Help re: measurement graph for S4700s

[Dave_72]

I wondering if one of yous can explain this to me. Thanks a mil!

64071

[pos]

look at the insert on the 2nd page here: http://www.jblpro.com/pub/recording/lsr32.pdf

[Dave_72]

Thanks. That kind of helps! lol. I just wanted to see how low the bass goes going by this graph.

[ivica]

look at the insert on the 2nd page here: http://www.jblpro.com/pub/recording/lsr32.pdf

Hi POS,

It would be nice if You can give use much more words about the exact meaning and the measurements procedure that would correspond to each of the curves, starting from the characteristics of the room (or place) where the measurement have been done .
I can imagine that at JBL lab there is a kind of "standard listening room, (or rooms)".
I can imagine that in such room there are number of measuring places, from where directivity data can be measured, and a kind of averaging can be calculated,
but it is for me difficult to understand the meaning of "DI of First Reflections" or "DI of On-Axis Response".

I have found something here:

http://www.acousticfrontiers.com/20131129controlled-directivity-speakers-open-up-your-acoustic-treatment-options/

regards
ivica

[pos]

Hi Ivica,
I think these curves are generated based on multiple measurements in an anechoic chamber, using an empirical "typical room" model for the first reflexion response and di curves.
It is all explained In Toole's in Sound Reproduction (http://www.amazon.com/Sound-Reproduction-Acoustics-Psychoacoustics-Loudspeakers/dp/0240520092).

[ivica]

Hi Ivica,
I think these curves are generated based on multiple measurements in an anechoic chamber, using an empirical "typical room" model for the first reflexion response and di curves.
It is all explained In Toole's in Sound Reproduction (http://www.amazon.com/Sound-Reproduction-Acoustics-Psychoacoustics-Loudspeakers/dp/0240520092).
Hi Pos,

so You believe that these curves have been measured in the lab. anechoic chamber and then "mathematically converted into the virtual reality", if that would be the truth, then almost any speaker would be "unbeliavable real"

regards
ivica

[audiomagnate]

What's causing the huge suckout in the on-axis response around 14 K?

[ivica]

What's causing the huge suckout in the on-axis response around 14 K?

Hi audiomagnate,
I believe that it is "comb effect" VHF and UHF driver time (spatial) misalignment
Look at the VHF and UHF position.

regards
ivica

[pos]

so You believe that these curves have been measured in the lab. anechoic chamber and then "mathematically converted into the virtual reality", if that would be the truth, then almost any speaker would be "unbeliavable real"
Hi Ivica,
Not sure what you mean.
I think this approach is perfectly reasonable, and allows high resolution measurements and some amount of room behavior prediction (as much as can be done, really).

It shows defects that would not be visible in normal in-room measurements, because these would be either windowed (with an inherent loss of resolution down low) or blurred with multiple reflexions plus mandatory frequency smoothing...

The only thing that has to be evaluated on-site in the low frequency response (boundary reenforcement and mode behavior, room pressure support... aka room gain), and possibly the high frequency slope (X curves, behind screen compensation, etc.).

[Mr. Widget]

Hi audiomagnate,
I believe that it is "comb effect" VHF and UHF driver time (spatial) misalignment
Look at the VHF and UHF position.

regards
ivicaThat's exactly what it is... as it turns out while far from ideal it isn't particularly harmful to the sonics of the system. All of the JBL systems with mid horns and super tweeter have similar "issues". In the DIY world where folks have attempted to "time align" their speakers, they typically still have similar issues that are simply moved around on the graph.


Widget

[4313B]

That's exactly what it is... as it turns out while far from ideal it isn't particularly harmful to the sonics of the system. All of the JBL systems with mid horns and super tweeter have similar "issues". In the DIY world where folks have attempted to "time align" their speakers, they typically still have similar issues that are simply moved around on the graph.In the case of the Everest they just unhook the 045's and take their measurements. No sense letting a super tweeter muck up the curves of a $75k system. Can't really get away with that in a system where the super tweeter has a crossover frequency low enough to matter.

[ivica]

In the case of the Everest they just unhook the 045's and take their measurements. No sense letting a super tweeter muck up the curves of a $75k system. Can't really get away with that in a system where the super tweeter has a crossover frequency low enough to matter.

Hi 4131B,

Do You think of something like :
http://www.audioheritage.org/vbulletin/showthread.php?35028-S3900-S4700-dbx-DriveRack-260&p=354856&viewfull=1#post354856

regards
Ivica

[ivica]

What's causing the huge suckout in the on-axis response around 14 K?


Hi audiomagnate,

"Comb filter effect"
http://en.wikipedia.org/wiki/Comb_filter
http://www.roger-russell.com/columns/combfilter2.htm
"....Comb filtering can be defined as "the frequency response caused by combining a sound with its delayed duplicate. The frequency response displays a series of peaks and dips caused by phase interference. The peaks and dips look like the teeth in a comb, with very narrow, deep notches where signals are attenuated." ........."

so such effect is present when there is two sources of sound that are physically apart, while reproducing the same frequency region.
Such distance would introduce a relative time delay between sound waves produced from each of the drivers, and at some frequencies these sound waves would add, and on some they would subtract, so producing a kind of 'bounce' in the FR measurements, and some sharp notches.

".....Some frequencies will cancel and others will be reinforced, which can dramatically change the tonal color of the sound........"

I think that on S4700, as VHF and UHF drivers are physically apart (looking towards z-axis, perpendicular to the baffle plane) , that a kind of time delay is present at some high frequency ( that corresponds to Lambda/2 length) relative to the measurement microphone, and as filers are not 'ideal (brick-wall)'. Even the filers are ideal, and a kind of time compensation has been done, on some other (off-axis) angles such behavior would be seen, as driver are apart in Y-axis too.
But, hopefully, owing to the reflections, and other "irregularities" such comb-filter effects at high frequencies are of less importance.

regards
ivica

[Mr. Widget]

".....Some frequencies will cancel and others will be reinforced, which can dramatically change the tonal color of the sound........"
I think this is clearly audible in the middle and lower frequencies. The narrow band peaks and troughs that these JBL systems exhibit are high up in frequency where we are less sensitive to this effect and the peaks and troughs are quite narrow in frequency making them difficult to hear.


Widget

[Dave_72]

That's exactly what it is... as it turns out while far from ideal it isn't particularly harmful to the sonics of the system.


Widget

Ok, cool. Good to know. Thanks.

[Mctwins]

Hallo!

I found this....Interesting!

http://www.embedded.com/print/4211634

:)