Jump to content
The Klipsch Audio Community
Sign in to follow this  
Chris A

Toward a Specification for Home Hi-FI Loudspeakers

Recommended Posts

Great topic Chris and I look forward to following this one.

Share this post


Link to post
Share on other sites

Just as an aside and then I will stop going off-topic.

In many ways, this is a form of project planning.

 

The solutions are along a "cheap to expensive" dimension. Where the ingredients include "simple, reliable, easy, plug and play to difficult to use, expensive, high maintenance, size etc"

 

Performance gains vary along a somewhat independent dimension and also vary in degree (small benefit to dramatic benefit). Unfortunately not everyone agrees on what is important or less important. Additionally, these benefits may be very application dependent. 

 

So the Project Management perspective comes in (and reflects most of life's decisions). Things that are cheap and offer a big bang for the buck in terms of performance (low hanging fruit), by all means do those. While things that are expensive and offer little benefit, then shy away from those. The other two combinations (expensive but with high impact or cheap with small impact), are those that need to be calculated or debated. 

 

In my world, not unlike yours, folks have difficulty prioritizing cost in its various forms (cheap, easy, reliable, size, maintenance,  plug and play etc). They also have difficulty in judging performance impacts (big, huge, sufficient, small and trivial). So the solution can degenerate into throwing their hands up in the air and declaring "everything is important". Thus, not making a good decision (since that would require a more thorough understanding of the the above two dimensions). 

 

Good Luck,

-Tom

Share this post


Link to post
Share on other sites

In sort of a preemptive manner, I offer the following for those who would argue what it is that is being discussed by myself in the past few posts...so I can move on with the thread topic in a more focused manner:

 

Quote

Distinguishing between architecting and engineering...

 

Generally speaking, engineering deals almost entirely with measurables using analytic tools derived from mathematics and the hard sciences; that is, engineering is a deductive process.

 

[System] Architecting deals largely with unmeasurables using non-quantitative tools and guidelines based on practical lessons learned; that is, [system] architecting is an inductive process.

 

At a more detailed level, engineering is concerned with quantifiable costs, [system] architecting with qualitative worth. Engineering aims for technical optimization, architecting for client satisfaction. Engineering is more of a science...architecting more of an art.  In brief, the practical distinction between engineering and architecting is in the problems faced and the tools used to tackle them.

 

This same distinction appears to apply whether the branch involved is civil, mechanical,chemical, electrical, electronic, aerospace, software, or systems.  Both architecting and engineering can be found in every one. [System] Architecting and engineering are roles which are distinguished by their characteristics. They represent two edges of a continuum...Individual engineers often fill roles across the continuum at various points in their careers or on different systems...

 

Note that I added the word "system" in front of the word architecting so as to distinguish the role from the commonly held notion of "building architect"--the profession that is associated with architecture schools for physical buildings and dwellings.  But it doesn't mean that system architects only design systems.

 

Note that definition of originating requirements sets and their relative precedence is usually a system architecting role.

 

Chris

Share this post


Link to post
Share on other sites

I find a great parallel with the Watch market... Modern Digital watches are imminently accurate and customizable and yet hand made timepieces are extremely popular (and accurate)... and the genre is more than alive and well despite the great quartz fiasco of the 1970's.

 

No one is arguing that a <$100 Casio watch is not MORE ACCURATE than a >$10,000 Patek Philippe, and yet there are no shortages of love for Manual movement timepieces... mainly it revolves around craftsmanship and HERITAGE.

 

The craftsmanship and handmade qualities of some of these 'distasteful' boutique pieces are non the less appreciated, function at a VERY high level and are very desirable.

Share this post


Link to post
Share on other sites
20 hours ago, Chris A said:

I find that most people haven't been exposed to effective design processes (even engineers)

 

I didn't know you were familiar with where I work.  I swear to God, if they came up with a good idea they'd implement it to conform with what they see when holding that idea facing a mirror.

 

Great thread.  Thanks.

Share this post


Link to post
Share on other sites

"If it measures good, and sounds good, it is good.  If it measures good, but sounds bad, you've measured the wrong thing." H.H. Scott.

Share this post


Link to post
Share on other sites
1 hour ago, DirtyErnie said:

"If it measures good, and sounds good, it is good.  If it measures good, but sounds bad, you've measured the wrong thing." H.H. Scott.

