That is an easy one. Use a barrier to keep them away. Get a wind screen, preferably a long one that looks like part of the microphone. Then they can snug up to it all they want. Or use two microphones. The singer cannot mouth on both of them at once.
As a sax player, I sometimes have to duel with a sound man who insists on sticking the mike right into the bell of the horn, which utterly ruins the sound of low stack notes in two octaves. Also, I have to beg them - if I am too loud, do not turn my mike down; Turn my monitor UP.
Back to the topic at hand... I am really pleased with the result of using a good quality equalizer, but not to override the recording-engineer's judgement. I have some equipment for measuring the response, and I try to get it dead flat. I removed the baffle step correction circuit from inside the box, so the EQ takes care of that too.
The point is, at a given listening location, if the response is not flat, something will sound wrong. Vocals particularly need flat EQ, assuming the recording is good. Treble affects not only "sparkle" but also transients and sharp attacks. Too much treble makes for overshoot and the resulting cringe-factor.
"I have some equipment for measuring the response, and I try to get it dead flat."
The problem is that flat needs to be flat from where the sound reaches your ears all the way back to the recording microphone. That's a lot to ask but anything less can mean audible FR distortion. Since there is no standard way to make a recording, you can expect overall record/playback system equalization to be different to obtain truly flat overall response for every recording. I've got two with consecutive catalog numbers, same musicians, same arrangements, same studio, same equipment, different recording engineer and they are very different eq wise. I've also got several recordings made by the same engineer of the same instrument in the same place at different times, differenent eq. BTW, reflections in the listening room count too! Just in case it all seemed too easy.
The problem is that flat needs to be flat from where the sound reaches your ears all the way back to the recording microphone. That's a lot to ask but anything less can mean audible FR distortion. Since there is no standard way to make a recording, you can expect overall record/playback system equalization to be different to obtain truly flat overall response for every recording. I've got two with consecutive catalog numbers, same musicians, same arrangements, same studio, same equipment, different recording engineer and they are very different eq wise. I've also got several recordings made by the same engineer of the same instrument in the same place at different times, differenent eq. BTW, reflections in the listening room count too! Just in case it all seemed too easy.
Another problem is that most tweeters piston range stops around 5KHz, causing severe uncontroled break up modes, that gives high distortion, right in the frequence range where the ear is most sensitive. That distortion, combined with the other tings already mentioned, is really bad for sound quality.
My remedy to this so fare, is to design a midrange (maybe fullrange) driver, that can cover the frequence range up to 9 KHz. Before X-over to the tweeter.
Even though dispersion of sound is less at high frequences, I dont feel it is a problem in a stereo system in a normal living room condition. Maybe even an advantage, since less high frequence energy is reflected on walls and roof/floors.
My remedy to this so fare, is to design a midrange (maybe fullrange) driver, that can cover the frequence range up to 9 KHz. Before X-over to the tweeter.
Even though dispersion of sound is less at high frequences, I dont feel it is a problem in a stereo system in a normal living room condition. Maybe even an advantage, since less high frequence energy is reflected on walls and roof/floors.
A self correcting problem
Don't worry, the problem at least to some extent goes away with time.
When I was younger and listening to older 'consultants' demoing hifi the systems generally sounded all to tizzy. Now I am older and the salesmen and recording engineers somewhat younger it's all OK.
I just need younger ears
Don't worry, the problem at least to some extent goes away with time.
When I was younger and listening to older 'consultants' demoing hifi the systems generally sounded all to tizzy. Now I am older and the salesmen and recording engineers somewhat younger it's all OK.
I just need younger ears
Some interesting and varied opinions on what sibilance actually is and what causes it, so I can't help but throw my 2c into the mix
I don't agree with Lynn's comment that sibilance is in the 3-5Khz region, that's the presence region and peaks caused by resonances in this range will cause harshness and aggressiveness in the midrange not sibilance.
The frequency range table posted by twinter is pretty much spot on - sibilance is in the ~6-8Khz low treble region, and is something you hear on "essess" in speech.
I'm surprised nobody else has suggested it, but in my opinion although the spectral balance through the 6-8Khz region will affect how obvious or objectionable sibilance is, (with too much output here being worse) just adjusting the frequency response balance on a speaker which has this problem will not cure it because the root cause is a poor, lingering decay response in this frequency region - typically due to cone/dome breakup.
In other words a very fast/clean cumulative spectral decay response in the treble is the key to eliminating sibilance while still maintaining crispness and a sense of air.
