Moog Music Forum

Hi FreqOut,

>> Is it possible to build a "passive" voltage inverter?

No, an inverter is always a so called active electronic circuit, consisting of one op amplifier and (some) resistors, maybe some capacitors near to the power lines ...

>> ... does the power supply need to match Moog's +5V/-5V scheme?

No, again. The output voltage of the inverter follows exactly to the input value (from the MMV), the sign of the voltage is turned of course. So that output voltage is always inside the proper range ...
It could be useful to limit that output voltage to +5V and to -5V by using diodes (switched between the op amp output and those +5V and -5V respectively). To supply the op amp you'll need two voltages, about 1V "higher" than the max. needed output signal, that means about +6V and -6V respectively (but +12V / -12V will work fine too).

Frank,

Thanks for the info!

It sounds like this might be more difficult than I was hoping for...
(since I really don't have any idea what I'm doing)
I'm going to keep looking for schematics that might work.
I'll post something here if I find anything that works!

FreqOut

I could be wrong (and often am), but I don't think that a voltage inverter applied to the pitch CV from any given oscillator will produce an inverted waveform... only an inverted voltage (hence, frequency, in this example).

Now, if I'm wrong, and the LFO actually outputs CV, then of course it's possible. Or, put another way, inverting the polarity of the LFO waveform itself will be successful, but inverting the voltage of the CV controlling its frequency will fail.

Just an attempt to clarify.

-Hoax

Hi Hoax, hi FreqOut,

please excuse me that I'd only answered to the very specific questions inside that post

>> Is it possible to build a "passive" voltage inverter?
and
>> ... does the power supply need to match Moog's +5V/-5V scheme? .

Of course, a simple voltage inverter won't help if the aim is to build a falling ramp from a raising ramp and vice versa, if both ramps are (or should be) always above 0V, that means are (or should be) always positive.

A voltage inverter does the mathematical operation of changing the sign of any input signal. That means for example, that if the input voltage (the original signal) is +1V then the output voltage will be -1V, and so on.

But changing a raising saw tooth (e.g. starting from 0V and ending at +5V) into another wave form (e.g. starting from +5V and ending at 0V) isn't done by inverting the sign of that input voltage!!!

In that case (I mean the example above) you need to add another +5V to the output signal of the inverter circuit.

Fortunately (!) it is very easy to do all these "mathematical operations" using only one op amp. You have to build an inverting adder that adds
1. the original wave form (to be inverted)
2. a constant negative voltage (because it's inverted too) (from a simple voltage divider consisting of two resistors or better of a small trim potentiometer, switched between ground voltage [0V] and the negative power supply [e.g. -12V or so]).

Sorry for my very short answer before, but I didn't trace all your needs.

Please ask further questions if you need any further help ...

Dang!! I thought I had it all figured out!
What I'm looking for is a way to create a rising ramp waveform to use as an LFO like on the old Model D minimoog.
Frank, can you help me figure out what I need to build? I know nothing about electronics (as you probably noticed), but I've built a few things from kits and can follow a schematic.
If it's possible, I'd like to build something that is powered either from the voyager itself, or a 9v battery, or a 12v wall-wart type power supply.
All help will be greatly apprectiated!

FreqOut

Hi FreqOut,

I did my best to understand your "problem" and investigated the different behaviours of the MMV LFO's, using an oscilloscope ...

As moogmusic had written in his message (Mar 15, 2004) you'll need a VX-351 to get the Osc. 3 as an output for modulation purposes. And connected this way, you'll get a saw tooth voltage, if the Osc. 3 wave form pot is adjusted correctly.

This voltage is a bipolar one (!!!), "symmetrically" around the time axis, and therefor (but only in this case) you'll need only a simple voltage inverter (as GregAE posted before, Apr 07, 2004), to get a rising saw tooth from the original falling saw tooth of that Osc. 3 wave form.

I'd tested it using the inverter of the CP-251 (offset turned to the middle position, "0") and observed it using an oscilloscope. Turning the offset knob of the CP-251 can move the whole LFO wave (the inverted Osc. 3 output signal) to more negative or to more positive mean values. It depends on your needs what resulting wave form is useful for you ...

