DynaPleat Line Array Low Frequency Impedance
On December 30, 1998 an informal study of the in-room response of the DynaPleat line array was performed in my livingroom. This study was intended to further the understanding of real world dipole room response using a loudspeaker with greater low frequency capability than the RD-75 planar-magnetic driver. Under consideration is a plan to use the array in the frequency range between ~60Hz to 300-500Hz. Obtaining this information takes the measurements involved out of the realm of the formalism of semi-anechoic measurements into the messy situation of reveberant room measurements.
The independent variable was the distance of the dipole line array from the closest wall, the front wall of the livingroom. Three conditions of baffle distance from that primary refecting surface were tested. Each condition was documented by five measurements across the 6' width of the listening postion. A more detailed explanation of the experimental setup is given below.
Each DynaPleat driver has a 4.5" square piston (not a cone) driven by 11 individual voice coils. There are 12 DynaPleat drivers in an array. The driver column is approximately 78" high, the speaker structure is about 82" high. The DynaPleat driver is apparently no longer manufactured, and is not being marketed to my knowledge.
The DynaPleat is nominally a 4 ohm driver; and, in the array groups of four drivers are wired in parallel. Those three groups are then put in series. At 1KHz the impedance of the array is about 3.3 ohms. The low frequency resonance is at about 43Hz with an impedance of about 4.6 ohms. The high frequency impedance rises to about 6.5 ohms at 20KHz. Shown below is the high resolution low-frequency impedance plot for the system. Note the unusual secondary and teriary resonances seen at approximately 175Hz and 420Hz. There is no indication in the frequency response that these resonances are significant.
DynaPleat Line Array Low Frequency Impedance
On March 1, 1998 a semi-anechoic frequency response was obtained from the DynaPleat array at a measurement distance of 3m and height of 1m. That day, Rudi Blondia and I each measured a number of systems at a large gymnasium as part of the RD75/Woofer Integration study. On a whim, knowing that access to the gym was becoming problematic, I bought a DynaPleat array for measurement. The high resolution low-frequency measurement from that day is the anechoic reference for the reverberant in-room measurements of this study.
The system configuration of the DynaPleat array at the time of the gym measurement is shown at the left. The DynaPleat drivers are contained within the framing rectangle of 2" wide wood strips adjacent to the triangular baffle section. The section to the left of the DynaPleat array contained the framework for the planned true ribbon driver project. The rear of the line array, at that time, was covered with a 2" thick acoustic foam sheet - which could be expected to reduce the dipole effect at higher frequencies due to sound absorbtion. The width of the dipole baffle was about 27" at the base of the system.
Given the difficulty in obtaining listening distance semi-anechoic low frequency measurements from loudspeakers due to the required size of the measuring venue when using time-gated measurements there is usually an open question regarding a loudspeaker's actual low frequency performance. Standard use of MLS or impulse based FFT measurement systems requires that the data sample be truncated to exclude room reflections in order to simulate anechoic conditions.
An 'in-room' measurement will include all the sound reflections and other
extraneous signals in the data sample. As such, the unsmoothed frequency
response plot will be quite jagged. Below are the reference frequency response plots from the gymnasium
measurements.
This data was processed without any FFT window or additional sub-octave smoothing. The 63.5mS duration of the semi-anechoic response implies a low frequency limit of about 15.75Hz for the measurement. The 2.124 second duration of the gym 'in-room' response implies a valid low frequency limit for the measurement of about 0.47Hz
Note how the 'in-room' response at the gym (a 93' square building) is contained predominently within a +/- 6dB window around the semi-anechoic (on a ground plane) response of the line array. Several additional features of interest are: a resonant peak in the frequency response at approximately 43Hz, which is the frequency of the impedance resonance; over the span 30Hz to 120Hz the amplitude response falls 12.9dB; over the span 60Hz to 120Hz the fall is 5.4dB; and, excess low frequency energy from the reverberation in the gym appears below 20Hz in this measurement.
The semi-anechoic low frequency response is generally what one might expect from a dipole speaker, with a gradual falloff of the low frequency response below the presumed baffle support frequency of about 130Hz (based upon inspection of the response plot). Theory predicts a -6dB/octave fallrate from that point down to the resonance of the driver, at which point a -18dB/octave fallrate is expected. The amplitude fall in the octave below the resonance, 43Hz to 21.5Hz, is 10.2dB. The octave from 31Hz to 15.5Hz has an amplitude reduction of 18.97dB.
Clearly the DynaPleat is not operating as a theoretically perfect dipole speaker. The specified sound absorbtion of the acoustic foam on the rear of the driver in this configuration is so minimal in the low frequencies that it is difficult to believe that it is introducing a significant effect on the frequency response in the low frequencies. The amplitude increase around 43Hz is consistent with the low frequency resonance of the system seen in the impedance plot.
Page 2 - Next Page
Page 3
Page 4
Acoustic Line Source Research - Table of Contents.