The biggest offenders are soft domes IMHO which are typically already in breakup modes at 6-8Khz but they're by no means the only culprits, full range drivers are another, much as I love them in other ways, and most tweeters have the issue at least to some extent.
I find the typical characteristic of a tweeter that has poor CSD in the sibilance region is that the frequency response balance seems unduly critical - just a tiny bit too much emphasis above 6Khz and it will sound sibilant, possibly grainy and eventually fatiguing, not enough emphasis and it may not sound sibilant but it will just sound dull with no sparkle or crispness that the original poster alludes to. The difference between too much and not enough can be 1dB or less.
You're in a no win situation trying to adjust the frequency response, especially when the treble balance of the recorded material can vary a lot more than this - causing some recordings to sound bad but not others, depending on small variations in spectral balance of the treble, despite there being no actual quality problem or inherent sibilance in the recording.
One approach which I think truly does eliminate harsh sibilance but still maintain excellent crispness is a true aluminium foil ribbon tweeter. (Not to be confused with a planar tweeter with a hybrid diaphragm) I started using them about 8 years ago and would never go back to a dome now.
A good ribbon tweeter is almost entirely devoid of sibilance artefacts, and if you look at it's CSD plot you can see why, they have the best decay response of just about any tweeter and are almost entirely resonance free in their operating range, not to mention extremely flat to well beyond 20Khz.
Their mode of operation means that they are not operating in cone breakup mode at all because the entire diaphragm mass is directly driven.
Of course they do need crossing over higher and steeper (minimum 18dB/oct) than most domes, but that's a good idea anyway because as Lynn pointed out IM problems with tweeters are a lot more serious than many realise, with a lot of the harshness of some domes being due to IM distortion from being over driven at and below their resonance...
A secondary factor which I think is important with sibilance is directivity patterns. Again, picking on the dome tweeter a typical dome tweeter will have very wide dispersion in the sibilance region but with a considerably reduced dispersion in the top octave from 10-20Khz.
This, along with the typical reduced dispersion of the midrange driver below the tweeter crossover frequency will produce power response bloom from the crossover frequency up to about 10Khz - so there will be some emphasis of both the presence "harshness" region and the sibilance region in the reverberant field of the room, so even though the on-axis response may be dead flat, that range may seem elevated in the far field, so if the driver also has poor decay response this will tend to sound sibilant.
A driver with deliberate directivity control, for example a waveguide with a constant horizontal beamwidth of 90 degrees over it's entire operating range will allow you to still design for a flat on axis response but without a crossover to ~10Khz bloom in the power response, making any tendency towards sibilance less noticeable. (This alone is not a cure though, just a mitigation)
A wave guide loaded ribbon tweeter has the best of both worlds - extremely clean CSD, flat frequency response and a relatively constant ~90 degree horizontal beamwidth over it's entire working range. (Depending on model)
What I notice when using a ribbon tweeter of this type is that the balance between the upper treble and the rest of the frequency range is far less critical. It can be out by as much as +/- 1 dB or so and still sound smooth and un-fatiguing, and without either having problems with sibilance or lack of crispness in one extreme or the other, making it much more tolerant of variations in the tonal balance of the recording.
If you want to eliminate sibilance but retain sparkle, look at the CSD response first and foremost...
I don't agree with Lynn's comment that sibilance is in the 3-5Khz region, that's the presence region and peaks caused by resonances in this range will cause harshness and aggressiveness in the midrange not sibilance.
The frequency range table posted by twinter is pretty much spot on - sibilance is in the ~6-8Khz low treble region, and is something you hear on "essess" in speech.
I'm surprised nobody else has suggested it, but in my opinion although the spectral balance through the 6-8Khz region will affect how obvious or objectionable sibilance is, (with too much output here being worse) just adjusting the frequency response balance on a speaker which has this problem will not cure it because the root cause is a poor, lingering decay response in this frequency region - typically due to cone/dome breakup.
In other words a very fast/clean cumulative spectral decay response in the treble is the key to eliminating sibilance while still maintaining crispness and a sense of air.
The biggest offenders are soft domes IMHO which are typically already in breakup modes at 6-8Khz but they're by no means the only culprits, full range drivers are another, much as I love them in other ways, and most tweeters have the issue at least to some extent.
I find the typical characteristic of a tweeter that has poor CSD in the sibilance region is that the frequency response balance seems unduly critical - just a tiny bit too much emphasis above 6Khz and it will sound sibilant, possibly grainy and eventually fatiguing, not enough emphasis and it may not sound sibilant but it will just sound dull with no sparkle or crispness that the original poster alludes to. The difference between too much and not enough can be 1dB or less.
