or Connect
AVS › AVS Forum › Audio › Speakers › List of Reference Level, High Sensitivity & SPL Speakers
New Posts  All Forums:Forum Nav:

List of Reference Level, High Sensitivity & SPL Speakers - Page 14

post #391 of 820
Quote:
Originally Posted by Bill Fitzmaurice View Post

If they have differing sensitivity then the driver with lower sensitivity will have to be driven to a longer excursion for the same dB level, and therefore will have higher THD.
Isn't it that the lower sensitivity driver will require more power to achieve the same excursion as the higher sensitivity one? Other things being equal that is.
post #392 of 820
Thread Starter 
Quote:
Originally Posted by popalock View Post
Question: Does speaker sensitivity have anything to do with dynamic capability? If one was to have ample power to push any speaker of their choosing, would sensitivity matter?
Context: For instance, say I have a 1000 watt x 2 amp and the speakers I am contemplating are 200 and 400watts respectively. Both have varying sensitivity, but both are "dynamically capable" of providing 125db of sustainable output. (not even sure if that is feasible)
If the power is available, would sensitivity matter? If one was to hook up one of those speakers to each channel of my 1000watt amp, would the only difference be having to adjust the gain of my amp for the different power requirements of each speaker to reach their full dynamic capabilty (125db)?
 

Where sensitivity comes into play is:

  • Was the 200w (let's say we're talking average levels, since you said "sustainable") speaker "sensitive enough" to yield the 125dB within the 200w (and you established hypothetically that is is)
  • But also is it in "loafing mode" with low distortion? Or "straining mode" where there's high distortion? If the

 

We can use examples in the simple SPL calculator.

Since you said "sustained" and chose a high SPL, let's assume we're talking about an outdoor concert application and 125dB is a meter away, where we're trying to deliver at least 85 dB continuous level to people sitting far away from the speakers.

Barely sensitive enough

A speaker of 102dB sensitivity (easily plausible) that can handle 200w, one meter away, outdoors, would yield 125dB. Our audience sitting 330ft away would get 85dB.

But this is theoretical. In probability, since 200w was the max for the speaker, there's going to be 4-5dB of compression, so we're only doing 120dB at one meter.

The engineer, noticing a loss of 5dB and not achieving his desired level, stupidly feeds more watts to the speaker: Another 440w to overcome the 5dB loss for a total of 640dB. But that just heats the voice coil up more, and probably it's breaking. So, a quick look at specs showed the speaker could barely do what we needed, but due to thermal compression effects, we'll either be short several dB or blow the speaker.

 

Not to mention that probably at max rated power there's a lot of distortion. The same SPL with distortion will sound what we call "loud/bad" whereas if we could eliminate the distortion from stress we might call it "loud/good". The former ellicits winces. The latter: grins.

 

A little more than sensitive enough

Take the same case, but a 105dB speaker. The calculator shows that feeding 100w should yield our 125dB desired SPL at a meter. Yay! But really (if we follow the JBL Pro data I posted in another thread), there's 2dB compression at 50% power. So we get 123dB in practice. We need to send it 160w or more to yield our desired 125dB.

Because we're not at the ceiling we'll have less distortion.

 

Way more than sensitive enough

Next, we'll use a 110dB speaker and feed it 32w to get 125dB. Huzzah! But we know at 1/10th power we get .5dB compression loss. So maybe we need to send it 40 watts.

Way less distortion.

 

105dB sensitive but higher power handling

Lastly, let's take the second 105dB speaker who was rated at 400w RMS power handling.

100w theoretically yields 125dB that you wanted. That's 1/4 of the rated power handling of this pro speaker, and it'll have between .5dB and 2dB of compression, so let's call it 1dB.

In practice we'll have to send it 130w, which is still only a third of its max capability, so it'll exhibit less distortion: happy audience.

 

And that last is the answer to your question. We vary sensitivity so that the speaker can do more work with that same power. We vary power-handling so that whatever we do send it is a smaller percentage of its max capability (of destruction!) so that it sounds better (or doesn't break). Both numbers relate in tandem. All things being equal, you can vary either, I think, and achieve your goal.

