Sensitivity VS headroom

[JBL 4645]

This one is for bop/Widget

Does a speaker with say 95db sensitivity means it sounds better played at 85db SPL with headroom of 10db if that makes any sense as, 85db is loud enough?

[Allanvh5150]

Nope. The sensitivity is actually taken using only 1 watt! It is basically a measure of how loud the speaker is compared to another. If one speaker with a 95dB rating is compared to another of 92dB, we know that it is much louder.

Allan.

[pos]

a 95dB/1W/1m speaker will only need 0.1W to achieve a 85dB/1M sound level, whereas a 92dB/1W/1m speaker will need 0.2W for the same sound level.
That means that you can use a lower power amp, of better quality (maybe class A, or tubes...), and also that the voice coil of the driver will see less power and will have less heat to disspiate. That means lower power compression, more dynamics...

But then if you sit 2m away from your speaker you will need 4 times the power you needed at 1m, that is 0.4W to achieve 85dB with your 95db/1W/1m speaker...

But the logic remains: the more sensitive the speaker the less power you need to obtain a given SPL level at your listening location.

[whizzer]

a 95dB/1W/1m speaker will only need 0.1W to achieve a 85dB/1M sound level, whereas a 92dB/1W/1m speaker will need 0.2W for the same sound level.
That means that you can use a lower power amp, of better quality (maybe class A, or tubes...), and also that the voice coil of the driver will see less power and will have less heat to disspiate. That means lower power compression, more dynamics...

But then if you sit 2m away from your speaker you will need 4 times the power you needed at 1m, that is 0.4W to achieve 85dB with your 95db/1W/1m speaker...

But the logic remains: the more sensible [sensitive] the speaker the less power you need to obtain a given SPL level at your listening location.

And, as it follows, the more sensitive the loudspeaker. the less power it takes to produce a given listening level at a given listening distance, and the more headroom there is remaining for cleanly reproducing peak transients. Another way of saying this is that highly efficient systems have a greater dynamic range than less efficient systems. There are some remarkably good sounding systems that sound "remarkably good" only at relatively low volumes because they may have the capacity to produce only (for instance) 87 dB at 1 watt at one meter; as much as ten times the input power may be required to produce a relatively realistic volume only a few feet away at the actual listening position. In actual live music, peaks may be perceived as four or more times as loud as the "average" level. Psycho-acoustically speaking, an increase of ten times as much power is required to produce an output that will be perceived as "twice as loud." The inefficient system soon runs out of steam, regardless of the amount of input power that is applied, for it will reach its maximum output level at some point far below its rated "maximum input power," and any increase is simply converted to heat inside the enclosure. For example, in Bell Labs, Klipschhorns were long used to reproduce sonic booms at over a measured 140 dB, while the typical "audiophile" system maxes out around 110 dB. A symphony orchestra, listening to it from a few rows back in the center, can produce in a typical performance, dB readings averaging 90 dB, with peaks 20-24 dB above that level. Most low efficiency systems cannot accurately produce this wide of a dynamic range given a 90 dB level at the actual listening position. This is perhaps the most serious shortcoming of modern "audiophile" type designs. That symphony orchestra will sound different, tonally, from one venue (listening environment) to another, but of course sounds no less "real" in any one of them than in another--that is, tonality is the principle variable--but all within the "musicality range," but, on the other hand, the dynamics of the playing are remarkably constant. Any hack can learn to play the right notes in the right order, but the virtuoso knows what to emphasize. The virtuoso musician understands dynamics, but, unfortunately, many loudspeaker designers evidently do not.

[MikeBrewster77]

The virtuoso musician understands dynamics, but, unfortunately, many loudspeaker designers evidently do not.

Nor do record companies, unfortunately. A bit OT and a rather long article, but worth the read and relevant to a discussion of dynamics in music reproduction.

http://www.stylusmagazine.com/articles/weekly_article/imperfect-sound-forever.htm

You can scroll down to the section titled "Loudness War" to get straight to the point.

