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

Better circuit choices for Silent Way AC Encoder
MUFF WIGGLER Forum Index -> Expert Sleepers Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next [all]
Author Better circuit choices for Silent Way AC Encoder
os
At a guess you want to use a TRS plug at the box end, not using the ring.

As a quick test though just plug a TRS-TRS jack lead into it and use a continuity tester/multimeter/whatever to see what gets connected to what, with and without anything plugged into the sockets that are going to the synth.
bsmith
Thanks Os.
Interesting. It was actually the opposite of that - I hadn't tried using a TS plug into the box, just the trs. Using my cable (which is 1/8 TS on the synth end) with a 18>1/4 TS adapter into the box works fine.
Thanks again- just hadn't poked around far enough.
PolishMilk
os wrote:
PolishMilk wrote:
Is there a better solution to this than just strapping a resistor across the tip & sleeve? Using the Axxe as a CV master for the VCO-sub has always worked fine so there must be a workaround?

I think the resistor is the workaround. The Axxe might have a non-capacitative output (e.g. direct out from an op-amp) or might have a resistor across the output internally.

The better solution for all of this is to build an active AC Encoder interface circuit - that's still on my plan, but it will be a while before I have time to try it properly.


OK, so I looked into this active circuit stuff and built the following circuit (ripped off from the dc mixer circuit on the doepfer DIY page and also from the MFOS dc mixer)



The input is from my modified ac-encoder cables with the voltage doubler circuit built in. The circuit uses a tl072 chip which has 2 op-amps built in, both of which are used in an inverting configuration.

Using this circuit I could control the Analogue Solutions VCO-sub module perfectly.

However I ran into problems when I tried to add a DC offset to the circuit. At the moment I'm getting 0-12 volts output so I wanted to offset by -6v to get -6 to +6v. The part of the circuit in red is my attempt to do this. I was measuring -6v on the 1M resistor before I soldered it in but then once I did I only got -1v offset on the output. I tried loads of different resistor values but all of them just give -1v offset!

Does anybody have any ideas about whats going wrong?
os
This is the (similar) circuit I was going to base my prototype on:

http://www.musicfromouterspace.com/analogsynth/oddsandends.html#LEVELS HIFTER

PolishMilk
Amazing. Thanks a lot os. I stuck the offset voltage into the + input of the second op-amp and it works great. Now getting -6v to +6v which gives me the total audible range of the synth. The circuit I used is...



Next step is build 7 more and stick it in a module!
os
I started a new thread for my active prototype:

https://www.muffwiggler.com/forum/viewtopic.php?t=15928
thenapking
Bit of a noob question here. To check: i'm going to build an unbalanced version of either the half-wave or voltage doubler cable. Since the ground is unconnected on the TRS version it's irrelevant, right?
os
It depends to some extent on the exact output circuitry of the balanced output that's driving the TRS.

But if you're building an unbalanced version, you only have T & S to play with, so it's a moot point.
thenapking
thanks. that's what i'd thought. I've got an old m-audio usb interface which doesn't get much use these days which i thought i'd press into service for this. I'll let you know if it works!
Navs
I just tried the half-wave rectification circuit with an unbalanced audio interface - an Emagic emi26, which uses RCA/ cinch jacks.

It's output is pretty weak, but the signal also seemed to droop over time. Switching channels got me the maximum output again, but this too would fade gradually.

Is this maybe down to the unbalanced outputs i.e. is the interface not suitable or would the ES-1 solve this? A square-wave LFO sounded slewed, which shouldn't be the case, should it?

I tried both a 1uF and 0.1uF cap and various AC Encoder settings.
os
RCA outputs will probably be quite weak, so that's expected, but I can't think why the output would droop. Does it fade away completely, or just drop a bit? Is this with the output unconnected, or driving a module?

An ES-1 ought to help, by presenting the audio interface with a proper high-impedance input, and by applying some gain to the CV. I believe others have used the ES-1 with RCA outputs successfully.
Navs
Thanks, Os.

