Why not just use the terms directivity, and directivity index? These terms are already defined.All sound energy that is a direct path from the source.
Direct sound- Sound waves arriving at the listening location directly from the source. Differing from reflected sound, which arrives at the listening location after bouncing off the surrounding surfaces.
Direct energy = Direct Sound. Sound Energy
"Sound energy is the result when a force, either sound or pressure, makes an object or substance vibrate"
I must not be saying it in a way that is commonly said in the field.
Does increasing Radiation area, increase Direct Sound, to the listener??? For some reason I am under the impression the answer is yes.
A larger driver does have greater directivity. Horns control directivity.
Last edited:
Hi again, my reply last night was a quick attempt to respond before bedtime, without fully reading all the replies exchanges. Reading them all this morning I see you and and I agree on all the technical points/considerations.
I still stand by my recommendation to avoid EQ boost on sealed, and recommend getting more displacement to reach desired SPL and low corner....if attaining the highest SQ possible is the goal.
Like you said, using EQ will limit the maximum SPL and use more power and excursion to get there.
That fact requires we derate the maximum linear SPL we think we can reach by the amount of boost used....if in fact we are concerned with maintaining linearity all the way down.
Typically, when a person exercises their system to the upper end of their SPL listening pleasure, my bet is one of two things happen when using a Linkwitz transform or low shelving (both without a hpf):
The sub has excessive excursion, and either distorts (and perhaps eventually fails mechanically).
It's often hard to hear the distortion though, simply due to high volume masking throughout the spectrum, unless we listen to the sub alone (which is always a good subwoofer reality check/test imo)
Or, and probably more likely, folks get saved from over-excursion by amps that simply can't produce the very low freq power required.
Many folks are aware that it takes 4x the pistonic displacement to maintain equal SPL for each octave decrease. I think it's seldom recognized that means 4x the power is also needed for each octave decrease.
Anyway, I'm not talking about PA..... (I haven't been involved with PA for over 6 years, during which time all my DIY builds have been entirely about maximum attainable SQ....often forgoing max SPL )
I'm talking about having a speaker that stays unstressed sounding at whatever maximum average SPL is desired....be it high or low too.
I think maintaining linear SPL across the spectrum at a person's volume preference is super important to SQ....
And I think in reality, volume preference is often heavily influenced by what our system can actually put out while staying linear.......
I mean who doesn't turn down the volume when it starts sounding stressed? Would you necessarily turn it down if it stayed clear and clean, louder?
When a speaker maintains linearity across the spectrum, including headroom for dynamic range and transients...it greatly adds to clarity and a sense of realism (to my ears).
When i run the numbers on what it takes to do that, I find that even at moderate volume, it takes more than is often acknowledged, ....possibly because folks seem to be OK with the idea "it will work for home audio levels". I think that idea often costs SQ.
Anyway again, hopefully back to more directly on current topic...EQ of sealed....
I use the same thing for sealed that you and b-force described....hope to be testing my 18"s sealed again, soon.
Thanks for those papers.
This first concludes similar to comments about low frequency phase I that I've seen from others making comparative listening tests.
5.1. Results The result has been evaluated in an informal listening test, where the correction was clearly audible for all participants. The lowfrequency reproduction gets tighter and more defined. Rhythmic instruments like bassdrums have a better coherence of bass and subbass frequencies and thus are fusing more into one sound. Due to the change introduced by the equalization, the resulting sound is also a little unusual since the listener is in most cases used to listening to uncorrected speakers for a long time. Another observation is, that the crestfactor of the output signal of the correction filter can change due to the phase shifts in the lowfrequency range. To avoid clipping, the level of the output signal may have to be reduced or limited according to the capabilities of the signal processing system.
Second paper shares some of my doubts about accepted science...I really hope more folks continue on with truly good science regarding phase......i'd really like to know what's generally audible and what's not. (with loudspeakers and extended low freq)
In today's age of cheap watts, does efficiency still matter? I have a PE Executive box with drivers in the low 80dB range. It plays more than loud enough for long enough on a small battery charge. I guess it depends on your thresholds and intended use cases.
