## Max/MSP Motu 828 to CV

Dec 1, 2006 at 10:04pm

# Max/MSP Motu 828 to CV

I just had the crazy idea to use the audio outputs from my 828 as a CV… only to find a post about people who have had the same idea on this forum.

I read that the Motu 828 mkII outputs +/- 5v. So I’m guessing sending a signal of 1.0 to the dac~ should output +5v.

My main question is, should i be concerned about damaging my hardware?

Should I build a circuit to isolate the soundcard from the synth? And is it damaging to a sound card to leave it at a DC voltage for a length of time?

Has anyone that has done this (or something like this) found any pitfals that should be avoided?

Thanks

Joel

I want my granular synth to have real VCFs!

#29003
Dec 2, 2006 at 4:18pm

Hi,
I am affraid that 828, or any other piece, can’t generate contiuous voltage. A solution could be to send sinus an then in the real world to convert the sinus in continuous value.
Good luck

#89704
Dec 2, 2006 at 5:52pm

I thought it seemed too good to be true.

I’ve yet to do an actual electrical engineering course, but I’ll see how far I can get.

So something like:

Full wave rectifier –> Capacitor –> Opamp with adjustable gain –> resistor –> output

I’m assuming that higher frequencies work best… maybe 1/2 nyquist to avoid phasing with the sample rate?

Would a circuit like this be enough isolation from the synth to avoid damage/work properly?

Joel

I wonder how you would do this with FM instead of AM? …nevermind.

#89705
Dec 2, 2006 at 5:53pm

Hi There,

I’ve done this using a sig~ to send DC voltages out through an 828 and synched msp to incoming square wave LFOs. The only issue I encountered was a loss of linearity due to the impedence mismatch, but this was resolved using an opamp working as a voltage follower or buffer. If you’re not trying for accurate pitch, linearity shouldn’t be an issue. (I was interfacing with 1v/oct Moog stuff which requires that the voltage be linear for accurate subdivisions of the octave)

I never had any problems running CV into or out of the 828, though using a buffer would also serve to protect each piece of gear from one-another electrically. My 828 still works like new.

BB

#89706
Dec 2, 2006 at 8:25pm

Then perhaps the legends are true! Perhaps it really will output a continuous DC voltage.

I guess I should really be pulling out my voltmeter and seeing what actually comes out of the 828 for myself. Did you find it output +/- 5v or did you have to amp it up?

So you were able to get a linear DC voltage using an opamp buffer then? Enough to get accurate pitches on the moog without too much mucking around?

So making an opamp buffer would be enough to safeguard against votages going the wrong way or other electrical disasters? And would there be any problem in connecting the ground of the synth to the ground of the motu?

My electrical engineering projects have been very basic and not all successful so please forgive the asking of too many questions.

Joel

#89707
Dec 3, 2006 at 9:46am

> I guess I should really be pulling out my voltmeter and seeing what
> actually comes out of the 828 for myself. Did you find it output
> +/- 5v or did you have to amp it up?

Please search the archive… I’ve posted a message last year saying
that I’ve tested it for a 828mk1 and mk2. The mk2 produces a better
5v. I have not tested it on analogue synth dough, in my to-do list
since then…

pa

#89708
Dec 3, 2006 at 3:11pm

hi,
it’s bizzar,i think about that two days ago,
my question is a little different:
i would use my motu 828mk2 to route signals and cv on my modular analogue system synth….
i would send to the motu’s ins: my two oscillators,ring mod,lfo and use max/msp to route them to hpf,lfp,ring mod in…..
there is no risk for gear??? (because i love my synth and i don’t want broken it!!!!)
and no risk with signal feedback???
thanks for all

#89709
Dec 3, 2006 at 5:31pm

I’ve been working late these last few days, so I only managed to test a few things last night.

I put the voltmeter to the audio output of the 828 mkii and found that a [sig~ 1] gave me about +/- 4.5v … I’ll have to keep fooling around with this to see if it will go all the way to 5v.

I tested some of the voltages on my MS20 just too see what they were doing. I hadn’t even realized the mod wheel went from -5v to +5v… you have to learn sometime i guess.

Threw caution to the wind and plugged the motu straight into the LP filter cutoff. It works!!! Filter responded like it would if the mod wheel was plugged in.

I hadn’t realized the trigger was a low = on (S-Trig) type so when I plugged it in with a [sig~ 0.] it triggered the synth and it and made me lose my nerve, so i quit for the day.

I’ll post most stuff as I make progress with this.

Joel

#89710
Dec 12, 2006 at 10:54am

I made a bit more progress on this project. I’ve put a few progress reports up at http://www.nononancy.com/cvcontrol/index.html

I emailed Motu and told them what I was doing asking them if they thought I could cause damage to the soundcard or synth… so far no reply.

