any lfo tutorial??

[swt]

Hi!. Do you guys know of a good url or tutorial to take a look at some lfo info?. Maybe something basic, like what affect what, or maybe changing depths, speed, etc, by changing specific components. Thanks a lot!!

[R.G.]

I haven't written one yet, but I can give you the quick guide.

Probably 75% of all LFOs are the integrator+comparator style. This is a relaxation oscillator (that is, it uses positive feedback). The basis of the I/C style is that an integrator produces a linear ramp up or down which has a speed that varies with how much current it's fed. The integrator may be a single capacitor (example: the Phase 90 LFO) or an opamp with a capacitor for feedback. The comparator is wired as a schmitt trigger, which has hysteresis. The schmitt trigger is fed the capacitor voltage and it in turn feeds the integrator through some resistance. The schmitt trip points are the places where the output of the schmitt tells the integrator to reverse direction, and this determines the output signal level of the LFO. The resistor from the output of the schmitt determines how fast the capacitor ramps and this determines speed, so this resistor is always the speed control. Depth is done by using a pot as a "volume control" on the output of the LFO. I/C LFOs have a wide frequency range and reliability - they oscillate first time, every time in the face of component drift and varying gains. They are also simple to control with only the one speed pot. The two disadvantages are that the speed pot usually needs to be a reverse log pot taper and that they produce a triangle and a square wave, not a sine wave.

Probably 24% of LFOs are gain-phase oscillators, sometimes called "Nyquist" oscillators. These are inverting amplifiers which have a feedback network that causes the feedback phase to change from negative to positive at some frequency. This makes the amplifier oscillate in classical gain-phase oscillation. This produces a sine wave which becomes as big as it can get before something limits it. Often the limit is the power supply swing of the amplifier. The limiting factor introduces a little distortion, but this is usually negligible in an LFO. Resistors in the feedback network are used to vary the phase shift of the feedback network and therefore the frequency which has a phase shift that is correct for oscillation. Phase shift oscillators are more finicky than I/C oscillators. The gain has to be just right, the attenuation of the feedback network has to be just right. They generally work best with dual gain speed pots, although you can use a single speed pot and sacrifice some frequency range. The most common example of the phase shift LFO is the Fender tremolo oscillator.  As I noted in The Technology of the Univibe, that LFO is a phase shift oscillator, but a very oddly set up current mode LFO, and not a good example. It's the only one of that kind I've ever seen.

The other 1% are a motley assortment of stuff. The Pulsar uses an astable multivibrator, for instance. Very odd.

In all of these, the oscillator determines the waveform and speed. There is a depth control after the LFO which is in general unrelated to the LFO.

[Stompin Tom]

Thanks, RG. Informative, as usual. Now I look up hysteresis...

[Joe Kramer]

Cool stuff RG!  I look forward to that "Technology Of. . . ." paper on LFOs.

SWT: As a bit of an addendum, here are some more examples of RG's descriptions.  IIUC, the LFO in the Boss CE-2 is the I/C type, and produces both triangle and square waves.  I think the Boss VB-2 is an example of the gain-phase type, which produces a sine wave.

IMHO, both of these LFO's are good places to start experimenting.  I've used both to great success in a number of projects--phasers, trems, vibes.  In the case of the CE-2 LFO, only a few component changes will give you a VERY wide range LFO. . . .

Joe

[swt]

wow R.G. Thanks a lot!!. Really good info. I appreciate the time and explanation.
What i'm after is a good lfo for filters. unipolar, and bipolar supplied, because one is for a simple filter at 9volts, and the other one is for a korg ms20 type clon, at +-15volts. The key here is to get a really deep swing with slow rates up to tremolo ones. that's all. Any suggestions?
PD: Joe...what do you think of that ce2 for filters?. will i get enough range and width?.
Thanks you all for the replies!!

[calculating_infinity]

Aha nice post R.G. another one for the bookmarks!  "The Technology of ..." is definately great stuff. 

