RECORDING the Hammond Organ

One of the most important improvements that we can make to the sound of many electronic organs is to destroy their precise tuning. At first this might seem strange, but it makes a huge improvement. In real-life musical situations, such as groups of singers, bands, orchestras, pianos, pipe organs; when you have a number of different musicians or organ pipes or piano strings all producing the same note, each one has a slightly different frequency, because in real-life, absolute tuning precision is unachievable; sort of like trying to locate the beginning or end of a rainbow. And yet, as long as tuning is very close, the discrepancy or lack of super accurate precision not only doesn’t sound bad, but it actually adds life, motion, spaciousness and interest to the overall sound, which is why choirs of singers, symphony orchestras and other normal sources of music sound so good.

But in a typical electronic organ, no matter how many stops you have on, if you play for instance, the note A440, in most cases, they all originate from the same A440 tone source and must therefore by definition be in exact tune. [The same of course is true for all of the other notes as well.]

In traditional Hammond organs, where the harmonics of some tones are actually other notes out of the tempered scale, tuning perfection is even greater. Although seemingly an ideal to strive for, tuning perfection makes such music sound uninteresting and boring unless we do something else to it; which is probably why invariably music played on a traditional Hammond usually has vibrato added, and why Leslie speakers are so desirable.

However, by digital signal processing, we can take a signal from a recorder track or a mixer channel [or even from the instrument itself] and very slightly alter its frequency. We make such a subtle change that the result does not sound out of tune in the least; and yet, when you add the slightly pitch-shifted results with the direct sound from the instrument, all of a sudden….the tuning precision has been destroyed. Now the sound takes on a lot more interest. It has a slight unsteadiness to it that is very subtle. It almost even appears to move around the room a little. The result, if done correctly, is very subtle. It does not sound out of tune at all, but it adds considerable life and interest to the tone.

When we do a typical digital pitch shift, the pitch difference is proportional to frequency, so that the rate of motion or “beat” to use the technical term, although slight, is different for every single note. Most people who hear the result have no idea what was done, nor do they care. But the invariable reaction is one of praise, or sometimes outright wonder why what they hear sounds better than what they are used to hearing, particularly if they own an instrument and play, but are not aware of what is possible. For a more detailed explanation, see the page about heterodyne vibrato in our article on vibrato.

Very slight pitch shifting is one of the effects that DSP can do, so we may add this to certain specific tracks on either a mixer or multi-track recorder.

We’ve already mentioned reverberation. Reverberation is the prolongation of a sound after the source of sound has stopped. It’s caused by multiple, complex reflections of sound from walls, ceilings and even some objects in a typical room or hall. Reverberation time is measured by how long it takes the reverb level to drop to 60 decibels below the original sound, which is referred to as RT60.

Reverberation should not be confused with ECHO, which is a single distinct repeat of a sound from a specific reflecting surface. Reverb is an infinitely complex combination of individual echoes in a room. It is possible in certain types of rooms or enclosures to get a distinct echo, sometimes even several successively quieter repetitions. This effect is very noticeable under the arches of many stone-arch bridges, and also in some rooms with a domed ceiling if you stand in exactly the right “sweet spot” under the dome. The echo effect is easily produced both by tape recorders (tape echo) and also by digital signal processing.

The graph on this page, figure eleven, shows two very different types of reverb. Both decay exponentially, but the first starts out just 10 dB lower than the original sound and then decays very quickly, hitting RT60 in just over 0.3 second. The second reverb is much less prominent, starting out already at 40 dB below reference but lasting almost a second and a half before hitting RT60. These are two very different reverb types; between the two an infinite variety is possible.

Digital reverb can also be adjusted to affect different frequencies differently. A short, fast, reverb like the one represented by the black curve could be useful if set to emphasize low frequencies to add a little pseudo pedal sustain to a traditional Hammond, although you’d have to experiment with it to make sure that it would not sustain so long that it could blur a rhythm pattern or make the pedal sound like thunder in the background. If set to emphasize higher frequencies, this reverb would add presence and emphasis to short, fast percussive sounds, such as that of a xylophone.

The above illustrates the importance of using mixers and/or multi-track recording, where we can add specific types of reverb to different sounds, rather than relying on one compromise reverb to affect everything as was typical of most analog electronic organs with their various spring type delay units, some of which were quite good and others almost useless.