• HifiVoice
• Audio
  • Milestones
    • Inspiration Stage
    • Ceramique Stage
    • Wooden Stage
    • Acoustical Stage
    • Measurement Stage
      • Anechoic
      • Tony
      • Rob
      • Marc
      • Thiel
      • Downloads
    • Electrical Stage
    • Financial Stage
    • Information Stage
    • Conclusions
    • Acknowledgements
  • Classic 25
  • iPod
  • DIY DAC

Anechoic measurements

The units have been measured inside the Milestones cabinet by Okke Ouweltjes in a large anechoic room, using Bruel & Kjaer measurement equipment. The measurement distance was 1.8m, the measurement height halfway the midrange and tweeter. The measurements have been performed with a felt grill on the baffle.

I've processed the measurements using Speaker Workshop as follows. Firstly, the measurements files were specified in dbV, so they have been converted to dBs first, and imported in SPW (see the impressive 85dB range of the measurement...).



Secondly, the dB values had to be converted to absolute dB levels. I took Thiel's C^2 79 own measurement at 1kHz as reference, which implied 106dBs had to be added to all above measurements.



Thirdly, the group delay caused by the rougly) 1.8m measurement distance was removed. For each measurement a correction of 5.5ms (indicating a 1.89m measurement distance) has been applied.



Fourthly, the impedance measurements were performed from 5Hz to 20kHz, and therefore the acoustic measurements have been cropped to that range as well.



Last but not least, smoothing has been applied. Smoothing is kind of dangerous, as applying the wrong type of smoothing affects the validity of the curves for design puprose. For instance, absolute smoothing also smooths the +180/-180 degrees peaks, which leads to strange phase responses! Too much smoothing (low order) attenuates the high frequency response too much. Hence, 1/24th order smoothing has been chosen to smooth as little as possible, only removing the small peaks.



See the measurement page for an overview of all curves, or download the final curves yourself.

Directivity

I've also calculated the unit distance (from the conus) with respect to the measurement position of 1.8m, and the typically listening spot, in my case roughly at 3.5m distance.





As one can see the relative offset between tweeter and midrange is constant, and the woofer shifts about 1.5 cm compared to the midrange. This is a negligible shift compared to the wave lengts in the operating range of the woofer (200Hz = 1.72 m). This implies that the anechoic measurements are a good representation of the unit response at the listening place, and no additional considerations are required for the polar response of the speakers.

next->