When working on a enclosure design for my 15" driver the differences found between equalized sealed and SBB4 vented (same FR) basically yielded a 6db differential in overall max SPL taking excursion into account. The group delay ranged from a differential of 18ms (with a high pass applied to the vented alignment) to 8ms with the high pass defeated. Peak GD occurring in the 25hz range of about 28ms which still seems below audibility from what I understand. So, it seems to me what the choice comes down to is how much SPL do you need in room.
From a driver perspective, it always will be a compromise between a lower Fs, sensitivity and BL -> size of the magnet.
(well that is an extremely simple explanation).
SPL vs power goes with with 10log, so it will require 4 times more power.
Meaning the speaker will also have to dissipate 4 times more power.
Sometimes (often?) this also means more cone excursion as well.
Unfortunately, manufactures make such wildly different driver designs, that good direct objective comparisons are sometimes almost impossible, but wasting more power in general is never a good thing, also not in a sense of distortion.
It maybe starts to sound boring, but in the end it all depends on context.
For subwoofers for example, this is not really a big deal and it's mostly all about just xmax and Sd
That being said, from a driver development point of view;
Because power is not an issue these days, it's a compromise that's going out of the window extremely quickly.
50W was unheard of 50-60 years ago, and very "special" 30-40 years ago.
Nowadays this is not even a little special anymore at all.
The result is that speakers can be made a lot more compact, since this power can now be provided by the amplifier, instead of the "acoustic power" by the cabinet.
Greetings all who responded to the sealed box equalization issue.
I am still not sure, what the problem with equalization is and why a high pass filter is required.
Consider the below attached image:
For the pupose of discussion, the modeled response has been made parallel to he excursion limit by choice of the enclosure parameters and a (small) series resistor between the amplifier and the speaker. Thus, as long as the required equalization is less or equal the difference between the modeled repose and the excursion limit, the excursion is withing the rated - not maximum - limits.
If such an amount of equalization does not satisfy the required SPL over the frequency of interest, of course, different displacement either by choosing different driver or additional driver(s) may be needed. By the way, this has never been disputed, cf. e.g., excerpt form fluid's post 10,973:
I also disagree with the notion that equalization is a small signal analysis. The attached image shows allowance of 9 dB of equalization.
Kindest regards,
M
I am still not sure, what the problem with equalization is and why a high pass filter is required.
Consider the below attached image:
For the pupose of discussion, the modeled response has been made parallel to he excursion limit by choice of the enclosure parameters and a (small) series resistor between the amplifier and the speaker. Thus, as long as the required equalization is less or equal the difference between the modeled repose and the excursion limit, the excursion is withing the rated - not maximum - limits.
If such an amount of equalization does not satisfy the required SPL over the frequency of interest, of course, different displacement either by choosing different driver or additional driver(s) may be needed. By the way, this has never been disputed, cf. e.g., excerpt form fluid's post 10,973:
I also disagree with the notion that equalization is a small signal analysis. The attached image shows allowance of 9 dB of equalization.
Kindest regards,
M
Last edited:
@b_force psychoacoustics def play a part too from what I've seen as do typical room acoustics. Not sure if equal loudness contours are controversial here but the variation in SPL perception across the spectrum probably factors into speaker design too. And of course source material, intended use yadda yadda
For me though outside of the thrill of achieving super high sensitivity the cheapness of watts kind of puts sensitivity on the back burner.
For me though outside of the thrill of achieving super high sensitivity the cheapness of watts kind of puts sensitivity on the back burner.
To me, the thrill of higher sensitivity is lower compression. It's not the only parameter, but more heat = more compression.
Hi M, would you post the drivers T/S parameters and the box size you're using? Along with the particular EQ (s)?
I'd like to take a look at excursion in Hornresp. thx
Hi mark100,
Of course, I will be happy if more people looked at the simulation, and point out any possible mistake(s) I might have made. Please check your p.m.
However, please note that this simulation was done as an exercise to support the points I made and no equalization had been applied. I actually built a different enclosure for a different driver, and I intend to measure it once the weather permits.
Kindest regards,
M
Of course, I will be happy if more people looked at the simulation, and point out any possible mistake(s) I might have made. Please check your p.m.
