Jump to content
The Klipsch Audio Community
Chris A

"Why Horn-Loaded Sounds Better Than Direct Radiating" FAQ

Recommended Posts

This Frequently-Asked Questions (FAQ) thread discusses some reasons why horn-loaded loudspeakers sound more realistic than direct-radiating loudspeakers, such as cone-type and planar type drivers.

Several manufacturers currently make or have made horn-loaded loudspeaker designs: Klipsch, ElectroVoice, JBL, Altec, and several smaller manufacturers.

 

"Why do horn-loaded loudspeakers sound better than direct radiating loudspeakers?"

 

Chiefly, the reason is due to low modulation distortion (i.e., not harmonic distortion). Horn-loaded electro-acoustic drivers typically have 25 dB lower frequency modulation (FM) distortion levels and 15 dB greater efficiency than when using those same drivers without horns to produce the same sound pressure level (SPL).

 

"What is Modulation Distortion, and Why is It Important?"

 

Frequency modulation (FM) distortion, sometimes called Doppler distortion since it is largely caused by the movement of the driver's cone/diaphragm at lower frequencies, is caused by simultaneous modulation of higher frequencies that are also being reproduced by the same driver at the same time.

 

Amplitude Modulation (AM) distortion is primarily due to driver nonlinear response when the cone/diaphragm is operating near its extent of maximum movement under high-load conditions. Figure 1 gives a visual representation of the two components of modulation distortion vs. time:

 

Amfm3-en-de.gif

Figure 1 AM and FM distortion visualization

 

Both of types of modulation distortion are very objectionable for listeners due to their non-harmonic frequencies that are produced.

 

Many people are familiar with harmonic distortion, which is integer multiples of the input frequencies greater than the input frequencies into the loudspeaker that are being reproduced on the output of the loudspeaker under higher load conditions. Figure 2 gives a view of frequency harmonic amount vs. relative input frequency (ignoring the effects of subharmonic distortion effects)

 

hardis.gif

Figure 2 Harmonic distortion visualization

 

Harmonic distortion is not as audible as modulation distortion due to the internal signal processing of the human hearing system, particularly the lower harmonics like second, third harmonics. Higher-order harmonic distortions (fourth, fifth, sixth order harmonics, etc.) are more easily detected by human hearing. Some sources call this human hearing effect "harmonic masking".

 

Contrasting the above harmonic distortion, modulation distortion (AM, FM, intermodulation, etc.) produces non-harmonic frequencies not found in the input signal driving the loudspeaker. Because these modulated frequencies are not related in integer multiples of either the lower or higher frequencies being reproduced, these distortion-produced frequencies are much, much more audible and objectionable than typical harmonic distortion. Figure 3 shows a visualization of both major types of distortion (harmonic and modulation distortion) versus frequency.

 

csm_distorted_two-tone_signal_24b8d2b0fc.jpg

Figure 3: Visualization of harmonic and modulation distortion

 

Note that modulation distortion shows up on the higher frequencies reproduced, which is typically more audible than lower frequencies due to the frequency response/acuity of the human hearing system.

 

Additionally, the modulation distortion frequencies shown in figure 3 are not integer multiples of either the lower fundamental frequency or the higher one. These non-harmonic frequencies are much more objectionable to listeners compared with harmonic distortion at the same relative amplitudes.

 

It can be seen that harmonic distortion will also modulate the upper frequencies making the effects of harmonic plus modulation distortion much more objectionable to listeners. The effect of these types of mixed distortions can be described as the speakers sounding "loud" and "opaque" while responding to high input signals.

 

"Why is Modulation Distortion So Much Lower in Horns?"

 

Modulation distortion is produced when the acoustic driver's cone or diaphragm moves - and the more it moves, the greater the modulation distortion. Horn-loaded drivers reduce the amplitude by a factor of ~5-10 (relative to using that driver as a direct radiator) that the driver has to move to produce a certain SPL output level. Less cone/diaphragm motion equals less modulation distortion.

 

Any acoustic driver that is horn loaded will experience a dramatic decrease in required motion in order to produce output SPLs.

 

  • Like 2
  • Thanks 1

Share this post


Link to post
Share on other sites

I'm always amazed how nice fully horn loaded speakers sound playing classical music. it's almost impossible for convectional speakers (price tags aside) to deliver life like presentation as horn speakers do. MY late LS had no that low frequency in below 50hz region but the dynamics and impact of the sound transition was just fantastic, never heard anything better. I'm looking forward to this subject

Share this post


Link to post
Share on other sites

Another important factor in favor of horns is their directivity. This "pushes" the direct sound field further into the room than cones can. Hearing less of the reverberant field improves clarity greatly

  • Like 1

Share this post


Link to post
Share on other sites

I've often wondered whether Frequency Modulation distortion and Intermodulation distortion were the same thing, but one used in regard to loudspeakers, and the other in regard to electronics. Are they the same? Are they related?