 

I don't think Chris is advocating otherwise. BTW, if you have the time, you might want to read the relevant chapters in Floyd Toole's book. He has made a good start on some of the more important physical measures that correspond to the psychological response of "hey, that speaker sounds good". His work, along with others, has made a big impact on me. 

 

  • Thanks 1

Share this post


Link to post
Share on other sites

This is going to be an evolutionary effort--like most specifications are.  So you're going to see an instance of how a specification is assembled from various sources of information, requirements, and rationale.  Readers following along with the process will increase their understanding of the material and its relevance in the end product: the draft specification.  The intent of this effort is to get people thinking about requirements and subsequent design/manufacturing. 

 

We'll start with performance requirements (the ones dealing with the major functions or capabilities of the product or system), then move on to the non-behavioral requirements listed above (e.g., reliability, maintainability, safety, etc.).  We'll first focus on the "whats"--definition of the capabilities of the system, and then move later to start defining the thresholds of performance for each.  That way we can focus on the important stuff now (the "whats") and leave the smaller stuff for later. 

 

If there are issues that arise that are outside of the loudspeaker design itself but are related to other parts of the home Hi-Fi Audio Reproduction System (the acronym chosen above was "HFARS" to identify the total system, including the electronics and room acoustics), those requirements, capabilities and rationale can be set aside for that higher level specification--to be assembled after or in parallel to the present loudspeaker specification. 

 

After looking at the Sean Olive's patent "Method for Predicting Loudspeaker Preference" (US 8,311,232 B2), I extracted the performance capabilities and their relative weightings, then placed them into the developing loudspeaker capabilities hierarchy that I've assembled thus far.  The added green-background items are from Mr. Olive (these items are not patented, but identified in the patent as results of the patented process and are provided in the patent with relative weights):

 

image.png.f6ace7078963c4e9c65f713d4e763bcf.png

 

The items in green are from Olive's patent, while the two light blue background items (also from Olive) are correctable using electrical filtering and/or a DSP crossover, so therefore is much less important.  Notice how few items from the Harman patent have affected the hierarchy--items that were advertised by Toole and Olive as being both necessary and sufficient to completely design superior sounding hi-fi loudspeakers.

 

Perhaps you are now beginning to see my problems with Harman's assertions about design requirements for loudspeakers--based on other information from PWK, Wolfgang Klippel, etc. and personal experience.  Toole, in his book chapter 20.1 (The Klippel Experiments), makes a comment extracted from Klippel's original 1990 article "Multidimensional Relationship between Subjective Listening Impression and Objective Loudspeaker Parameters" where Toole apparently takes out of context Klippel's comment and quotes the following:

 

Quote

“All dimensions perceived in the performed listening tests correspond with features extracted from the sound pressure response of the diffuse and direct field at the listener’s position. There was no hint of a relation to phase response or to nonlinear distortion.”

This is actually not what Klippel implies (as evidenced by his later papers).  Klippel makes a living, among other pursuits, of selling NFS (near-field system) measurement setups for $10K--$100K USD.  These specifically measure nonlinearities of loudspeakers and drivers, etc. in addition to the factors that Olive picked up, above.  Klippel says that listeners can hear nonlinear distortion (i.e., harmonic, modulation, compression, phase, etc.) clearly, and the threshold of detection for these types of distortion are typically at very low threshold levels.

 

Chris

  • Like 1

Share this post


Link to post
Share on other sites

Here are Olive's top four capabilities with their relative weights (and the main objective of the patented process):

 

image.png.f62635d2659a8ad14ab4df0a90a31eae.png

 

You can see what's missing from this, and I think it was deliberately designed like this to benefit Harman/JBL's products--I'm sorry to say.  Both Toole and Olive have pretty much permanently traded in their integrity (in my book) and like the discussion on the audibility of loudspeaker phase distortion that Toole also obfuscated in his book, I can't really trust their conclusions. 

 

Others may forgive this type of "brand support" marketing that appears often in this domain, but for me it's a lost cause for me to trust these guys again.  I will always look at the sources that are quoted by either of these guys before forming an opinion on this subject domain.

 

Chris

Share this post


Link to post
Share on other sites

By the way, the first anniversary of the date where I finally heard the difference between flattened loudspeaker phase response vs. higher-order crossover filter response passed last Friday.  That is another contributing reason for which I believe this type thread is timely.

 

Chris

Share this post


Link to post
Share on other sites

Chris, an interesting topic.