Best,
- Mike

[BMWCCA]

http://www.stylusmagazine.com/articles/weekly_article/imperfect-sound-forever.htmGreat article. Thanks for the link. We all feel the pain!

[toddalin]

Nope. The sensitivity is actually taken using only 1 watt!
Allan.

Maybe at one time, but I don't think so anymore. I think that sensitivity is taken at 2.83 volts input to the speaker.

The amount of wattage is a function of the impedience and this varies with each speaker so the actual wattage also varies.

[JBL 4645]

a 95dB/1W/1m speaker will only need 0.1W to achieve a 85dB/1M sound level, whereas a 92dB/1W/1m speaker will need 0.2W for the same sound level.
That means that you can use a lower power amp, of better quality (maybe class A, or tubes...), and also that the voice coil of the driver will see less power and will have less heat to disspiate. That means lower power compression, more dynamics...

But then if you sit 2m away from your speaker you will need 4 times the power you needed at 1m, that is 0.4W to achieve 85dB with your 95db/1W/1m speaker...

But the logic remains: the more sensitive the speaker the less power you need to obtain a given SPL level at your listening location.

Good point there kinder like when I heard Rich JBL diy 43 series a few years ago. He was running less power and they where TUBE and yet they where of high sensitivity the JBL, the slam kick range was uniform.


Given my distance for JBL control 5, from front is about 2.4 meters only 5 watts to achieve 85db. Sensitivity 89db leaves only 4db out for minimal safe headroom for peaks no?

For the JBL 4645 diy sub only 1 watt required for 85db and that leaves 13db headroom no?

Surrounds JBL control 1 about 1.5 meters watt required for 85db 3 watts for sensitivity 87db leaves 2db, no its half that sorry, 82db plus 82db = 85db for the two sides, so that would 1 watt not 3 watts, so that would leave only 5db headroom no?

Also the high end, the tweeter for the JBL, control 5 what is the sensitivity of that individual speaker?

[timc]

The headroom is not linked to sensitivity. If i have understood correctly it is the difference between the playback level and the peak output lvl of the speaker.


so if you play at 85dB and you speakers can put out 100dB, then you have 15dB of headroom. At the most, music can have a crest factor of about 30dB i think.



-Tim

[JBL 4645]

30db at what distance 1 meter or other?

Hmm, I’m going to do a few tests with pink noise measured at 1m/1w then listen to few moments form Quantum of Boredom it was Bond Rubbish! :barf: The plane VS helicopter has assorted use of ranges of loudness

[timc]

30db at what distance 1 meter or other?

Hmm, I’m going to do a few tests with pink noise measured at 1m/1w then listen to few moments form Quantum of Boredom it was Bond Rubbish! :barf: The plane VS helicopter has assorted use of ranges of loudness


Crestfactor doesn't have anything to do with distance. If you increase volume by xdB at one distance you also increase it by the same in every other distances.


-Tim

[JBL 4645]

Crestfactor doesn't have anything to do with distance. If you increase volume by xdB at one distance you also increase it by the same in every other distances.


-Tim

Can you, please show me an example I catch on fast.

[JBL 4645]

Great article. Thanks for the link. We all feel the pain!

I agree that was a fascinating read, “The Loudness War” interesting subject.

There are some films/CD music that I wouldn’t dare push too far, as the dynamics is too LOUD, Jarre In China is one musical example that I have on DVD.

[timc]

Can you, please show me an example I catch on fast.


If you play at 80dB/1m. Then you play at 74dB/2m 68dB/4m and so on.

The crestfactor say something about the difference between nominal/average level on the source material and the peak level. You then need a headroom equal or greater than the crestfactor at any given time.


-Tim

[JBL 4645]

Hmm, I have a few other speakers I can use on the centre channel to see which one sounds the best?

JBL Control 5 sensitivity 89db
JBL HT1F sensitivity 87db
JBL MR Centre sensitivity unknown
JBL control 1 sensitivity 87db
AWIA SX-77 sensitivity unknown
AWIA SX-N7 sensitivity unknown
Ariston acoustics MSX-07 sensitivity unknown

I’ll see which one sounds best, I have feeling the JBL control 5 will win LOL.

