Electric Organ

BasslineBoss

I think I stumbled onto a way to get that “moving air” feel from an electric organ without a Leslie, and I’m buzzing to see if anyone else has tried something like this or can help refine it. I routed my organ through a simple 2-way crossover: lows to a small sealed sub tucked behind the bench, highs to two tiny nearfield speakers mounted on a silent, belt-driven cake turntable with curved foam baffles in front of each speaker. The baffles aren’t horns, just shallow S-shaped deflectors. With the rotor creeping at “chorale” and then ramping to “tremolo,” the sensation in the room actually feels like air is moving, even at neighbor-friendly volumes. It also changes how I play, which is the magic bit I’ve always missed with pedal-only rotary sims.

What I’ve tested so far:

850 Hz crossover point sounds most Leslie-like here; 600 Hz was muddy, 1.2 kHz got phasey.
A gimbal BLDC motor with sine-drive ESC is dead quiet and ramps smoothly; steppers microstepped were okay but had a faint tick at certain speeds.
A 3D-printed lazy Susan with rubber isolation grommets killed most cabinet buzz.
Recording with a single mono mic a meter away gave that classic “folding image” when blended with a bit of close DI.

Questions I’d love input on:

Is there a known “sweet spot” crossover for this kind of physical rotary proxy? I’m hovering around 700-900 Hz, but would love measured opinions.
Any rules of thumb for the baffle geometry to mimic the HF horn vs LF drum? My shallow S-baffles seem to throw highs nicely, but I’m guessing surface curvature matters a lot.
Accel/brake curves: what time constants feel right to you for 122 vs 147 behavior? I’m around 4-5 seconds up to fast, 6-7 seconds down to slow, but not sure if that’s psychoacoustically ideal.
How are you taming Doppler vs amplitude modulation in DIY builds? Mic distance and directivity help, but I’m curious about diffuser materials or asymmetrical baffles to bias the effect.
For live use, would dual counter-rotating baffles (lighter mass, lower speed) be more convincing than one heavier rotor? I tried counter-rotating at half speed and it got ultra-wide but maybe too chorusy.
Best way to integrate expression so that speed ramps are linked to pedal position without feeling gimmicky? I’m considering a “hold at chorale unless pedal >60%, then ramp to tremolo” mode with a long brake tail.
Any EMI gotchas with BLDC controllers around vintage tonewheel preamps? I shielded and grounded the motor assembly, but I’d love proven layouts to keep hash out of the preamp.

If anyone has measured polar plots of a Leslie horn/drum or has built a “quiet rotor” rig, I’d be thrilled to see geometry, materials, and speed profiles that worked. If this can get 80% of the vibe at 20% of the volume, it could be a game-changer for apartment-dwelling organ nerds.

guitargenius

This is awesome. You’re basically recreating the psychoacoustic bits that sims gloss over. A few hard-won notes from measuring/tearing down 122/147s and building a quiet rotor:

Crossover: 800 Hz is the classic Leslie handoff (12 dB/oct). Your 700-900 window is right. If you want a more “swimmy” overlap like the real thing, use 12 dB; if you want cleaner phase and more focus, try LR4 at 800 Hz. Sometimes nudging HP up ₅₀ Hz vs LP helps the “whoosh” without mud.
Baffles: For HF “horn-ish” throw, a shallow S with radius ≈ 1-1.5× driver diameter and a 25-35 mm round-over on the leading edge works. Add felt on the trailing half to tame rear radiation and bias AM. Make the two baffles slightly different (10-15% size/curvature) to avoid a static comb. A tiny lip (10-15°) on the leading edge increases AM without harshness.
Speeds/ramps: Targets that feel 122/147-ish: horn ≈ 48 rpm slow / 400 rpm fast; drum ≈ 40 / 340. Ramps: horn up 2-3 s, down 1.5-2.5 s; drum up 5-6 s, down 6-7.5 s. Use exponential (or S) curves, not linear.
Doppler vs AM: Closer mic = deeper AM, less Doppler; 0.8-1.2 m is the “folding image” sweet spot, slightly off-axis. Perforated diffuser (30-40% open area) in front of the HF baffle adds Doppler while softening level swings. Asymmetry and some absorptive foam on the rear edge help a ton.
Counter-rotating: Contra motion feels wider, but keep speed ratio near 1:1 with a tiny offset (<5%) or it gets chorusy. One light HF rotor usually reads more “Leslie” than two at half speed.
Expression linking: Add hysteresis and dwell. Example: go fast above 60% for >200 ms; fall back below 45%. Map pedal delta to ramp rate (bigger shove = snappier accel), not just a binary toggle. A long brake tail on fast→slow feels authentic.
EMI: Put the ESC in a metal box, twist and shield the three phase leads, ferrite each phase at the motor, and run PWM >25 kHz. Star-ground the motor chassis separately from the preamp, keep 20+ cm spacing, and if possible balance the audio run or isolate it with a small line transformer. LC snub (≈10 µH + 1 µF-to-chassis per phase) helps kill hash.

If you ever post a clip and a pic of those S-baffles, I bet a bunch of us will try cloning this.

rockinrhapsody

A few extra tweaks that have helped here without changing your core concept:

Crossover: on real 122/147 networks I’ve measured the electrical handoff lands ₇₈₀-820 Hz but shifts with load. If you use LR4 at 800 Hz, add 1-2 ms of HF delay to align acoustic centers; it noticeably tightens the image.
HF baffle voicing: make it mildly cardioid. A shallow rear slot stuffed with felt (resistive port, ₃₀-40% open area) plus a 20-30 mm round‑over on the leading edge biases AM forward and reduces “phasey” off‑axis wash.
Ramps: use jerk‑limited S‑curves with different asymmetry per rotor. Model belt slip by capping accel torque and adding a tiny static→dynamic friction step; you’ll get that “lazy drum, eager horn” feel without changing times.
Counter‑rotating: keep speeds nearly equal with a slow random phase dither (±10-15° at 0.1-0.2 Hz). It preserves width while avoiding obvious chorus.
EMI: hall‑sensored FOC, PWM >25 kHz, ESC in a bonded metal can, twisted/shielded phase leads with ferrites at both ends, and a common‑mode choke on the DC feed. Star ground the motor chassis separately from the preamp.
Mic’ing: for the “folding image” in mono, an M/S pair 0.8-1.0 m out, 15-20° above rotor plane, then blend a touch of DI. Live, send M/S mid to FOH mono and keep side local to your cab to avoid phase grief.