Why ask anything about what companies do? There are so many priorities and compromises involved that its never ever will start with "what XO should we use in our speakers". It just seems foolish to start that way.
Not all audio companies are concerned with the ultimate SQ or pleasing audiophiles because there is a little thing called profit and survival. There is enough volume in the crowed audiophile market either, hence the HUGE price tag on designs that are not all that special in the first place.
A 3rd order XO design can sell as many if not more speakers then a 1st order design if two conditions exist.
1. Its looks very nice
2. Its price is right
Its not rocket science here. Audiophiles do not control the audio business in real terms of volume sales (that is a good thing too!!). Business success has little do with trying to make the ultimate speaker. Business success is more about making the speakers that ultimately sell.
Look at the answers in this thread, how replies are simply "I like this sound" type subjectivity? The problem with those replies is those individuals forget to mention their speaker design goals, drivers used, etc. They are not exactly drivers that would work in my world and 1st order would absolutely suck but it works for them. Its all about specific applications and priorities. Geddes speakers I think use much higher XOs and everyone that has heard them loves them so XO order has no correlation back to great SQ. A first order XO can sound horrible as much as a 3rd order XO.
Define the application, define the priorities. define the box design, pick the drivers, measure drivers then figure out the XOs. Notice how far down the chain XO design happens?
Last edited:
1st Order What?
A loudspeaker driver, when modeled as a filter, has poles as well. So, the notion of a "first order" speaker system is at best, a conundrum.
An assertion of "not true" does not change the physics of the matter. Which I will again repeat here:
"As displacement is increasing inversely with the square of frequency, high pass filters with slopes of 6 db/oct do not afford the required protection."
There is undeniable relationship between driver efficiency and bandwidth. The product of these parameters is constant. If a high frequency driver must sustain relatively large excursions that are imposed by the lack of adequate filtering, then its efficiency and related rise times will have to be sacrificed to sustain the increased bandwidth requirement. In such systems, output capacity will be constrained by the high frequency drivers (with small [Vd])rather than those (with high [Vd]) covering the lower bands, where most of the power exists in the signal.
Regards,
WHG
A loudspeaker driver, when modeled as a filter, has poles as well. So, the notion of a "first order" speaker system is at best, a conundrum.
An assertion of "not true" does not change the physics of the matter. Which I will again repeat here:
"As displacement is increasing inversely with the square of frequency, high pass filters with slopes of 6 db/oct do not afford the required protection."
There is undeniable relationship between driver efficiency and bandwidth. The product of these parameters is constant. If a high frequency driver must sustain relatively large excursions that are imposed by the lack of adequate filtering, then its efficiency and related rise times will have to be sacrificed to sustain the increased bandwidth requirement. In such systems, output capacity will be constrained by the high frequency drivers (with small [Vd])rather than those (with high [Vd]) covering the lower bands, where most of the power exists in the signal.
Regards,
WHG
Yesterday I burned a ScanSpeak 13M kevlar mid 🙁 .... I think I will be inclined to think a little more about power handling in the future ....
I'm running 2. order filters at 400 Hz and 2.5 kHz.
I was testing my new "small" class-d amplifier, and with such clean sound, you don't really realise how loud you play until it's too late .....
First order .... don't think so ...... at least not for the main speakers ....
I'm running 2. order filters at 400 Hz and 2.5 kHz.
I was testing my new "small" class-d amplifier, and with such clean sound, you don't really realise how loud you play until it's too late .....
First order .... don't think so ...... at least not for the main speakers ....
Well I don't have a clipping indicator on the amp .... so that is hard to say .... but I don not believe so, from the sound of it ....
About 150 W in 8 ohm
About 150 W in 8 ohm
No, no kit as far as I know. X-over was .22mH (MDM) on the woofer and 2.2 uF on the tweeter. Surface mount the woofer and rebate the tweeter. Enclosure is up to you!
Terry
Last edited:
That depends on what drivers you are using! As I said earlier, driver selection is the most important part of any design. You are after all marrying the drivers. You wouldn't marry a nun to a member of the sex pistols.
Terry
That is not a good way to design a speaker IMHO. Define the application, Driver selection, Driver selection, Driver selection! Most of what I read here is "This is a good tweeter, this is a good mid, this is a good woofer, now how do I put them all together???
You must pick a good driver, say the woofer and then ask what is a good mid that will work well with this woofer and then will it work with this tweeter. All too often your x-over is a compromise because you start to try to shoe-horn drivers together when in fact they don't want to be together. You effort then goes into making the best of the bad situation.