 

But it's more hazy than you think. Was the sensitivity spec'd as anechoic? In-room? In a corner? I've even seen it spec'd as "with two speakers"! Is it really a 4ohm speaker, so you need to adjust their number by 3dB to compare apples to apples? At what distance was the sensitivity measured? Is it measured when the speaker has been stressed a long time? or just warming up? Over what frequency range: just an easy-to-drive 1000Hz? Does it include 100Hz? Does it include 30Hz? Do you care if includes 30Hz if you're going to crossover at 100Hz? How many watts longterm, shortterm, will make it distort too much?

 

Not to mention all the other aspects that go into speakers: coverage, keeping the same response over the desired angles, how fast does the spl drop (cylindrical patterns from line arrays maintain twice the volume over distance as traditional speakers), the lobing, the i/o jacks, the looks, the weight, the size, etc.

post #393 of 820
Thread Starter 
Quote:
Originally Posted by JerryLove View Post

Because it's back to the OP, and because I'm hoping it will do more to *not* reignite the feud that ignite it:
Individual speakers are going to be individual. If you are concerned about distortion being an issue: find out of the speakers that interest you distort heavily at the volumes you intend to play them. No matter who is right on the general rule: having accurate data on the specific speakers you are looking at will not lead you astray.
I would really like to believe that people on both sides of this debate can agree with that.

Sounds good.

I do wish the speaker manufacturers would publish distortion characteristics for average levels and peak levels of frequency range.

 

Not realistic, but I do believe the pro reviewers should give us this information. With the advances software like EFT and REW, isn't the technology accessible to large publications? Bill, if you had to, couldn't you measure distortion over some RMS stress test and use shaped burst tones to show distortion at peak levels or other's speakers (theoretically from a capability standpoint)?

 

I know Stereophile used to show distortion, but that wasn't for peaks but just average levels, I think?

 

If y'all were going to write the standard/template for such speaker reviews, what thresholds would you want to see? What types of distortion at what percent?

And beside that, would you pick a reasonable number to test to?

Like...

"We at XYZ Publication feel, and studies have shown, that 10% of ABC distortion is noticeable and objectionable. The tests of this speaker showed that:

  • At 10% of ABC distortion the speaker output 109dB at one meter in half space. (show graph varying frequency and watts, because it may unintuitively rise exponentially and 104dB was at 3%). This required 85w. [notes on how the tweeter is capable of more, but as we see the distortion was below crossover and the woofer is weak]
  • At 100w, ABC distortion was at 30% [notes on how reviewers ran screaming from the room] (this is a common number for receiver RMS)
  • At 200w, ABC distortion was at 70% [notes on how reviewers ran screaming from the room] (this is a common number for receiver peak w for .5s or external amps)
  • Shaped burst tones around:
    5000Hz at 105dB from 12 feet away (which was 18w) had .5% of ABC distortion (this behaved admirably)
    1200Hz at 105dB from 12 feet away (which was 58w) had  2% of ABC distortion (this is okay, as the woofer comes into play at crossover)
      100Hz at 105dB from 12 feet away (which was 88w) had  4% of ABC distortion (this has enough output for ref level @ 12 feet)
  • This speaker's output capability is enough for reference level at a 12 foot seating distance with modest amplification. "
post #394 of 820
Thread Starter 
Quote:
Originally Posted by JerryLove View Post
Does anyone here believe that distortion suddenly went down at the same power level?

I think we'll use less power now because it's more sensitive?

Quote:
Originally Posted by JerryLove View Post
Does anyone believe that my crossover taking away efficiency has made the drivers distort more?

Bill said we should look at sensitivity across the frequency range. You're right, we shouldn't disregard a speaker because it lacks sensitivity in a range we don't care about. JBL cinema speakers will say their band is up to 16kHz so they don't have to worry about 20kHz. They must figure the customers for that line (theaters) don't care either.

