Hammond Organ Emulation

Prepare yourself for remake the sound of the most popular electromechanical instrument in the world. In this article we will explain you how to emulate the famous Hammond organ on your synthesizer.

The Hammond Organ is the most famous electromechanic instrument we know. It is an electric organ designed to give a low cost alternative to church pipe organs.

How it works

The Hammond Organ uses additive synthesis to generate its sound via tonewheels instead of electronic oscillators. Each of the 91 tonewheels generates one of the pitch of the fundamentals or overtones, of the many stops. Each stop can be controlled using the 9 drawbars which represents the most important harmonics, from left to right.

Pic. 1 – Click on each stop to hear how it sound.

Pitch	Interval	Note
16′	Sub Octave	Octave under root
5 1/3′	5th	Fifth over root
8′	Unison	Root
4′	8th	Octave over root
2 2/3′	12th	Octave and a fifth over root
2′	15th	Two octaves over root
1 3/5′	17th	Two octaves and a major third over root
1 1/3′	19th	Two octaves and a fifth over root
1′	22nd	Three octaves over root

The simplest way to emulate an hammond organ is to use Additive Synthesis, since almost all modern keyboards allows to stack many single sounds in a big combination. Each manufacturer may call it differently (Combi for Korg, Performance for Roland, Multi for Yamaha, Setup for Kurzweil and so on) but they do all the same thing. Now we will start to emulate all nine hammond stops, without taking care (by now) of clicks and leslie that will be covered after. This will be your master template, if your keyboard don’t have all physical controllers needed (sliders, buttons, etc).

Full Hammond Emulation

For a full nine drawbars hammond emulation we need to create nine single programs, each one tuned for a single stop. All programs must use those common parameters:

• Sine Wave as waveform (only one per program)
• Wave amplification level to 80%
• Filter wide open and no resonance (set Frequency to max value allowed)
• Amplitude Envelope with a short attack, full decay, full sustain and shortest release

Pic. 2 – Hammond Drawbars

Now you can tune each program for each stop. Use the image below as reference.

Once you’ve completed all programs, group them into a combination and play some notes.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Audio 1: Full Organ

If you have enough physical controllers, you can easily assign each controller to each program amplitude level. In this way you can control each stop as you’re playing with real drawbars. Set the volume level for first three stops (16′, 5 1/3′ and 8′) to 80% and silence all other. Now we have the standard 888000000 preset, a classic hammond sound. We will use this preset as reference.

A note on drawbars:
Bear in mind that on a real Hammond you can only push and pull the stops by increments of 1/8, with 0 being “all pushed in” (silence) and 8 being “full out”. To get this back to percentage, you should consider 8:100=1:x which results in 100/8=12,5%. If you can, set the controller to work on increments of 12,5%. It should also be said that each drawbar on a real Hammond does not exactly sound the same at different positions, meaning it’s not just a matter of waveform amplitude..but that’s way out of hand for the nice’n'easy approach.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Audio 2: 888 Preset

Vibrato and Chorus

A real Hammond organ (and 99% of the clones) has a Vibrato/Chorus knob, along with two on/off buttons. The knob is used to select which kind of Vibrato or Chorus is used. Available choices are: V1, V2, V3 – Vibrato (1,2,3) C1, C2, C3 – Chorus (1,2,3). The number generally indicates the (increasing) intensity of the effect, although everybody agrees there are more differences than just the intensity.

Explaining these is pretty simple: – vibrato is a form of pitch modulation, where the cycle of a sine waveform (different waveforms could be used but it’s generally close enough to a sine wave) affects the pitch. This translates into a regular, rhythmic, symmetric oscillation of pitch around the main value. – chorus is an another form of modulation where the same waveform gets duplicated and slightly detuned. The difference in frequency causes a sort of internal oscillation which gets perceived as an increase in “roundness”, “depth” and “spatial field” of the sound. Some people use the term “fat” for chorus-ed sounds and “thin” for non-chorused ones.

The vibrato/chorus knob affects both the lower and upper manuals. You can then choose to turn it on or off on each manual (4 combinations: none, lower only, upper only, both) by using the two vibrato/chorus on/off switches. What you cannot is actually have two different vibrato/chorus effects between the two manuals.

Creating Percussion

Now we can start to create the typical percussion sound.