Seems to convert what I put out in max/msp into the correct CV fairly well, but I need to find access to an oscilloscope and a better multimeter before I can say conclusively. I know there are some frequency limitations to the opamps, but I’m not sure that will affect this. Still trying to figure out if there should be more to the circuit to safeguard the hardware.

And I still have to test if I can reproduce the correct CV for the V/Hz on the MS20.

Learning as I go.

Joel

#89711
Dec 13, 2006 at 8:47pm

Got an email from motu tech support,

Hi Joel,

We would not recommend sending control voltages through your 828. Outputting DC voltages through your 828 will damage your 828 and void it’s warranty.

#89712
Dec 14, 2006 at 7:23pm

Joel Rich wrote:
> We would not recommend sending control voltages through your 828.
> Outputting DC voltages through your 828 will damage your 828 and void
> it’s warranty.

Did they explain why? I would guess they only care about their own
security, any unusual use is something they never test, and thus it’s
safer to not allow it officially.
I didn’t do this myself, but as I recall some on the list did it over a
long time send DC without problems.
Even in a “normal” way of usage it could happen that you send out DC by
accident, the warranty would cover that, because the user can’t know. If
in the manual isn’t any warning about opening a soundfile with DC and
playing it with the editor of your choice, I guess it still should be fine.

Disclaimer: This is no professional advice and if you still do send DC

Just a humble oppinion…

Stefan

Stefan Tiedje————x——-
–_____———–|————–
–(_|_ —-|—–|—–()——-
– _|_)—-|—–()————–
———-()——–www.ccmix.com

#89713
Dec 14, 2006 at 8:38pm

I tried to follow up and get some more details from motu but got a one line reply again.

Hi Joel,

The 828Mkii is not designed to pass DC voltages. We do not recommend doing this since there is a good chance that it will damage your 828Mkii.

Thanks for writing,
Ronan
MOTU Tech

So I’m trying again from a different approach as it seems like they are covering their butts a little. I emailed them as someone who realized they were accidently inputting and outputting a DC voltage and was concerned to avoid damage asking what it would take to wreck the motu. Using someone elses email and name of course ;)

I wonder what the problem with DC voltages is? Does it send out too much current causing something to overheat? I’ll send him another email with more specific questions stating that I’m definitely not trying to do this experiment… just curious.

hehe thanks for the disclaimer Stefan, I’ll remember that

Joel

Oh yeah… i tried this with a couple other soundcards and no luck. Seems the motu is unique in its ability to output DC (as far as my collection of sound cards go).

#89714
Dec 14, 2006 at 9:02pm

Well you could always output AC voltage into some sort of rectifier or enveloper follower circuit.

http://en.wikipedia.org/wiki/Full_wave_rectifier

Note the section about Rectifier Output S. This will give you a fairly consistent control voltage.

This circuit is fairly simple and wouldn’t be too hard to build yourself. However there are also commercial envelope followers available, which might be better. Of course you would need one follower per control channel….

Here is a good example I found with a quick google search. Its been discontinued but I’m sure there are other similar units out there-

http://www.oakleysound.com/follower.htm

#89715
Dec 14, 2006 at 10:18pm

The same thought crossed my mind. Though I was worried that if the voltage was smoothed too much it would be acting like a lag/portamento generator. Though maybe this isn’t a big concern, I’m not sure.

The other problem with doing this is, like you said, other people have already done it better. Its a little bit like reinventing something.

One slightly crazy thought I had was to figure out how do this like an FM radio. Maybe outputting a carrier signal through an additional channel then have the frequency deviation act like a + or – voltage. And I have no idea how this is done, but it is kind of a neat idea.

Joel

#89716
Dec 16, 2006 at 12:56pm

Joel Rich wrote:
> Oh yeah… i tried this with a couple other soundcards and no luck.
> Seems the motu is unique in its ability to output DC (as far as my
> collection of sound cards go).

I tried my Quatafire, and its not outputting DC. I should open it and
get rid of the output capacitors… (I’d better not ask if that would
void my warranty ;-)

But more interesting for me would be DC input…

Stefan

Stefan Tiedje————x——-
–_____———–|————–
–(_|_ —-|—–|—–()——-
– _|_)—-|—–()————–
———-()——–www.ccmix.com

#89717
Mar 6, 2011 at 9:59pm

Hello all! (:

This may be 4 years ago.
But what happened with this experiment? (:

If it’s possible and you made a patch for it , would you like to share?

Greetings,
Cengiz

#89718

You must be logged in to reply to this topic.