AC Encoder active circuit prototype

Discussion and support for Expert Sleepers' Silent Way software and hardware.

Moderators: Kent, os

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

AC Encoder active circuit prototype

Post by os » Sun Apr 25, 2010 7:51 am

I finally got around to building an active AC Encoder circuit:

Image
http://www.flickr.com/photos/expertslee ... 926880008/

Image

This is based on the MFOS design at http://www.musicfromouterspace.com/anal ... VELSHIFTER
and very similar to that built recently by PolishMilk (viewtopic.php?p=210351#210351).

The two trim pots give an adjustable gain from 1x to 6x and an adjustable DC offset. (If you just want positive voltages (e.g. for a gate) you can do away with the DC offset pot.)

I tested this with my Mobile I/O 2882 interface. In +4dB mode this outputs ±8V anyway to I put it into -10dB mode, which gives just over 2V output with the passive rectifier circuit. Using the active circuit I got a nice ±5V or 0-10V, and it was rock solid driving ASys and Doepfer VCOs. Calibration with Silent Way covered the full (audible) range of the oscillators.

The quoted component values give an RC time constant for the rectifier of 0.2ms. I certainly couldn't hear any slew when doing 5 octave jumps on the keyboard.

The circuit was powered from the ASys case bus.

Total component cost for one channel is about £2, about half of which is the trim pots.

Next step is to solder up a few of these channels into a module!

Edit - see below for a slightly updated version of this circuit.

Edit2 - see further below for a DC-coupled version of this circuit.
Last edited by os on Thu Apr 29, 2010 7:40 am, edited 2 times in total.

User avatar
Bryan B
Super Deluxe Wiggler
Posts: 1377
Joined: Wed Sep 23, 2009 6:09 pm
Location: Minneapolis, MN

Post by Bryan B » Sun Apr 25, 2010 12:06 pm

Thanks for posting this!

What happens when you throw normal a normal audio signal into it?
doctorvague wrote:I like hiding out at Muff's. Let's build a fort out of some chair cushions and a blanket.
JP wrote:Hell I want to be able to turn my toaster on with a CV trigger.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Mon Apr 26, 2010 3:01 am

I guess you'd have a kind of envelope follower, but with a much faster response than would be useful.

Why would you want to do that?

User avatar
bsmith
Super Deluxe Wiggler
Posts: 2158
Joined: Thu Jan 07, 2010 11:33 am
Location: Dallas

Post by bsmith » Mon Apr 26, 2010 7:28 am

Really exciting os! Are the trim pots something that would ultimately brought out front of the panel on a pot or would they be a set and forget kind of thing? How does the different results that folks have with different capacitors for their interfaces get handled in this active model? Sorry if these are ignorant questions, am a simple cave man....

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Mon Apr 26, 2010 7:38 am

The trim pots ought to be 'set and forget'. You'd just need to set them up once to suit the output of your audio interface, and whether you want bipolar CVs or not.

The capacitor choice can be fixed because it depends only on the input impedance of the circuit, which is fixed. In the passive case, the input impedance is determined by whatever module you happen to have it plugged in to.

User avatar
bsmith
Super Deluxe Wiggler
Posts: 2158
Joined: Thu Jan 07, 2010 11:33 am
Location: Dallas

Post by bsmith » Mon Apr 26, 2010 7:46 am

os wrote:The trim pots ought to be 'set and forget'. You'd just need to set them up once to suit the output of your audio interface, and whether you want bipolar CVs or not.

The capacitor choice can be fixed because it depends only on the input impedance of the circuit, which is fixed. In the passive case, the input impedance is determined by whatever module you happen to have it plugged in to.
Again excuse if this is a silly question - why wouldn't one want bipolar?

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Mon Apr 26, 2010 7:59 am

You might have a Buchla :)

It's possible that some modules really don't like negative CVs. I don't know of any myself though.

User avatar
Arnoid
Wiggling with Experience
Posts: 284
Joined: Mon Jan 25, 2010 4:32 am
Location: Belgium

Post by Arnoid » Mon Apr 26, 2010 8:34 am

Nice one !

This may be the solution for me because with the passive AC encoder circuit the calibration was a bit unstable on certain oscillators...

Gonna check this out when I have some spare time does this circuit also work on 15V ?
Last edited by Arnoid on Mon Apr 26, 2010 8:45 am, edited 1 time in total.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Mon Apr 26, 2010 8:39 am

Should work fine at 15V.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Mon Apr 26, 2010 2:52 pm

A slight update:

Image

- the output protection resistor (R7 in the first pic above) was redundant since the TL082 has this internally, so I've removed it.

- since the first op-amp is amplifying a +ve only signal, it has a max range of 0-12V. Therefore the DC-offset op-amp (the second one) in the original circuit was producing a max ±6V signal, since it had unity gain. I've therefore doubled the value of R2 to give 2x gain, allowing a bipolar signal which covers the whole possible output range (in theory - in practice slightly less).

Using these mods I've just got a nice ±10V and tracking of the oscillator over the whole MIDI note range.

User avatar
Thonk Support
aka Monobass
Posts: 8867
Joined: Fri May 29, 2009 5:39 pm
Location: Brighton, UK

Post by Thonk Support » Mon Apr 26, 2010 5:36 pm

This is really exciting.... I'm very keen to see if I can use a circuit like this to turn the audio outputs of my Nord Modular G2 into CV outs.. does it still definitely require a balanced output?

