Variable LFO

Here is a link to a simple circuit for producing a tremelo effects pedal which involves a LFO. Bear in mind, though, that to get a true Leslie sound requires reproducing a slight Doppler effect on the sound caused by the spinning rotors.

** This is your best bet.

With a dual 50kohm pot and a pair of 3.3uF caps, the frequency range is from 1Hz to 20Hz.

If you use +/-15 volt supply rails, the four diode network can be changed to a pair of back to back zeners for greater output.

... Phil

Thanks. That has some good possibilities and it's real cheap to implement. The Doppler will be implemented via steering said LFO to slightly vary filt ering as well as some panning and tremolo. I can see phase, freq. and ampli tude changes would all be involved with a Leslie so I am trying to replicat e it as closely as possible electronically.

OSC (sine) that can be varied from zero to approx. maybe 20 cycles continu ously. Been busting my brain to implement it. This is for a source into a k eyboard to act as a Leslie. Any ideas guys. THANKS.

I think it's all done digitally now.

Those are really low frequencies, made worse since you say you need it to go down to zero. Obviously not, but still I can see why you want a slow rate, and that will make things complicated, big value resistors and capacitors, and maybe problems gettting down that low with one turn of the knob.

Why not a stepped sinewave? WOuld that be good enough? You use a ripple counter, and some weighted resistors, and you get a stepped sinewave that persumably is easier to filter out the imperfections. Since you will be dividing the oscillator frequency in the ripple counter, the oscillator can be at a higher frequency, making a wider range easier, and the component values more manageable.

And speaking of imperfect sinewaves, things like the Intersil 8038 and Exar 2206 generated a triangle wave, and used diode shaping to get sinewaves. You wouldn't want it for testing distortion somewhere, but perhaps that's good enough for this purpose. Though, you are then back to large value capacitors for a very low frequency oscillator, but at least the oscillator is an RC one.

Michael

p

e

o

Ideally it would start from zero and ramp up to speed and then slow back to zero upon turn off. Then while on it needs to be somewhat variable as a re al Leslie is. The Leslie being mechanical has a lag when started and moment um when stopped. Would love to replicate that effect instead of a constant speed. Not seeing an easy way to do the speed up/down without overly compli cating the thing. However being able to continuously vary the speed and tur n it on and off would be key.

A Stanford paper called "Doppler Simulation and the Leslie"

DDS in a cheap microcontroller, pwm putput, low-pass filter.

zero Hz is hard to do in analogue.

--
umop apisdn

OSC (sine) that can be varied from zero to approx. maybe 20 cycles continuo usly. Been busting my brain to implement it. This is for a source into a ke yboard to act as a Leslie. Any ideas guys. THANKS.

I've done a Wien bridge down to 3 Hz.

You could push it to 1 Hz (maybe)

Or a stepped sine thing with square wave into a CD4017 and artfully chosen resistors on the output... into an opamp summer.

Could it be a triangle wave?

George H.

p OSC (sine) that can be varied from zero to approx. maybe 20 cycles contin uously. Been busting my brain to implement it. This is for a source into a keyboard to act as a Leslie. Any ideas guys. THANKS.

n resistors on the output... into an opamp summer.

Upon researching it further it appears that a triangle would probably in fa ct work close enough to be unnoticeable. I might have over analyzed it. Doe s make the design a whole lot easier. The ramp up also might be a little ov er kill. Went to an old friend who works on Hammonds and has a Leslie corre cted me on that. It appears they have either off, slow and high speeds and the ramp up is purely a mechanical lag which I have determined is way more difficult to implement per some of the responses received. Thanks for all t he help guys. Will let ya'll know what happens.

Thanks and WOW!! CCRMA. Now that's a oldy but goody. Had an old CD produced by them way back in the 80's called the Digital Domain that was really coo l. It was made specifically to show the benefits of digital audio. Some of the cuts on it were planes landing that really pushed the dynamic range to the max for the CD standard. Also some of them were just plain weird. They also included a version of deep note that would blow speaker cones right ou t of the basket.