What ever happened to TIM? Was it ever measured in speakers?

Since you have a lot of information, here's another question. If someone ran frequency response curves on a speaker at relatively low SPL, then higher SPL, then very high SPL, but still within the speaker's capability, would the three curves look the same (smoothness, where peaks and valleys are, etc.)?

Share this post


Link to post
Share on other sites

This Frequently-Asked Questions (FAQ) thread discusses some reasons why horn-loaded loudspeakers sound more realistic than direct-radiating loudspeakers, such as cone-type and planar type drivers.

Several manufacturers currently make or have made horn-loaded loudspeaker designs: Klipsch, ElectroVoice, JBL, Altec, and several smaller manufacturers.

"Why do horn-loaded loudspeakers sound better than direct radiating loudspeakers?"

Chiefly, the reason is due to low modulation distortion (i.e., not harmonic distortion). Horn-loaded electro-acoustic drivers typically have 25 dB lower frequency modulation (FM) distortion levels and 15 dB greater efficiency than when using those same drivers without horns to produce the same sound pressure level (SPL).

"What is Modulation Distortion, and Why is It Important?"

Frequency modulation (FM) distortion, sometimes called Doppler distortion since it is largely caused by the movement of the driver's cone/diaphragm at lower frequencies, is caused simultaneous modulation of higher frequencies that are also being reproduced by the same driver at the same time.

Amplitude Modulation (AM) distortion is primarily due to driver nonlinear response when the cone/diaphragm is operating near its extent of maximum movement under high-load conditions. Figure 1 gives a visual representation of the two components of modulation distortion vs. time:

Amfm3-en-de.gif

Figure 1 AM and FM distortion visualization

Both of types of modulation distortion are very objectionable for listeners due to their non-harmonic frequencies that are produced.

Many people are familiar with harmonic distortion. This type of distortion is due to integer multiples of input frequencies greater than one than the input or recorded frequency(ies) being reproduced on the output of the loudspeaker under higher load conditions. Figure 2 gives a view of frequency harmonic amount vs. relative input frequency:

hardis.gif

Figure 2 Harmonic distortion visualization

Harmonic distortion is not as audible as modulation distortion due to the internal signal processing of the human hearing system, particularly the lower harmonics like second, third harmonics. Higher-order harmonic distortions (fourth, fifth, sixth order harmonics, etc.) are more easily detected by human hearing. Some sources call this human hearing effect "harmonic masking".

Contrasting the above harmonic distortion, modulation distortion (AM, FM, intermodulation, etc.) produces non-harmonic frequencies not found in the input signal driving the loudspeaker. Because these modulated frequencies are not related in integer multiples of either the lower or higher frequencies being reproduced, these distortion-produced frequencies are much, much more audible and objectionable than typical harmonic distortion. Figure 3 shows a visualization of both major types of distortion (harmonic and modulation distortion) versus frequency.

e06f652d9d.jpg

Figure 3: Visualization of harmonic and modulation distortion

Note that modulation distortion shows up on the higher frequencies reproduced, which is typically more audible than lower frequencies due to the frequency response/acuity of the human hearing system.

Additionally, the modulation distortion frequencies shown in figure 3 are not integer multiples of either the lower fundamental frequency or the higher one. These non-harmonic frequencies are much more objectionable to listeners compared with harmonic distortion at the same relative amplitudes.

It can be seen that harmonic distortion will also modulate the upper frequencies making the effects of harmonic plus modulation distortion much more objectionable to listeners. The effect of these types of mixed distortions can be described as the speakers sounding "loud" and "opaque" while responding to high input signals.

"Why is Modulation Distortion So Much Lower in Horns?"

Modulation distortion is produced when the acoustic driver's cone or diaphragm moves - and the more it moves, the greater the modulation distortion. Horn-loaded drivers reduce the amplitude by a factor of ~10-30 (relative to using that driver as a direct radiator) that the driver has to move to produce a certain SPL output level. Less cone/diaphragm motion equals less modulation distortion.

Any acoustic driver that is horn loaded will experience a dramatic decrease in required motion in order to produce output SPLs.

You forgot one word in your last statement.....properly.

