Hello.
I know that the distance from the center of the speaker to the floor is an issue especially at low frequencies, but the idea here is to discuss how reflections from the floor affect all frequencies (bass, midrange, treble).
As a matter of theory, let's say a bookshelf type speaker is 80 cm from the acoustic center of the speakers (woofer and tweeter) above the floor, this should give a cancellation of 1/2 of the wavelength at 214 Hz, and if my calculation is right, it takes 2.33 ms from the center to the floor.
Now, if that same speaker is 150 cm from the acoustic center to the floor, that would give a cancellation at 114 Hz and a time of about 4.35 ms from the acoustic center to the floor.
Really, I'm not interested in the theory, but rather the practice of how raising the speaker relative to the floor can alter the accuracy of tones, the soundstage and the response of the speaker, and also, different rooms and speakers can affect this in different ways.
I know that the distance from the center of the speaker to the floor is an issue especially at low frequencies, but the idea here is to discuss how reflections from the floor affect all frequencies (bass, midrange, treble).
As a matter of theory, let's say a bookshelf type speaker is 80 cm from the acoustic center of the speakers (woofer and tweeter) above the floor, this should give a cancellation of 1/2 of the wavelength at 214 Hz, and if my calculation is right, it takes 2.33 ms from the center to the floor.
Now, if that same speaker is 150 cm from the acoustic center to the floor, that would give a cancellation at 114 Hz and a time of about 4.35 ms from the acoustic center to the floor.
Really, I'm not interested in the theory, but rather the practice of how raising the speaker relative to the floor can alter the accuracy of tones, the soundstage and the response of the speaker, and also, different rooms and speakers can affect this in different ways.
I think you are over-analyzing this and the cases you describe would not really be audibly noticeable.
It general, a bookshelf speaker should have its acoustic center at about your ear level in your seated listening position.
Moreover, the floor material will have a bigger effect on the sound than the speaker height. Hard tile vs wooden floor vs carpet is much more of a factor.
It general, a bookshelf speaker should have its acoustic center at about your ear level in your seated listening position.
Moreover, the floor material will have a bigger effect on the sound than the speaker height. Hard tile vs wooden floor vs carpet is much more of a factor.
Yes and no, I have already tested it with my speakers and increasing the height can audibly change the soundstage, at least to my ears and in my room with my speakers increasing the height of the center of the tweeter to 85 cm to 135 cm made the sound stage higher, like at the right height.
Yeah, I did some tests and one of them that impressed me was in the film Gemini Man in the scene where the train passes from one direction to another, in the lowest position it looks like the train is coming from below (the height of the boxes) and then it goes further up and then down again, at the highest level it seems to just pass without any strange height errors.
Other: You are right that the flooring material can affect the sound.
In my case the floor is concrete and tile.
Yeah, I did some tests and one of them that impressed me was in the film Gemini Man in the scene where the train passes from one direction to another, in the lowest position it looks like the train is coming from below (the height of the boxes) and then it goes further up and then down again, at the highest level it seems to just pass without any strange height errors.
Other: You are right that the flooring material can affect the sound.
In my case the floor is concrete and tile.
If you are watching movies and want the sound to match the screen images then 53" might be a good choice. Depends on how high your screen is. And you are not listening for maximum musical pleasure anyhow. It's mainly for the movie dialog.
But if you are just listening to music and sitting in an easy chair then 53" is probably going to be too high for the center of a bookshelf speaker. That would put the tweeter even higher and you are going to lose some of the sound due to the increased directionality of the upper frequencies. While at the same time the effects on the lower frequencies that you mention in the initial post will be minimal.
A floor distance of 34" to the center of the speaker is more in line with what most people would prefer. You won't be losing as much of the higher end that way.
But if you are just listening to music and sitting in an easy chair then 53" is probably going to be too high for the center of a bookshelf speaker. That would put the tweeter even higher and you are going to lose some of the sound due to the increased directionality of the upper frequencies. While at the same time the effects on the lower frequencies that you mention in the initial post will be minimal.
