugs at a buck fifty a pop. Used some rat shack 6 foot 3.5 TS cables that were laying around here which are 4 dollars or something.
Klipspringer
I think this is working OK for me on the FA-66. I'm not sure I'm calibrating the oscillator properly as it seems to only play 1 octave before it peters out. Maybe that's correct?

Regardless, I'm able to do things with Live and the modular!

So I'll fiddle around with this tonight so more, but it's amazing and just saved me about 700 bucks.

By the way, if you fit the cap just so and trim your leads just so, you can put all the components inside a standard 1/4" TRS casing! nanners
Klipspringer
Just an update: I'm getting about 2.5v at the output. Does that seem right for this?

And here are some total n00b questions:

How do I gate this stuff? I guess I need more than one cable made up to send to the gate. Correct? If so how do I tell Silent Way what audio channel the gate signal is on? I can't figure it our from the tutorial videos.

Edit: Just found the PDF manual. Output 2 is Gate I think.

Edit edit: I figured out the gating stuff. I'll have to make an additional cable and I can gate the Borg in VCA mode. The Envelator will need more juice.

Regardless, I don't have a VCA on my rack yet, just an ASol VCO 6, ASoL MX61, Borg, Envelator, Wogglebug. So if I want to gate the ASol, I was going to try and use the Envelator, but there isn't enough signal I think to do this. Was that your observation too bsmith?

I was sort of getting something going that sounded gated with the Borg in VCA mode, but then my demo timed out and my computer screwed up and I had to eat my dinner and forgot wtf I was doing very frustrating

Edit: Solved This is fun!

So I guess I need to buy a little VCA for all this to work properly. Correct?

Edit: Yes, unless you want to use the Borg in VCA mode.

Finally. Even without a gated signal when I play a scale in Live it's not right. It's almost as if every second note sounds the same and one only hears ascended pitch every third note or so. Is this what is referred to as tracking? As in "your ASol VCO 6HP don't track good, man"?

Thanks for your patience (struggling to learn quick)!
os
I kinda lost track of your internal monologue there smile

So - the one outstanding issue is the tracking?

What value did you use for the capacitor? Can you save and post (or email) the calibration file that you get from this setup? (instructions in the User Manual).
bsmith
Poke around with your oscillator to see what calibration results you get, make sure you're just giving the input audio a solid level of unfiltered and un-whatever-elsed output. Using the square wave set at 16' range with the new cable on my digi 002 I just got 8 octaves on the kraftzwerg! And make sure the two sides of the encoder plug in are lit up on 'enable' obviously.
As far as the gate, yes am finding so far that the envelator works fine if I juice it up (with the cable I use a .47 uF capacitor on) - setting that side to unipolar in the ac>dc plugin. Based on seeing pretty different results between the .47 uF and a 10 uF (.47 showing the best results), I want to try others and see how they do. I've got a couple .22 uF's, we'll see how it goes for them later today....


***edit
Went and got some .1 uF capacitors and they work great on pitch calibration and both gates I have to poke at (envelator and the kraftzwerg's). The envelator's AD mode seems to act better with it boosted - you need to pause between note on's otherwise not sure what up with that, AR mode is fine (and fast without the need to pause deal). Anyway, enough solder fumes today, time to play with all these newfound cv outs!!!
os
I find Silent Way often calibrates best with a triangle wave, but YMMV.

I just did some tests using my modified cable, my 2882 and some more modules:
- using the cable to control a gate, on an RS-60 envelope generator. Works fine.
- using the cable to control a VCA (an RS-180) using Silent Way's envelopes. Also works fine.

So, bsmith, I think the kraftzwerg gate issue you mailed me about might be a kraftzwerg thing. If you're getting 8 octaves of pitch CV, you ought to be able to generate a +4V gate, which is enough for most modules - although not enough for some older synths. On the other hand, that's about what you get from a lot of DC-coupled MOTU interfaces, so it's not too bad for a $2 cable!
josefranco
I'm thinking of building a box with 8 channels of the circuit and jacks for I/O so I can use regular unmodified cables. Any thought or ideas for this?
os
Sounds like an excellent idea. If your DIY electronics skills extend to it, I would consider building in an amplifier/DC offset per channel too.
Klipspringer
os wrote:
I kinda lost track of your internal monologue there smile

.


I was too excited!

os wrote:


So - the one outstanding issue is the tracking?

What value did you use for the capacitor? Can you save and post (or email) the calibration file that you get from this setup? (instructions in the User Manual).


Yes that and triggering the Envelator. I'm using a 1 mirofarad 50V cap. Just like what's in the picture. The output is about 2.5V at the lead's end.. I've attached the Calibration data for a square wave from my ASol VCO 6HP.

I don't think I have any other radial caps to experiment with. I have some ceramic caps. Would they work?
Klipspringer
bsmith wrote:


As far as the gate, yes am finding so far that the envelator works fine if I juice it up (with the cable I use a .47 uF capacitor on) - setting that side to unipolar in the ac>dc plugin. Based on seeing pretty different results between the .47 uF and a 10 uF (.47 showing the best results), I want to try others and see how they do. I've got a couple .22 uF's, we'll see how it goes for them later today....

***edit
Went and got some .1 uF capacitors and they work great on pitch calibration and both gates I have to poke at (envelator and the kraftzwerg's). The envelator's AD mode seems to act better with it boosted - you need to pause between note on's otherwise not sure what up with that, AR mode is fine (and fast without the need to pause deal). Anyway, enough solder fumes today, time to play with all these newfound cv outs!!!


I'm finding similar behaviour now with the Envelator with the note ons in AR mode. However, I can get that mode to work now. To do this I had to decease the max out in the acencoder (.615 in my case).

AD mode still doesn't seem to work yet. (probably because I can't boost the signal).

Still, all is much cleared to me now!
Analog>Plugins
Does this mean that the cheapy Behringer adat thing would work?

http://www.behringer.com/EN/Products/ADA8000.aspx

Also how would this work with say the Arp Odyssey which I am sure uses -5 to +5v?

OS you constantly amaze me with your findings dude....just don't forget about the UBER slow tuning mode for silent wau (you know for added accuracy for those dodgy older synths that track a bit funny and non linear).

Cheers dude
os
Klipspringer wrote:
I'm using a 1 mirofarad 50V cap. Just like what's in the picture. The output is about 2.5V at the lead's end.. I've attached the Calibration data for a square wave from my ASol VCO 6HP.

Mmm, that's pretty messed up. Could you try setting the min/max output levels on the AC Encoder to, say, 0.4-0.6, to try to focus in on a range of values where the tracking is behaving better?

While 1uF works well for me, others seem to be finding smaller values track faster/better. Try something around 0.1uF if you have it.
MUFF WIGGLER Forum Index -> Expert Sleepers Goto page 1, 2, 3, 4, 5  Next [all]
Page 1 of 5
Powered by phpBB © phpBB Group