IMPULSE RESPONSE COMPARISION: With Baffles.

Comparative inspection of the impulse responses of drivers, made under similar conditions, can reveal much about differences between the drivers. It is a fact that, mathematically, impulse and frequency responses are two different views of the same event - one is a time-based and the other a frequency-based view of the same data. This is the essence of the Fourier Transform, time and frequency are convertable with proper mathematical tools.

The first two comparisons shown also contain impulse responses from the 'Without Baffles' condition.

Below is the plot of the complete 4096 point data sets of the acquired impulse responses (without notch filters) of both of the Carver and BG planar drivers. Note the major reflections in the data appearing past the midway point on the plots. The thickened 'Green' trace impulse spike is apparently an artifact of the plotting. Expanded views are shown below.

Damaging evidence, in regards the question of driver damping, is already seen in this display. Note that the impulse 'tail' on both Carver drivers extends well past the end of the impulse tail on the BG drivers.

Below is shown the 432 point data domains from each of the four drivers, without notch filters installed, that were used in the frequency response analysis portion of this comparison.

With both Carver drivers, under both conditions, with and without baffles, low-level 'ringing' in the impulse responses is seen to the midway point in the data set. Both BG drivers seem to have an end to their impulse tails in about 1/2 the time of the Carvers! An "underdamped" driver continues to 'ring' longer than a better damped driver.

Below is shown a 50 point data domain for each of baffled driver conditions. For the first time it is seen that the driving impulse is attempting to simulate a rectangular pulse. Clearly it has finite risetime and a small, but measurable undershoot on the return to zero amplitude.

Three things to note in comparison of the Carver and BG drivers without notch filters: 1. The risetime of the RD75 driver is quicker than the Carver driver. 2. The Carver driver has no 'flat top' corresponding with that on the driving impulse, as does the RD75 driver. 3. The Carver driver is clearly 'ringing' at greater amplitude for a longer time than the RD75 driver. The Carver driver used in the unbaffled condition shows much better impulse response correspondence with the BG driver in the unbaffled condition. Carver driver number 2 maybe defective?

When the notch filter is inserted in series with the RD75 driver a fair simulacrum of the driving impulse is seen. A dramatic reduction in the amplitude and duration of the impulse tail is also seen when using the notch filter.

Below is shown a 25 point data domain for both drivers (no notch filter) and the driving impulse. It is provided in order to show very high resolution detail of the impulse responses.


Sampling Rate: 61200Hz - Point resolution = 0.000016S
Microphone at 1 meter distance
Microphone height: 1 meter for RD75s; 1.21 meter for Carver 60s

CARVER60 vs RD75 UNDERDAMPING COMPARISON

NO BAFFLE Impulse Response Frequency Response Waterfall FR
BAFFLED YOU ARE HERE Frequency Response Waterfall FR
OTHER Calibration

The RD75 Dipole Baffle Study - Table of Contents.
Acoustic Line Source Research - Table of Contents.