National Music Museum Logo   National Music Museum  
Home  Collections
Virtual Tour
Calendar Gift Shop FAQ Site Index Maker Index

 

Measuring Sound:  BIAS Aids Understanding of Brass Instruments

By Sabine K. Klaus
Joe R. and Joella F. Utley Curator of Brass Instruments
with Contributions from Robert Pyle (Cambridge, Massachusetts)

When cataloging brass instruments, some important questions that need to be answered include:  What is the key of this instrument and at what pitch does it play? Do the notes speak easily and are they in tune with each other? How do the available notes compare with any desired musical scale? How high can one play on this instrument? All these questions could be answered subjectively by a player after some practice, but the results will vary according to his or her skills. Moreover, playing an historic brass instrument is not always desirable in a museum setting, where conservation takes precedence. Therefore, in order to answer such questions (particularly when faced with the task of cataloging hundreds of instruments), an objective tool is required, such as an artificial device that generates acoustical properties.

Experiments with artificial sound generation for brass instruments go back to the 1960s, when Earle Kent at the Conn Research Department developed a machine for tone analysis, nicknamed “Hot Lips Harry.” This artificial embouchure device was created to improve brass instrument design based on objective acoustical measurements. A similar device, called BIAS (Brass Instrument Analysis System), was developed in the 1980s by the Institut für Wiener Klangstil in Vienna with the goal to study objectively the unique sound of Viennese orchestras. BIAS became commercially available in the 1990s to help manufacturers improve their instruments. A similar system was also developed for bowed instruments: VIAS (Violin Analysis System).

Although developed primarily for use by instrument makers, these computer-generated acoustical measuring devices are also ideal for use in museums as they produce relatively objective results without placing an instrument under the stress of being played. For the cataloging of the Joe R. and Joella F. Utley Collection of Brass instruments, BIAS was, therefore, an obvious choice to elicit from the instruments vital information about key, pitch, playing qualities and condition.


How Does BIAS Work?

In the test setup, the instrument’s mouthpiece is attached to an apparatus that pumps air into and out of it. The apparatus used to measure the instruments in the Utley Collection injects a signal that simultaneously possesses all frequencies up to 4096 Hz in 0.5 Hz increments and measures the resulting sound pressure in the mouthpiece cup. Using rubber rings, a self-centering device and a bayonet locking ring, the mouthpiece is sealed, airtight, to the apparatus before the instrument is attached.

BIAS measuring head
 
BIAS measuring head components
 
Mouthpiece affixed to measuring head

BIAS measuring head with two sound transmission lines and one microphone.

 

BIAS measuring-head components with which the mouthpiece is attached to the apparatus.

 

Mouthpiece affixed to the measuring head.

Note:  Click on any photo above or below (except for graphs) to see an enlargement.


Acoustical Profile:
Input Impedance, Pulse Response and Harmonicity

With the help of BIAS, the acoustical profile of an instrument may be illustrated in various ways, including the input impedance, pulse response, and harmonicity.

• Input Impedance

The acoustic input impedance is the ratio of sound pressure to the oscillating air flow that produced it. To visualize measurements of the acoustic input impedance of a brass instrument at a particular frequency, one might imagine an oscillating piston pumping air in and out of the mouthpiece in a pure tone (sine wave). The microphone in the mouthpiece cup senses the sound pressure and illustrates it as a peak in the diagram.

The figure below shows the input impedance of a natural trumpet made by Wolf Magnus Ehe, Nuremberg, ca. 1720 (NMM 7410), illustrating the position and strength of the harmonics that are playable on this instrument. The player can with relative ease produce notes whose fundamental frequencies lie on (or very near) the frequencies of the impedance peaks. At moderate or loud playing levels, tone production is also aided by the presence of higher impedance peaks near the harmonics of that particular tone. For example, if the player sounds the fourth harmonic of the Ehe trumpet, the eighth and twelfth peaks will contribute in addition to the fourth peak. The closer the frequencies of the fourth, eighth, and twelfth peaks are to the ratio 1:2:3 (which is the same as 4:8:12), the greater their mutual cooperation in tone production. On a well-made instrument, the frequencies of the tall impedance peaks throughout the playing range will closely follow the harmonic series.