You're in a no win situation trying to adjust the frequency response, especially when the treble balance of the recorded material can vary a lot more than this - causing some recordings to sound bad but not others, depending on small variations in spectral balance of the treble, despite there being no actual quality problem or inherent sibilance in the recording.
One approach which I think truly does eliminate harsh sibilance but still maintain excellent crispness is a true aluminium foil ribbon tweeter. (Not to be confused with a planar tweeter with a hybrid diaphragm) I started using them about 8 years ago and would never go back to a dome now.
A good ribbon tweeter is almost entirely devoid of sibilance artefacts, and if you look at it's CSD plot you can see why, they have the best decay response of just about any tweeter and are almost entirely resonance free in their operating range, not to mention extremely flat to well beyond 20Khz.
Their mode of operation means that they are not operating in cone breakup mode at all because the entire diaphragm mass is directly driven.
Of course they do need crossing over higher and steeper (minimum 18dB/oct) than most domes, but that's a good idea anyway because as Lynn pointed out IM problems with tweeters are a lot more serious than many realise, with a lot of the harshness of some domes being due to IM distortion from being over driven at and below their resonance...
A secondary factor which I think is important with sibilance is directivity patterns. Again, picking on the dome tweeter a typical dome tweeter will have very wide dispersion in the sibilance region but with a considerably reduced dispersion in the top octave from 10-20Khz.
This, along with the typical reduced dispersion of the midrange driver below the tweeter crossover frequency will produce power response bloom from the crossover frequency up to about 10Khz - so there will be some emphasis of both the presence "harshness" region and the sibilance region in the reverberant field of the room, so even though the on-axis response may be dead flat, that range may seem elevated in the far field, so if the driver also has poor decay response this will tend to sound sibilant.
A driver with deliberate directivity control, for example a waveguide with a constant horizontal beamwidth of 90 degrees over it's entire operating range will allow you to still design for a flat on axis response but without a crossover to ~10Khz bloom in the power response, making any tendency towards sibilance less noticeable. (This alone is not a cure though, just a mitigation)
A wave guide loaded ribbon tweeter has the best of both worlds - extremely clean CSD, flat frequency response and a relatively constant ~90 degree horizontal beamwidth over it's entire working range. (Depending on model)
What I notice when using a ribbon tweeter of this type is that the balance between the upper treble and the rest of the frequency range is far less critical. It can be out by as much as +/- 1 dB or so and still sound smooth and un-fatiguing, and without either having problems with sibilance or lack of crispness in one extreme or the other, making it much more tolerant of variations in the tonal balance of the recording.
If you want to eliminate sibilance but retain sparkle, look at the CSD response first and foremost...
Last edited:
I think what Lynn is saying (and I agree with him 100%) is that the problems originate in the crossover region (3-5kHz), but manifest themselves at higher frequencies. For example, if a tweeter is driven past linear excursion at 3k because of a low slope crossover, the principal harmonic distortion components will be at 9 and 15k.
That seems reasonable.
In all likely-hood "any crap that shouldn't be there" in the 6-8Khz sibilance region will contribute to sibilance in some way - whether slow decaying tails of resonances due to poor CSD, or harmonic/IM products from other frequencies that fall within that 6-8Khz region - at any given moment in time both are extra frequency components that aren't in the original signal, which are "over stimulating" the ear in that frequency range.
I still think the CSD is the more important of the two though, as ribbon tweeters have drastically better CSD's than domes but tend to have somewhat higher distortion figures than the best domes, despite that they still manage to sound far cleaner and free of sibilance, which makes me think the former property is more important.
If you have both poor decay response and harmonic distortion then harmonics of the lower frequencies (like the 3rd harmonic of 2Khz) would not only fall in the sibilence region, but would stimulate those slow decaying resonances into action, compounding the problem...
Last edited:
Ok, so now we're almost on to something!
As I was saying some time ago, the issue is "bad harmonics". IM distortion is *one* source. Lynn pointed out the issue of transistor amps switching between A and AB and a sliding bias point. This speaks to what I said earlier about the "sibilance" sounding different with different amps (amps that are not highly flawed... maybe even almost perfect!), especially solid state amps with global feedback.
