Adding a Rotary Speaker to a Hammond M-111 Organ

December 29, 2008

Early in 2008, I was bitten by the Hammond organ bug, resulting in the acquisition of a 1962 Hammond M-111 spinet. Like all Hammonds of this vintage, the M-111 features an electro-mechanical tone generator, harmonic drawbars, and a vacuum tube amplifier.

Top-end Hammond console organs like the B-3 and C-3 did not have built-in speakers, requiring an external amplifier and speaker in order to produce any sound. Hammond sold suitable tone cabinets (containing the amplifier and speakers) for use with their organs, but many Hammond owners elected to purchase a speaker from the Leslie company instead. What set the Leslie speaker apart from Hammond’s offerings was the rotating treble horn and bass deflector instead of the fixed speakers in the Hammond tone cabinets. This rotating motion brought the sound alive and is now an integral part of the “Hammond sound”.

It is possible to disable the built-in speakers of an M-100 series organ and add an appropriate connector for a Leslie speaker, but I decided against this because it would have meant finding a Leslie speaker and figuring out where to put it.

Rewind to my childhood, when I took organ lessons for a few years and my parents bought me a brand new Yamaha BK-20B organ. I stopped playing by the early 80s, but my parents kept the organ and gave it to us when my wife and I moved into our house. It sat in the living room for a while, but I rarely played it so it was relegated to the basement. The Yamaha was a great organ feature-wise, but it just doesn’t sound like a Hammond. One thing it did have going for it is a rotary midrange/treble speaker. Unlike a Leslie where the speaker driver is fixed and plays through a rotating horn, the entire Yamaha speaker actually rotates.

Most Leslie speakers rotated both the treble and bass components of the sound. The treble played through the aforementioned rotating horn, whereas the bass was produced by a large fixed speaker playing downward into a rotating drum containing a deflector and having an opening on one side. The two rotating components turned in opposite directions. However, some Leslie models had only a rotating treble horn and fixed bass speakers.

I finally decided that I would decommission the Yamaha and transplant the rotary speaker into the Hammond, hoping to produce a somewhat Leslie-like sound.

Mechanical Aspects

The Yamaha organ is a model of modularity. The entire rotary speaker assembly and its control circuitry is mounted on a piece of 3/4-inch (19mm) plywood. The control circuitry is connected to the rest of the organ with a 6-pin connector, while the speaker has its own 2-pin connector.

Removing the assembly from the BK-20B was a matter of unplugging a few connections, undoing a few cable ties, and removing enough bolts.

The next step was to determine how best to fit the assembly into the Hammond. My initial inclination was to install it in the same way as it was in the Yamaha, rotating in the vertical plane and playing through a grill in the side of the organ. There wasn’t enough room to make this work though, and it would have required cutting a hole in the side of the organ. Furthermore, part of what makes a Leslie sound like it does is the motion of the sound around the room. Having the speaker rotate vertically would not give the same effect.

Making Room

I came to the conclusion that the only way this rotary speaker was going to fit was horizontally. To make that work, it would be necessary to remove the AO-35 reverb amplifier, reverb spring tank, and 8-inch reverb speaker. I did not hesitate to do this because the reverb wasn’t working, and the room that the organ is in provides very nice natural reverberation. Removing the reverb also freed up two tab switches to control the rotary speaker.

It proved necessary to relocate one 12-inch main speaker to the opposite side of the organ where the 8-inch reverb speaker used to be. I was able to cut a new 12-inch opening without removing the speaker grill cloth. By wedging some pencils between the grill panel and the cloth, I could get the cloth away from the panel and then use a router to cut the enlarged openings (making several passes). With the cut-outs completed, I thoroughly cleaned everything and then made the grill cloth black using upholstery spray paint.

With the reverb amplifier out of the way and the left hand (as viewed from the front) main speaker relocated to the right, the entire left hand side of the M-111 was now empty. The next step was to construct a hardwood frame to which to attach the Yamaha speaker assembly. This frame was securely fastened to the floor and tone generator shelf of the organ.