However, please note that this simulation was done as an exercise to support the points I made and no equalization had been applied. I actually built a different enclosure for a different driver, and I intend to measure it once the weather permits.
Kindest regards,
M
Last edited:
With "small signal analysis" we mean behaving according to T/S parameters (which are measured with a small signal) as well on a linear fashion.
In reality speakers don't work that way, depending on the system and solutions that were chosen.
How good or bad this behaves, completely depends on how a certain speaker with a certain Fs, Qt , cone excursion etc is being "EQ'ed".
An average 8 inch subwoofer in something like 10-20 liter with a LT at 35Hz will most definitely reach its maximum excursion pretty soon without any HP filter.
Choosing a different speaker is one of the solutions, but definitely not the only solution.
In the end it's all about compromises, having a teeny tiny bit more group delay is pretty low on my priority list to be perfectly honest. There are so many factors far more noticeable that could be otherwise be compromises. Often extremely practical factors.
Based on my experience with quite a lot of different companies and individuals.
I can't say anything here in general.
There are speakers with (very!) high sensitivity but a teeny weeny voicecoil and no cone excursion.
Which will (audible) go much quicker into compression, in fact they are even developed that way for that reason.
Hi b_force,
Kindest regards,
M
Thank you for the clarification. So, from a practical perspective, how can one determine the extent to which they can be relied upon? For example, if the driver is used within some of the limiting parameters, e.g., below power rating, within Xmax, etc.
Kindest regards,
M
I can see why you are confused. I didn't ask the question correctly, the 2nd time after clarifying my use of "direct energy" that is, direct sound...
The part that mattered was that I am pointing to the omnidirectional part of the spectrum for each source
Source A; 4" 1m 50hz 86db
Source B; 18" 1m 50hz 86db
Source C; 24" 1m 50hz 86db
All are omnidirectional at this frequency, I believe. Will the larger radiation sources deliver higher Direct Sound levels, to the 1m listening point, in a room, is the question?
@gedlee
That's a good question, LOL!!!
If Klippel information is available, that really really helps.
Otherwise my only proper answer is, by experience and trying.
Still, I have seen drivers (PA in this case, but doesn't matter in general sense) were the cone literally kinked under heavy loads. This was even from a pretty known brand actually.
In my opinion, a better approach is the other way around.
So build in some safety margin, or use something like a simple 1st order HP filter (or a compressor/limiter in some cases, although doesn't really apply to home hifi).
Not a crazy amount, but just enough to prevent the speaker going into kamikaze mode.
Hi b_force,
thank you for the reply.
As I understand, it implies proper selection of enclosure/driver parameters, so the the required max SPL is reached only occasionally, as noted several times throughout the discussion.
Kindest regards,
M
thank you for the reply.
As I understand, it implies proper selection of enclosure/driver parameters, so the the required max SPL is reached only occasionally, as noted several times throughout the discussion.
Kindest regards,
M
yes, in PA / sound-reinforcement land, a good engineer takes that as like worst case mark.
And often puts an hard limiter on it (or 3dB lower), as well as a compressor that's a little more lose.
In home audio, this is rarely the case, unless you're trying to squeeze out to much from a couple of poor little 4 inch woofers or so. That being said, I have seen (and developed) active speakers for home audio that like some kind of protection against overly enthusiastic (and often drunk) consumers, lol 😀
(actually that's not even a joke, but a serious concern for companies)
It depends on the system, but reaching the limits for quality woofers >= 6 inch above 100Hz, is quite the challenge inside a living room. I think most neighbors or other family members would have reached their limits already before that 😀
In general it implies proper selection of all requirements as well.
To reach a certain maxSPL, sometimes one just has to move up the lower -3dB point a little.
Or spend more money on a driver, or has to go for a BR instead, or has to go for a bigger woofer in a bigger cabinet etc etc
I think you get the point here 😉
Although there were 2 rooms with Klipsch speakers at the Dutch Audio Event, the Jubilee was unfortunately missing.
Fwiw, Steve Guttenberg's video about the Jubilee.
In my humble opinion - as recently reaffirmed - efficiency (lack of signal loss) is perhaps the single most determinant factor for the final sound quality.