I hope to get an ES-1, but want to make sure the emi26 is capable first and will troubleshoot this tomorrow using my RME interface as a control.

The signal doesn't fade away completely. The droop is noticeable whether the output is connected to another module or just the multi-meter.

One thought I did have was whether it might have something to do with using the emi26 as an aggregate device. AudioMIDISetup displayed something about 'up-sampling' to match the FF400 which was providing the master clock.
os
Do you have a simple buffer module you could pass the emi26 through before applying the rectifier? That would simulate the es-1's input buffer stage.
Navs
I've just had another idea: if all the circuit does is rectify the signal, I could send it to an A-119 or VCS/ Maths, boost it and take the envelope follower (rectified) signal out, right?

Guess there's only one way to find out ...
os
You'll want to smooth it too.
Sablerock
Hi guys,

I'm really excited to get things going with Silent Way! OS confirmed for me yesterday that my M-Audio Delta 1010LT card will work (the version with the RCA harness, not the breakout box).

Because it uses the RCA (unbalanced non-TRS) connections, how would that change the circuit design? Also, should I be building the amplification circuit and how would the wiring differ there as well? I plan on using Silent way to control the Korg Monotron as well as other analog synths. Any help would be greatly appreciated....

UPDATE: I have come across a device called 'The Matchbox' by Henry Engineering. http://www.henryeng.com/matchbox.html the version I have isn't the HD but has near identical specs. Would this overcome the need for any custom circuit building to work with my AC Coupled device? Can I directly connect my RCA's to the device and expect a sufficient output voltage? On the front of the device it is labeled as "Direct Coupled" also "IHF to PRO". It is powered by 120vac. I'm hoping that this might be an easy, off-the shelf solution to those of us with AC Coupled Devices that also use unbalanced RCA jacks... OS, any input on this one?

Update #2: With the Matchbox, I'd have 2 unbalanced input (-10db) from RCA (Delta1010lt) impedance adjusted to 2 Balanced outputs (+4db) with adjustable output gain. I'm hoping that 2 channels is all I'd need to accomplish Pitch and Gate functionality from Silent Way. Could this device possibly eliminate the ned for the AC Encoder as well?

Erik Hines
os
The Matchbox would certainly boost your levels. However it's not going to remove the need for the AC Encoder - your 1010 is still AC-coupled (presumably - do we know this for sure?) and it only takes one AC-coupled link in the chain to break the DC-coupling.

The simplest thing to try is this:

but where you see 'ring' on the TRS jack use the sleeve on the RCA jack.

If you then need more voltage, take it from there.
Sablerock
Great! Thanks Os.

I'm going to give it a shot once I head over to Electronics City and get the needed diodes and caps. I'll be testing both with\without the 'Matchbox' to see what kinds of voltages I get. Stated earlier I'll only be focusing on Pitch CV and Gate (2 channels) initially, but if it works, than I'll dedicate a few more channels for modulation sources...EXCITED!!!

Erik
Sablerock
Os and gang,

So I put together my cables last night for use with the Delta 1010 (actually, ended up using spare outs on the Delta 410 PCI instead) for use with the AC Encoder Plugin.

The first test I did was without the 'Matchbox' from Henry Engineering, I was able to squeeze about 3.8 volts max when feeding a saw wave sweep to 44khz. Because I expect to use more than 3.5 Octaves, I added the matchbox between the Delta outputs and the circuit and got EXCELLENT voltages!! Because the 'Matchbox" has an adjustable output, I calibrated to 12V when feeding the Saw wave at 44khz. I didn't max the output gain but am expecting that I can obtain up to 24V with the Matchbox.

2 questions arise from this test:

#1: What should be my maximum value for CV output? Should I use the full 24V output if I have it?

