I think I may have just had a moment of blinding insight....

[G._Hoffman]

So, as you may remember, some time ago I was working on making a Leslie/Vibrotone cabinet for myself.  It turns out that while I'm a very fine woodworker, and can stumble through the whole electronics part fairly well, the whole mechanical engineering part of a rotating speaker is more work than I expected.  If I turn on the motor for my unit, the damn thing walks across the room.

So, I go to eBay, and I find a Leslie OEM thing from some old home organ (LIKE THIS if you are really interested, but mine is a single speed and has a 10" speaker and rotor that I should be able to make work with a 12").  I'll build a cabinet around it.

Now, my only problem is I want to have programmable and switchable variable speeds.  As I search around for a way to do this with the AC motor (which is preferable to a DC motor in that they have already figured out how to damp the vibrations), I find a few uninteresting options on the market (they all use methods which are really hard on the motor, they don't offer the kind of variability I want, and they are really hard on the motor - there was one decent option by a guy who was a friend of friends, but he died a few years ago!), and the idea of a Variable Frequency Drive.  Basically, a VFD uses a PWM (Pulse Wave Modulation) circuit to drive an inverter, and by varying the output of the PWM you get both variable frequency and variable voltage (important, as the motors reactance changes with frequency, so if you don't lower the voltage the motor will over heat and you'll eventually wreck your motor).  

All well and good so far.

The thing that interests me is the following fomula:

n=120f/p

where n is the speed of the motor, p is the number of poles in the motor, and f is of course frequency.


From here, it is not all that great of a stretch to imagine being able to program a microcontroller to do the PWM, and it is then easy (at least for me, who has such things on the mind) to arrive at:




A TAP TEMPO LESLIE!!!!!!!!



 



Obviously, you would have to program in the pulley ratio, and find some way to allow for the load on the motor, and for all that it's probably never going to be as precise as you would really want, but it could be done fairly simply.

Even better, you could use some kind of optical encoder to count rotations, and get even more precise.  Though, at that point you would probably be better off using stepper motors, I suppose.


OK, I'm going to stop now, because I think my head is starting to hurt.



Gabriel

[stingray_65]

I've given MUCH thought on this as I've collected a half dozen of the style you mention plus a couple 2 speed models.

the cost and complexity of VFD's left MY head reeling!

I keep wondering if a simple variable transformer wouldn't work.

altho I love the thought of a 2 pedal controller, one for vibe speed and the other for volume (speed and intensity?)

I also have quite a few pedal assemblys from conn organs, the way they are designed they accept standard pots (3/8 bushing 1/4 shaft and 300 deg rotation) and would work well for me for this application.

do let me know what you come up with!
My AC 30 ish project just got put on to indefinite hold due tot he fact my stupidity reared it ugly head to its fullest tonight putting the Leslie's further up the ladder on the project list.

I'd also like to know more about your cabs. I havent given these a bit of thought yet figuring I'd just make them on the fly.

Thanks

[jjasilli]

Maybe google some old-fashioned ceiling fan circuits with 3 speeds.  They derive different motor speeds by switching different caps in conjunction with the motor.  I think it works something like this, with a 4 position SW:  OFF, no cap, one cap; 2 caps together (parallel or series I don't know).  This causes the motor to run at 3 specific speed settings, and off.  Presumably more caps would give more speed choices. 

I know the "cap thing" works to start motors, or to run motors at various specific speeds.  Unfortunately, I don't understand it.

[PRR]

> I know the "cap thing" works to start motors, or to run motors at various specific speeds.  Unfortunately, I don't understand it.

Caps are used in different ways.

A single-phase AC motor can spin either way, and at start-up usually "can't decide" and won't spin at all. A small cap can give it a clue. A larger cap and an added winding can give it a big kick the right way; so big that if left on, it would burn up, so there is a switch to break the cap after the motor is up to speed.

Capacitor-run motors use a cap partly for starting but also to give quasi-2-phase operation from 1-phase power, to change the torque and smoothness.

There are about a dozen ways to muck around with capacitors and AC motors. I had a 400 page book on the subject, read it, still don't understand. Richard has it, I'm sure he won't learn much practical insight neither. Wise men let motor companies figure this stuff out, and read the installation sheet.

> old-fashioned ceiling fan circuits with 3 speeds.

I'm pretty sure those had many-pole motors which could be re-rigged for different numbers of active poles. However a fan load is pretty steep, so you can "change RPM" by actually changing torque and allowing the motor to run sub-speed (most motors won't like this, torque motors are special). Without the fan-load they may rise to quite high RPM.

> n=120f/p

Yeah, but... as you say, slippage may be uncertain. And you don't want to control motor speed, you want rotor speed. Put a sensor on the rotor. Monitor that and feedback so it triggers at the tap-tempo rate.

But you don't want tap-tempo. You want it to run at playing-rate. Use a beat detector, possibly just a peak-detector, on your main or rhythm signal. Compare that to the rotor sensor. adjust to equality.

[G._Hoffman]

> n=120f/p

Yeah, but... as you say, slippage may be uncertain. And you don't want to control motor speed, you want rotor speed. Put a sensor on the rotor. Monitor that and feedback so it triggers at the tap-tempo rate.

But you don't want tap-tempo. You want it to run at playing-rate. Use a beat detector, possibly just a peak-detector, on your main or rhythm signal. Compare that to the rotor sensor. adjust to equality.

Yeah, I thought of that.  Well not so much the beat detector, but the sensor on the rotor. 

The whole tap tempo thing will probably never happen though.  I'm already working on a tap tempo trem, which should be plenty enough, and then I've got the Leslie for when tempo doesn't matter so much.


Gabriel