Input impedance of natural trumpet in E-flat by Wolf Magnus Ehe, Nuremberg, ca. 1720

Input impedance of a natural trumpet in E-flat by Wolf Magnus Ehe, Nuremberg, ca. 1720 (NMM 7410), showing which notes are easily playable and at what pitch. The higher the peaks and valleys, the easier it is to play the notes. This diagram also shows how high one can play on this trumpet; strong resonances are available up to the twelfth harmonic.*

•  Impulse Response Diagram

The player’s breath and vibrating lips inject repeated pulses of air into the mouthpiece (simulated artificially by BIAS). As each pulse arrives at the bell, some acoustical energy radiates out to the listener, but some is reflected and returns to the lips (and in the case of BIAS to the microphone), delayed by the amount of time the sound wave requires to travel back and forth within the instrument. The round-trip time is one period of the fundamental frequency and is primarily determined by the length of the instrument. The Impulse Response Diagram visualizes the reflection of the impulse at the end of the bell (at 201.9 cm on the Ehe trumpet, as charted below), thereby determining the acoustical length of an instrument.

Impulse response of natural trumpet in E-flat by Wolf Magnus Ehe, Nuremberg, ca. 1720

Above:  Impulse response of the Ehe natural trumpet (NMM 7410). The acoustical length of the trumpet, as the impulse is reflected at the end of the bell, is 201.9 cm (indicated by red arrow).

•  Harmonicity Diagram

Another means of visualizing the pitch and quality of individual notes of an instrument is the harmonicity diagram that can be derived from BIAS measurements. This diagram shows how closely the resonance peaks follow the harmonic series and therefore is an indicator as to how well the instrument is in tune and how well related harmonics support each other.

The harmonicity diagram of the Ehe natural trumpet (NMM 7410) shows that the harmonics are fairly well in tune with each other. As on most natural trumpets of the Baroque era, the frequencies of the first two or three resonance peaks lie substantially below the corresponding harmonics of the pedal note. A good player can, nevertheless, play the second and third harmonics of the instrument in tune because there are higher-frequency peaks that are close to integer multiples of the desired pitch. However, the pedal note or first harmonic is, for this reason, difficult to produce and was described as Flattergrob—meaning a fluttering and rough note—in German treatises of the seventeenth and eighteenth centuries.

Harmonicity diagram of natural trumpet in E-flat by Wolf Magnus Ehe, Nuremberg, ca. 1720


BIAS as an Investigative Tool for Historic Developments

Using BIAS to analyze an entire collection of historic and modern brass instruments, such as the Utley Collection, provides ample opportunity both to assess developments in brasswind design and evaluate statements that have been made throughout history to promote new inventions. Since the early nineteenth century, the development of brass instruments has been driven in particular by improvements in valve designs. Makers and inventors have frequently claimed that their new models homogenized the playing quality of an instrument with and without the valves.

For a lecture presented at the conference, Vienna Talk 2010, held at the Hochschule für Musik in Vienna in September 2010, acoustician Robert Pyle (Cambridge, Massachusets) and I examined the validity of such claims based on a few selected instruments from the Utley Collection.

Périnet Valves in a Modern Trumpet

Front view of NMM 6787.  Trumpet by Vincent Bach, Elkhart, Indiana, 1985.
Back view of trumpet by Vincent Bach, Elkhart, Indiana, 1985.

In a modern trumpet with Périnet valves (such as NMM 6787, a Stradivarius model trumpet in B-flat by Vincent Bach, made in Elkhart, Indiana, in 1985), the impedance diagram for the instrument, measured without the use of valves, shows sharp peaks and valleys that are regularly spaced (see graph below). No peaks stick up or fall below their neighboring peaks. This feature was defined as essential for a good instrument by the Conn research department in the 1960s.

Left:  NMM 6787. Trumpet in B-flat by Vincent Bach, Elkhart, Indiana, 1985. Stradivarius model.

Impedance diagram of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Impedance diagram of the Stradivarius model B-flat trumpet (NMM 6787), without use of the valves, showing regular, equidistantly-spaced peaks and valleys.

This profile does not change much when using the valves, except when the first and second valves are combined. In that case, the eighth harmonic falls below its neighbors. When all three valves are used simultaneously, the height of all peaks is reduced, indicating that they speak slightly less easily.

Impedance diagram for first valve of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Impedance diagram for the first valve of the Stradivarius model trumpet (NMM 6787).

Impedance diagram for the second valve of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Impedance diagram for the second valve of the Stradivarius model trumpet (NMM 6787).

Impedance diagram for first and second valve of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Impedance diagram for the first and second valve of the Stradivarius model trumpet (NMM 6787). Note that the eighth harmonic falls below the neighboring ones.

Impedance diagram for all three valves of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Impedance diagram for all three valves of the Stradivarius model trumpet (NMM 6787). All the peaks are at a lower level than in the open instrument.

Harmonicity diagram, without the valves, of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Harmonicity diagram, without valves, for the Stradivaius model trumpet (NMM 6787).

When looking at the harmonicity diagram for this trumpet, tested without the valves, we see only minor deviation from the theoretically ideal pitches. In a good instrument, this should not change much when the valves are used. This is indeed the case in the Stradivarius model trumpet (NMM 6787), except for the second harmonic which becomes increasingly flat.