#2: When I put the output on the O-Scope and fed a pure sine wave, it comes back with what looks like the effects of capacitor discharge. My sine ended up looking more like a saw, with a slight sine ramp up and logarithmic tail after the peak. The capacitor I used is 1uF Tantalum Cap. Should I try for a lower value? Should I go Electrolytic instead? I'm concerned that any LFO's fed from Silent Way will be affected...

Before I go ripping my cables apart, I wanted to get your input on this one....

Erik
os
Don't feed your synths with CVs more than the synth's supply voltages (e.g. 12V for Eurorack).

Not quite clear where this 44kHz sweep is coming into things. Is that your input to the AC Encoder? Or are you testing your cables with an external oscillator?

If you're viewing the output of the rectifier cable on your scope, you should be looking at a flattish DC value if you feed it a constant amplitude signal.
Sablerock
10-4 Os on the voltage limits,
The 44khz signal is being fed from a waveform generated by Soundforge (audio editing program) to the outputs I've allocated for CV use. I do obtain a flat(ish) DC level at constant amp signal, solid as a rock and tunable to 10mv.

I just picked up some .47uf and .22 uf caps and will bread-board them tonight to see if the time constant decreases on discharge. I've been reading posts re: Tantalum vs. common Electrolytic and seems that the Tantalum type are NOT recommended for this type of application.

I'll post results once complete. Perfect sine wave or not, it was fun being able to modulate my Electrix Filter Factory with my 1-44khz sweep thru CV last night. Can't wait to test it on the Korg Maxi and MonoPoly!!!

Erik
Sablerock
Os and Gang,

So I have AC Encoder running into my M-AudioDelta 410, Half Rectifier Cables with the voltage doubler circuit, and all this connected to the Korg Monotron CV Mod (www.dinsync.info/2010/06/how-to-modify-korg-monotron.html).

I am able to run the calibration process and trigger notes via Midi, and get about 5 Octaves. Unfortunately, it seems that the capacitors in the Half_Rec cables I made have a high time constant for capacitor discharge, and so the dischgarge causes notes pitch-fade down after each keypress releases. How do I need to modify the circuit to allow for a much shorter time constant at key release?

BTW, the caps are .47uf 50v

Thanks,

Erik
os
Shorter time constant just means a smaller capacitor value.

However I'm slightly confused by what you describe, since the pitch shouldn't change at all if you just release a key. Too large a capacitor will give you a 'glide' effect between notes, but not just at the end of one note played and released.
Sablerock
Thanks for the response OZ. I'll do a capacitor swap once I get my gruuby finger on some lower value caps. As to the question of this affecting pitch: Apparently with the Monotron (per other users), the Gate voltage is tied to the pitch, so as long as the gate value stays constant there isn't any issue. I don't know why Korg did this, maybe as some sort of 'tuning' abitlity. As a workaround I did the following:

1) calibrated the Korg with the gate voltage in ACEncoder at max (1.000), ditto for VoiceController Gate output voltage.
2) after calibrating (I get about 5 octaves), I adjust the Gate output voltage down to the lowest thereshold the Monotron recognizes (approx 1.5 V). Since the voltage drop in the Gate on key-release causes the signal to ramp down from cap discharge, it falls below the threshold, in effect eliminating the symptoms.
4) Because I had to adjust the Gate (affecting pitch on the monotron), I either transpose the midi notes to tune, or use the VoiceControllers Detune to get it perfectly in tune.

I know this is only a temporary workaround, but until I get the Cap Discharge to a much smaller time constant, it is what I'm doing...

I'm probably looking at going down to a .1uf cap. I hear people talk about the smaller values having side effects when it comes to range. Will I lose Octaves going too low in Cap value?

Erik
os
Interesting info, thanks.

If you use too small a capacitor the whole thing just becomes unstable. You'll know it if you hear it!

The right capacitor value varies so much from synth to synth, it's impossible to predict what the ideal choice will be. You just have to experiment.
MUFF WIGGLER Forum Index -> Expert Sleepers Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next [all]
Page 2 of 7
Powered by phpBB © phpBB Group