OB speakers and room acoustics

[Barleywater]

Juhazi,

In your waterfall plot:

AC power signal and harmonics along with IMD below fundamental appear to peak at 2nd harmonic only about 25-30dB below test sweep. Hope this was really quite sweep in what otherwise appears to be very quiet room. A well used Auddessey setup could really improve your surround setup.

Thanks for sharing.

 

[Barleywater]

Rudolf,

That reminds me, must update my REW; yes current version does have smoothing and rise time features, that, yes, can make for some really unbelievably good looking results. Not that you haven't got the nuts and bolts drilled down.

 

[Rudolf]

Sorry, I got the x-axis wrong.
It was the first CSD in REW for me.
This should be more realistic:

 

[Barleywater]

Rudolf,

Really quite nice looking. What is FR with <3ms window/gate look like?

 

[Rudolf]

Measured from the listening position (which is just 1.2m distance from the speakers) with a 3ms gate applied:

And now with mic compensation and 1/24 smoothing applied too

 

[markus76]

Regarding the waterfall graphs:
I think we're looking at data that is hard to interpret because relevant information is missing.
What is the noise floor in each measurement?
What window size was selected? Window size is always a trade-off between time and frequency resolution. If time resolution is good, frequency resolution is bad and vice versa.

 

[Barleywater]

Measured from the listening position (which is just 1.2m distance from the speakers) with a 3ms gate applied:
View attachment 349062
And now with mic compensation and 1/24 smoothing applied too

Thanks Rudolf,

Yes this result is consistent with floor/ceiling reflections/modes to what I get with Pluto Clone at similar distance.

 

[Rudolf]

Yes this result is consistent with floor/ceiling reflections/modes to what I get with Pluto Clone at similar distance.

I'm a bit confused. I could hardly imagine a floor/ceiling interaction more different from the Pluto than my design. So where do you see the consistency in our "results"? Would you care to explain?

Rudolf

 

[markus76]

"floor/ceiling reflections/modes" within a 3ms window? There are none. 3ms is just a 1m bubble around the speaker/mic.

 

[natehansen66]

Measured from the listening position (which is just 1.2m distance from the speakers) with a 3ms gate applied:

How are you getting data down to 70hz with a 3ms gate?

 

[Rudolf]

How are you getting data down to 70hz with a 3ms gate?

You don't get it. ARTA would show a "no value" bar up to 300 Hz for that 3 ms window, but I would consider anything below 1 kHz as not reasonable enough for discussion.

 

[markus76]

You don't get it. ARTA would show a "no value" bar up to 300 Hz for that 3 ms window, but I would consider anything below 1 kHz as not reasonable enough for discussion.

REW does similar. Your window is probably longer than 3ms. For example this is a 4ms window - REW correctly calculates 250Hz:

 

[Barleywater]

I'm a bit confused. I could hardly imagine a floor/ceiling interaction more different from the Pluto than my design. So where do you see the consistency in our "results"? Would you care to explain?

Rudolf

Sorry to distract from discussion of reflections.

Yes, the conditions for seeing floor/ceiling require longer window and I see nulls with 12ms window; so what causes 400Hz, 800Hz, 1600Hz notches in dipole 1m, 3ms measurements? Dipole with your design doesn't achieve driver integration at 1m listening distance. Drivers at 400Hz are more out of phase than they are in phase. Doubtless this is audible listening to sweep; but likely difficult to pick out with listening program.

Pluto Clone response at 1m, 186ms window gated to 3ms:

An externally hosted image should be here but it was not working when we last tested it.

At this listening distance direct/reflected ratio makes getting strong contralateral reflection that Markus is exploring difficult/impossible to get without very specialized/dish-like reflector.

 

[markus76]

186ms window gated to 3ms

Changing window size with frequency?

 

[lolo]

I measured the response (to one side only) with a reflective panel and with 7 cm thick Basotect covering that panel

Rudolf

Hi Rudolf, what subjective difference does the basotect make to you?

Markus, do you mean full range contralateral reflexion?

 

[markus76]

Markus, do you mean full range contralateral reflexion?

Yes, as much as possible. The whole idea is not new. It was known as "Haas kickers" in LEDE control rooms. While providing some spaciousness they have been abandoned in control room design because of the imaging/transparency tradeoff.

 

[lolo]

I see! So you absorb the ipsilateral reflexion and use the contralateral one?

 

[markus76]

The Nathan's have a rather narrow directivity so there's no need to attenuate the ipsilateral reflection. Toe-in takes care of that. I do absorb as much low frequency around the speaker as possible though.

 

[Rudolf]

Your window is probably longer than 3ms. For example this is a 4ms window - REW correctly calculates 250Hz:

Thanks for explaining, Markus. And yes, my REW window must have been 7 ms or longer.

Yes, the conditions for seeing floor/ceiling require longer window and I see nulls with 12ms window; so what causes 400Hz, 800Hz, 1600Hz notches in dipole 1m, 3ms measurements?

For the curious I add the 3 ms (left) and 10 ms (right) gated measurements for the woofer part, middle-high part and combined of the Swing dipole:

The result should be better, I know .

 

[Rudolf]

Hi Rudolf, what subjective difference does the basotect make to you?

lolo,
the contralateral reflection is weaker than -12 dB compared to the initial impulse. That is too faint for me to make any judgement with authority. And I see no way to increase the contralateral reflection in my situation at all. In fact I now have the Basotect instead of the reflectors at the contralateral angle. Just experimenting.