OSC (sine) that can be varied from zero to approx. maybe 20 cycles continu ously. Been busting my brain to implement it. This is for a source into a k eyboard to act as a Leslie. Any ideas guys. THANKS.

eap OSC (sine) that can be varied from zero to approx. maybe 20 cycles cont inuously. Been busting my brain to implement it. This is for a source into a keyboard to act as a Leslie. Any ideas guys. THANKS.

sen resistors on the output... into an opamp summer.

fact work close enough to be unnoticeable. I might have over analyzed it. D oes make the design a whole lot easier. The ramp up also might be a little over kill. Went to an old friend who works on Hammonds and has a Leslie cor rected me on that. It appears they have either off, slow and high speeds an d the ramp up is purely a mechanical lag which I have determined is way mor e difficult to implement per some of the responses received. Thanks for all the help guys. Will let ya'll know what happens.

Yeah, compared to a sine a triangle is dead easy. A current into a cap, with something that changes current direction at the limits... Heck might as well use a 555, where is John F.?

George H.

1. VCO that varies from 1kHz up to 1020Hz, as iput into an analog multiplier with the other input 10kHz, don't recommend that though
2. 16 bit 1024 looped register [can be implemented in C/C++] containing

1024 digial values of a sinewave. Vary the shift clock from 0 to 20480. Output into a DAC. Definitely continuous waveform. no idea of the 'costs' for this compared to the reliability ot get what you want.

typos!!!

There was a project to build an electroninc Leslie in a magazine in the '70s. I think it was Popular Electronics but I no longer have my collection. It was also sold as a kit. I assemebled a couple for someone who couldn't solder very well. It might have been one of Don Lancaster's designs, he had a lot of them published back then.

--
Anyone wanting to run for any political office in the US should have to 
have a DD214, and a honorable discharge.

fact work close enough to be unnoticeable. "

Well, it would most likely be percieved as a tweeter with a narrower disper sion I would think. Which of course, if you ask "Leslie" he would probably say that's good.

I suggest you have a look at the circuit of a Waketek function generator ma de back in the 1960s. The 100 series. I have them but you can get them easi er on BAMA. I think their design kicks f cking ass ! A buddy of mine has a

103 which I fixed, and I have a 111 which I fixed. They are actually VCOs t hat go damnear to zero, but of course have limits. But the great part of th e design is how they do it. It is a voltage, there is no AC on the pot that controls the frequency. These things, in 1965 (!!!) are state of the art a nalog today, but with those old devices.

The impressive part is that the triangle wave they generate is the first on e and the basis for all others which gives them the ability to use an enhan ced compression sscheme dependent upon transitor gains and all tis to make a sine wave. Just making a triangle wave was not all that hot. thye decided to start with that because it was the best way to a sine wave, which they did down to 0.5% distortion.

In 1965. And there ain't a coil to be found in any of them. Superb.

Look at the designs of those, and I think you have found it.

Another thing about it being voltage controlled is that yyou can load it up with a big cap to emulate the motor rotor mass...

Mwahahaha...

Sorry, that is WAVETEK

They are there on BAMA.

A triangle is pretty easy to convert to a quasi-sine (via an overdriven differential pair with feedback) if the simple triangle is not close enough. The sine approximation is about 1% THD with a bit of tweaking (lower with a *lot* of tweaking). The beauty of this approach is that it doesn't care about frequency, and you can make voltage-controlled triangles down to *very* low frequencies. This was a typical way to make VCO LFO's on old analog synths.

The problem with the stepped sine would be that when you stepped it slowly to get low frequencies, you would hear each step.

Best regards,

Bob Masta DAQARTA v7.60 Data AcQuisition And Real-Time Analysis

Scope, Spectrum, Spectrogram, Sound Level Meter Frequency Counter, Pitch Track, Pitch-to-MIDI FREE Signal Generator, DaqMusiq generator Science with your sound card!