It does decrease but not as much as higher poles. However the protection needed is also dependent on the tweeter you use. How do you know their protection requirements are all the same? It even depends on the type of music you play.
But you cannot change this? No speaker has got infinite bandwidth.
Looks like that paper is based on test tones. This is not new info if that's the case. The study I linked a few times showed us that. There are many studies that show audibility on test tones and improved discernment through headphones. Problem is I don't enjoy listening to that stuff especially through headphones and I suspect I'm not alone. I mostly listen to music, TV, or movies through my stereo. Studies done with music show no real difference especially in a room. That's where we are at this time.
Dan
Where are the studies? I have only seen the same old study done by someone at miami, quoted over and over and that is inadequate to be conclusive.
Sorry, it's the only link I've got of its nature. I was just pointing out that restating was has already been shown in the Daisuke Koya University of Miami study does not invalidate a separate and more important conclusion from a that study which also happens to uphold your study's same results and doesn't refute Daisuke Koya's findings.
Weighting priorities,
Dan
Weighting priorities,
Dan
There was a paper in JAES this year where they tested the audibility of crossovers of several orders (and also including FIR) with percussive instruments.
Regards
Charles
Regards
Charles
This is not a case for 1st order crossovers but an interesting read anyway.
Troels killer note article
Troels killer note article
QUOTE=Professor smith;2304365]It does decrease but not as much as higher poles. However the protection needed is also dependent on the tweeter you use. How do you know their protection requirements are all the same? It even depends on the type of music you play.
Not true! With a constant input signal being halved per octave of frequency decline (-6 db/oct) displacement [Vd] of a linear HF driver is still doubling. When the driver's non-linear region is reached, then the differences go to heat the voice coil and generate increased distortion products. Because HF drivers have small and light moving systems with limited excursion [Xmax] capabilities, this is not a desirable situation for them to sustain. The differences between available HF driver candidates is at least a magnitude below the dynamics of the issue addressed here. This assumes that driver selection is focused on audible performance, rather than such being sacrificed for the lack of adequate filtering of the driving signal.
But you cannot change this? No speaker has got infinite bandwidth.[/QUOTE]
Precisely my point! The notion of a first order loudspeaker system is a conundrum.
When you consider all the polls of filtration in the reproduced music delivery chain, the issue of crosover order, at best, is in the noise of the signal being delivered.
For those readers who want to gain important insights into crossover design, please study the information provided at
Linkwitz Lab
Crossovers.
There are other sites as well, but this one comes to mind as I type this post.
If your seeking ultimate loudspeaker performance, stay in the digital domain until it is time to drive a voice coil. When there, you will be doing it with a dedicated power amplifier tailored to suit the driver chosen.
Regards,
WHG
Not true! With a constant input signal being halved per octave of frequency decline (-6 db/oct) displacement [Vd] of a linear HF driver is still doubling. When the driver's non-linear region is reached, then the differences go to heat the voice coil and generate increased distortion products. Because HF drivers have small and light moving systems with limited excursion [Xmax] capabilities, this is not a desirable situation for them to sustain. The differences between available HF driver candidates is at least a magnitude below the dynamics of the issue addressed here. This assumes that driver selection is focused on audible performance, rather than such being sacrificed for the lack of adequate filtering of the driving signal.
But you cannot change this? No speaker has got infinite bandwidth.[/QUOTE]
Precisely my point! The notion of a first order loudspeaker system is a conundrum.
When you consider all the polls of filtration in the reproduced music delivery chain, the issue of crosover order, at best, is in the noise of the signal being delivered.
For those readers who want to gain important insights into crossover design, please study the information provided at
Linkwitz Lab
Crossovers.
There are other sites as well, but this one comes to mind as I type this post.
If your seeking ultimate loudspeaker performance, stay in the digital domain until it is time to drive a voice coil. When there, you will be doing it with a dedicated power amplifier tailored to suit the driver chosen.
Regards,
WHG
Last edited:
While it is true that the displacement must increase with decreasing frequency for first order slope, second order being constant displacement, there are factors you are ignoring.
Since the focus is on tweeters, let's consider them. A key factor is whether or not a driver exceeds it's Xmax under usage. The driver's output can be designed so that it is first order down to some reasonable point below Fc, where it will begin to transition to its raw slope. As long as that point and below is such that under usage it does not exceed Xmax, it will not be an issue. In all liklihood there will be a highpass that sufficiently attenuates the signal to prevent it exceeding its thermal limitations in the stop band as well. I have yet to have any tweeter blow under even somewhat heavy usage with music signals and I have used them first order and very low Fc with higher orders. One would actually have to measure the distortion to know if any signals in the stop band were creating any significant distortion due to insufficient attenuation in that region. Just saying it's bad has no merit.