[Joe Kramer]

Hey SWT,

No reason why the CE-2 LFO wouldn't work with a filter.  As far as the speed range, you can get everything from nearly DC to audible frequencies all with one control pot, depending on how you tweak the parts values.  The pk-pk swing is more a function of the supply voltage and the type of op amp you use.  In my experience using it with 9V single-supply, the swing is a little more than 5V with a TL062.  I chose the TL062 for low current consumption, but it also seems to give the widest swing (compared to BJT op amps) without ticking/thumping problems.  Rail-to-rail op amps will give a wider swing, but they also seem to tick or thump for me.  You can easily set up the CE-2 LFO for bipolar supply use, and in that case, the swing will be within about 1-2 volts of the power rails.  Also, just to be clear, I have mostly used this LFO to drive a LED/LDR combo.  If you plan on driving the VC input of a synth filter, you may need extra signal conditioning, like a buffer, and some kind of voltage offset control.  For instance, the LFO on the MS-20 puts out a ramp wave of +/- 2.5 volts, but the squarewave is 0-5 volts.

Speaking of the MS-20, Korg once made a stand-alone LFO/modulation pedal that was supposed to be a companion to the MS-20.  I believe the model number was MS-04, and it output triangle, square, random, and glide control.  The schemo is somewhere out there on the web. . . .

Regards,
Joe

[caress]

i know someone who has that pedal!  pretty amazing it seems, although it is not functioning...

[Joe Kramer]

Hi Caress,
I'd be happy to take that pedal, working or not, off your friend's hands if he doesn't want it!   

[puretube]

[Joe Kramer]

That's the one--sure looks nice!      Thanks for the pix Ton. . . .

[swt]

hey Joe!! thanks for the info. What do you think it's better to change in the ce2 lfo and why?. I know the cap varies the range of the frequency. What does changing the resistor in the feedback loop, or the other feedback resistor do?. And changing the resistor from the freq pot? or even the freq pot?. Thanks for your help. I'll breadboard this as an experiment.

[Joe Kramer]

Hey SWT,

My knowledge on this is not exhaustive by any stretch, but I can tell you what's worked for me.  Here's a schemo for reference (thanks to Aron & Chris):

http://home.hetnet.nl/~chrisdus/download/ce2.gif

You can see the LFO in the lower lefthand corner, built around IC 2.  The minimum speed is set by R31 (the smaller, the slower), and the maximum speed is set by R32 (the smaller, the faster).  The overall speed is more-or-less set by C19 (the smaller, the faster); although other factors also affect the overall speed, C19 is the go-to component.  Here's my default set-up for a wide-range LFO using this circuit:

R31 = 470 ohm
R32 = 470 K
C19 = .047 uF

With these values, the slowest speed is in the sub-1Hz region, and the fastest speed gives an almost-ringmod type of tone.  You can get pseudo-ringmod sounds from a chorus/trem/phaser using a much smaller cap for C19 (.001 or so), which will put the LFO in the audio range.  At that point it's not strictly an LFO anymore. . . .

Hope that helps get you started.  Have fun at the breadboard!

Joe

PS: Hey Ton, that modulation pedal would sure be welcome company to my lonely MS-20.  Would it like to come over and visit?  Possibly shack up permanently?   

 

[puretube]

sorry, Joe: finally sent it to Ashley (ZVex`s right hand), last Christmas...
dunno if this thread is still helpful for you?

[Joe Kramer]

Hey Ton,

Lucky Ashley!  What's interesting is that she was really looking for the Traveler control pedal, not the MS-04.  I wonder if she's still looking, since I have a Traveler control pedal.  I'd be up for a trade. . . .

Thanks for the links to the MS-04 schemo.  I've had that in my files and pored over it many times.  Looks do-able but for that pesky center-tapped pot.  Then again, there's probably a workaround.  At the moment I have more money than time (alas, not much more), so building is not the best option today. . . .

Sorry for veering off-topic here. . . .

Joe