[Nederlandstalige versie]

<Ribby>


No string gives an adequate sound if not coupled to a resonator of some sort. This truth is caused by the fact that a string never has enough surface area to set the surrounding air into substantial vibration directly. Therefore, in string instruments the vibrations of the string are transmitted to a soundboard with a large surface and/or a resonator. If we could only achieve to increase the surface of the string itself, it would be possible to obtain a well sounding string without recourse to further acoustic amplification techniques. This insight has led us to the idea of designing an instrument using ribbons instead of circular strings. If the surface of the ribbon is only large enough, we would thus obtain a string instrument without soundboard. Although we played around with this idea for a long time, and actually used ribbons under very low tensions in electromechanical instruments in our <Hex> project (1980) , only in 2010 we started the practical design of an instrument along these lines. The very first problem to tackle was how to bring the ribbon under sufficient tension. It will be clear that standard tuning pegs or screws are of no use in this case. Bending the ribbon over a cilinder invariably breaks the ribbon as soon as appreciable tension is applied. Therefore we designed a flat clamping system sliding on a solid frame and adjustable with a trapezoidal thread.

We got the ribbon under high tension this way, but the problem was that the slightest mismatch between the forces excerted on both sides leads to the ribbon sounding a quite distinct interval clearly demonstrated by striking the ribbon at alternating edges. Obviously proper resonance and a full sound could not be achieved this way, although as a proof of concept it seemed to work quite well: we got it to sound pretty loud without using any kind of soundboard. This called for a solution to make it possible to remove all torsion components in the excerted force. Either two separate trapezoidal threads or a sideways adjustment came up as solutions.

After these experiments and their evaluation we thought that considering to make an instrument with one ribbon for each note, although imaginable and possible leading to a much improved 'piano' design, would call for a factory to make us the required cast iron frame and the tuning mechanisms. The costs involved appeared unsurmountable. Then the idea came up to design the instrument as a monochord -a few of the same could be combined to an extended and even polyphonic instrument. In this case different pitched could be played by using a slider over the ribbon, as used on the slide guitar. This slider could than be driven with a long linear motor such that on a single ribbon, an ambitus of about two octaves would become possible.

As to the excitation of the ribbon, we considered and experimented with the possibilities: striking with a piano-like hammer mechanism, bowing, or, applying the e-bow technology as we developped it earlier for our <aeio> robot. The experiments made clear also that a damping mechanism would be a musical requirement. Without damping, all notes would be sounding with a nasty portamento, Hawaian guitar style...

Midi implementation and mapping:

• sizes: height: 1400 mm, depth 200mm, width: 200mm. (first estimate)
• weight: estimated ca. 60 kg
• power: 235V / 200W
• tuning: strictly required before playing. The instrument can be used for tuning since it has built in references. The GMT-ribby test code can be used for easy tuning.
• ribbon-string: Hasberg spring steel, width 25mm - thickness 0.2 mm
• Ambitus: to be determined
• Loudness level: to be determined. (tentatively 0 - 90 dBA)
• Insurance value: 15.000 Euro (first estimate)

• Johannes Taelman (firmware PIC controllers)
• Troy Rogers (mechanical and acoustic research, test software and lookup measurement)

Back to Logos-Projects page : projects.html

Back to Main Logos page:index.html

To Godfried-Willem Raes personal homepage...

To Instrument catalogue

te doen.

Omdat ons vaak wordt gevraagd hoeveel werk en tijd kruipt in, en nodig is voor, het bouwen en ontwikkelen van een muzikale robot, hebben we ook voor <Ribby> een beknopt bouwdagboek bijgehouden. Omdat we de bouw tot in de laatste details graag illustreren, kan het ook voor anderen die ons op dit pad willen volgen en/of verbeteren, van praktisch nut zijn.

• 01.12.2009: First design sketches for applications of long-throw high force linear motors.
• 09.03.2010: Start of an experimental setup for bringing steel ribbons under sufficient tension to sound like strings.
• 11.03.2010: Proof of concept worked out with Troy Rogers. A Hasberg ribbon, thickness 0.2mm and length ca. 110cm was brought to sound. Striking, bowing as well as e-bowing were tried out. A steel slide works well to obtain different pitches. The sound when bowed is very good.
• 12.03.2010: Further experiments into getting the force equal over the width of the ribbon , by Troy Rogers.
• 15.03.2010: Project postponed.

by Godfried-Willem Raes

Further reading on this topic (some in dutch):

• Berekening van snaren
• Strijken van snaren
• Ebows en feedback snaarbesturingen
• Elektromagnetisch aangestreken snaren
• Propagatie van trillingen in isotrope en anisotrope media
• Omzettingtabel midi noten naar frekwenties
• Expression control in musical automates
  

• 24 Electromagnetic string drivers: modified loudspeaker line transformers (I-mu-metal removed from transformer, the original transformer has alternating E-I and I-E core metal) . Original manufacturer: Philips, type LBC 3174/01 . NC 8901 317 40101 (Nr.082256 - No.082293). Windings 0-4-8 Ohms, secondary. Primary 1.5W, 3W, 6W / 100V. The DC resistance of the 8 ohm speaker winding is 1.65 to 1.9 Ohms, depending on temperature. Continuous DC current should be limited to 2A if the power rating is not to be exceeded. The original inductance of the coil at 1kHz was 1mH, but after removal of the I-core, this value dropped to 0.56mH. The combined ac/dc impedance of the original 8 Ohm winding now becomes 5.3 Ohms. The maximum ac rms voltage that can be applied to the winding now is 5.64V. With 50% duty cycle square waves this leads to a maximum voltage of 8V. For 25% duty cycle the maximum voltage becomes 11.3V, hence our choice for a 12V power supply.
• Alternative transformer type: Eagle, 100V line transformer, type MT-8W, ref.nr P037T Primary 1W,2W,4W,8W / 100V. Secondary 8 Ohms and 15 Ohms. (Obtainable from Farnell). Inductance of the 8 Ohm winding after removal of the I-core: 1.3mH. DC resistance of the winding: 0.92 Ohm. Inductance of the 15 Ohm winding: 2mH. DC resistance 1.13 Ohm. The presence of the steel string does influence the inductance. The inductance of the 8 Ohm winding of the original transformer was ca. 7.3 mH. This type is used for the two highest strings.
• Exciter hammer solenoids: August Laukhuff, 12V action magnet. DC-resistance 75 Ohms. Current 190mA. These solenoids are driven with variable duration pulses on 24V. Same power supply as the dampers, since hammers and dampers are never driven together, the rating of the 24V power supply is unaffected.
• Exciter hammers: taken from an old upright piano, central octave. Hammerhead felt sligthly brushed up and hardened again.
• Wheels: large wheels: diameter 400mm, axis 25mm, width:73mm. Small wheels: diameter 144mm, axis 10mm, width 50mm. These wheels are mounted on the flanges with M10 x 20 stainless steel hexagonal bolts.