 

But conversely, we shouldn't see a 96dB number and calculate that'll be enough, when that number was the manufacturer really saying, "Well, the max sensitivity is for our horn driver. Yup: 96dB. Pay no attention to the 6" woofer that'll yield a fraction of the horn when pushed with more watts." The FR chart with a 10w stimulus might show "flat," but at 60w or 120w, you see the woofer's response drop off precipitously. We'll never know that unless we the chart of how the speaker does when pushed. (or through experience know what a 6" low displacement driver can do).

 

I swear, it's as we were seeing the specs (and often reading the reviews) for cars who were only tested around the block. But never corner g-forces, max acceleration 0-60, top speed in straightaway, etc.!

 

Oh, yay, reviewer. You verified or contradicted what the speaker did when fed a watt. How about what it does with 100w, 300w that the manufactuer says it'll take?

Imagine if tires were rated for a m.p.h. where exceeding that guaranteed explosive failure?! "Here we go...103, 104, 105...kaBLAM right on cue." That's what specs that only indicate destruction are like. They have their place, but I also care about what 90% of those max watts will do. 90% distortion isn't acceptable either.

post #395 of 820
Quote:
Originally Posted by Eyleron View Post

Bill, if you had to, couldn't you measure distortion over some RMS stress test and use shaped burst tones to show distortion at peak levels or other's speakers (theoretically from a capability standpoint)?
Anyone can, with a laptop, free software and $100 worth of hardware. When you needed $5k or more of testing gear to do so there was some justification for manufacturers not providing detailed data, but not much, since the pro-sound guys have been doing so since the 70s. Today there's no excuse.
post #396 of 820
Quote:
Originally Posted by Eyleron View Post

I think we'll use less power now because it's more sensitive?

Likely yes: but go back to the claim I was disputing. The claim was that, for a given percent of maximum power (say 80%), a more sensitive speaker will always have less distortion than a less sensitive speaker.

By adding the baffle I increased sensitivity. Did I decrease distortion at the same number of watts? Of course not. Therefore the claim which I am disputing is proven wrong.
Quote:
Bill said we should look at sensitivity across the frequency range. You're right, we shouldn't disregard a speaker because it lacks sensitivity in a range we don't care about. JBL cinema speakers will say their band is up to 16kHz so they don't have to worry about 20kHz. They must figure the customers for that line (theaters) don't care either.

If the discussion has been about ripping out the raw drivers and measuring them without crossover and then generating some sensitivity graph; then I missed that.
Quote:
But conversely, we shouldn't see a 96dB number and calculate that'll be enough, when that number was the manufacturer really saying, "Well, the max sensitivity is for our horn driver. Yup: 96dB. Pay no attention to the 6" woofer that'll yield a fraction of the horn when pushed with more watts." The FR chart with a 10w stimulus might show "flat," but at 60w or 120w, you see the woofer's response drop off precipitously. We'll never know that unless we the chart of how the speaker does when pushed. (or through experience know what a 6" low displacement driver can do).

Usually in my experience: speaker (enclosures with one or more drivers and, if more than one, some sort of crossover) manufacturers do not give you sensitivity of each driver but rather of the entire assembly over it's stated frequency response range and with a maximum deviation (usually 1.5db or 3db).

Conversely, driver manufacturers cannot be expected to know what other drivers or what enclosures you might put their product in. As such, they can only spec for their own driver.
Quote:
Imagine if tires were rated for a m.p.h. where exceeding that guaranteed explosive failure?! "Here we go...103, 104, 105...kaBLAM right on cue." That's what specs that only indicate destruction are like. They have their place, but I also care about what 90% of those max watts will do. 90% distortion isn't acceptable either.

Maximum watts don't indicate guaranteed and immediate destruction, and certainly not to that level of precision. (and what does this even have to do with the topic?)

And tires *are* rated for speeds and *will* explode (fail catastrophically) if you exceed that speed. Not instantly and not by 1%; but that's not true for speakers either.
post #397 of 820
Claim: More sensitive = less distortion @ same percent of max power.
Disproven by example: Shove a baffle-less speaker into an infinite baffle and you get +6db sensitivity with zero change in distortion.