Some things to remember:

1. On a real Hammond organ the percussion is only available for the upper manual, not for the lower one;
2. The percussion sound is monophonic, meaning that if you play legato you will only hear the percussion sound on the first note you play, not on the others;
3. The percussion sound has two main modes: 2nd percussion and 3rd percussion. The 2nd percussion is an octave above the note you are actually playing while the 3rd one is an octave and a fifth (a perfect 12th interval) above;
4. You can hear the percussion sound by itself when you push all the drawbars in. This is also called the “vibes” organ sound, or vibe-y sound. You probably want to play with the 2nd percussion rather than the 3rd in order not to play transposed;
5. The percussion sound can sound in different ways because of two parameters that can be switched between two modes each: the first one is the percussion decay setting which can be set on SLOW or FAST, the second one is the percussion volume, which can be set on NORMAL or SOFT. This basically means you can have 8 different percussion modes;
6. The percussion sound is monophonic but is triggered on all the notes of a cluster of notes if these notes are played simultaneously (no roll);
7. On a real Hammond organ the percussion sound uses the electric contacts used for the 1-foot drawbar (the so-called “whistler”). Some people re-wire this to use a different contact, thus allowing the 1-foot drawbar to sound;
8. When the percussion is set to on the overall organ volume generally becomes -3dB softer. I am not sure but I think this can be re-wired as well and most of the time happens only on real Hammond organs (although good clones might take this into consideration).

Duplicate 4′ program (for 2nd percussion) and 2 2/3′ program (for 3rd percussion). Modify their amplitude envelope in that way:

• Faster attack (its minimum value)
• Decay to 300ms (for FAST percussion setting) or 630ms (for SLOW percussion setting).
• Sustain and Release to zero.

Add those new made programs to your Hammond combination and set their volume to zero. Raise the volume of one of them and play some notes. This is an example with our 888 preset, 3rd percussion and fast decay.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

3. 888 Preset with fast decay 3rd percussion

Cheap rotor speakers

Now the hard work. We need to emulate a classic Leslie rotor speakers. If your synthesizer have an effect for that, use it. Anyway, here’s a cheap way to create a “rotor”.

Edit your “stops” programs and assign an LFO to panning stage. Set the LFO amount to 23%, LFO wave to Sine and Frequency to 0.81Hz. Assing a controller (Mod Wheel or similar) to LFO Frequency and set the depth in order to reach around 6Hz value (for Chorale) or above 10Hz (for Tremolo).

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

4. Cheap Leslie rotors simulation

Final Touch

Your Hammond organ is almost complete. Now you can add an equalizer, to boost bass frequencies, and an Overdrive. Feel free to experiment with those settings and share your results.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

5. Horace Silver – The Preacher

Challenging aspects of Hammond organ emulation

A little background

(The first part is a quote from the Albert-o-pedia)

“..Organ clones and software emulations have improved over time, greatly. In the ’90s, with the development of PCM sampling and its constant improvement companies and developers started to use sampling as a method to reach new heights of musical instruments emulations. Software samplers led the way to new, unprecedented results but the equation “more Mb=more quality” didn’t last long. In the last few years developers re-aligned their research from “dozens of Gigabytes” and “640 notes polyphony” to “better algorithms” for emulating different aspects of acoustics. This change of approach influenced organ clones, as well.

There is, in fact, a vast number of events that cannot be technically captured and re-created by using sampling. Sampling by definition means capturing an acoustic event into a digital equivalent (the sample). The sample can then be sent to a speaker in order to convert digital information back into physical acoustic vibrations of air (= sound). Since the content of a sample (=binary data) cannot be changed but only played back by the sampler (that is, during the playback phase, of course), it’s easy to understand that events which dynamically change or which “might” get affected by other variables are not free to “behave” and react like they would in nature, once they are captured in a sample.

One of the most famous examples on this subject is the resonance happening in an acoustic piano. When you play the piano the sound you generate vibrates and travels through the surrounding air pretty much in all directions. These vibrations affect the wooden furniture, other strings and elements inside the piano. When you press the sustain pedal down and play exactly the same thing, a lot more resonance will happen since the dampers will be lifted up and won’t push on the strings anymore, leaving them open just as it happens in a guitar or in any other string instrument (the opposite is a “muted string”). If we had to sample such a phenomenon we would have to duplicate the information into two separate groups: “pedal up” and “pedal down” groups. This was pretty much impossible with first-generation samplers because of their limited memory but it later became a common method when software samplers were created based off modern, more powerful computers (an example was Nemesys Gigasampler with the famous 1Gb GigaPiano library).