User avatar
Thonk Support
aka Monobass
Posts: 8867
Joined: Fri May 29, 2009 5:39 pm
Location: Brighton, UK

Post by Thonk Support » Mon Apr 26, 2010 5:47 pm

definitely a TL082P required? will any other TL082 also work?

edit: I have some TL082CN
Last edited by Thonk Support on Mon Apr 26, 2010 6:00 pm, edited 1 time in total.

User avatar
Bryan B
Super Deluxe Wiggler
Posts: 1377
Joined: Wed Sep 23, 2009 6:09 pm
Location: Minneapolis, MN

Post by Bryan B » Mon Apr 26, 2010 5:47 pm

os wrote:Why would you want to do that?
People ask me that all the time :)

I figured it out before you answered though.

You could use a full volume loop (white noise?) and use Ableton's envelope editing to create CV outs maybe? I have to noodle on this some more, but it seems like you could do some fun experimenting with something like that.

A filter on the sound would probably slow the envelope follower response time...
doctorvague wrote:I like hiding out at Muff's. Let's build a fort out of some chair cushions and a blanket.
JP wrote:Hell I want to be able to turn my toaster on with a CV trigger.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Tue Apr 27, 2010 2:58 am

@monobass - the prototype uses TLO82CN. Ignore TL082P - that's just what the schematic software brought in as default.

Probably does not need a balanced out, but I need to confirm that.

@Bryan B - that would probably work. You could use Silent Way DC and draw parameter curves to achieve the same thing though.

User avatar
Bryan B
Super Deluxe Wiggler
Posts: 1377
Joined: Wed Sep 23, 2009 6:09 pm
Location: Minneapolis, MN

Post by Bryan B » Tue Apr 27, 2010 7:26 am

My cat gets skinned differently every day.
doctorvague wrote:I like hiding out at Muff's. Let's build a fort out of some chair cushions and a blanket.
JP wrote:Hell I want to be able to turn my toaster on with a CV trigger.

User avatar
Arnoid
Wiggling with Experience
Posts: 284
Joined: Mon Jan 25, 2010 4:32 am
Location: Belgium

Post by Arnoid » Tue Apr 27, 2010 7:52 am

Is the Half-wave rectifier implented in this circuit ?
Because I only get good voltages if I use the voltage doubler or even tripler...or isn't this important for this circuit ?

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Tue Apr 27, 2010 7:57 am

This circuit is using the full-wave rectifier.

The op-amps as configured above provide up to 12x gain, so the passive voltage doubler isn't required.

Cryptowen
Learning to Wiggle
Posts: 3
Joined: Wed Apr 28, 2010 12:20 pm

Post by Cryptowen » Wed Apr 28, 2010 12:23 pm

Any plans to sell a pre-assembled version of this? I've been holding off buying Silent Way until I could get it working, & I can't seem to get any of the AC encoder circuits to work properly with my MS-10. I'm assuming it's just me not doing circuits right.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Wed Apr 28, 2010 1:02 pm

Yes, there's a plan, though a fairly long-term one.

Stay tuned!

Cryptowen
Learning to Wiggle
Posts: 3
Joined: Wed Apr 28, 2010 12:20 pm

Post by Cryptowen » Wed Apr 28, 2010 1:09 pm

Awesome.
Out of curiosity, how long term might it be? I've got an album coming out in August & I need to record good versions of the synth parts some time before then. Depending on when Silent Way hardware is released I may or may not go ahead & buy a synth with midi in-cv out already included to use as a converter for now.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Wed Apr 28, 2010 1:29 pm

Hard to make any guarantees, but it should be well before August. I guess I have to say, wait until you can't wait any more, and go out and buy the other synth. You can always sell it later.

My initial plan is for a Eurorack module, though the PCB will be usable in a DIY kit to be mounted/powered how you like.

Cryptowen
Learning to Wiggle
Posts: 3
Joined: Wed Apr 28, 2010 12:20 pm

Post by Cryptowen » Wed Apr 28, 2010 1:30 pm

Sounds good. I'll keep following the forum & the Expert Sleepers website for updates.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Thu Apr 29, 2010 3:06 am

It's probably worth mentioning that this circuit makes a very good voltage booster for DC-coupled interfaces too, simply by omitting the rectifier:
Image

Just tried this with my Ultralite mk3, and with lowest gain and no dc offset, turned the Ultralite's ±4V into ±8V. Nice.

PolishMilk
Learning to Wiggle
Posts: 6
Joined: Wed Mar 31, 2010 5:20 pm

Post by PolishMilk » Thu Apr 29, 2010 5:51 am

This is looking great. I particularly like your idea of x2 gain on the second op amp to get the full voltage range.

Just wondering if you think there will be a difference between basing the circuit on 20k resistors like your circuit, or on 100k like I used? Possibly current draw or response speed?? Just wondering which values would be best to go for when I make the circuit for real, or whether it makes no difference.

User avatar
os
Super Deluxe Wiggler
Posts: 15344
Joined: Tue Jun 30, 2009 6:55 am
Location: UK
Contact:

Post by os » Thu Apr 29, 2010 6:00 am

I don't think it makes a great deal of difference, but my electronics chops are a little rusty so I'm happy to be corrected.

The key things are:
- the ratio of resistor values where they differ, as this determines the gain
- the relationship of the resistor values to the capacitor values, as this determines the time constant of the smoothing filter.

Post Reply

Return to “Expert Sleepers”