The Advantages of Floor Coupled Up-Firing Speakers

[el`Ol]

Originally posted by Etienne88

If I remember well, Stig Carlsson said in his paper that the drivers should be pointed so that they cross behind the listener. It is somewhere in this paper... From that you can construct speakers adapted to your own listening room!

Thanks, Etienne!

It seems to depend to the driver as well. The fullrange driver I use have to cross in front of the listener, independent from the vertical orientation. I think I will build an equal-sided triangle for the front baffle. What seem to be essential is the absorber in the back of the driver. The difference is astonishing. I have a thick carpet in my workshop, so there is no need for a bottom absorber.

Regards,
Oliver

[Etienne88]

Oliver,

Carpet is not the best absorber. It absorbs quite well at 4 kHz with absorption coefficient around 0,6 – 0,8 but as the frequency goes down, so does the absorption coefficient… They yield around 0 – 0,1 at 125 Hz, which is very close to nothing!

Nevertheless and according to the following quote from Lynn Olson website, floor reflections would assist localization.

the floor reflection actually assists in localization and the natural perception of timbre. (The BBC did a series of experiments of listening to stereo in an anechoic chamber, and found to their surprise that adding a plywood "floor" panel between the speakers and auditioner improved localization and gave a substantially more natural sound quality.)

I did not find any information about the set up used by the BBC in the anechoic chamber, but I guess that they used traditional front firing loudspeakers.
From that you can draw your own conclusions!

Regards,
Etienne

[Etienne88]

Graaf,

I did not forget you. It is summer here and I prefer to spend some time outside before darkness falls again on Sweden!

I think you want to be right because it the set up is your idea… No offence intended but sometimes it is hard to admit you own errors!

My hobby, obviously yours as well, is called high fidelity. I like to listen to music very much but as an engineer I like technique as well, so that I make my life a little bit more complicated to ensure that the equipment I use to listen to music does that with the highest accuracy possible to the recorded support (not to the original performance!). I also include the room in the reproduction chain. This means that if I send an impulse to my preamp, an impulse should come out of the speakers. Of the 2 impulse responses I posted earlier, one almost respect that criteria and the other one is very far from it. I agree that both situations are a long way from being perfect impulse reponses. In the first chart you see an impulse (which actually look like an impulse), then some problems before the first millisecond. After that it is “more or less” quiet. On the second chart, I do not recognise the impulse typical shape right after 0 millisecond… and the rest is far less quiet than the other impulse response.

The analysis of the impulse responses is quite straight forward to me. One of the things that might be questionable is the way I measured them since I am still in the learning process of how to use a microphone. Nevertheless I followed the instruction I get with the measurement software and the only thing I changed between both measurements was the position of the loudspeakers.

Regards,
Etienne

[el`Ol]

Originally posted by Etienne88

Carpet is not the best absorber. It absorbs quite well at 4 kHz with absorption coefficient around 0,6 – 0,8 but as the frequency goes down, so does the absorption coefficient… They yield around 0 – 0,1 at 125 Hz, which is very close to nothing!

I forgot to say it is a 2cm persian.

[graaf]

Originally posted by Etienne88
I think you want to be right because it the set up is your idea… No offence intended but sometimes it is hard to admit you own errors!

Dear Etienne,
how can I assure You that this is not the case?
I am too old for this

what "errors"? what "my idea"?

the setup is essentially borrowed from Carlsson, this is Carlsson setup combined with Beveridge idea of speaker positioning at the opposite walls

I am tired I have to admit
so I decided to leave this discussion and this thread as it appear to be pointless, there is simply no communication, no genuine interest

maybe someone will pick this discussion up sometime but I doubt...

my last word or You Etienne - do Yourself a favour, contact a typical Carlsson speaker (perhaps the OA 52.3 - Stig Carlsson's personal favourite) owner (it should be easy for You in Sweden, perhaps through carlssonpanet?), listen to it in a typical setup recommended by Carlsson and then take an impulse response measurement of this setup

look at Your results and then ask Yourself some questions - starting from "was Stig Carlsson competent sound engineer and scientist?" and so on

good bye and good luck,
graaf

[Jim G]

graaf, I am a diy'er following this with interest. I have been experimenting and I am impressed with the results so far. Please don't assume there is no interest. I am sure that there are others like me as well. I have no technical ability to contribute. I intend to build various arrangements to try out. OB firing vertically combined with a compression tweeter on a wave guide more or less toward the listener but off axis and angled toward the ceiling seems very promising.( all this low to the floor and against the wall) Thanks for your efforts. Jim

[Etienne88]

Oliver,
A 2 cm Persian might make a difference, yes!

Graaf,
I agree with you that our last discussion (from when I posted the impulse response) was not very productive/constructive.
Nevertheless, I found most of the discussion we had very interesting and instructive. I am thankful you started this thread, it opened my eyes to the reflexions issues which we don't seem to tackle from the same angle!
I sincerely hope those will not be your last words to me and I am sorry if I expressed myself too harshly.

Regards,
Etienne

[graaf]

Originally posted by Jim G
graaf, I am a diy'er following this with interest. I have been experimenting and I am impressed with the results so far. Please don't assume there is no interest. I am sure that there are others like me as well. I have no technical ability to contribute. I intend to build various arrangements to try out. OB firing vertically combined with a compression tweeter on a wave guide more or less toward the listener but off axis and angled toward the ceiling seems very promising.( all this low to the floor and against the wall) Thanks for your efforts. Jim

thanks Jim
sincerely I think that almost everything has been already said in this thread
I mean everything needed for interested diy'er like You to start His own experiments, to have fun and to find the sound He likes better
I choose not to persuade anybody anymore
after all I am not marketing any product
just sharing my experience and thoughts

I wrote above "almost everything has been said" because there is one more thing that should be "hypothesised" to make the reasoning behind my proposition complete
early lateral reflections are indeed detrimental to the sound quality in case of most typical front firing speakers
they seem not to be detrimental to the sound quality in case of omnidirectional speakers, especially of the Carlsson type
why? this is the question

best regards,
graaf

[graaf]

Originally posted by Etienne88
I found most of the discussion we had very interesting and instructive. I am thankful you started this thread

and I thank You for Your contribution, especially for the measurements as I am not able to post any

Originally posted by Etienne88

I sincerely hope those will not be your last words to me and I am sorry if I expressed myself too harshly.

no problem, no offence taken

best,
graaf

[pjpoes]

Originally posted by Etienne88
Oliver,

Carpet is not the best absorber. It absorbs quite well at 4 kHz with absorption coefficient around 0,6 – 0,8 but as the frequency goes down, so does the absorption coefficient… They yield around 0 – 0,1 at 125 Hz, which is very close to nothing!

Nevertheless and according to the following quote from Lynn Olson website, floor reflections would assist localization.

I did not find any information about the set up used by the BBC in the anechoic chamber, but I guess that they used traditional front firing loudspeakers.
From that you can draw your own conclusions!

Regards,
Etienne

As I recall once reading, the BBC Dip was developed with the LS3/5 monitor. While testing the design and developing its crossover they came up with this theory. Actually, I had always assumed it was the LS3/5, but given that they were just doing the research and development for the overall concept of a BBC standardized studio monitor, it could have really been anything. None the less, I agree that it probably was a normal front firing speaker, and would add that it probably was a very early form of the LS3/5.