Even with this new trick up our sleeves, things are far from being realistic: hit the piano hardly with a chord you like, hold the notes down and THEN press the sustain pedal. You’ll hear resonance filling in the sound and that’s because you lifted the dampers and the open string became stimulated by the ones that where playing already (thanks to your fortissimo chord). This is impossible to achieve with sampling because, in all truth, it’s like having a mega-jukebox that pushes “play” on each sample depending on what and how we trigger it. It is not able to change things on the fly, just play it back. “but, wait.” developers said: “we can use crossfading to bring the pedal up samples down and the pedal down samples up in amplitude.” The result was not bad but still not natural as it is not what happens in a real piano.

This made people go back to their 4Mb stock piano ROM sound from their beloved keyboard from the ’80s. It just had more character and vibes than a huge amount of pre-recorded events being triggered.

When the tide went down again, developers, musicians and technicians realized sampling was not that much of a deal without physical modeling. That’s the word and it means: turning physical, natural events into mathematical expressions. For as easy as it sounds, it’s not. In theory, everything should be mathematically explainable but practice is going to give us the bad news. The world we live in is so complex and full of variables that even the most powerful calculator would go against what’s commonly called a “combinatorial explosion”..too many variables, boom…”

Back on track

Don’t worry, no developer actually gave up and they’re still here, trying to figure out the perfect equation. In the meantime, let’s see what makes our Hammond sound so beautiful and hard to replicate, so that you know your enemy in advance and either start loving it or plan things so that you can nail it down.

this section needs expansion, for the moment I can just write down a nice list and then go to bed.

* Tonewheels leakage;

Generally related (but not always) to the age and condition of the organ itself, “leakage” is term used to describe the quantity of electromagnetic information a pickup gets NOT from the tonewheel it is meant for but from others nearby.

* Tubes in the pre-amplifier;

The Hammond organ uses real tubes to drive a pre-amplified signal out to the speaker. This introduces what is commonly called “tube distortion”.

* Tubes in the rotary speaker;

Vintage and high-quality rotary speakers use small tube amplifiers (e.g. the 10W amp in Leslie amps) to amplify the input signal. Tubes are generally used, meaning: more tube distortion.

(note: “Tube Distortion” does not mean armageddon distortions. It has more to do with “saturation” and gentle waveform compression than that. Unless you crank everything to 11.)

* Age and general conditions of the organ (aka: mouse shit in your garage-parked B3);

* Horn rotation in the rotary speaker: high rotor rotation VS low rotor rotation and combinations between the two (only happens in a dual motor rotary speaker);

* Power fluctuations when more notes are played;

* Impossibility to predict where the (single or dual) horn/s will stop when you put the rotary on “brake”;

That’s all for now, there’s way more but I will try to explain the most common ones, at least the ones that organ clones tried to replicate and use to make the instrument something more than a piece of electronic devices.

Famous organs, clones and organists

Tonewheel organ clones became very popular in the last years. It should be said that – exception made for the original, real tonewheel organs – the Hammond company became the first company to develop organ clones when they stopped manufacturing tonewheel organs in the mid ’70s in favor of the – much more futuristic at that time – electronic versions. They basically became the first company to clone the original Hammond sound: they cloned themselves.

Famous Hammond organ models:
- Hammond B3
- Hammond C3
- Hammond A100

Famous Hardware organ clones:
- Hammond-Suzuki XK-1 and XK-3
- Korg CX3
- Clavia Nord Electro
- Clavia Nord C1
- Roland VK-7, V-Combo and VK-7m
- Voce V5 and V5+
- Viscount DB3

Famous Software organ clones:
- Native Instruments B4 and B4 II
- Linplug Organ 3
- Emagic EVB3
- Guido Scognamiglio’s Organized Trio and VB3
- Ultimate Sound Bank Charlie

Famous organists: (a partial list in random order)
- Jimmy Smith
- “Brother” Jack McDuff
- Jimmy McGriff
- John Medeski
- John Lord
- Joey De Francesco
- Lonnie Smith
- Sam Yahel
- Richard “Groove” Holmes

Thanks to Alberto Rizzo Schettino and Carlo Castellano for their collaboration to this article. Hammond organ big picture by Keary Ortiz, Barjack Photography.