The IMP/MLS provides for very quick and easy measurement of impedance, Fs, and Qt. This ability allowed me to glit-the-lily as it were once I decided to attempt repair of the foam surrounds of the drivers. The "Lambda Acoustics" "Moyen" damping compound seems to be acrlyic-based. I started with the 'black' version on the outside of the surrounds. The first coats soaked right in, and after many more coats a smooth surface was built up. Eventually I coated the entire front of the cone with the black "Moyen" - a number of times.
Inspecting the rear of the surround showed that the black "Moyen" hadn't penetrated completely through, so if a little was going to be good, more should be better. I started coating the rear of the surrounds with the white/clear "Moyen." It seemed clear that this process was going to make the surrounds less compliant, which proved to be the case when I finally measured the Fs of all the drivers at this point in the process.
The Fs values varied about 10Hz at this stage. A note on the measuring, I quickly found that I had to measure all 28 drivers in rapid sequence - each and every time - since Fs seemed to temperature dependent. So, as they say, in for a penny in for a pound, I decided to mass load the higher value Fs drivers to match the entire set to a lower Fs value. Mixing the white "Moyen" with 80 grit silica sand seemed to work quite well for this purpose. The 'goop' mixture was painted on the rear of the cones.
After many, many iterations all 28 drivers measured with an Fs between 39.55Hz
and 40.82Hz. In those drivers which required the greatest amount of mass loading
the Qt value would steadily rise after each application of goop. The Qt values for
the 28 drivers ranged from a low of 0.605 to a high of 0.847. 15 of the drivers had
Qt values in the range 0.700 through 0.799.
Typical near field frequency response of a single driver.
The plot at left shows the typical near field response of an individual driver
measured at 1" from the front of the cone. The mass loading presumably did affect
the sensitivities within the driver set. The drivers were selected for sensitivity
in their intended line array (left or right), and the array was tapered with the
most sensitive driver at the bottom, with the lowest sensitivity at the top.
The frequency 54.84Hz was chosen as the reference frequency for sensitivity
comparison, since all the drivers appeared to peak at that point.
The left line array had a total variance in driver sensitivity of 2.68dB from
bottom-to-top; and, the right line array had 2.53dB change in sensitivity
bottom-to-top. The variance between the 54.84Hz peak and the 400.3Hz low point
was between ~2.98dB and ~4.9dB across the the driver set. The lowest sensitivity
drivers had the largest variance.
Line array impedance and zobel development.
Shown at left is a comparison impedance plot for the completed left line array.
This comparison resulted from my decision to apply a zobel to the line arrays. There
are four high frequency impedance plots (low resolution), starting with a no zobel condition,
and
followed by three with increasing size capacitors used with an 18 ohm resistor to create
a simple zobel. The "IMPzobel" program from the "IMP-AID" suite of programs once
available through "Old Colony Sound" gave a good indication for a proper starting
point.
The green plot trace represents the 0.47 Hz high resolution, 1.92KHz sampling rate, impedance of
the array at resonance using the 18 ohm - 20.1uF zobel. It is the compliment of the brown
trace, made at 61.2KHz sampling rate (15Hz resolution).
An 18 ohm resistor and a 20uF capacitor seem to provide about the best impedance
correction possible with a simple zobel.
The New Line Array for 2003 - Page 1.
The New Line Array for 2003 - Page 2.
The New Line Array for 2003 - Page 4 - Next Page.
Acoustic Line Source Research - Table of Contents.
