There's something nasty coming up off the floor!
There's something nasty coming up off the floor!
Rudi Blondia has estimated that the beam dispersion of the RD75 line-source driver is 2-3 degrees, which might seem insignificant. Yet in the situation where there is a hard reflective floor directly in front of the loudspeaker system there are some interferring reflections or bounce.
This is a troublesome situation since raising the driver farther up off the floor will generally move the listening axis farther from the center of the driver. At ~76" tall with the driver resting on the floor, that puts it's midpoint at 38" off the floor, which just happens to be a good approximation of a usual listening height which is often specified at 37".
In the section on 'Tapering' is seen a possible fix for the situation in which one does not have the 4" thick wool carpet in the listening area.
The first look at the situation is a comparison of the first 4.88mS of the impulse responses at 3m distance/1m height of the 1S6T baffle with and without the layer of mattress pad foam on the floor (as in the photograph above). The impulse responses have been magnified to reveal the detail after the main impulse section. Note that in the second plot (Brown) that there is much less 'grit' after the main impulse. In the top plot (White) there is excess energy after the main signal that is not part of the signal.
Astute observers will ask "What are those two little ripples at about 2.4mS and 3.8mS, that show up so nicely in the 'with' foam response?" Good for you! This may be your first note of Virtual Diffraction Line-Source Signals (VDLS). Much more on this in the 'VDLS' section, now for the foam!
Comparison of first 4.88mS after impulse, with & without foam on floor.
If you've been surfing through this website you may have run across other sections in which I've attempted to outline problems with smoothed frequency response data. There is good use for smoothed data, but it does obscure a greater reality. Get tough! Here's the 1/12 octave smoothed frequency response comparison from the FFTs of the impulse data above (but at a data length of 57.8mS).
3m comparison with & without foam on floor. 1/12 Octave smoothing.
So you say "Ho Hum" to the 1/12 octave smoothed frequency response differences. How about the 'No Smoothing' overlay below! A concept that is entirely relevant here is that of the 'envelope' of the 'grass' in the responses ('grass' being the euphemism for the jaggy plot). We've dealt with this before in the 'Listening Room' data where the 'with and without' conditions for Chifon fabric covering of the front of the driver were compared.
It seems quite clear to me that the condition with the smaller envelope or 'variance' sounds the best. I've subjectively compared these effects in the Listening Room and the benefits of a reduced 'envelope' are quite easily heard. Do you 'see' the 'with' foam condition as having an overall smaller envelope than the 'without' condition?
3m comparison with & without foam on floor. No Octave smoothing.
The Gymnasium Measurements - From the Rear.
The Gymnasium Measurements - Theory: Tapering and Shading the Line Source.
The RD75 Dipole Baffle Study - Table
of Contents
Acoustic Line Source Research - Table of Contents.