It's also quite common to have a driver reach its thermal limitations in the passband rather than its stop band near or below Fc. We do this sort of decision making all the time with woofers. We don't highpass them in most cases other than with the box. In that case the woofer is being fed a non-attenuated signal below Fb and we must ensure that we do not over-drive it. That most certainly does not disqualify its usage in that manner, only that we take that condition into account.
In addition, your argument is akin to saying that no one should ever consider making a 2-way system since it's very likely that the woofer is likely to be displacement limited below Fb before it is thermally limit above it. It's nothing more than a design decision, so your position does not hold up under scrutiny. Every system is nothing but a set of tradeoffs. You have yours, others have theirs.
With regard to the increase in poles somewhere beyond Fc (highpass and lowpass), yes, that does occur. It's unavoidable. The key factor in this case is whether or not it matters in the course of intended usage. Those first order systems such as those of Dunlavy were well implemented and had excellent measured results in areas such as square wave response. The fact that a highpass or lowpass may increase order far down into the stop band is largely irrelevant since at that point its contribution to the response can be made very small. Again, it's up to the designer to select drivers appropriately for the desired task and implement a crossover that satisfies the conditions. Your concern is not necessarily misplaced here, but appears to be a total rejection without due consideration of other real-world factors. Again, nothing but a set of tradeoffs.
It seems to me that you've made your points, but those points are not definitive with regard to whether or not someone's design goal is attainable from a realistic standpoint. Yours is just one take of many and does not preclude a very realizable and good sounding system. First order has its limitations, but so does every system extant. Including any of yours.
Dave
p.s. As thoriated pointed out, there are other methods available to improve the response by shaping in the stop band that are benign in perceived response and minimal in measured response with regard to system summed response.
Last edited:
I see the objective as approximating a first order system crossover characteristic with the fewest out of band, dynamically related or lobing artifacts using the simplest practicable xover implementation, not being overly concerned with the implementation 'purity' of standard xover topologies.
My feeling is that conventional 'brick wall' filter approximations or adding DSP processing are of the most interest to audio professionals in sound reinforcement applications who stand to gain significantly by having adjustable dispersion control and squeezing the ultimate 2 or 3 db of dynamics from a given driver implementation that justifies the loss of low level detail and imaging integrity.
My feeling is that conventional 'brick wall' filter approximations or adding DSP processing are of the most interest to audio professionals in sound reinforcement applications who stand to gain significantly by having adjustable dispersion control and squeezing the ultimate 2 or 3 db of dynamics from a given driver implementation that justifies the loss of low level detail and imaging integrity.
I was not referring to infrasonic signals.
Similar concerns apply to MF and even HF for highpass.
Academically of course, drivers cannot be filtered to maintain first order to their limits. Within a useful bandwidth, the response can be made close enough that it's nothing more than an academic discussion. What you're saying is not news. One would have to measure the distortion due to out-of-band signals to say anything meaningful on that aspect. It is different for every system, so blanket statements are not sufficient.
But if we're going to get academic, how the poles are configured is an issue. What has been described is more appropriately staggered poles. That is a useful method that can provide most of the benefits that are attributable to true first order. If you don't like the description as a first order system, you're entitled to that. I consider one that does a respectable job of passing a square wave roughly intact to be close enough that I will call it first order. You're certainly free to continue to repeat that you do not consider it such.
Most of this is academic, since if it were possible to filter a driver to maintain first order to zero or infinity, as soon as one moves off of the design axis it's going to fall apart anyway.
But there is one aspect that is still a valid factor regardless of what one may want to call a system. That's power response. It is closely tied to the crossover type, staggered or not, regardless of how well it is maintained in the stop band. It is most important at and near Fc, so staggered poles or not, it is a consideration, or should be. Maybe you would reject that assertion, that's your choice. Other factors change such as the off-axis response aside from the power response, but I see no need to go there.
In any case, I suspect that we understand each others position.
Dave
I linked this in the OP's other thread on the subject but I think it applies here too. Rather than letting the extra poles roll off the response in a haphazard manner in the stopband, you can shape the curves to get 'transient perfect' response using the Duelund XO concept. The catch is you need lots of drivers to build a full-range system.
HTGuide Forum - Duelund meets Dunlavy (aka Duelund meets transient perfect)
Fuzzy-Up!