Claim: Less sensitive speaker = greater XMax (for given frequency and SPL).
Disproven by: formula for computing X-Max (which does not include a "sensitivity" variable at all)

Claim: Greater XMax = greater distortion.
Rendered irrelevant by: showing the lack of relationship between XMax and sensitivity.
Disproven by example: ribbon tweeters distort despite near-zero x-max.
post #398 of 820
Thread Starter 
Quote:
Originally Posted by JerryLove View Post

Claim: More sensitive = less distortion @ same percent of max power.
Disproven by example: Shove a baffle-less speaker into an infinite baffle and you get +6db sensitivity with zero change in distortion.
Claim: Less sensitive speaker = greater XMax (for given frequency and SPL).
Disproven by: formula for computing X-Max (which does not include a "sensitivity" variable at all)
Claim: Greater XMax = greater distortion.
Rendered irrelevant by: showing the lack of relationship between XMax and sensitivity.
Disproven by example: ribbon tweeters distort despite near-zero x-max.

More sensitive = less distortion @ same percent of max power.
I thought this should be:

  • More sensitive = clip amp less and compared to a less sensitive speaker with the same power handling, you'll use a smaller percentage of max power.
     

Less sensitive speaker = greater XMax (for given frequency and SPL).

  • Less sensitive speaker = would need greater XMax to achieve same SPL @ freq
    Analogous in physics: to achieve the same momentum while reducing speed, you need to increase mass (mass times velocity = momentum). 
    This is not disproved by: "The formula to calculate mass does not include 'speed' variable at all."
post #399 of 820
I can only deal with the assertion put in front of me. That assertion was that distortion would necessarily be higher on a 80db sensitive 100W speaker running 90W than it would be on a 90db sensitive 10W speaker running 9W given unlimited amp power... a claim which I fund both unsupportable and unsupported (though I make not comment as to what might "often" be the case as I don't know).
Quote:
Originally Posted by Eyleron 
Less sensitive speaker = would need greater XMax to achieve same SPL @ freq

A single 12" cone driver in a sealed box in an open space.producing a 40Hz wave at 100db would have an X-Max of 4.7953mm

Doesn't matter if the driver was 80db sensitive or 200db sensitive.

SPL = 112 + 10 * log(4 * pi^3 * Ro / c * (num * (.83 * diam)^2 * pi / 4 * Xmax)^2 * f^4)

If you want an example: just throw a -10db crossover in front of it. Won't change the x-Max even a little. All it will do is increase the power required to achieve it. The air doesn't know or care how hard it is to move those 113 square inches, only how far they are moving.
Quote:
Analogous in physics:

Analogies get you somewhere for a while and then they break... like cars.

And you can't have "an analogy in physics" because this *is* physics.
Edited by JerryLove - 11/30/12 at 2:08pm
post #400 of 820
Thread Starter 
Quote:
Originally Posted by JerryLove View Post
Analogies get you somewhere for a while and then they break... like cars.

Oh sure, I can't use analogies but you get similes?!  wink.gif

post #401 of 820
Quote:
Originally Posted by JerryLove View Post

I can only deal with the assertion put in front of me. That assertion was that distortion would necessarily be higher on a 80db sensitive 100W speaker running 90W than it would be on a 90db sensitive 10W speaker running 9W given unlimited amp power... a claim which I fund both unsupportable and unsupported (though I make not comment as to what might "often" be the case as I don't know).
A single 12" cone driver in a sealed box in an open space.producing a 40Hz wave at 100db would have an X-Max of 4.7953mm
Doesn't matter if the driver was 80db sensitive or 200db sensitive.
SPL = 112 + 10 * log(4 * pi^3 * Ro / c * (num * (.83 * diam)^2 * pi / 4 * Xmax)^2 * f^4)
Horn-loading increases the effective size of that driver, right? (The required xmax to achieve a given SPL is less because of the increased coupling efficiency)
Many HE speakers are horn-loaded.
post #402 of 820
Quote:
Originally Posted by psgcdn View Post

Horn-loading increases the effective size of that driver, right? (The required xmax to achieve a given SPL is less because of the increased coupling efficiency)
Many HE speakers are horn-loaded.