If you want to use the original MMV LFO (but there is no saw tooth of course), then you'll see that the original wave form is always positive (not bi-polar, always "above" the time axis, always positive). Any voltage inverter therefor changes the direction of any ramp (falling to raising and vice versa) but also "shifts" the whole wave form below the time axis (the signal is always negative), so the modulation result doesn't meet your requests, may be. Using an additional offset (adding another constant voltage to the original wave form) can adjust the resulting wave form, so that it comes back to the "always positive" range ...

This experiment can also be done using a CP-251.

FreqOut, you always repeat that you want to get back the LFO of your original Minimoog D. I can't answer if a simple voltage inverter (Osc. 3 output signal as input for this inverter) will sound the right way if you use it to modulate any destination of the MMV.

As I know, the 25 pin connector of the MMV delivers all the signals and all the voltages to use it as a power supply for such an inverter. But I would be very carefully to avoid any defect of my MMV when using that connector for experiments ...

Summary:
If you already own a VX-351 it's very easy to build a raising saw tooth wave form from the Osc. 3 either by
- using the CP-251 inverter (offset turned to "0")
or
- applying any voltage inverter to that signal coming out of VX-351, supplied by the MMV's 25 pin connector (you have to use/build an "Y" cable to connect the VX-351 and to get access to the power supply at the same time) or by any external power supply.

Remark:
External power supply means two batteries connected this way:
(+)______(-)(+)______(-), where the middle connection point '(-)(+)' has to be connected to the ground of the MMV and has to be used as ground voltage for the inverter circuit. The voltage between '(+)______(-)' must be at least 6V to achieve full 5V output range of the outcoming signal. So, the use of a single 9V battery needs additional components to make the circuit working proper and isn't recommendable therefor.


Again, I don't know if these explanations will help you with your own demands. If you want to hear my advice, I would recommend to buy a CP-251. This would result in very useful modifications of different sound designs and avoid any trouble or damage ... But it costs a lot more than a simple voltage inverter, that could achieve the same purpose in a smaller range of applications.

Frank,
Thanks a million for your great response to my question!
I'm going to order a CP-251 right now....
I'll post here to let everyone know how it goes when I get it all working!

FreqOut

I've always kept this post in the back of my mind.

Can a simple u741 based voltage inverter solve this problem or am I shooting too low? With properly choosen rail voltages and component values I'm thinking this can reverse the ramp.

I'm working in a second semester circuits class building fundamental opamp configurations in the lab and i've been thinking about this and other interfaces between my 351 and Voyager.

But, I am very cautious about smoking stuff (in an electrical sense).

BTW, I'm just about finished with my PAiA 9700. I've recently been looking into ways to interface this modular with my Voyager setup. Has anyone else used a 9700 with a Voyager??

The paia uses a radical IMHO method of temperature compensation in the exponential converter of the V/Oct inputs on the oscillators. They use diodes instead of tempcos which translates to only about 2.5 to 3 maybe four octaves of total good linearity. The diode versus tempco design has more of an effect in the high freq range of the VCO. Its more of a what goes in comes out sideways oscillator in the higher octaves. So don't expect really good like voyager linearity across 10 oct. It is a shame, though. The essence of a modular is its voice ala VCO. They took a chance with a little cost reduction and IMHO crippled it . Typically the most expensive and complex part of any modular is the VCO. The filter is another beast. Please be careful when using high resonance as it will go into powersuppy rail to rail oscillation which is not gentle nor a sinewave. I have smoked speakers and ears accidently with high resonance with the paia. Overall, understand it does many things and is a good value for what it does. In the bass area, adding the 9700's two osc. adds a huge fatness. You get a ring modulator in VCA module so that is very fun. I usually just have the voyager drive the Piaia via midi and it works very well. The Paia MIDI-CV triggering is a little funny but usuable.

Sorry for the ramble, Hope this helps
stefan

ive tried both the lfo within, the 251, mmv lfo and osc3 as lfo sent to bus and then output via 351 and can verify that you NEED to add offset voltage (i used the 251) to be able to make it modulate a destination and sound like the internal lfo routed internally.

tried with my ms20 too and the cv put in to mod2 input ( or mod1 you can use both) need to be in the positive range hence not bipolar