Harmonicity diagram for the first valve of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Harmonicity diagram for the first valve, compared with no valves in use, for the Stradivarius model trumpet (NMM 6787).

Harmonicity diagram for the second valve of the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Harmonicity diagram for the second valve, compared with no valves in use, for the Stradivarius model trumpet (NMM 6787).

Harmonicity diagram for all three valves on the Stradivarius model trumpet by Vincent Bach, Elkhart, Indiana, 1985

Harmonicity diagram for all three valves on the Stradivarius model trumpet (NMM 6787).

The harmonicity diagram for NMM 6787 shows very little deviation from the theoretical multiples of the fundamental pitch.


Stölzel Valves

In the earliest and most simple valve type, the so-called Stölzel valve, the tubing lies in a plane. The windway enters or leaves the piston from below and proceeds through the valves in ninety- and forty-five-degree angles. Late nineteenth-century makers frequently criticized these sharp angles and regarded them to be the main defect of this design. François Périnet tried to eliminate them in 1838 by using tubular porting that substituted for Stölzel’s open passages.

In the Stölzel valve, the bore is first abruptly reduced as the windway enters the piston, because the main tubing also serves as the outer valve casing. Sudden bore enlargement occurs when air travels through the piston, due to the angular windway described above. The pulse response diagram clearly indicates such bore irregularities. In an unsigned French cornopean or cornet à pistons (NMM 6816), a sudden bore irregularity can be clearly seen in the impulse response graph (below) when the second valve is used.

 
NMM 6816.  Cornet a pistons in B-flat, France, ca. 1840.

NMM 6816. Cornopean (cornet à pistons) in B-flat, France, ca. 1840.

A bore irregularity occurs at about 85 cm when the second valve is used.

Impulse response for cornet a pistons in B-flat, France, ca. 1840.

The impedance diagram of the open instrument (NMM 6816) is less regular and the peaks are at a lower level than those of the modern trumpet discussed above.

Impedance diagram with no valves for cornet a pistons in B-flat, France, ca. 1840.

When using the second valve, the impedance diagram of the cornet à pistons (NMM 6816) becomes even less regular.

Impedance diagram with second valve for cornet a pistons in B-flat, France, ca. 1840.

Harmonicity diagram of the cornet à pistons (NMM 6816), with no valves in use.

Harmonicity diagram with no valves for cornet a pistons in B-flat, France, ca. 1840.

Harmonicity diagram for the first valve of the cornet à pistons (NMM 6816), compared with no valves in use.

Harmonicity diagram with first valve for cornet a pistons in B-flat, France, ca. 1840.

Harmonicity diagram for the second valve of the cornet à pistons (NMM 6816), compared with no valves in use.

Harmonicity diagram with second valve for cornet a pistons in B-flat, France, ca. 1840.

Harmonicity diagram of the cornet à pistons (NMM 6816), showing how the combined use of the two valves versus the use of no valves negatively influences the intonation of certain harmonics. Intonation clearly worsens as valves are pressed down, suggesting that this early valve type indeed had some problems.

Harmonicity diagram with both first and second valve for cornet a pistons in B-flat, France, ca. 1840.


Double-Piston Valves

NMM 7077.  Trumpet in G, Austria (possibly Vienna), ca. 1840.
Back view of trumpet in G, Austria (possibly Vienna), ca. 1840.

A valve type of great longevity is the double-piston valve or “Vienna valve.” NMM 7077 is a typical Viennese or Austrian trumpet that resembles Leopold Uhlmann’s 1830 patent in major details.

NMM 7077. Trumpet in G, Austria (possibly Vienna), ca. 1840.

Double-piston valves on trumpet in G, Austria, ca. 1840

In the double-piston valve, the wind channel goes straight through the valves and the leadpipe is very short. In the impulse response for this particular trumpet (NMM 7077), the location of the valve cluster at only about 90 mm along the tubing can be observed in irregularities at the beginning of the curve, indicating inconsistencies in the bore diameter.

Impulse response for trumpet in G, Austria, ca. 1840

The impedance diagram is more regular than that of the Stölzel valve cornet à pistons (NMM 6816), but some of the peaks are not as sharp, providing the player with less support for locking into certain harmonics. This same feature, on the other hand, allows for more flexibility in bending the notes.

Impedance diagram for trumpet in G, Austria, ca. 1840

Introducing the second valve seems to improve this instrument with regard to the definition and sharpness of the peaks. One can conclude, therefore, that the valves do not worsen, but rather improve the quality of some notes (by locking into their harmonics) in this particular trumpet.

