Reyburn CyberTuner

Innovations in Visual Tuning

(Piano Technicians Journal/January1997)
By Kent Swafford, RPT(*1)
Kansas City Chapter (PTG)

As computers have become simultaneously more powerful and more affordable, professionals have begun to carry their "offices" with them in the form of notebook computers, keeping customer and financial data at their fingertips throughout their daily travels. CyberTuner, designed for the professional piano technician, is a new visual tuning device (VTD) in software. CyberTuner is cool, definitely fun, and is giving piano technicians a unique opportunity to make a portable computer an integral part of their workdays. With CyberTuner, one can tune a piano with a stock Macintosh [or Windows] computer. (*2)(See Photo)

PowerBook 2400./RCT

Exploiting the built-in digital audio recording capability of newer Macintosh Powerbooks [and Windows laptops](*2), CyberTuner requires no expansion cards or other equipment beyond the computer itself. To some, integrating a visual tuning device into a notebook computer will seem natural and perhaps even inevitable. Wonderful new capabilities are possible in a visual tuning device that can take advantage of the speed and processing power of a modern computer. Reyburn CyberTuner (RCT), by Dean Reyburn, RPT, brings some genuine innovations to visual tuning. The program is the result of years of development effort and is a remarkable accomplishment.
Comparisons to the Sanderson Accu-Tuner (SAT)(*3) will be unavoidable. As the author of the August 1994 Journal article entitled "Frequently Asked Accu-Tuner Questions," I believe the Accu-Tuner is a very fine VTD worthy of the praise it has received from among the finest piano technicians. The SAT's compact, durable design and its proven performance are just as good as they ever were, which is very good indeed. Dean Reyburn has created a new visual tuning device which is, in my opinion, clearly worthy of a place alongside the Accu-Tuner. Happily, CyberTuner is compatible in a number of ways with the Accu-Tuner, and can actually enhance the usefulness of an SAT.
The SAT remains unsurpassed as a "dedicated" VTD, that is, a device designed from the start to be for musical instrument tuning only. Compared with using a dedicated device, the drawbacks of using an off-the-shelf computer, even a small notebook, as a visual tuning device are few, but notable. There are some vertical pianos with no place on which to put a notebook computer(*4), in which case one may need to place it to one side of the piano's keyboard. Power management will be more complex with a notebook VTD than with an SAT which can tune many pianos on a single charge of its batteries. Notebook computers can run on batteries, but only for a limited time. A charge of the standard notebook battery should easily last through any single RCT tuning and perhaps through as many as three fast tunings, but a charge may not last through two difficult tunings. Powerbook batteries can be recharged while the notebook is in use, and can apparently be fully recharged in the time it takes to do a tuning. Therefore, a conservative approach might be to plug the notebook into AC during every second tuning. Others might choose to be less conservative and plug in only when absolutely necessary. Additional batteries are available, including ones with higher capacity than the standard ones. Batteries being rather expensive, one might just decide to plug into AC on most tunings and save the battery for a tuning where no AC is available.
The benefits of using a personal computer as a VTD are many. As long as the computer program is still in development, revisions to the program can be easily obtained through email or on floppy disk. An almost unlimited number of tuning records can be stored on the notebook's hard drive with complete, automatically alphabetized names, allowing tunings to be easily recognized and loaded back into the program at any time in the future, with no need for an external printed directory. The notebook will run any other computer application available, so the piano technician can use his "visual tuning device" to do anything else a computer can do. I am writing this article on my "VTD."
Reyburn CyberTuner consists of several integrated program "modules" which perform the different functions necessary to calculate, perform, and store tunings. The visual tuning device portion of the program with which one actually tunes is called CyberEar (See Figure 1).

CyberEar

Figure 1 - CyberEar computer screen display.

