Physics of downward firing woofers


Ok ... this question will show my complete lack of knowledge about physics ... but here goes anyway:

Every now and then I come across a speaker with a downward firing woofer. I wonder: why don't the sound waves bounce off the floor back towards the speaker, rattling the bejeezez out of it and / or messing up the woofer itself?

~Jim
128x128jimspov
Sound travels at about 1087 ft/second so the length of a 40 hz bass wave in 1087/40=27 feet.  Carpeting won't have any significant effect on a wave that long.  It's why bass traps have to be so big. 
The Vandersteen 2Wq sub has 3 eight inch woofers facing downward at the bottom of the cabinet with no holes for woofer exposer. It seems to have slots cut around the bottom outer edges of the cabinet where sound comes out from the back side of the internal woofers. All you have to do is listen to these subs properly setup and make your own decision as to how they sound.
I have 2 REL's and have wondered if carpet, or a pad under the drivers is better for sound (to reduce relection) or not. I have not expermented with it yet to see. Any thoughts?
I’m not sure here, but I’ve always assumed that the idea is to get the driver as close to the primary reflective surface - in this case, the floor - as possible. An equivalent strategy for a front firing subwoofer would be to place it facing away from the listening position and close to the wall opposite the listener.

In either case, the theory is eliminating quarter wave interference in the output from a subwoofer. This is the most destructive cancellation and will cause dips in output at the frequency where the first (powerful) wall reflection encounters subsequent output from the subwoofer.

Imagine that you hold a synth key to continuously generate output over the octave between 25 hz and 50 hz. The sub is continuously pouring this output into the room. Also bear in mind that these long waves are omnidirectional.  As you first hit the synth key, the the initial wave migrates across the room and bounces back from the nearest reflective surface (a round rip measured in milliseconds). The reflected wave then encounters the subsequent output from the sub because you’re still holding the key down and generating new (subsequent) output at 25-50 hz. If this reflected wave is at the same frequency as that output, the reflected wave will cancel the new wave and create a dip in frequency response.

If you search for it, you can find the math to calculate how the frequency for quarter wave cancellation changes with distance to the nearest reflective surface. It "shifts" with time/distance and won’t "come back" at the same frequency it left the sub.

In a nutshell, the closer the driver is to the primary reflective surface, the higher the frequency of the quarter wave destruction. If you get the distance low enough, the destructive interference occurs above the subwoofer’s high cut and the sub’s output should be very smooth. In a downward firing woofer, the distance will always be small (inches from the floor) and the quarter wave cancellation will be high enough in frequency to spare the sub’s subsequent output.

I think the idea is to save you from yourself. In theory, it should be harder to place a downward firing sub "improperly". I can’t swear that this is the best theory to employ in sub design, and I might be wrong in assuming that this is behind the idea of a downward firing sub, but it’s always been my assumption.