Share this post


Link to post
Share on other sites

Another important factor in favor of horns is their directivity. This "pushes" the direct sound field further into the room than cones can. Hearing less of the reverberant field improves clarity greatly

Yes, and this is why cone speakers are typically away from corners so they don't spray their sound all over the side walls for better imaging, but increase the IM distortion. Big trade off.

Share this post


Link to post
Share on other sites

Modulation distortion is produced when the acoustic driver's cone or diaphragm moves - and the more it moves, the greater the modulation distortion.

Isn't that wonderful...

so what happens if I build a bass horn using a 15" cone driver and, in fact, realize a reduced cone excursion but the problem is that the reduction occurs over a range of frequencies that's narrower than the frequency range I'm trying to operate my bass horn over?

This is what happens, you end up with a horn that has a high sensitivity over that region where you've realized this reduced cone excursion. In other words you end up with a Klipschorn folded bass unit...

post-864-0-78820000-1404951031_thumb.jpg

Edited by John Warren

Share this post


Link to post
Share on other sites

Below the red plot is the Z-mag of the K33E in free-air, the green is the K33E mounted in the Klipschorn placed into a concrete corner.

The increase in Z-mag relative to the red is the so-called "horn loading". The black ellipse identifies the region where the horn is operating at it's highest sensitivity. Compare that range to the plot above.

The loudspeaker motor is working hard, but the cone excursions are low. It's working against the air-load at the throat.

The pay-off is higher sensitivity the downside is it's over a limited bandwidth.

Nothing comes for free.

post-864-0-52700000-1404954753_thumb.gif

Edited by John Warren

Share this post


Link to post
Share on other sites

John, I'm trying to understand your concern with the quoted statement. I don't want to get the answer wrong, so perhaps you should be more explicit with your comment or question? I said nothing about sensitivity.

If a decade of usable pass band performance isn't good enough for your needs then I assume that you'd do whatever it is that you want to do...regardless of how it sounds?

Share this post


Link to post
Share on other sites

"Distortion is inversely proportional to efficiency."

It's the increased sensitivity that's responsible for the reduction in modulation distortion. The two are linked together, and one cannot be examined or discussed without the other.

There is no free lunch with any of this stuff. In this.case, the reduction in modulation distortion results in a nasty frequency response and limited bandwidth.

Okay John. Some of the K-55-V and K-55-M drivers give solid performance for almost four decades, and the FR is, well, umm - acceptable. I personally think they sound great - at least in my average sized rooms.

Edited by DeanG

Share this post


Link to post
Share on other sites

Perhaps you two should start a thread on the efficiency of horns? This FAQ (not really a typical thread) is concerned with why horns subjectively sound better, a subject clearly difficult enough to tackle without also trying to discuss the engineering-related subject of acoustic efficiency at the same moment.

Share this post


Link to post
Share on other sites

"I've often wondered whether Frequency Modulation Distortion (FMD) and Intermodulation Distortion were the same thing, but one used in regard to loudspeakers, and the other in regard to electronics. Are they the same? Are they related?"

 

As it turns out, slew-induced distortion" (SID) and is most often discussed in relation to electronic amplifiers. The implication is that, if a sound reinforcement setup experiences slew-induced distortion, then all the issues are credited to the amplifiers(s) rather than the loudspeakers.

 

"Transient intermodulation distortion, or TIM, occurs in amplifiers that employ negative feedback when signal delays make the amplifier incapable of correcting distortion when exposed to fast, transient signals."

 

"DIM/TIM (dynamic/transient intermodulation distortion) A procedure designed to test the dynamic or transient behavior, primarily, of audio power amplifiers. The other IM tests use steady-state sine wave tones, which do not necessarily reveal problems caused by transient operation. In particular, audio power amplifiers with high amounts of negative feedback were suspect due to the inherent time delay of negative feedback loops. The speculation was that when a rapidly-changing signal was fed to such an amplifier, a finite time was required for the correction signal to travel back through the feedback loop to the input stage and that the amplifier could be distorting seriously during this time."

  • Like 1

Share this post


Link to post
Share on other sites

"If someone ran frequency response curves on a speaker at relatively low SPL, then higher SPL, then very high SPL, but still within the speaker's capability, would the three curves look the same (smoothness, where peaks and valleys are, etc.)?"

 

Horn-loaded loudspeakers will typically outperform direct-radiating loudspeakers in this area. In general, "power compression" or "amplitude compression" in loudspeakers is related to thermal heating of driver voice coils and passive crossover components.