The final step was to attach the speaker assembly to the back of the frame with wood screws. After I did this and reinstalled the grill panel, I found that the speaker barely grazed the edge of the opening in the panel. I had to add some shims between the speaker assembly and frame to move it back about 1/4 inch.

Electronics

There were two problems to solve in the electronics department: switching the sound between main-only and main-plus-rotary speakers, and controlling the speed of rotation. Here is a schematic of the circuit I designed and built to meet these requirements:

The parts of the schematic in the shaded area are all on a circuit board.

Both the audio and speaker speed control switching is done by relays. RELAY1 is a 4PDT relay (only 3 of the 4 parts are used) and controls whether or not the rotary speaker is to be used. RELAY2 is a SPDT relay (although I used one half of a DPDT relay instead) and selects the rotation speed.

Notice that there are three electrically separate circuits here as shown by the magenta, blue, and red connections. The relays isolate each part of the circuit from the others. Under no circumstances should the grounds of each part be interconnected, since they are at different AC potentials (don’t ask me how I figured this out).

Audio Switching

A real Leslie speaker has rotating bass and treble components, with the sound divided between them by a passive 2nd order (12dB per octave) crossover network at about 800Hz. In order to get as close as possible to a Leslie-like sound I designed a similar network, taking into account the different speaker and amplifier impedance compared to a Leslie speaker. I compared the performance of my network and that in a Leslie speaker using the MapleSim physical modeling software.

When a real Leslie speaker is stopped, the two parts could be facing in any direction, but since the cabinet has openings on all sides, sound will still come out. In my M-111, the speaker could come to a stop facing the side of the organ or the back of the speaker assembly, which would mute the treble sounds. Therefore, I decided that the sound should all come through the main speakers when the rotary speaker is stopped.

The magenta circuit takes care of audio switching and speaker crossover. When RELAY1 is de-energized (as shown), the output of the AO-29 amplifier is connected directly to the main speakers. When RELAY1 is turned on (via the ROTARY ON) switch, then the AO-29 output is re-routed through a crossover network consisting of C1, L1, C2, and L2.

The output to the Yamaha speaker is made to the black and white wires from the speaker assembly. These wires are terminated by a two-pin connector, and I wired the matching connector from the organ into my circuit. Notice that the Yamaha speaker is connected to the crossover with reversed polarity (white speaker lead to black amplifier output lead). This corrects for the 180° phase difference between the two outputs of a second-order crossover.

Speed Control

The blue blue part of the circuit performs speed control switching. The Yamaha control circuit has a very simple interface. A single speed control input expects a voltage between 0 and 6V, resulting in rotation between about 0 and 7 revolutions per second. There is a fixed output that provides the appropriate voltage for slow (chorale) speed, and there are 6V and GND outputs from which one can derive the appropriate voltage for fast (tremolo) speed. In the Yamaha organ, this voltage could be varied by a knob, but I elected to have a single fast speed. Resistors R1 and R2 produce a voltage at their junction to give about 6.7 revolutions per second.

When RELAY1 is off, the speed control input is connected to the Yamaha circuit’s ground output, causing the speaker to remain stationary. When RELAY1 is turned on, the speed control input is connected to RELAY2, which selects between the two voltages. The chorale voltage is selected when RELAY2 is off and the tremolo voltage when RELAY2 is on.

All of the connections to the Yamaha speaker unit are made via its 6-pin connector and the matching connector and wiring harness taken from the organ. Four of the terminals (violet, red, white, and orange wires) are the speed control connections. The other two terminals (yellow and green wires) must be connected to 120VAC. I wired them to the black and blue wires that originally supplied power to the AO-35 reverb amplifier, ensuring that the speaker assembly only receives power when the organ is turned on.

Power Supply and Relay Control

The relays require power to turn on. I originally considered adding a small power supply within the organ for this, but then noticed that there was already 6VAC available from the main transformer. This powers the tube heater filaments in the amplifier, and the front panel power indicator lamp.