Yes. Putting the speaker in a corner would also lower the x-max required for a given SPL at a given frequency.
So would embedding the speaker in a wall.
So would porting, or adding a TL tuned near the frequency in question.

Lots of things will change the X-Max / SPL / Freq relationship.

Some (many? most?) of them will *also* change efficiency. The three I just mentioned would raise the efficiency (because all three are used to raise SPL).

But as you point out: that's not a casual relationship.

This speaker (88db) http://www.madisoundspeakerstore.com/approx-12-subwoofers/sb-acoustics-sb34swnrx-s75-12-woofer/
and this speaker (84db) http://www.madisoundspeakerstore.com/approx-12-subwoofers/aurasound-ns12-794-4a-12-subwoofer/
and this one (90db) http://www.madisoundspeakerstore.com/approx-12-woofers/goldwood-gw-1244-12-woofer-4-ohm/

Have the same x-max at a given SPL/frequency (until failure begins; we are discussing within operating parameters.).

This one has a largeer x-max, despite being more efficient (89.5db) than two of the three above http://www.madisoundspeakerstore.com/approx-8-woofers/goldwood-gw-8024-8-woofer-4-ohm/
post #403 of 820
Quote:
Originally Posted by psgcdn View Post

Horn-loading increases the effective size of that driver, right?
By as much as a 10:1 ratio.
post #404 of 820
Next thing you know JerryLove is gunna argue the point that the sun is not really hot even though all signs point to it as being hot lol.biggrin.gif
post #405 of 820
OK, I will try to give an example of two speakers playing the same volume but one is clean and the other is not. This happened to me so it is just and experience.

Speaker one(Paradigm) rated to handle 250 watts and peak of 114 dBs. Speaker 2(JTR triple 8) rated at 1200 watts and 129 dBs. I play the Paradigm until it starts to sound bad and measure the peak spl. It measured 105 dBs. I then turned the other speaker to 105 dBs and it did not sound bad at all. The Paradigm has a sensitivity of 92 dBs and the JTR has a sensitivity of 99 dBs. I used a 400 watt amplifier into both. The difference here is that the low sensitivity speaker falls apart at a much lower level requiring too much power that it can not handle. The paradigm is supposed to play loud enough but it just does not. My question is this, how do you know which low sensitivity speaker will play loud enough? That paradigm was distorting like crazy at 10 dBs less than specs.
post #406 of 820
Quote:
Originally Posted by MKtheater View Post

My question is this, how do you know which low sensitivity speaker will play loud enough? That paradigm was distorting like crazy at 10 dBs less than specs.
Chances are the Paradigm maximum SPL was a calculated figure, as are virtually all maximum SPL claims. They take the base sensitivity, add the dB increase that would result from driving it at the maximum thermal power rating, and that's what they claim. But as the SPL increase is not linear with respect to the power increase, and the thermal power rating only says how much power the speaker should handle before it burns out, not how much it will handle before THD becomes excessive, it's a meaningless spec. They should measure the actual output of the speaker at the industry standard of 10% THD, but AFAIK no one does.
post #407 of 820
Examples are useful in disproving rules but not in proving them (they are anecdotes).

What is the THD of an 8" 95db@1w@1m speaker with 100w max power being run at 10w with a 300Hz sine wave?

Can't answer that?

Know why? Because distortion is effected by more then efficiency, volume, percent max wattage, frequency, and driven wattage. It is affected by things like the design and materials of the driver, and the design of the enclosure.
An infinitely ridged cone material, for example, would never experience cone breakup (one cause of distortion). It would have other problems (ignoring that it's impossible to being with), but not that. One cause of distortion gone completely.

There can be trends. I couldn't comment to trends as I lack statistical data (see anecdotes comment above); but there's no obligatory relationship that makes distortion a function of efficiency.

I have gotten the impression that this will shock people: but I'm not advocating a specific type of speaker. I'm merely dealing with provably false facts presented by other posters in support of their advocacy.