I’ll use a test program and microphone up the centre with spectrumlab at 1m distance and test a monaural program source, how about JAWS 2 chapter 4 time 19 minutes 13 seconds to 20 minutes 19 seconds.

I’d have to set the graph spectrum for a wide window from 0Hz to 400Hz I’ll also run a frequency sweep of each speaker, not sure what this would prove? It’s only fun.:)

[JBL 4645]

I’ve had to delay the test due to an unexpected emergency. One of the speakers wasn’t magnetically shielded while doing the sensitivity test on the Ariston acoustics MSX-07. The vibration started to move the speaker and it slid off the TV LOL and was hanging by the cable for only few seconds till I moved it away from the SONY while noticing some slight discolouring on the left hand low middle part of the screen.:banghead:

For the past hour I’ve been using a technique where you run a screwdriver from side to side of the screen. In most cases it clears the discolouring but not always and not often to full affect.

I’ve moved the JBL 4645 away from the SONY also other speakers tested it again with mild discoloured blob kinder like The Blob itself a purple blob! I’ve been scratching my head thinking what else could still be cursing the discolouring? Then it clicked due to the unexpected incident what other speakers where attached to the SONY!?

The TV speakers themselves! Yes so I had to remove the back covering unplug the back part of the SONY that houses the speakers. I powered it back-up and bingo the issue was resolved.

Now all I have to do is fully remove the speakers from the back part which remove quite easily.

Next up is to put the back-cover back on and place everything back and hope the discolouring stays away.

Yes, yes, I know I know about speakers near TV doesn’t do it any favourers and I should have known better to check first. I guess too many years, with magnetically shielded JBL control 5 has made me too comfortable and I simply forget.

Well that was close call now its time, to get the back-cover back on and get back to the tests which where going fine up until, sigh. I guess I’d have to place the other non-magnetically shielded speakers to one side away from the SONY and test at 1m.

[JBL 4645]

Brilliant screen is back to normal with JBL 4645 sat beside it no discolouring. Sigh the toughest part is getting the back-cover on that was nerve-racking.

Note the screen DVD main menu of JAWS 2 the bright blue colour just to one side left hand middle lower corner was where the purple blob was just over the bubbles coming out of Bruce’s mouth, yes the shark is called Bruce.:D

Right back to the test sigh, this has only taken 2 hours away :banghead: sigh.

[JBL 4645]

Here’s that test that I did a few days ago. The top part of the frequency graph is short frequency sweep from 20Hz to 400Hz why 400Hz, the graph is bit dicey to work with at higher frequencies, haven’t quit mastered that part yet.

The lower graph is record in-room at 1metre distance with monaural source program JAWS 2.

The JBL control 5 was playing as is in active crossover mode. The remaining speakers had the crossover set to full range expect for the LP low pass which was still set at 50Hz. I only needed to swap the LF lead around that’s all.

Note: the gaps or dips in the frequency response sweep on each one.
Note: the colours also in each sweep as well as the waterfalls spectrum some speakers produced a hotter sound over the other.


http://i282.photobucket.com/albums/kk279/SpectrumsubbassLFE13417/JAWS2centrechannelJBLcontrol5.jpg

[JBL 4645]

http://i282.photobucket.com/albums/kk279/SpectrumsubbassLFE13417/JAWS2centreJBLHT1F.jpg

[JBL 4645]

http://i282.photobucket.com/albums/kk279/SpectrumsubbassLFE13417/JAWS2centreJBLMRcentre.jpg

[JBL 4645]

http://i282.photobucket.com/albums/kk279/SpectrumsubbassLFE13417/JAWS2centreJBLcontrol1.jpg

[JBL 4645]

http://i282.photobucket.com/albums/kk279/SpectrumsubbassLFE13417/JAWS2centreAWIASX-77.jpg

[JBL 4645]

http://i282.photobucket.com/albums/kk279/SpectrumsubbassLFE13417/JAWS2centreAWIASX-N7.jpg