I choose, for the purpose of these discussions, to stay focused, on the core issue, that started this thread. That issue remains loudspeaker signal conditioning: 1st. order or higher. Obviously it is higher. Academia as well as the professional comunity is where the definitions of the terms we are all using come from. If we start to fuzzy-up the their meanings, to suit a particular position, then the conversation degrades into an exchange of meaningless jibber. Audio professionals are engaged in the design of loudspeaker systems of all types, not just those for sound reinforcement venues. If it cost a lot more and does not sound appreciably better to the client, you are out of a job. That notion is about as non-academic as it gets. Loudspeaker drivers remain the most expensive and problematic components of a sound system. The challenge remains to properly condition the signal fed to them. It is obvious that an approach the involves filters having more than on poll is required. This is confirmed by the text that proceeded this message. And, to this I maintain the position, that protection of the MF and HF drivers remains a paramont concern in the design of multi-way loudspeaker systems. The reader can take or leave this advice as he or she so chooses.
Regards,
WHG
I choose, for the purpose of these discussions, to stay focused, on the core issue, that started this thread. That issue remains loudspeaker signal conditioning: 1st. order or higher. Obviously it is higher. Academia as well as the professional comunity is where the definitions of the terms we are all using come from. If we start to fuzzy-up the their meanings, to suit a particular position, then the conversation degrades into an exchange of meaningless jibber. Audio professionals are engaged in the design of loudspeaker systems of all types, not just those for sound reinforcement venues. If it cost a lot more and does not sound appreciably better to the client, you are out of a job. That notion is about as non-academic as it gets. Loudspeaker drivers remain the most expensive and problematic components of a sound system. The challenge remains to properly condition the signal fed to them. It is obvious that an approach the involves filters having more than on poll is required. This is confirmed by the text that proceeded this message. And, to this I maintain the position, that protection of the MF and HF drivers remains a paramont concern in the design of multi-way loudspeaker systems. The reader can take or leave this advice as he or she so chooses.
Regards,
WHG
whgeiger,
But what about the issue of timbre? By using a high order crossover does this not change the relative harmonics? That timing is what characterises the timbre and sound of an instrument or voice. Can this be sacrificed?
But what about the issue of timbre? By using a high order crossover does this not change the relative harmonics? That timing is what characterises the timbre and sound of an instrument or voice. Can this be sacrificed?
Threads typically tend to diverge. Focus on what you wish. I suggest that you not presume that we should as well.
Most here are not concerned with clients. Have you not noticed the name of the site, DIYAudio? Please don't impose your conditions here. If you disagree, fine, but hammering your position is a benefit to one.
There is nothing being "fuzzied up", I believe that the majority here understand reality. I think that you'd have had a bit of issue from John Dunlavy from your position. He was as you likely know, rather involved as a professional designer and manufacturer. I doubt that he would accept your limited definitions of what one would consider a first order speaker system, given his preference. It really doesn't matter that there is some level of deviation from a perfect system, since perfect systems do not exist. It also doesn't change desires of those who might prefer them to make their best effort. Yours is a limited position I would wager.
In any case, this debate is wasted effort and I will waste my time no more. I will, however, help where I can, even to those who might desire a "first order" system. I doubt they will be in it for its commercial concerns. Real world concerns such as driver usage is valid, but limits that you would impose are not.
Dave
WHG, I guess I misunderstood what you were saying. I thought you were saying that a single electrical pole plus the poles of the drivers' natural roll-off yield multiple poles in the acoustic response. I certainly agree with that but....
Filters to me implies electrical response. A single cap may give a desired 3rd or 4th order acoustic response.
A single cap can provide plenty of protection if done right. See my earlier WinISD sim.
Your Presumptions
If a question was asked by me, it was rhetorical and not a solicition for a response. If you consider responding to my posts a waste of time, then simply move. You were not asked to do so. In the meantime, just call it what it is. The number of polls you use equals the electrical filter order. The frequency limits of each drivers acoustical response will approximate some order of high and low pass filter as well, and they certainly will not be 1st. order as much as you might like them to be. As the text in this thread clearly demonstrates, phase coherence can be successfully approached using multi-poll analog filters. End of story.
Regards,
WHG
If a question was asked by me, it was rhetorical and not a solicition for a response. If you consider responding to my posts a waste of time, then simply move. You were not asked to do so. In the meantime, just call it what it is. The number of polls you use equals the electrical filter order. The frequency limits of each drivers acoustical response will approximate some order of high and low pass filter as well, and they certainly will not be 1st. order as much as you might like them to be. As the text in this thread clearly demonstrates, phase coherence can be successfully approached using multi-poll analog filters. End of story.
Regards,
WHG
Last edited:
- Status
- Not open for further replies.