The Advantages of Floor Coupled Up-Firing Speakers

[markus76]

Originally Posted by el`Ol

Don't get me wrong, I'd love to see a meaningful investigation how room reflections influence spatial reproduction. But I don't see anybody trying anything in a meaningful way. Basic requirements for listening tests are ignored. That's why I bought a Smyth Realiser to hopefully gain some insight how my own spatial perception is affected.

[el`Ol]

Originally Posted by Elias

Would you advice the best arrangement for a such a test? I'll place the speaker on the floor facing upwards.

If drivers from the same brand and the same series just with different sizes work so differently and and a driver of an other brand again I am not sure if one can advice anything. And for a non-stereo-hearer maybe it is a different cup of tea again, even if listening mono. I agree with Oliver that one has to distinguish between concepts and implementations. I once told someone about the flooder in a german forum and it worked for him with the 8" Beyma in the "Viech". The "Viecher" were lying on the floor and he was sitting on the floor. The Beyma in a quasi-IB enclosure would be a cheap and simple try.
What interests me more: Can a "non-stereo-hearer" hear stereo with Sonabs?

[LineArray]

Originally Posted by graaf

...
it is absorption (togehter with directivity of the loudspeaker) that causes this fall of overall radiated power in function of frequency, isn't it?
...

Typically not. A pistonic acting dynamic speaker has a frequency range
where it's radiated power is more or less constant.

That range is above resonance and below the frequency where it becomes
directional. Falling VC velocity with frequency is in that range compensated
by increasing radiation resistance.

Above that frequency the radiated power falls off.

Radiation resistance does not increase anymore, falling VC velocity
is no more compensated, VC inductivity leads to driving force falling
with frequency.

A falling power output of 12db / Octave is the result above that
constant range.

A fullranger can keep ON AXIS pressure level constant above that
"constant power" range by beaming, which means radiating
the power into a narrower angle. Nevertheless, the total
power falls off.

The dynamic speaker itself is the reason for acoustic output power
falling with frequency above the non-directional frequency range.

The transitions between the frequency ranges are design dependent.

Non pistonic behaviour of the cone, VC inductivity etc. influence
the behaviour of a given driver, but physics governs the overall
pattern, giving an individual driver only small space for variation.


Best

[graaf]

Originally Posted by Elias

Hello Graaf,
- Elias

Hello,
actually el`Ol has already answered to Your questions

I would only like to add that I have never flooded such conventional two-way speaker, only fullrange 8 '' Fostex, various cheap fullrangers and two-way coincident UniQ KEFs
I had intuition that conventional multi-way would be unsuitable but I could be wrong of course.
It is terra incognita.
As regards placement - it all depends, if the loudspeaker is bass-reflex it can react unfavourably to near wall or near corner placement and so on

anyway I would start with near wall placement with the wall in the vicinity of the speakers left reflective without any absorption and listening position at least 3 m away from the speaker

what do You mean by side wall? In mono setup there is no side wall

happy experimenting!
graaf

[el`Ol]

Nice explanation, Oliver, but some fullrangers have tons of copper around the polepiece.

[LineArray]

The curves are the voltage transfer functions for the
lower and the upper trio of drivers in Dipol 08.

The rising voltage in the brilliance region is a least
partially motivated by compensation of output power
decrease.

Additionally the power is shifted towards the upper
trio, to make the array shorter effectively at high
frequencies.

[el`Ol]

Originally Posted by markus76

That's why I bought a Smyth Realiser to hopefully gain some insight how my own spatial perception is affected.

Strange product. Good for our discussion, but does it really made sense to simulate loudspeakers in a virtual room for getting ideal headphone playback? For example: Isn't it an advantage to have low crosstalk with headphones?

[LineArray]

Originally Posted by el`Ol

Nice explanation, Oliver, but some fullrangers have tons of copper around the polepiece.

That may reduce the VC inductivity but cannot compensate
the major effect of VC velocity falling with frequency due to
mass inhibition of the vibrational system.

VC inductivity just adds another 6db /Octave. You cannot
make inducivity zero, you can of course shift the full
12db/Octave power slope towards higher frequencies.

You can just shape the transition regions as said before.

Make a whizzer or a dustcap decoupling from the main cone,
(loosing mass with increasing frequency, yeah ...!)
keep VC inductivity low ... all the tricks used to make a
fullrange instead of "dullrange".

Physics governs. Btw. a whizzer cone e.g. and less stiff glue
for the main cone while attaching the whizzer hard to the VC is
de facto a mechanical 2-Way ... with all the consequences.
Often overlooked by fullrange purists.

But my reason for explaining was that high frequency power
cuttoff is typically due to the physics of the dynamic
fullrange driver itself and not due to the listening room.

To come back to the threads title ...

[markus76]

Originally Posted by el`Ol

Strange product. Good for our discussion, but does it really made sense to simulate loudspeakers in a virtual room for getting ideal headphone playback?

If you take into account the hourly rate of a professional studio, absolutely!

[el`Ol]

But isn't it the tons of equipment that makes a studio expensive, not the specific monitors in a specific room?