Now, what becomes of the IM distortion that the tweeter is producing? Does it just go out into the world? OR does it show up on the speaker cable as a signal? If so, does the amp "see" that as "distortion" in the output and try to correct it? Well, wait a second. Actually, wait a few uSec there cowboy! IF this happens, then the signal that is trying to be corrected by the feedback is LATE compared to the signal that is being sent to the tweeter, through the xover! It has to be late by definition because of the time constants and phase shift inherent in the xover!
I agree with what DBMandrake says, in essence it mirrors my argument regarding the quality of the HF driver(s) being a major factor in how much and if "sibilance" is heard.
The best direct radiator "tweeter" I ever heard is/was the now extinct Stage Accompany, slightly modified. The thing spec'd at 1% THD @ 128dbSPL! Work backwards to normal listening levels... things like the Focal T120 series sounded like "hash" in comparison. Truly. And those are the same tweeters used by Wilson in his Watt and bigger systems... I added Ferrofluid which improved it over the stock, fyi.
Soundminded - there is a problem with your illustration, HF falls off faster than does LF, so as you move away from the listener with a "flat" nearfield source, the highs will naturally roll off. Also, the sound that you perceive is not just the direct sound, it is the sum of the direct sound AND the reverberant field as long as it meets the Haas time limit... (which btw, is one reason that pretty large rooms can sound very terrific...).
There are horns that are deliberately somewhat narrow in their HF dispersion (check the JMMC thread here), that in an of itself does not produce sibilance. I still do not subscribe to your theory on this.
Lynn - I have horns that you can hear sibilance from, but that sibilance varies with the components in the signal chain from non-existant to somewhat objectionable, depending on what component is put in. This without any peaks whatsoever in the response - documented. So the source of the sibilance is almost certainly not the horn or compression driver, especially since I can report that with what I consider to be the best amps tried on my system (that includes a wide range) that the highs are anything but rolled off, they are very clear and open - and there is nil sibilance. In fact it makes you want to try every recording you have because it is just so great to listen!
_-_-bear
As I was saying some time ago, the issue is "bad harmonics". IM distortion is *one* source. Lynn pointed out the issue of transistor amps switching between A and AB and a sliding bias point. This speaks to what I said earlier about the "sibilance" sounding different with different amps (amps that are not highly flawed... maybe even almost perfect!), especially solid state amps with global feedback.
Now, what becomes of the IM distortion that the tweeter is producing? Does it just go out into the world? OR does it show up on the speaker cable as a signal? If so, does the amp "see" that as "distortion" in the output and try to correct it? Well, wait a second. Actually, wait a few uSec there cowboy! IF this happens, then the signal that is trying to be corrected by the feedback is LATE compared to the signal that is being sent to the tweeter, through the xover! It has to be late by definition because of the time constants and phase shift inherent in the xover!
I agree with what DBMandrake says, in essence it mirrors my argument regarding the quality of the HF driver(s) being a major factor in how much and if "sibilance" is heard.
The best direct radiator "tweeter" I ever heard is/was the now extinct Stage Accompany, slightly modified. The thing spec'd at 1% THD @ 128dbSPL! Work backwards to normal listening levels... things like the Focal T120 series sounded like "hash" in comparison. Truly. And those are the same tweeters used by Wilson in his Watt and bigger systems... I added Ferrofluid which improved it over the stock, fyi.
Soundminded - there is a problem with your illustration, HF falls off faster than does LF, so as you move away from the listener with a "flat" nearfield source, the highs will naturally roll off. Also, the sound that you perceive is not just the direct sound, it is the sum of the direct sound AND the reverberant field as long as it meets the Haas time limit... (which btw, is one reason that pretty large rooms can sound very terrific...).
There are horns that are deliberately somewhat narrow in their HF dispersion (check the JMMC thread here), that in an of itself does not produce sibilance. I still do not subscribe to your theory on this.
Lynn - I have horns that you can hear sibilance from, but that sibilance varies with the components in the signal chain from non-existant to somewhat objectionable, depending on what component is put in. This without any peaks whatsoever in the response - documented. So the source of the sibilance is almost certainly not the horn or compression driver, especially since I can report that with what I consider to be the best amps tried on my system (that includes a wide range) that the highs are anything but rolled off, they are very clear and open - and there is nil sibilance. In fact it makes you want to try every recording you have because it is just so great to listen!
_-_-bear
Last edited:
As mentioned, you need to separate system sibilance (your equipment) from systemic sibilance, i.e., swallowed microphones and EQ problems at the recording end.