Reddig: I miss the point of your post.
post #408 of 820
BTW, this is just one example, I have tried this a many speakers. The proof is in measurements. Take a speaker and measure when and where it compresses and how the THD is effected, in your room. Take the next speaker and do the same thing. See which one sounds better, maybe a speaker at 50% THD will sound better than another at 5%, who knows. I am not going to just believe that this speaker is expensive, uses exotic materials, so it is automatically better. I don't measure drivers, I measure the whole speaker and at my LP where it matters. If one speaker can't play loud enough at my LP then it goes away, I don't care what the materials are made of. The whole thread is about High sensitivity of speakers, I am sorry if your speakers are not listed but many are not because the OP is interested in reference level listening in his room so he created this thread.
post #409 of 820
What is your nominal SPL at your listening position MK and at what distance?
post #410 of 820
Quote:
Originally Posted by Chu Gai View Post

What is your nominal SPL at your listening position MK and at what distance?

I listen to music with 100 dB peaks and movies at reference when I compare. I sit about 13 feet away from the screen. My older Triads were listed at 92 dB sensitive and 200 watts power handling and could not play at reference during action flicks but they could play louder(cleaner) than the paradigms. I am just saying I don't trust specs and measure myself. My Klipsch THX KL-650's could play reference but they are not as sensitive as their specs say. This of course could be everyone else under rating their sensitivities. My M&K speakers had the same sensitivity as the Triads and Paradigms and could easily play reference.
post #411 of 820
Quote:
Originally Posted by JerryLove View Post

There should be no difference in performance between a hypothetical 90db efficient 100w speaker and a hypothetical 80db efficient 1000w speaker and a hypothetical 100db efficient 10W speaker; assuming you have enough power to drive them all.
Assuming the same distortion levels at the same fraction of total power a given speaker can withstand, this statement is correct. As a practical matter though there are probably no 87dB 6000W woofers to match JBLs 2226H 97dB/600W for example. I think it is safe to say that, on average, in order to reach the same output levels, lower sensitivity speakers are driven closer to max power handling levels than high sensitivity ones.
post #412 of 820
Quote:
Originally Posted by zheka View Post

Assuming the same distortion levels at the same fraction of total power a given speaker can withstand, this statement is correct. As a practical matter though there are probably no 87dB 6000W woofers to match JBLs 2226H 97dB/600W for example. I think it is safe to say that, on average, in order to reach the same output levels, lower sensitivity speakers are driven closer to max power handling levels than high sensitivity ones.

Yes, and closer to max you get you run into compression and higher THD.
post #413 of 820
Quote:
Originally Posted by JerryLove 
There should be no difference in performance between a hypothetical 90db efficient 100w speaker and a hypothetical 80db efficient 1000w speaker and a hypothetical 100db efficient 10W speaker; assuming you have enough power to drive them all.
Quote:
Originally Posted by zheka View Post

Assuming the same distortion levels at the same fraction of total power a given speaker can withstand, this statement is correct. As a practical matter though there are probably no 87dB 6000W woofers to match JBLs 2226H 97dB/600W for example. I think it is safe to say that, on average, in order to reach the same output levels, lower sensitivity speakers are driven closer to max power handling levels than high sensitivity ones.

As a rule of thumb, that would be obvious; but it cannot be applied to two random drivers as it is not universal (max power handling can vary considerably and is not tied to sensitivity).
Quote:
Originally Posted by MKtheater View Post

Yes, and closer to max you get you run into compression and higher THD.

On a given speaker, yes. You cannot compare two different speakers by "percent of max power used". THD cannot be computed solely from "percent of max power" without also knowing the speaker in question. Two different speakers will produce different THD at the same percent of max power.
post #414 of 820
Quote:
Originally Posted by JerryLove View Post

Quote:
Originally Posted by zheka View Post

Quote:
Originally Posted by JerryLove View Post

There should be no difference in performance between a hypothetical 90db efficient 100w speaker and a hypothetical 80db efficient 1000w speaker and a hypothetical 100db efficient 10W speaker; assuming you have enough power to drive them all.
Assuming the same distortion levels at the same fraction of total power a given speaker can withstand, this statement is correct. As a practical matter though there are probably no 87dB 6000W woofers to match JBLs 2226H 97dB/600W for example. I think it is safe to say that, on average, in order to reach the same output levels, lower sensitivity speakers are driven closer to max power handling levels than high sensitivity ones.
As a rule of thumb, that would be obvious; but it cannot be applied to two random drivers as it is not universal (max power handling can vary considerably and is not tied to sensitivity).

