How to calculate floor bounce reflection(cancellation)

[jtsaudio]

I have a 6" two way tower speaker that I may add a larger woofer to for more extended bass.

I read somewhere that there is a floor-bounce cancellation related to woofer height (Allison effect?).

The 6" driver is approximately 30 inches from the floor.

How do I determine the frequency at which this occurs, and how will/should this affect my choice of lower-end crossover frequencies?

Thank you.

Joe

[dantheman]

This will help:

Floor/Ceiling Reflection Calculator
audio blog: A little more into boundary conditions

Dan

[CharlieLaub]

This Excel spreadsheet based modeler written by Jeff Bagby does the same, and more:
Loudspeaker Design Software

-Charlie

[DBMandrake]

For a really quick, back of the napkin calculation, there is also this useful page:

Floor/Ceiling Reflection Calculator

[Michael Chua]

Originally Posted by jtsaudio

I have a 6" two way tower speaker that I may add a larger woofer to for more extended bass.

I read somewhere that there is a floor-bounce cancellation related to woofer height (Allison effect?).

The 6" driver is approximately 30 inches from the floor.

The floor bounce is real. I picked this up while I was designing a 2-way with Zaph's ZA14W08. Speaker was on a 30" stand.

The cancellation doesn't show up in near field measurements. However, when the mic was placed 1 meter away, an in-room 20-20K ungated sweep registered a deep notch at approx 150Hz.

That's one reason why mid-bass is missing sometimes.

More measurements can be found at SWIFT

[CharlieLaub]

Originally Posted by Michael Chua

The floor bounce is real. I picked this up while I was designing a 2-way with Zaph's ZA14W08. Speaker was on a 30" stand.

The cancellation doesn't show up in near field measurements. However, when the mic was placed 1 meter away, an in-room 20-20K ungated sweep registered a deep notch at approx 150Hz.

That's one reason why mid-bass is missing sometimes.

More measurements can be found at SWIFT

A 3-way with the woofer located at or very near the floor, and the woofer-midrange crossover point at or above 200Hz, can completely eliminate this problem.

-Charlie

[Wayne Parham]

Originally Posted by Michael Chua

The floor bounce is real. I picked this up while I was designing a 2-way with Zaph's ZA14W08. Speaker was on a 30" stand.

The cancellation doesn't show up in near field measurements. However, when the mic was placed 1 meter away, an in-room 20-20K ungated sweep registered a deep notch at approx 150Hz.

That's one reason why mid-bass is missing sometimes.

More measurements can be found at SWIFT

Floor bounce definitely is real. So is the notch from the wall behind the speakers. Vertical axis room modes also tend to fall in this frequency range. Really messes up the lower midrange.

I tend to suggest flanking subs to mitigate this problem. Run a "helper woofer" just a smidge on the high side, like just north of 100Hz. It will smooth the lower midrange, sort of filling in the holes from the self-interference notches.

• Room modes, multisubs and flanking subs

[Wayne Parham]

Originally Posted by CharlieLaub

A 3-way with the woofer located at or very near the floor, and the woofer-midrange crossover point at or above 200Hz, can completely eliminate this problem.

-Charlie

Definitely. I either run a three-way with mid and woofer blended between 100Hz and 250Hz or, if I run a stand-mounted two-way, I run flanking subs, which effectively do the same thing. I don't usually run flanking subs quite that high, but I do let them blend in the octave above 100Hz.

[bbggg]

Originally Posted by jtsaudio

How do I determine the frequency at which this occurs, and how will/should this affect my choice of lower-end crossover frequencies?

It's dead simple both to visualize and to calculate. You have to imagine a mirror image of the driver deep inside the floor. When sound emission from the driver itself and from this subterranean image are in antiphase, there is cancellation. This happens when path length difference is half a wavelength. You calculate the path length difference by calculating two hypotenuses and subtracting one from the other, and then divide the speed of sound by the result: this number is twice the frequency in question.