On the equipment side, a "hot" sounding MC cartridge overdriving a solid state phono stage is an awesome way to create sibilance, especially when using a speaker prone to sibilance in the first place. For speaker-induced sibilance: this can be cured by using a planar driver that is run thru the entire mid range and into the lower-to-mid treble region. After that, you can use any tweeter you like, because you've just by-passed the problem.
On the equipment side, a "hot" sounding MC cartridge overdriving a solid state phono stage is an awesome way to create sibilance, especially when using a speaker prone to sibilance in the first place. For speaker-induced sibilance: this can be cured by using a planar driver that is run thru the entire mid range and into the lower-to-mid treble region. After that, you can use any tweeter you like, because you've just by-passed the problem.
Last edited:
"HF falls off faster than does LF, so as you move away from the listener with a "flat" nearfield source, the highs will naturally roll off. "
Not that quickly. I've got data for about 200 concert halls. Very typical is at 8 khz, RT is about 50% as 1khz RT. So for a hall with 2.0 sec RT @ 1KHz, it's around 1 sec at 8 khz. This represents sound traveling for about half a mile and includes the relative spectral loss of dozens of collisions with the boundaries which are selectively more reflective at lower frequencies. For a few feet in a small room this adds up to an insignificant difference.
I'm glad you mentioned MC cartridges with hot high ends. Can't find old data but what I had for them from ancient lab reports for example for Ortophon showed uniformly that they had HF peaks of around 5 db centered around 15 khz. We've discussed this with great disagreement on other threads, notably on one regarding JC's Blowtorch preamp. I contend that the best MM cartrdiges were flat, he claims they were rolled off and the MCs are flat. Never saw a CD-4 cartridge that was an MC. Requred FR is to 40 Khz and beyond and tracking force needed to be low to avoid shaving off the outband rear channels between 20 and 40 khz. I've got an Empire 4000D/III myself and it sounds identical to me to the 999VE. Both track very well at /4 gram in Empire arms. The Shure V15 Type V MR seems to work very well too. BTW, I'm not a fan of vinyl phonograph records.
The answer to that is essentially no. Distortion and cone breakup resonances are not reflected back to the amplifier electrically and thus can't be controlled by it.
Even if the amplifier had precise control of the voice coil in the tweeter (which it doesn't due to voice coil inductance, passive crossover components etc) it could only control the voice coil movement.
In a dome tweeter the voice coil is only attached to the dome at the edge and the dome is not infinitely stiff, so above some frequency the middle of the dome starts moving independently of the edge. At some frequencies it will even be moving in the opposite direction direction to the edge. (Imagine a saw blade held at both ends - move it back and forth quickly and at the right speed the middle will be moving the opposite way to the ends, same if you hold it only in the middle...)
At even higher frequencies small pizza slice shaped sectors of the cone will be moving back and forth in complex semi-random relationships - it's now well into the cone breakup region, which starts as low as ~6-8Khz on soft domes.
Without proper damping a wide spectrum of narrow band resonances form, but even with damping (doping of the fabric on soft domes) the resonances are only reduced, not eliminated. Because the material doesn't bend perfectly linearly non-linear IM and harmonic distortion are also introduced by the bending of the cone.
Titanium domes try to make the dome so stiff and light that it doesn't bend at all until above 20Khz, but metal is inherently poorly damped so at some frequency between 24 and 30Khz the diaphragm resonates extremely strongly. It's open to debate which approach is the lesser of two evils - damped but chaotic resonances and distortion in the audible range or a very powerful single frequency supersonic resonance...
The beauty of a ribbon tweeter on the other hand is the voice coil and diaphragm are one and the same, not separate components. The entire moving and radiating mass is under direct magnetic control rather than just being moved from the edges.
Equal controlling force is being applied to all parts of the ribbon foil diaphragm at once, (within the limits of equal field distribution in the gap anyway, which is never quite perfect) so it follows the signal more or less precisely as a piston without any bending or chaotic resonant behaviour, right through and well beyond the audible spectrum, with no high Q super sonic resonances either.
As a result the CSD is exceptionally clean and on a well designed ribbon the response is very flat too. (Non-flatness of response on some ribbons is largely due to less than optimal wave-guide designs which introduce some diffraction effects or unwanted frequency dependant horn loading, but it's not an issue on a well designed model, and doesn't really affect the CSD)
"It's open to debate which approach is the lesser of two evils "
I like the mylar soft domes used in arrays. With a phase plug dispersion is quite good and can be made even better in arrays. Arrays are practical because these units are now so cheap. I've paid as little as five cents each. In fairly large arrays each one consumes so little power and produces so little sound that the chances for breakup and for dynamic compression due to voice coil heating is greatly reduced. Usually however they are used singly in low cost models and are pushed too hard IMO.