that's why I used the "on average" qualification. i could probably say "in the vast majority of the cases" instead and still be accurate.
post #415 of 820
Quote:
Originally Posted by zheka View Post

that's why I used the "on average" qualification. i could probably say "in the vast majority of the cases" instead and still be accurate.

Sorry if I missed that. I could not comment on a statistical claim as I lack data that isn't anecdotal.
post #416 of 820
Quote:
Originally Posted by MKtheater View Post

BTW, this is just one example, I have tried this a many speakers. The proof is in measurements. Take a speaker and measure when and where it compresses and how the THD is effected, in your room. Take the next speaker and do the same thing. See which one sounds better, maybe a speaker at 50% THD will sound better than another at 5%, who knows. I am not going to just believe that this speaker is expensive, uses exotic materials, so it is automatically better. I don't measure drivers, I measure the whole speaker and at my LP where it matters. If one speaker can't play loud enough at my LP then it goes away, I don't care what the materials are made of. The whole thread is about High sensitivity of speakers, I am sorry if your speakers are not listed but many are not because the OP is interested in reference level listening in his room so he created this thread.

So, can we talk a sec on compression and how it relates to THD?

Once a speaker initally starts to compress it does so at a specific fq, not necessairly across the spectrum, correct? So, though your db meter reads 105db, if you were to read your omnimic measurements, as you increase the volume other parts of the spectrum can still increase as well right? Does that make sense?

Using Omnimic/REW/similar can one easily pinpoint when a speaker starts to compress even if THD may be at a tolerable level?

THD starts to rise rapidly at the onset of compression, correct?
post #417 of 820
Quote:
Originally Posted by JerryLove View Post

Sorry if I missed that. I could not comment on a statistical claim as I lack data that isn't anecdotal.

That's funny. I guess you would also find the following statistical claim dubious.
Quote:
In Northern Hemisphere, on average, there are more warm days during summer than during winter
post #418 of 820
Quote:
Originally Posted by zheka View Post

That's funny. I guess you would also find the following statistical claim dubious.

Your guess would be wrong. Is there a point?
post #419 of 820
Quote:
Originally Posted by popalock View Post


Once a speaker initally starts to compress it does so at a specific fq, not necessairly across the spectrum, correct? So, though your db meter reads 105db, if you were to read your omnimic measurements, as you increase the volume other parts of the spectrum can still increase as well right? Does that make sense?
Perfect sense. Once the coil hits xmax additional power applied in the low end of the passband does not result in additional output, additional power applied at higher frequencies does result in additional output, and the signal is therefore clipped, with very high THD the result, the same as what happens when an amp tries to produce output that the rail voltages won't support.
post #420 of 820
Quote:
Originally Posted by popalock 
Once a speaker initally starts to compress it does so at a specific fq, not necessairly across the spectrum, correct? So, though your db meter reads 105db, if you were to read your omnimic measurements, as you increase the volume other parts of the spectrum can still increase as well right? Does that make sense?
Quote:
Originally Posted by Bill Fitzmaurice View Post

Perfect sense. Once the coil hits xmax additional power applied in the low end of the passband does not result in additional output, additional power applied at higher frequencies does result in additional output, and the signal is therefore clipped, with very high THD the result, the same as what happens when an amp tries to produce output that the rail voltages won't support.
Perhaps I've failed to understand your post.

Are you stating that *the* cause of compression is a speaker hitting X-max?

Cause I'm pretty sure compression is caused by the temperature of the voice-coil changing its DC resistance.

Not that XMax isn't a concern and doesn't affect distortion. Of course it does. But I have not heard it associated with the term "compression" before.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Speakers
AVS › AVS Forum › Audio › Speakers › List of Reference Level, High Sensitivity & SPL Speakers