The display portion of CyberEar is a spinner which not only moves clockwise to mean "sharp" and counter-clockwise to mean "flat," but the entire spinner moves toward the left of the computer screen to denote very "flat" and to the right of the screen to denote very "sharp." The CyberEar spinner display is reliable throughout the scale and, in my experience thus far, is able to read every note with equal and remarkable clarity, even in very noisy environments.
The rotational speed of the CyberEar spinner is normally an indication of how many cents off a note is, not the more familiar one revolution per second per beat. To put it simply, the spinner is sped up and more sensitive in the bass and slowed down and more stable in the treble, creating a more realistic visualization of piano string frequency across the scale. The faster spinner rate in the bass is a real eye opener, showing that bass strings can be very unstable, perhaps as unstable as high treble strings. This suggests that some bass tuning problems come from unstable strings previously undiagnosed as such. CyberTuner may help us piano tuners train our ears to hear better for bass tuning, just as the Accu-Tuner has helped many of us train our ears for high treble tuning.
Switching from note to note as one tunes with CyberEar is automatic in many instances. As long as the technician is moving up or down the keyboard in chromatic fashion, CyberEar will "automagically" (as Dean puts it) detect each move to the next note on the piano and switch itself accordingly, freeing the technician from physically switching notes on the VTD. This feature is luxurious and makes tuning from note to note smooth and seamless.
Additional and very exciting automatic functions come into play for pitch raises and lowerings. CyberEar does pitch corrections with speed and accuracy that is remarkable. In the pitch raise modes, each note as you come to it on the piano is automatically measured to determine its deviation (flat or sharp) from the value that the tuning record shows for that note. An over-pull amount is figured for the note, based on how far each note deviated from the tuning record value, how far the previous five notes deviated from their tuning record values, and where the note lies in the scale. The percentage of deviation which CyberEar uses to calculate the over-pull varies through the scale, being relatively small in the bass, more in the mid-range, and much more in the tenor and high treble(*5). Once the over-pull is calculated for an individual note, CyberEar offsets itself so that the spinner will stop when the string has been properly over-pulled, all of this happening automatically for each note in little more than the blink of an eye.
The portion of CyberTuner which takes measurements from a piano and then calculates the proper pitch for each note is called Chameleon 2. CyberTuner dramatically increases the amount of data taken from each piano. The original stretch calculator used only one measurement of just one note of a piano. FAC, the tuning calculator built into recent Accu-Tuners, improved matters by taking measurements of three notes of each piano. Chameleon 2 measures as many as 21 partials, four partials each from A1, A2, A3, and A4; three partials from A5; and, optionally, two partials from A6. Because Chameleon 2 collects so much data from the bottom to the top of the scale, Chameleon 2 tunings will be more accurately "customized" for individual pianos.

Chameleon 2

Figure 2 - Chameleon 2 computer screen display.