 

It's also related to nonlinearities of the driver such as [magnetic] force factor Bl(x), inductance L­e(x) and compliance Cms(x) varying with displacement x. At low frequencies where the amplitude of the displacement is high these mechanisms produce the highest compression.

  • Like 1

Share this post


Link to post
Share on other sites

"Horns direct their acoustic output into angles of space (i.e., they have much greater "directivity" than direct-radiating loudspeakers). Does this have anything to do with why they sound better?"

 

Yes, it does. Because of listener effects, such as the Precedence Effect and a special case of the Precedence Effect, called the Haas Effect, near-field acoustic reflections around the loudspeakers cause listeners to hear distortions in the sound stage of stereo and multi-channel acoustic images. This subject is a bit more complicated, but one that can be simplified into a few basic concepts:

 

1) Splashing sound around in an enclosed space (a typical characteristic of many direct-radiating loudspeakers) can make the loudspeakers sound much larger, but at a high price in terms of loss of soundstage imaging performance. Many refer to this as the "Bose effect". Refer to the Corner-Horn Imaging FAQ.

 

2) Early reflections around horn-loaded loudspeakers, from the room boundaries and furnishings close to the loudspeakers (less than 6 feet, or 2 metres), tend to affect soundstage imaging of horns even more than direct-radiating loudspeakers. The reason for this is significantly better horn-loaded loudspeaker imaging performance if these early reflections are controlled/reduced in amplitude. This is the reason why many audiophiles recommend placing direct-radiating loudspeakers out into the floor of a room away from the corners or walls.

images.jpg

Since horn-loaded loudspeakers control their acoustic output much more effectively into angular sectors of space, they can also be placed in closer proximity to room walls or corners without losing their stereo or multi-channel imaging performance if simple acoustic treatments are used near the horns' mouth(s) to attenuate midrange early reflections, something that direct-radiating loudspeakers have much more difficulty doing without a large amount of absorptive acoustic treatments being used to absorb off-axis acoustic energy reflections off of near field surfaces.

 

3) The Haas effect (perception of direct+reflected sound paths as single acoustic images, centered on the direct sound path direction) only comes into play when the reflected midrange sound is greater than 0.7 ms (.0007 seconds), corresponding to 9.5 inches (24 cm) of path length difference between the direct sound and reflected sound surface. Any reflected sound delays less than 0.7 ms or 9..5 inches are perceived by listeners as altering the direction of the sound toward the reflected sound direction.

gallery_26262_6_4037.jpg

4) The Haas effect of listeners begins to break down after about 20 ms of delay in reflected sound, corresponding to ~23 feet (~7 metres) of delay between the direct midrange sound and the reflections. Many home listening spaces do not have primary reflections delayed by this amount due to the smaller dimensions of the listening rooms. However, larger and acoustically live listening spaces that have delayed midrange reflections greater than 23 feet will experience "garbling" of spoken words by the reflected waves.

  • Like 1

Share this post


Link to post
Share on other sites
The pay-off is higher sensitivity the downside is it's over a limited bandwidth.

thank goodness for 3 and 4 way wetups with Active Xovers, time delay, and PEQ. Taking that "systems" approach with horns is what makes them and the ROOM become closer to sonic nirvana.

Edited by ClaudeJ1

Share this post


Link to post
Share on other sites

Thank goodness for 3- and 4-way setups with active Xovers, time delay, and PEQ. Taking that "systems" approach with horns is what makes them and the ROOM become closer to sonic nirvana.

Three- and four-way loudspeakers can sound good, but I've found that particular care must be used in the crossover filters (including time delays) used, and matching the horizontal and vertical off-axis polars of the horn/drivers being crossed at the crossover frequencies.

The crossover must be inaudible

The crossover must be inaudible on program material. This also implies that the power response of the two drivers must be similar in the crossover region, and that requires special attention during the loudspeaker's concept and design phases.

Crossovers may be implemented either as passive RLC networks, as active filters with operational amplifier circuits or with DSP engines and software. The only excuse for passive crossovers is their low cost. Their behavior changes with the signal level dependent dynamics of the drivers. They block the power amplifier from taking maximum control over the voice coil motion. They are a waste of time, if accuracy of reproduction is the goal.

Siegfried Linkwitz

It seems like many implementations of multi-way speakers don't pay enough attention in these areas. Certainly using electronic active crossovers will help a great deal in locking in the crossover performance even under high-load conditions when thermal heating can cause crossover filter performance drift in passive designs, and over time when passive components can drift in value with age.

But I digress from the topic of this FAQ.

Edited by Chris A

Share this post


Link to post
Share on other sites

"So again, what does 'controlled directivity' have to do with why horn-loaded loudspeakers sound better?"