Regarding the detectability of one form of a phase "anomaly", group delay, the work from Moore's lab is probably considered the best (see,  Flanagan, Sheila; Moore, Brian C. J.; Stone, Michael A. (2005) and published in JAES).  The detectability could be as low about 2 ms, however, when presented in a reverberant space the thresholds were higher (more difficult to  detect). For various reasons, I am unable to use the word "distortion" when referring to a process that changes phase. That is why I stated it as an "anomaly". 

 

Good luck and I meant this comment as an aside rather than a topic in itself,

-Tom

Share this post


Link to post
Share on other sites

Thanks Tom.  I found the following figure based on the search for that article:

 

320008693_GroupDelayDetectionAudibilityLimitsvs_freq.thumb.JPG.92cda250332f7ce94529b5bcb80b90bb.JPG

 

Generally speaking, the detection thresholds shown correspond to my personal experience, and to some extent to the article that I linked to in the thread that I referenced, above (Greenfield and Hawksford, 1990) that Toole quoted in his book.  I've found that most of the group delay issues that I've heard and dealt with are above 200 Hz and below 2 kHz, with increasing thresholds below 200 Hz, especially deep bass response, i.e., the effects of bass reflex ports on group delay.  I actually detected the difference in group delay and phase shift at 5-6 kHz using a BMS 4592ND dual diaphragm driver (albeit in a sighted experiment), once I figured out how to flatten the phase response of the crossover filters between the two diaphragms.  This was apparently audible, but it was in the realm of subtle--removing a definite edge to the sound quality.  The group delay removal at that frequency was at the sub-millisecond level, but represented at 90 degrees of phase growth through the crossover region.

 

But in general, less than 1ms group delay is inaudible above 200-300 Hz, and up to somewhere around 2 ms or 1 cycle, and it becomes audible using Klipsch Jubilees with flattened phase in a treated room for early reflections--the listening room that I'm currently in.  Directivity of the loudspeakers and the effects of nearfield reflections cannot be separated from these hearing thresholds.  The problem with headphones is the eardrum bounce issues.

 

One of the things that I've found with all these phase and group delay audibility threshold experiments is that the researchers are using loudspeakers that do not control their polars below about 1 kHz, i.e., they are mostly using "little" two-way direct radiating studio monitors, so unless the subjects in the experiments are in an anechoic chamber, they are also dealing with nearfield reflections at the same time as they are listening for the direct arrival group delays.  This seems to be systemic.  Using the K-402 and Jubilee bass bin seems to improve GD threshold discrimination capability...rather significantly (anecdotally).

 

But I digress.

 

Chris

Share this post


Link to post
Share on other sites

A few issues are getting convolved here.

 

In terms of working toward a specification, it is probably best to ignore those "data points" to the left of 300Hz in the graphic above.  I might be wrong, but I am not convinced those are actually detection thresholds. (or ones that I would give much credence to) 

 

In terms of incorporating audibility into an eventual specification for an "acceptable amount of group delay", then 2 ms is a good approximate limit. This course can be refined when you add the assumption that the speaker is in a somewhat reverberant field (like a real room). In that case, detection of group delay would be poorer (i.e., larger than 2 ms), 

 

With other sorts of phase anomalies ( reversed polarity, or a variety of phase shifts), then a simple summary statement would be difficult and necessarily have a host of qualifications. 

 

Good Luck,

-Tom

Share this post


Link to post
Share on other sites

I'll take your comments as suggestions, but note that this (group delay) is in an area where experimental results of threshold of audibility definitely lags personal experience for many people.  Bass reflex audibility is apparently a group delay phenomenon rather than a phase delay issue, so taking a 1 cycle or 1.5 cycle (wave periods) threshold for this type of informal requirements list is definitely in-bounds--as far as I'm concerned.  Given more data on sub-500 Hz group delay audibility, I would be willing to consider those sources, if and when they appear.

 

Remember, the objective here is usability of the results for a wide range of "informed consumers", DIYers (of which this group is becoming quite sophisticated technically with the democratization of the measurement tools and engineering knowledge) and perhaps others, including perhaps practicing engineers working for very small or single proprietor enterprises without the means of establishing independent design thresholds for themselves. 

 

I'm sure that those wanting greater technical or experimental rigor (such as yourself) can apply their own filters to these lists when using the results for themselves.

 

Chris

Share this post


Link to post
Share on other sites

Sometimes formulating a plan can take longer than whipping a few spit-wads at the wall, so to speak.  Both methods have their place, I think.

Share this post


Link to post
Share on other sites
On 5/1/2020 at 11:46 AM, DirtyErnie said:

What I would like to see, more than anything, is the whole recording and reproduction industry getting together and agreeing that all recordings will be full-range and not manipulated, and playback devices engineered to protect themselves if need be.  

If it could be agreed on for overall compression, loudness, dynamic range, and frequency response, and then the devices can add whatever emphasis they want, great.  Home A/V receivers with room compensation routines & microphones can get most people close enough.  And then there's "us"...

Agreed

And, quite honestly, in this day and age, I don't see why "they" (the 'radio' stations, iPod/ear bud types, etc) can't have it their way, and we (quality, true to the original audio want the best reproduction types) have it our way too. Today it should be easy and cheap enough to release this music in an unadulterated form along with the "loudness wars" crap they send out for airplay.

Share this post


Link to post
Share on other sites

Continuing on with the core purpose of the thread, the following posts expand on some of the biggest loudspeaker requirements and design factors for the three major branches of the originating loudspeaker requirements/capabilities hierarchy.  First, the SPL and phase response hierarchy:

 

image.png.f482993c88a6a12b5ff80ac8d9aabd2d.png

 

Note again the light blue and green background items that were identified by Olive.  The light blue background items indicate those items that are correctable using DSP filtering.  A brief description of each area follows:

 

1.1.1 Crossover filter order: the higher the order of the analog or digital filters, the greater the phase shift that is experienced from the higher frequency to lower frequency drivers.  Generally the rule of thumb is 90 degrees of phase shift across both the high pass and low pass filters for each "order" of the electrical filters.  The higher the filter orders, the greater the phase shifts.  Some correction of the phase/time alignment of the loudspeaker is possible by positioning the higher frequency drivers further away from the listeners, but note that group delay effects (the plot of the instantaneous slopes of the phase curve) will remain even if HF driver setbacks are used.  This is the argument for using first order or "zeroth order" crossover filters.

 

1.1.2 Spacing of crossover filters: crossover filters in general are spaced apart in their breakpoint frequencies so that the summed SPL from the crossover drivers in the crossover frequency band sum to unity SPL relative to the rest of the SPL response.  This rule of thumb isn't necessarily followed in the actual loudspeaker design.  There are instances where greater or lesser frequency overlaps of the high pass and low pass crossover filters are used.  This will affect the phase and group delay curves in the crossover frequency bands, as well as the summed SPL response of both "ways" of drivers being crossed.

 

1.1.3 Driver Impedance: individual drivers in a loudspeaker do not have flat impedance vs. frequency, especially in the crossover frequency bands.  Frequently, the phase and/or SPL response of the driver is changing rapidly with frequency at crossover frequencies, and these electrical impedance phase and amplitude shifts show up as SPL and phase shifts at the output of the loudspeaker. 

 

1.2.1 Driver centerline-to-centerline distances (laterally):  Under the "radial spacing of acoustics drivers, if the drivers/horns are not coaxial, the spacing of them laterally across the front baffle of the loudspeaker or free space containing the driver diaphragms (direct radiating) or horn mouths (horn loaded) will create polar lobing issues with horizontal listening angle laterally in the listening room.  This is like having one really good loudspeaker pointed directly at the listener's ears, plus two poorer performing loudspeakers on either side of the good loudspeaker.  Severe frequency response issues appear based on the off-angles that the loudspeaker--listener angles.  Horizontal coverage issues are more pronounced that vertical coverage issues, because the range of listening angles and in-room reflections are greater in horizontal direction than in the vertical direction.  Additionally, the human hearing system is set up to hear lateral differences in SPL and phase response more easily than in the vertical direction (i.e., ears are on the sides of the head). 

 

1.2.2 Driver centerline-to-centerline distances (vertically): Like the preceding discussion, under the "radial spacing of acoustics drivers, vertical offsets of the driver centerlines create polar lobing in the vertical direction in-room.  While these effects are not as sensitively perceived as horizontally, the effects of vertical offsets in driver/horn centerlines have similar effects on the sound quality of the loudspeaker.

 

1.3.1 Box resonances: Under the "on-axis narrow band deviation in SPL response" (Olive), the smoothness of response vs. frequency in narrow frequency bands, especially of the bass driver (woofer), but also of the midrange driver in direct radiating midrange loudspeakers, is strongly affected by acoustic resonances inside the loudspeaker enclosure(s), including the structural resonances of the enclosure walls themselves.  The preferred approach is to have box materials that are so stiff that their panel resonance contributions to interior of the box are minimized.  Another approach is to move the resonances of the the enclosure to either a higher (usually the preferred direction) or lower (usually less preferred) frequency band in order to move the peak resonances out of the frequency band of the driver, usually the woofer(s). Additionally, placement of the driver within the front baffle relative to the walls of the box will strongly affect the amplitude of the enclosure-induced resonances, so the usual trial-and-error placement of the woofers within the loudspeaker front baffle or the closed back chamber volume of a horn-loaded bass bin are required. (Note that I'm not referring to the placement of accelerometers on the walls of the enclosure, something that is a trademark of a noted stereo magazine editor, and is of limited or dubious value.  The preferred method to measure box resonances is to place a small microphone inside the enclosure and then seal the enclosure again.)

 

1.3.2 Driver resonances: it is a poorly kept secret that drivers themselves experience self-resonances from internal reflections (i.e., cone to spider reflections, cone to basket reflections, diaphragm-phase plug resonances, back chamber resonances of compression drivers, etc.).  The best approach is to avoid drivers whose internal resonances were not fully controlled by the designers of the driver. 

 

1.3.3 Horn resonances: In similar fashion to the driver self-resonances, resonances within horns, either axial resonance modes or off-axis (higher order modes, or HOMs) are controlled by the design of the horns themselves.  In folded horns, there are many opportunities for resonances at the various horn folds and at the cross-channels (higher order modes), excited by the ends of the individual horn folds.  Very careful design of the fold geometries and distances is required for successful folded horn designs.

 

1.3.4 Front baffle step:  This is a concern for direct radiating loudspeakers where the width and height of the front baffle on which the direct radiating drivers are mounted becomes a 180 degrees "horn" that experiences a drop in horn loading below the 1/2 wavelength frequency associated with the dimensions of the front baffle,  There is usually a 6 to 9 dB loss in SPL output below the 1/2 wavelength frequency coinciding with the dimensions of the front baffle.  The same effect occurs in regular horns when the frequencies of the loudspeaker fall below the half-wavelength dimension of the horn mouth, either vertically or horizontally.

 

1.4.1 Woofer low frequency cutoff:  From Olive's patent, the bass extension of the loudspeaker determines about 30% of the subjective sound quality of the loudspeaker.  Low frequency cutoff of the woofer/enclosure is a function of the woofer itself (mostly the diameter of the woofer, but also the natural frequency cutoff). 

 

1.4.2 Bass horn axial length: In the case of a horn-loaded bass, the dramatically increased acoustic loading on woofer drivers enables frequency response much lower than the free air resonant frequency of the woofers themselves. 

 

1.4.3 Room boundary placement: The dominant effect in horn-loaded woofers is the mouth size of the horn and the horn's placement near room boundaries in order for the bass horn to pick up boundary gain (i.e., corner or floor-wall boundary gain), which can extend the effective length of the bass horn by more than 100% and increase the output of the horn below its axial length by as much as 18 dB.  In the case of direct-radiating woofers, the same effect occurs as in horn--loaded woofers is the woofers are placed well within 1/4 wavelength of the room boundaries at the higher operating frequency of the bass bin.  While the resulting bass bin response must be equalized to flatten any peaking response, the effects of boundary gain are real and are free to all loudspeaker types--except dipole radiators.

 

...To follow: phase and directivity control decompositions of requirements/capabilities.

 

Chris

Share this post


Link to post
Share on other sites

image.png.4dd763e3004a7088d4a5bfebb614f845.png- 

 

2.1.1 Acoustic loading of drivers--efficiencies:  As the acoustic loading of the driver diaphragms increase, the motion required for them to produce a certain SPL on-axis decreases to about 1/5th that of direct radiator operation.  The reason why this is important is because the modulation distortion spikes on the upper frequencies are controlled by the distance that the diaphragm moves.  If the diaphragm moves less, the modulation distortion sideband peaks are attenuated (see the following plot of harmonic distortion and modulation distortion products of a dual-tone measurement of distortion products from a single driver):

 