Impedance diagram for second valve of trumpet in G, Austria, ca. 1840


Berlin Valves

Front view of NMM 6861.  Cornet in B-flat, Prussia, ca. 1850/60.
Back view of Prussian cornet (NMM 6861).

Wilhelm Wieprecht’s (1802–1872) innovation, known as the Berlin valve, was explicitly aimed at eliminating the right angles of the Stölzel valve. He claimed in his patent petition of 1833 that in this design “the windway goes only in circles . . .  and the tone has therefore the original effect, . . . .” Wieprecht’s earliest instruments feature valve loops that are literally circular--a definite disadvantage in that they could not be fitted with valve slides. He later changed this configuration and developed a whole family of instruments with a uniform, barely flaring bell. The treble size of this family of instruments became known as the Prussian cornet, of which NMM 6861 is a typical example.

Left:  NMM 6861. Cornet in B-flat, Prussia, ca. 1850/60.

Impedance diagram without valves for the cornet in B-flat, Prussia, ca. 1850/60

The impedance diagram for NMM 6861, without valves, shows fairly regular peaks and valleys.

This instrument has regular impedance diagrams that do not change much with the use of the valves, suggesting that Wieprecht’s claim for homogeneity had some justification.

Impedance diagram with first valve for the cornet in B-flat, Prussia, ca. 1850/60

Impedance diagram for NMM 6861 with first valve only.

Impedance diagram with second valve for the cornet in B-flat, Prussia, ca. 1850/60

Impedance diagram for NMM 6861 with second valve only.

Impedance diagram with all three valves for the cornet in B-flat, Prussia, ca. 1850/60

Impedance diagram for NMM 6861 remains regular when all the valves are used.


Disc Valves

The most important English contribution to the development of valves was the so-called disc valve. In 1851, the London maker John August Köhler described the merits of this valve design with the following words:  “I entirely do away with the angularity both in the complementary and main windways, and I obtain, consequently, a fullness, clearness, and brilliancy in tone which surpasses that of all other valve instruments.” Today, conflicting views as to the quality of the disc valve exist:  Edward Tarr opines in his New Grove article that “this type of valve, with one disc rotating against another generated too much friction to work rapidly enough, and it never gained acceptance.” Crispian Steele-Perkins, on the other hand, noted “that a well-regulated disc-valve cornet responds well and has a big sound.”

The condition of many museum instruments, for which maintaining playability is not a priority, can sometimes give us a clue as to the durability of a design. The BIAS measurements of one of the Utley Collection’s disc-valve cornets (NMM 7063) gives some indication as to its present condition.

Right: NMM 7063. Cornopean by John August Köhler, London, ca. 1843, with disc valves.

Back view of cornopean by John Koehler, London, ca. 1843.
Front view of NMM 7063.  Cornopean by John Koehler, London, ca. 1843.

The open instrument shows a fairly regular impedance curve, and this is pretty much maintained when the second valve is used, indicating that Köhler’s claim that the smooth windway in his disc valve improved the sound may be correct.

Impedance diagram without valves for cornopean by John August Koehler, London, ca. 1843.

The impedance diagram of the cornopean (NMM 7063) with the second valve in use is fairly regular.

Impedance diagram with second valve for cornopean by John August Koehler, London, ca. 1843.

The impedance diagram of NMM 7063 with the first valve in use results in a very irregular curve on a very low level, which shows considerable disturbance.

Impedance diagram with first valve for cornopean by John August Koehler, London, ca. 1843.

Leakiness

The poor performance of the first valve in the Köhler cornet is due primarily to air leakage. The air tightness of this instrument normally amounts to only about 27%. However, when the two rotating discs are pressed together manually, the air-tightness improves to a respectable 95%.

The chart below shows the air tightness of the other instruments discussed here, as well as a modern rotary valve trumpet by Helmut Gantner, Munich, 1987 (NMM 6902). For both the modern trumpets—the Périnet and rotary valve trumpet—the air-tightness is one-hundred percent, whereas it is only 50% for the Stölzel valve, which explains its poorer performance.

Air tightness for all instruments under discussion.


Conclusion

This brief study allows only a glimpse into how valve designs can be evaluated using BIAS. Other factors, such as the overall bore and bell design, were not taken into account, and they, too, contribute to the quality and character of an instrument. This study seems to confirm, however, that eliminating sharp angles in the windway—and thereby bore irregularities and possible leakages—may improve the quality of a valve brass instrument.

* The BIAS graphs shown in this article were all generated by Robert Pyle.

Return to NMM Newsletter Index (December 2010)

National Music Museum
The University of South Dakota
414 East Clark Street
Vermillion, SD 57069

©National Music Museum, 2010
Most recent update: January 3, 2011

The University of South Dakota
Return to Top of Page