A floor distance of 34" to the center of the speaker is more in line with what most people would prefer. You won't be losing as much of the higher end that way.
Soft furnishings in a home can make a big difference.
Carpets, curtains, sofa's etc etc They kill the sound depending on quantity.
Hi, I'm sorry but you know it wrong, so better look into theory to get it right
Sound reflects specularly, like light, so if speaker acoustic center is at 80cm height, and listener ear is at 80cm height 2m away, the specular reflection point on the floor is half way between speaker and listener. And you don't just calculate the distance of reflection point, but relevant here is how much longer path length sound takes through reflection compared to path length of direct sound. Now you can calculate how much later the reflection comes after direct sound in order to figure out the interference at any frequency. The reflection interferes with direct sound, which makes interference pattern, a "comb filter". In this case, direct sound path length is 2m and reflection path length is easy with trigonometry, roughly 2.56m in this case. Thus, the reflected sound comes 56cm late to direct sound, about 1.6ms. When this 56cm is half wavelength, there is the first dip, that is correct, roughly 600Hz in this case though.
Most often floor reflection is not bass issue, this is very common myth still floating around, unfortunately. Floor reflection typically has the shortest path length of all first specular reflections, which means it has the null highest in frequency of all the early specular reflections. For example, if you have 2.5m high ceiling, speaker and ear 80cm from floor, they are 170cm from ceiling. At 2m listening distance ceiling reflection would make here almost 4m path length, which would make first interference dip around 176Hz, where the ~2m extra is half wavelength. If you raise speaker, or listening height, or both, the ceiling reflection interference moves up in frequency as it is now shorter, and floor reflection moves down as it is now longer. Front wall, is round trip so easily about 1m late. Also side wall reflections have quite long path length, unless room is very small and speakers close to sidewalls. Listener close to back wall? sometimes.
Sorry, even more theory
In a cubicle there are six first early reflections + very many second order early reflections, third order, and so on. Some second order reflections might come sooner than some first order early reflections, and so on. The usual twist in small room is that about all the earliest reflections pile on the low mids, ~200-800kHz or so, and all together form a huge dip. Typical speaker radiates toward the early reflections points almost as loud as direct sound is toward the listener, and 6 of them together easily overwhelm the direct sound, especially if listening distance is too far and the brain has not enough processing time before earliest reflections come in.It is very hard to avoid any of this other than by removing the boundaries, or reducing sound toward early reflection points, or shrinking the listening distance (stereo triangle). Here is where speaker directivity, toe-in and positioning in general is helpful. Height is best to be left so that the speaker design axis is at ear height, otherwise the top octave attenuates. If you wish to raise speakers higher, and the highs reduce too much you could try to tilt them down, but now your listening distance is locked.
In general, speaker positioning is very important freedom to have in order to find best sound, so please experiment as you wish and I hope you find a position where you find sound is best to you! I hope this wasn't too much of a theory, it's really not too complicated, but also not too simple, everything affects and multiple things at once, many things are interlinked.
If you want to experiment with speaker height and this stuff, here is simple tool for you to build intuition on this subject, specular reflections https://audioexperiment.com/tools/vertical-reflections/
Last edited:
Here is even more fun: https://www.desmos.com/geometry/npwfgcvsxc
Task, can you position the listening triangle so that there is no amplitude modulation if you sweep the frequency?
ps. also here is just two boundaries in addition to direct sound, and crazy modulation is observed. Add few more and it's quite clear how much work our own auditory system does, it is quite a miracle what it does. Also shows how speaker spacing also affects for phantom / center sounds especially on lows. Can't hear kick punch? adjust your positioning, size and location of your triangle.
Task, can you position the listening triangle so that there is no amplitude modulation if you sweep the frequency?
ps. also here is just two boundaries in addition to direct sound, and crazy modulation is observed. Add few more and it's quite clear how much work our own auditory system does, it is quite a miracle what it does. Also shows how speaker spacing also affects for phantom / center sounds especially on lows. Can't hear kick punch? adjust your positioning, size and location of your triangle.
Last edited:
