musical thoughts

Another article on Rhodes tone quality which I originally posted at the SuperSite.

If you haven’t already, please have a look at dimensions of tone in which I explain the background of what is going on here.

In that thread, the question came up:

It would be fascinating to see a comparison of the same note, same strike dynamic, but with the tine/pickup intonation adjusted between the extremes of “more overtone” and “more fundamental,” to see visually how the harmonic content changes with this basic Rhodes adjustment. Thanks for sharing your analysis!

We should all be familiar with this, but in case you aren’t here’s the relevant figure from the service manual. Each note is separately adjustable in the manner described in the diagram.

The manual gets it right: let your ear be your guide. I just aim to help you visualize what it is you’re hearing.

The procedure I followed was to take the same note F3 that we previously analyzed and sweep the tine position through the full range of motion while repeatedly striking a note with the same dynamics (within the limits of my ability). Strikes at the beginning are in the position shown at left, and the adjustment screw was turned to advance toward a position as shown at right. That procedure was recorded and normalized.

You can listen to the change in timbre:

Visualizing this as a waveform is useful to observe how the overall amplitude changes as well with the position of the tine.

(click to enlarge this diagram)

It’s important, because psychoacoustics research has shown that people prefer a louder sound in A/B tests (speaker salesmen have used this for years to their advantage). You need to be aware of that tendency as you make your voicing adjustments–you can get suckered into a false preference just because one position is louder relative to another (and you may prefer the other if volume were equalized).

Now let’s look at the spectrum changes with tine position.

(click to enlarge this diagram)

I can equate the descriptive terms in the service manual to reasonably empirical measurements in the spectrum graph. “Pure fundamental (deep)” is represented by a lack of harmonics and presence of the fundamental. The first three strikes or so fit this description well. “More overtone (ideal)” is represented by a strong fundamental together with noticeable overtones. The 10th through 15th strikes fit this description well. “Pure overtone (shallow)” is represented by the lack of fundamental, and strong presence of the first harmonic. The 21st through 24th strikes fit this description well.

But, you can see and hear that the “quality” or overtone profile is continuously variable–there’s not a well-defined transition point between these tones, although you’ll know them when you hear them. As the tine advances past the centerline of the pickup, the tone changes back toward a fundamental sound.

Observations:
1. “ideal” turns out to be the loudest too, luckily for us. That makes you most likely to land there if you “trust your ears” while doing voicing. The strikes from around 6-7 seconds were loudest, and most ideal from an overtone perspective.
2. “ideal” is probably actually ideal if you consider what we previously discussed, that it’s easy to take overtones out with EQ, but hard to put them in.
3. The overtones we’re hunting for are concentrated in the attack, so you should repeatedly strike a note while you adjust timbre, not sustain. (You can still perceive a difference if you do sustain.)
4. If no tine position sounds good, you may need to adjust strike line or conduct some repair procedure.
5. It’s probably better to do voicing with a straight-off-the-harp signal, but at least try to minimize any coloring from EQ as you work.

Let me know if you like this kind of thing and I’ll work on some more topics. (Origianlly posted to the SuperSite on Sun Jul 22, 2007)