 

Basically, in small listening rooms such as home listening rooms there is a great advantage to keeping excess midrange and tweeter energy off the walls, ceiling, floors and other furniture, directing that acoustic energy instead toward the listeners without first reflecting off of another surface. The advantage shows up as dramatically improved soundstage imaging and the perception of realistic playback.

 

"If that's true, then why do dipole loudspeakers have a large perceived depth of field?"

 

Planar dipole loudspeakers radiate approximately equal amounts of acoustic energy forward and rearward, but not much from the sides, top, and bottom of the loudspeaker. That acoustic energy radiating to the rear of the loudspeaker bounces off the front wall and returns to the listening position as delayed reflections, as described above. That delayed reflected energy doesn't exist on the recording so the effect of this "front wall echo" is perceived as increased depth of field.

 

radiation.gif

 

The problem with that of course is that this added acoustic reflection information isn't in the original recording. You can achieve the same effect by using a reverberation unit, where you can also control its intensity and delay time. The only way to control the front wall reflections using planar dipoles is to move the speaker either into the room more or back toward the front wall. Most successful setups using these type of loudspeakers usually have larger dimensions in room length, width, and height to accommodate adjustment of the front-wall reflection. This usually requires at least 6 feet (2 metres) of spacing, which intrudes significantly into the usable space of the room. Additionally, just like all other types of loudspeakers, it is very desirable to keep any near-field acoustically reflective objects out of the near-field of the loudspeaker in order to preserve their soundstage imaging performance.

 

Cone-type dipole loudspeakers typically have the added disadvantage (or advantage, depending on your point of view) of having increased off-axis energy that the reflects off of side walls and other close-by furniture. These early reflections can create havoc with your soundstage imaging as described in the Corner-Horn Imaging FAQ.

 

So if the standard by which we measure "sounding better" includes accuracy in recreating the original sound space of the recording itself, dipole loudspeakers actually intentionally create a different sound space image than the original in order give the listeners a feeling of depth that wasn't actually there to begin with.

 

"So what's wrong with that?"

 

Nothing, as long as the listener really doesn't care about accurate sound reproduction.

  • Like 1

Share this post


Link to post
Share on other sites

Perhaps you two should start a thread on the efficiency of horns? This FAQ (not really a typical thread) is concerned with why horns subjectively sound better, a subject clearly difficult enough to tackle without also trying to discuss the engineering-related subject of acoustic efficiency at the same moment.

The comment by our buddy Chris A puzzles me a bit. I'm on John's side. You can't talk about subjective "why" without talking about engineering "why."

Regarding "subjective" I heard my first K-Horn on Thanksgiving Friday on 1974 at a dealer in Ithaca, NY. Everything about music reproduction changed for me, right then. I daresay it is the same for others.

In this hobby (which concerns love of reproduced music), we're all trying to find a better way of doing things with the hope that better hardware or software makes things subjectively better. But what is the link between them?

PWK spent a lot of intellectual effort on linking music, horn hardware, analysis of distortion, efficiency, and what is essentially subjective enjoyment of music, in a scientific way. A lot of others at the time, and to this day, engaged in hucksterism.

WMcD

  • Like 1

Share this post


Link to post
Share on other sites

Modulation distortion is produced when the acoustic driver's cone or diaphragm moves - and the more it moves, the greater the modulation distortion.

Isn't that wonderful...

so what happens if I build a bass horn using a 15" cone driver and, in fact, realize a reduced cone excursion but the problem is that the reduction occurs over a range of frequencies that's narrower than the frequency range I'm trying to operate my bass horn over?

This is what happens, you end up with a horn that has a high sensitivity over that region where you've realized this reduced cone excursion. In other words you end up with a Klipschorn folded bass unit...

What??? You never heard of path length differences???

Share this post


Link to post
Share on other sites

"Distortion is inversely proportional to efficiency."

It's the increased sensitivity that's responsible for the reduction in modulation distortion. The two are linked together, and one cannot be examined or discussed without the other.

There is no free lunch with any of this stuff. In this.case, the reduction in modulation distortion results in a nasty frequency response and limited bandwidth.

Okay John. Some of the K-55-V and K-55-M drivers give solid performance for almost four decades, and the FR is, well, umm - acceptable. I personally think they sound great - at least in my average sized rooms.

Geesh!!! Horns don't have a low pass filter unless they are folded. And if you have a ragged freq response....DON'T BLAME THE HORN. So much for the experts on this forum.....

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.


×
×
  • Create New...