The procedure for collecting the data which Chameleon 2 needs to calculate its tunings is automated, freeing the technician from any "measuring." The procedure takes two minutes and requires a somewhat quieter environment than does CyberEar. The program prompts the technician to play the notes that it needs, digitally records each one, and then analyzes the recording, separating out the individual partials as they sounded simultaneously. All measurements are done internally, so the technician has no direct contact with the data collected. However, the piano technician has an unprecedented control over the amount of stretch calculated into the tunings because Chameleon 2 has ten gradations of general stretch levels, called "octave tuning styles," from which he may choose (See Figure 2).
Before CyberTuner, if one attempted to control the general stretch level by using stretch numbers higher or lower than measured, one ran the risk of introducing beat rate inconsistencies at those places in the scale where the VTD changed the partial being used. Chameleon 2 has been specifically designed to calculate wide or narrow tunings as the user chooses, and so minimizes such errors.
The concept of octave tuning styles was one of the more difficult parts of Reyburn CyberTuner for me to understand. (You might say it was a stretch for me.) Although the octave is in reality a tempered interval in piano tuning, we usually say that we tune "clean"-sounding octaves. I had not previously been accustomed to thinking about the specific speed of beat rates in octaves. But Chameleon 2 actually calculates and predicts the beat rates to the nearest tenth of a beat per second for certain octaves in its tunings! This "virtual direct-interval tuning," as Dean calls it, is possible because of the multiple partials measured for each note. The screen displays beat rate predictions for the A2-A3 octave, the A3-A4 octave, and A2-A4 double-octave. If one does not like the predicted beat rates, with a few mouse clicks one can have Chameleon 2 calculate a wider or narrower tuning. No measurements must be repeated; only the computer's calculations, which take less than a second, must be repeated.
This ability to calculate slightly wider or narrower tunings is useful for all pianos but is particularly valuable on small pianos with high inharmonicity. On these pianos, two contiguous single octaves just don't "fit" very well into a double-octave. Commonly, if octaves on these pianos were stretched, the double octaves would be too wide and would beat noticeably. It is a genuine advancement in visual tuning to be able to calculate successively wider or narrower tunings until a suitable compromise is reached between the single and double octaves.
I am very excited about the accuracy of Chameleon 2 tunings because of the specific way I like to use visual tuning devices. A VTD should provide a good overall framework for a tuning by properly adjusting for the inharmonicity of the piano, taking into consideration the general level of inharmonicity, the change in inharmonicity through the scale, and my stretching preferences. This visual tuning is a macro-tuning, if you will, fitting a tuning to the piano accurately across the scale, creating a uniform stretch throughout. Final refinement, or micro-tuning, can take place aurally. Since the visual tuning arranges one partial from each note into a uniformly smooth curve, the anomalies present in the macro-tuning can be assumed to be the result of inconsistencies in inharmonicity. These anomalies would show themselves as uneven beat rates across scaling breaks or uneven beat rates on "maverick" notes, such as notes that exhibit negative inharmonicity. After smoothing out these anomalies as much as possible, the resulting tuning may be superior because the tuning will have taken into consideration all three aspects of inharmonicity, that is, general level, change through the scale, and inconsistency.
Chameleon 2 has the flexibility to calculate its tunings so that the notes from A0 to G#1 can be tuned from their sixth, seventh, or eighth partials. Similar flexibility is available in the tenor by providing the option to tune A2 through G#4 from either the third or fourth partials, and special mention must be made here of Chameleon 2's ability to tune well in the piano's tenor section as a result.(*6)
The tenor area and the break between plain-wire and wound strings provide unique tuning challenges. VTD's must make "assumptions" about the amounts of inharmonicity a piano will have across its scale, but tenor strings sometimes conform particularly poorly to the assumptions. Generally, inharmonicity decreases as one moves down through the piano's plain-wire strings. On some pianos, however, as one moves down the scale into the tenor's lower plain-wire strings, the inharmonicity actually increases instead. It is common, particularly in small pianos, for the lowest plain-wire strings to be shorter than would be ideal, resulting in their inharmonicity being higher than ideal. Visual tuning devices commonly tune from the fourth partials in the tenor. When a VTD tunes the fourth partials to a uniform curve, notes with inharmonicity higher than expected will end up with first, second, and third partials that are too flat, with fifth partials too sharp, and with only fourth partials tuned "right." Intervals formed above notes like this will exhibit some very "wrong" beat rates. Flat second partials will form 2:1 octaves that are wide, even though the 4:2 octaves may be correct. The major thirds may be slow. So, in order to tune these notes that, because of high inharmonicity, measure both too sharp and too flat, depending on which intervals are checked, compromise is in order, but the solution will be different for different pianos, both because scales differ so widely from piano to piano, and because of the new flexibility of CyberTuner. On well-scaled pianos, it will make little difference whether notes in the tenor are tuned from their third or fourth partials. However, many piano scales are imperfect, and I am convinced that tuning the lower plain-wire strings from the third partial will by itself provide a big improvement in the tunings of many pianos. When a VTD tunes a note with unusually high inharmonicity from the third partial instead of the fourth, the first and second partials will be less flat, the third partial will be "right," and the fourth and fifth partials will be sharp. This means the 2:1 octave, while still wide, will be less so, and the 4:2 octave will be a bit narrower than if it had been tuned from the fourth partial. In other words, a VTD using the third partial to tune plain-wire tenor strings which have unusually high inharmonicity should provide a better compromise (than tuning from the fourth partial), by automatically splitting the inharmonicity-induced error between the 2:1 and 4:2 octaves. After all, a VTD that is tuning the lower plain-wire strings with the fourth partial will tune the 4:2 octaves just fine but will put the entire error caused by unexpectedly high inharmonicity on the 2:1 octave. Third partial tuning tends to produce clean fifths and octaves at the expense of the major thirds which will slow down on the lowest plain-wire strings; still, this may be the best compromise possible.
It is common for the highest wound string of the tenor to have significantly lower inharmonicity than the lowest plain-wire string right beside it. Tuning a string with unexpectedly low inharmonicity, as happens when tuning the highest wound strings of some pianos is the other tenor tuning problem. Tuning these notes from the third partial would cause the fourth partial to be unexpectedly flat, making the major thirds formed above such notes to be too fast. An extra tuning record can be prepared by Chameleon in less than five seconds. Perhaps the beat rates of such pianos could best be smoothed out across the tenor break by preparing two tuning records and tuning up through the wound strings above A2 with a tuning calculated to use the fourth partial, switching at the break to a tuning calculated to tune the plain-wire tenor strings from the third partial. This new dual-record RCT procedure appears to have clear advantages, especially on the lowest plain-wire strings, over the dual-page SAT technique that I used regularly for years to tune through the tenor break. RCT provides many new possibilities, including tuning with one partial and checking with another.
While it has certainly been fun and exciting to learn how to use a new visual tuning device, it has been a challenge, too, because of all the new possibilities that are built into RCT. The best techniques for tuning with CyberTuner will have to evolve over time.
The Sanderson Accu-Tuner is perfectly capable of storing tunings calculated by Chameleon 2 which use the third, seventh, or eighth partials. Chameleon 2 tunings are compatible with the Accu-Tuner and can be sent from CyberTuner directly to a MIDI-equipped Accu-Tuner, or entered manually into an SAT with no MIDI capability. It is worth noting the extraordinary foresight of Dr. Al Sanderson, developer of the Accu-Tuner. Because of the flexibility of the SAT's memory pages, and because of the SAT's use of the MIDI standard, the Accu-Tuner is now compatible with both a computer and a computer program not even dreamed of well over a decade ago when the SAT was originally developed.
Using a VTD and tuning from A0 up through the scale chromatically and tuning unisons on the way, yields very stable tunings. However, there can be good reasons for not wanting to tune chromatically. CyberEar allows the tuner to program CyberEar to step through any note sequence desired. Students using CyberTuner as an aid in learning aural tuning would want CyberEar to tune through an aural temperament sequence. Experienced tuners might want to step through an aural temperament sequence when learning CyberTuner to get a feel for the various levels of stretch available in Chameleon 2. And finally, one can imagine experienced users of CyberTuner using a short aural sequence at the beginning of each tuning to make sure the Chameleon 2 tuning is indeed appropriate for the piano in front of them.
There is more. Pianalyzer is another function of CyberTuner, and it measures the pitch and relative amplitude (loudness) of each partial in the partial series of any note on the piano(*7). Pianalyzer demonstrates the increase in inharmonicity as one goes up the partial series of "normal" notes, and easily documents the fact that notes with negative inharmonicity are relatively common. Besides being very neat, this function should be useful in voicing, education, and research.
I began using CyberTuner in March, 1996, and now use CyberTuner on a daily basis. There have been no disappointments; CyberTuner is "for real" and is advancing the state of the art in visual tuning devices.
Many thanks to Mitch Kiel, RPT, David Lamoreaux, RPT, Carl Lieberman, RPT, Dean Reyburn, RPT, and Don Rose, RPT, who all assisted in the preparation of this article.[PTJ]