The front panel lamp consumes about 150mA of current. By replacing the lamp with a high-brightness orange LED which requires only 15mA, I could make that current available to operate the relays. The relays consume only 40mA each, so plenty of power was available.

The parts of the circuit shown in red constitute the power supply, switching of the relay coils, and the new front panel LED.

Circuit Construction and Installation

I usually design and etch a printed circuit board for my projects, but since this was a “one-of” and compactness was not important, I opted to use a pre-etched strip-board and hand wire the circuit.

Do not use the photos as a wiring guide if you plan to duplicate this project, since the photos above were taken when there were still some wiring errors in the circuit.

The largest parts of the circuit are the crossover air-core inductors and polyester-film capacitors. It would have been possible to use smaller ferrite-core coils and non-polarized electrolytic capacitors, but general wisdom suggests that these are less than optimal for audio applications (although that is probably not critical in a lo-fi application like this).

I installed the completed circuit board on the side of the organ just above the AO-29 main amplifier. All the connections to the amplifier (audio and power), speakers (main and rotary), and speaker control used existing connectors and wiring harnesses (already in the M-111 or taken from the BK-20B). The connections to the switch panel and power LED were made using a multi-conductor cable and a screw-type terminal strip. The cable was routed up to the front panel, where the existing wires and resistors connected to the reverb switches were removed and the new wires connected.

I removed the power indicator lamp from its socket and cut off the wires to the socket. These wires now supply power to my circuit instead. I made a very small circuit board containing just the LED and its current limiting resistor, and used double-sided foam tape to attach this to the lamp socket. The 5V power wires from the main circuit board were then soldered to the board.

Finally, I used a Brother P-Touch label printer to relabel the former reverb switches and panel area. The white tape with black lettering is a bit whiter than the switch tabs (perhaps due to age) but doesn’t look out of place. The black tape with gold lettering is an almost perfect match for the panel. I used FRANKLIN GOTHIC MEDIUM CONDENSED for the labels, which matched the existing switch and panel labels almost exactly.

How Does It Sound?

Very good, I think. I have to admit I’ve never heard a Hammond organ played live through a real Leslie speaker, so I only know what it sounds like on recordings. I’m probably not as discerning as most Hammond afficionados (yet) either, but I like the way it sounds.

Here is a brief recording that I made. My musical talents are nothing to write home about yet, so I’m just playing a C chord (G-C-E inversion on the C above middle C). All the drawbars are out, and for the first 6 seconds, the sound is coming through the main speakers only. I then switch to the combined main and rotary speaker at chorale speed. At about the 16 second mark, I switch to tremolo speed and you can hear the speaker speeding up. At about 24 seconds, I switch back to chorale.

References

This project required a fair bit of research. Fortunately, there’s a wealth of Hammond information available on the Internet. Some of the sources that I used include The Hammond Wiki for general information such as Leslie speeds, and Captain Foldback’s Hammond and Leslie Page for Hammond and Leslie schematics.

If you want to make a modification like this to your own Hammond organ, the hardest thing will be finding a Yamaha rotary speaker assembly. Yamaha used these in a number of organs in the 70s and 80s, although I don’t know which ones (besides the BK-20 series). Many non-Hammond spinet organs of that era are a dime-a-dozen these days, so you should be able to obtain a rotary speaker cheap or for free. Yamaha also used the same speakers in their RA-50, RA-100, and RA-200 guitar amplifiers (containing one, two, and three rotary speakers respectively), but I suspect these will not be as cheaply available, and it would be a shame to dismantle one.

                    

Related Articles

If you've found this article useful, you may also be interested in:

• Rebuilding a Hammond AO-29 Amplifier from the Ground Up
• Hammond Organ Tonewheel Generator Capacitor Replacement and Calibration
• Overhauling the AO-29 Amplifier in the Hammond M-100 Series
• Window Seat Bookcase Tone Cabinets for a Hammond Organ
• Retronome – A Versatile Analog Drum Machine for My Hammond Organ
• Overhauling and Improving the Hammond M-100 Series Vibrato System