I like the mylar soft domes used in arrays. With a phase plug dispersion is quite good and can be made even better in arrays. Arrays are practical because these units are now so cheap. I've paid as little as five cents each. In fairly large arrays each one consumes so little power and produces so little sound that the chances for breakup and for dynamic compression due to voice coil heating is greatly reduced. Usually however they are used singly in low cost models and are pushed too hard IMO.
I'm afraid that I don't agree with some of what was just said...
But it's pointless to argue on all of these points...
I was not suggesting that the amplifier "controls" the back EMF from drivers, I'm suggesting that it can't and doesn't because what we have is destructive interference, albeit rather low in level due to the phase delays and time delays of the sources - those being IM and THD...
Try using the speaker as a microphone, it will produce output, yes?
But you don't have to believe this, and I can't prove it at the moment...
Soundminded - I'd be interested in your sources for "5 cent" mylar tweeters!
But any driver in a line source played at a given level is lower in distortion than a single driver played at the same level. No news there. Arrays offer some real benefits.
True ribbons have their own issues.
The ribbon has standing waves on the surface. The longer the ribbon the worse it gets.
The ribbon has limited power handling and SPL
IF the ribbon is long and narrow, the vertical dispersion is very limited.
The horizontal dispersion is narrowing as the freq rises, WRT the width of the ribbon.
No free lunch today...
_-_-bear
But it's pointless to argue on all of these points...
I was not suggesting that the amplifier "controls" the back EMF from drivers, I'm suggesting that it can't and doesn't because what we have is destructive interference, albeit rather low in level due to the phase delays and time delays of the sources - those being IM and THD...
Try using the speaker as a microphone, it will produce output, yes?
But you don't have to believe this, and I can't prove it at the moment...
Soundminded - I'd be interested in your sources for "5 cent" mylar tweeters!
But any driver in a line source played at a given level is lower in distortion than a single driver played at the same level. No news there. Arrays offer some real benefits.
True ribbons have their own issues.
The ribbon has standing waves on the surface. The longer the ribbon the worse it gets.
The ribbon has limited power handling and SPL
IF the ribbon is long and narrow, the vertical dispersion is very limited.
The horizontal dispersion is narrowing as the freq rises, WRT the width of the ribbon.
No free lunch today...
_-_-bear
Thank you Lynnn!! Now I have a clue as to why my speakers sound great on some material and terrible on others.
One point for clarification, when you are saying 1st order crossover, are you meaning 1st order electrical or 1st order acoustic? My existing crossover is 1st order electrical on the tweeter, but 2nd order bessel acoustically.
Is a resonant peak filter on the tweeter a possible remedy (I don't currently have one) , or should I just go to a higher order acoustic slope in all likelyhood?
Tony.
Tony, I don't want to speak for Lynn, but a "resonant peak filter" or notch filter ( I presume ) will not solve the problem Lynn is referring to. He is saying that although the acoustic output of the tweeter is falling, the excursion may not be falling sufficiently fast below the "xover point" so as to NOT cause significant IM distortion when there is program material with energy in that zone.
Hope that made sense to you, the writer always thinks he/she is clear...
_-_-bear
PS. classical xover theory/practice calls for drivers that have a flat response that overlaps by one or more octaves around the xover point! Of course in practice drivers like that rarely exist, so most systems consist of substantial compromise WRT the xover, the xover point and the filter slopes...
Hope that made sense to you, the writer always thinks he/she is clear...
_-_-bear
PS. classical xover theory/practice calls for drivers that have a flat response that overlaps by one or more octaves around the xover point! Of course in practice drivers like that rarely exist, so most systems consist of substantial compromise WRT the xover, the xover point and the filter slopes...
Perhaps a "DSS filter" is the way? I've been using this happily with some of my dipoles (but not others)
ORION-3
The shelving filter is centered at 1.8khz with depth of 3.3db and Q=1
SL's argument is that we're listening to the speakers at an angle and hence the HRTF is not flat.
Last edited:
I didn't know mylar soft domes were still being made
I know they were very popular in the 70's. My Dad had some early Philips ones (AD 0162 T8 ?) in some speakers he built back then and they have built in phase plugs held in place by a silk screen cover. (With the dome recessed behind it) Not bad sounding for their day, quite efficient at about 96dB/w/m but, they definitely still had sibilance issues, possibly even more than a doped fabric soft dome...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.