A) foot notes
(*1) Registered Piano Technician
(*2) As RCT is continually being improved and developed, after version 3.0 was released in November, 1998, RCT has become compatible with the Microsoft Windows computers, offering a wider choice of hardware and a greater advantage in its portability and battery duration. (*3) Sanderson Accu-Tuner's reference web site URL; http://www.cris.com/~fast440/index.html
(*4) Many users place the notebook on a chair, or purchase one of the newer micro-laptops such as the Toshiba Libretto or Sony VAIO.
(*5) One can deal with the different break points between plain wire strings and wound strings for different pianos, by setting up the program accordingly for each piano.
(*6) The current version of RCT provides more flexibility. Notes from A0 to G#1 can also be tuned from their tenth and twelfth partials, and notes from A1 to G#2 can be tuned from their fourth, fifth, and sixth partials.
(*7) Pianalyzer also measures and shows the sustain for the multiple partials.

Kent Swafford, RPT is a full-time piano-tuner technician in the Kansas City area and is regional vice president of the Piano Technicians Guild. He can be reached at:

email: kswaffordearthlink.net

Reprinted by permission (c) 1997 Piano Technicians Journal, January 1997 issue.
PTJ is the Official Publication of the Piano Technicians Guild

This article reflects the opinions of the author and is not an endorsement by the PTG

CyberTuner, CyberEar, Pianalyzer, and Chameleon are trademarks of Reyburn Piano Service, Inc.
Sanderson Accu-Tuner is a trademark of Inventronics, Inc.
Apple, the Apple logo, Macintosh are registered trademark of Apple Computer Inc.

Top of page