Audibility of floor bounce?

[mlee]

Hi All,

The following questions are directed towards a 2-way monitor on stands. I did a quick search on the "Audibility of floor bounce". I could not see any data that dealt with this issue in detail, but was wondering if anyone has done a bit of testing regarding this issue? For discussion purposes, lets say the ceilings and side walls are already dealt with.

1. Can anyone link a good discussion thread and/or paper or is it simply not audible. Example, 2-way monitor on stands has floor bounce.

2. We can calculate it, measure it, and sometimes deal with it via room treatments. If we cannot deal with it via room treatment, is there a suggested design format that avoids it for the most part? 3-way design, woofer at floor highish midrange (larger distance from woofer), Xover around 350-400Hz? Do distributed subs clean up the >150Hz range? Maybe the subs should not be run up that high?

3. Any other comments or suggestions.

Thanks,
Mike

 

[SY]

Google "Roy Allison." The Allison Effect is a bit more complicated than just floor bounce, but that's part of the equation. He wrote a freeware program called "Bestplace" which does an excellent job of simulating the effect with various speaker placements.

 

[mlee]

Hi SY,

Thanks for the suggestion. I quickly checked it out and need to do some reading on those articles.

One thing I noticed is the inputs to the program do not have a listening distance entry. I was under the impression that the floor reflection was dependant on listening distance but I know speaker placement it is more complicated then looking at one variable in a room.

Looks like you can eliminate the "room reflection" effects based on careful boundary placement. At that point you are probably at the mercy of room modes?

Unfortunately, my room at the moment is not symmetrical and I don't have the luxury of placing them optimally. I'm hoping to avoid some of the problems with careful speaker design, but maybe it's not worth the effort till I can change rooms? Alternatively, maybe handling floor bounce is not as much an issue; however, when I change my crossover from 80Hz to 300Hz to fill in the measured dip (suspect it's floor bounce as it behaves like it) I do like the sound better. Maybe placebo effect.......

Thanks,
Mike

 

[wowo101]

To my knowledge, the studies most pertinent to the topic were done in the 1990s by Søren Bech and published as "Timbral aspects of reproduced sound in small rooms I+II" and "Spatial aspects of reproduced sound in small rooms". (Just google them, PDFs should be available online.)

His conclusion is that the first-order floor reflection a.k.a floor bounce is the single most audible reflection, making countermeasures such as carfully choosing driver placement and crossover frequency advisable. A simple tool to help with that is Boxycad (sheet "floor bounce effects"). (Its calculations incorporate listening distance.)

NB: When deciding about driver placement, be certain to balance floor bounce mitigation (greater ctc distance) with avoiding lobing (smaller ctc distance, if possible < 1/2*wavelength at XO frequency)!

 

[bear]

Floor bounce, bad.

You can get a pretty good idea of what it is doing by measuring the speaker normally in the room, and measuring the response A) with the speaker on the floor, or better B) outside facing up (mic above).

If you can EQ the notch from the bounce back up, I'd say ur lucky. However likely ur working the speaker hard and losing headroom.

The usual solution is to go to a sub/woofer below the notch. Some speakers put the woofer low in a floor standing box and the other drivers high to solve this problem.

_-_-

 

[simon7000]

I work this issue from the other end. Every sound reinforcement system I have done has one filter set to cut 4db from the 400hz band to compensate for floor bounce into the microphone. If a lecture or pulpit is used with a miniature gooseneck mic, that also gets a cut. These days with the better loudspeakers that is often most of the eq.

Although I have been doing this since the mid 70's it took me a few years to figure out why all sound systems needed those cuts. There also used to be one at 8k but changing the brand of microphones fixed that.

 

[vacuphile]

Fortunately, the brain is a clever box of tricks, which mitigates many issues which show up large in measurements. The Haas effect is one of those, but there are other afaik as of yet unnamed mechanisms which allow us to identify timbre and understand speech in highly compromised environments. Therefore, it is understandable that Ed has to compensate a microphone for a floor bounce that in all likelyhood goes unnoticed by the human ear at the same location.

I am happy with this good provision of nature, since I need a floor for all sorts of reasons, so have to live with a bounce.

Don Keele's banana array is the only solution I know of that does away with floor bounce by actually using it to its advantage.

 

[Michael Chua]

The effects of floor bounce with speaker (ZA14+XT25SC90) at 1 meter height. Below 500Hz, the reflection from the floor comes into play. Most distructive is the deep notch at about 140Hz.

Regards
Mike

 

[chris661]

Ah, but the curve above is a steady-state measurement.

Since there's a half-cycle delay between the constructive and destructive waves, surely you'll get half a cycle at that frequency before its cancelled out.

So, what does that do for the audibility?

Chris

 

[SY]

Ah, but the curve above is a steady-state measurement.

Impulse and MLS show the same anomalies. When you hear a speaker designed to remove these effects, there's a certain "rightness" to the bass on good recordings that you don't hear with conventional speakers. The boundary dips already exist on a good recording, so no need to add extra ones.

On not-as-good recordings (e.g., where the mikes are flown high overhead), the listening room boundary effects can sometimes be beneficial to "anchor" the sound. Tony Faulkner wrote a nice article about this in Audio Amateur about thirty years ago.

Anyone seriously interested in speaker design should read the Allison articles on the subject: "The Influence of Room Boundaries on Loudspeaker Power Output", JAES 22:5, June 1974 and "The Sound Field in Home Listening Rooms", JAES 24:1, January/February 1976.

 

[speaker dave]

Anyone seriously interested in speaker design should read the Allison articles on the subject: "The Influence of Room Boundaries on Loudspeaker Power Output", JAES 22:5, June 1974 and "The Sound Field in Home Listening Rooms", JAES 24:1, January/February 1976.

One thing to note about the Allison paper is that he really is calculating the boundary impact on the radiated power curve. He models 1, 2 or 3 boundaries (a corner of a room rather than a total room) and integrates around all included angles. The upshot of this is that the null frequencies he shows are always higher than you will see at a typical listening position.

The power integration includes response out front in a typical position and also angles much closer to the boundary walls. Those typically won't be listening positions, but, due to the shallow bounce angles, will have the bounce nulls at much higher frequencies.

David S.

 

[Michael Chua]

Impulse and MLS show the same anomalies. When you hear a speaker designed to remove these effects, there's a certain "rightness" to the bass on good recordings.....

That's true. The bass doesn't sound right because the floor bounce killed a lot of the mid bass.

I recall an article by Peavey regarding missing bass when bands perform on stage which is traditionally 3ft high. When the bass bins were relocated down to the floor, the bass was restored.

 

[mlee]

Hi All,

Thanks for the posts and information! They point me to do some more reading. Seems like the general consensus is that floor bounce is indeed audible and that if possible, should be designed around.

Some of the comments of "rightness" seem to ring true but it's hard to trust the ears sometimes. Seems like it just makes sense to design for a real room as much as possible.

Mike

 

[mlee]

I don't know if it's looking at the curve but back to back listening of bounce there and bounce not there seems to present a thinner sounding speaker (the one with bounce included). The low bass still there but still seems like there is no meat to go with those potatoes.

 

[mlee]

wowo101,

Thanks for the readings list. Those were good reads.

 

[Michael Chua]

.....seems to present a thinner sounding speaker (the one with bounce included). The low bass still there but still seems like there is no meat to go with those potatoes.

Yes, that's exactly what it sounds like. Most of the information in the upper bass is lost due to the floor bounce. You can easily verify this by comparing the playback with a good pair of headphones.