585407185_KlippelHDandmodulationdistortionspectrumplot.thumb.JPG.64ab71fe3d6c33b61563954c5835ac0f.JPG

 

In a horn-loaded driver those spikes labeled "IMD2" and "IMD3" (intermodulation distortion sidebands around the upper frequency) above are dramatically reduced in amplitude relative to the driver being used as a direct radiator.  Note that horn loading of the drivers reduces the magnitude of these those sideband spikes by 20-25 dB (90-95% of their peak values) relative to the same drivers being used without horn loading.  This is the difference between horn loaded loudspeakers and direct radiating ones in terms of how they sound.  Horn loaded drivers typically sound crystal clear and extreme transparent (due to the absence of the modulation sidebands), while the direct radiating drivers sound opaque---like mud has been introduced. 

 

So the higher the efficiency of the drivers, the lower the modulation distortion (the type of distortion that is most audible and objectionable to listeners because the sideband spikes are not at harmonic multiples of either input frequency.

 

In the same way, because the driver diaphragms don't have to move as far, the heating of their voice coils is similarly reduced--dramatically--so that even under very high output SPL conditions, the horn-loaded drivers generate much lower heat.  This leads to much lower levels of compression distortion--sort of like fade in automobile brakes when they get hot.  Additionally, the higher efficiency of the driver diaphragm-air coupling produces a much more positive air coupling, and therefore reduces the compression distortion of woofers, etc. when being driven to high output levels.

 

2.1.2 Width of driver frequency bands: Modulation distortion sidebands come from two sources--amplitude modulation distortion (AMD) due to nonlinearities within the driver itself and the amplitude of the diaphragm movement, and frequency modulation distortion (FMD) due to the difference in frequencies handled by the individual drivers ("ways") of the loudspeaker.  This is also known as Doppler distortion. 

 

2.1.3 Linearity of driver with displacement: This performance capability affects not only modulation distortion quantities of AMD and FMD, but also harmonic distortion, shown in the table below (Klippel) and includes the sources of where the nonlinearities occur in drivers: diaphragm suspension stiffness nonlinearity, length of voice coil in the magnetic gap (i.e., zero-lap, underhung, or overhung voice coil--Bl nonlinearity), "flux modulation" nonlinearity, cone geometry and suspension nonlinearities, diaphragm and suspension bending modes, Doppler effect (discussed above), and wave steepening (horns):

 

Sources of driver nonlinearities (Klippel).JPG

 

2.2.1.1 Driver self-distortion: the harmonic distortion portion of the driver nonlinearities (discussed above)

 

2.2.1.2 Port Flow (ported only enclosures): This is also shown in the table above.  This nonlinearity contributes to harmonic distortion only of the woofer.

 

2.2.1.3 Horn-related (wave steepening): Also shown in the table above.  This phenomenon is associated with very high SPL levels within horns, notably non-straight-sided horns.  This nonlinearity describes the effect of wave fronts to become steeper as they progress down the horn path length, and results in both harmonic distortion and modulation distortion.

 

2.2.1.1 Driver-to-driver time misalignments (correctable using DSP crossovers): This distortion results in phase misalignments, resulting in degraded impulse response and perceived timbre shifts.

 

2.2.1.2 Crossover network phase shifts (correctable using DSP crossovers): This distortion adds to the time misalignments of drivers within a loudspeaker, and results in the same degraded impulse response and timbre shifts.

 

2.2.1.3  Driver diaphragm flexibility: This is the apparent movement of the acoustic center of individual drivers within their covered frequency bands due to diaphragm breakup and nonlinearities of air (thermodynamics of moisture in the air partially condensing due to extreme SPL). 

 

2.2.2.1  Driver voice coil heating: this is the nonlinearity of driver SPL output vs. power input over time to the voice coils of the driver.  This is a significant factor in direct radiating drivers, and results in compression distortion of the loudspeaker, as well as temporary shifting of the crossover network frequencies (passive crossover networks only).

 

2.2.2.2 Undersize driver diaphragm/horn mouth: this source of distortion results in compression distortion of bass SPL, notably for direct radiating bass bins.

 

Chris

Share this post


Link to post
Share on other sites

image.png.2ae5f216b05082137d203af1e82a12c5.png

 

  

Share this post


Link to post
Share on other sites

So much for "free market" 😘

 

Try sending that stuff to the marketing department

🤑

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...