Think driver selection first and formost!
Terry
Thats because there are too many variables. But would Geddes speakers preserve the waveform which goes in? If not, its not high fidelity.
what abut phase shifts in crossovers though? Once you have phase shift you cant fix it which is why you might start with the crossover first.
the protection is less but it also depends on what drivers you use? Some are more robust than others.
Not everyone that has heard them loves them. But that's a topic for a different thread.
Clipping shouldn't be a problem with that power unless it's insanely loud. I thought it may have been a 5W-10W amp which could have a problem with clipping.
Well - this statement does not tell the full truth.
I can certainly agree that there is less protection with 1st order filters than with higher order ones. So far so good.
But then there are always those who say that 1st and even 2nd order crossovers would give no displacement reduction at all, which is definitley not true.
The fact is that they simply offer less protection, but that might still be adequate for the right drivers (so the expression "required protection" is quite vague). Also keep in mind that the driver's displacement doesn't continue to rise below its resonant frequency - at least not for a closed box (which is the case for most tweeters and mids).
In a design that is using the right drivers for the task the only disadvantage that is left is the problem of lobing.
Regards
Charles
And this is only a problem if the acoustic centres are not aligned in the verticle plane.!!! Also does not occur in MTM arrangement.
Terry
No!
The inverse relationship between frequency and volume displacement for a constant power response is independent of the driver used. As frequency decreases, more volume displacement from any driver will be required.
[vd](max) = [sd]*[xmax] is the displacement limit of the particular driver used. Some drivers are incapable of reproducing low frequencies because they become displacement limited at these frequencies. However their improper use does not change the requirement to displace an adequate amount of air when required.
Regards,
WHG
The inverse relationship between frequency and volume displacement for a constant power response is independent of the driver used. As frequency decreases, more volume displacement from any driver will be required.
[vd](max) = [sd]*[xmax] is the displacement limit of the particular driver used. Some drivers are incapable of reproducing low frequencies because they become displacement limited at these frequencies. However their improper use does not change the requirement to displace an adequate amount of air when required.
Regards,
WHG
We are talking about tweeters and mids, not woofers! "Some drivers are incapable of reproducing low frequencies" Yes, some are, others are not. Choose drivers carefully and 1st order is not a problem!!
Terry
Gossamer vs. Beefcake
The relationship applies to all drivers across the entire audio frequency range! For a tweeter vs. midrange the rule applies equally to that of a midrange vs. a woofer. The high-pass filter needs to be greater than 6db/oct to adequately protect the driver receiving the signal from it. Please not that when a signal is feed to a driver, and it does not respond proportionally to it, the difference gets translated into additional heat in the voice coil. If it does respond, then at a minimum, increased amounts of IM distortion will be introduced into the output due to the Doppler effect and operation of the driver suspension in its non-linear range. While these conditions may be characterized as "not a problem", there are other far superior design regimens available to the loudspeaker designer that will maximize performance of a given driver assemblage by proper signal filtering. In the case of most HF and MF units, if they are operated by a signal not filtered by, 12db/Oct minimum sloped filter, the manufacturer's warrantee will be voided.
Regards,
WHG
The relationship applies to all drivers across the entire audio frequency range! For a tweeter vs. midrange the rule applies equally to that of a midrange vs. a woofer. The high-pass filter needs to be greater than 6db/oct to adequately protect the driver receiving the signal from it. Please not that when a signal is feed to a driver, and it does not respond proportionally to it, the difference gets translated into additional heat in the voice coil. If it does respond, then at a minimum, increased amounts of IM distortion will be introduced into the output due to the Doppler effect and operation of the driver suspension in its non-linear range. While these conditions may be characterized as "not a problem", there are other far superior design regimens available to the loudspeaker designer that will maximize performance of a given driver assemblage by proper signal filtering. In the case of most HF and MF units, if they are operated by a signal not filtered by, 12db/Oct minimum sloped filter, the manufacturer's warrantee will be voided.
Regards,
WHG
I am sick and tired when people knock something without being sensible! According to your stance every speaker system ever designed with first order xovers was doomed to failure and would destroy their drivers.
Hey, news flash! It didn't happen! Most were quite successful because the designers knew how to design with first order xovers.
What do you say to all the designers that use first order?
First order does have a slope to protect drivers, it is at 6dB per octave. Use the drivers wisely. Choose the drivers carefully!
The ability of 1st order xovers to be phase coherent is not a piddling traight. It is in fact the purest form of reproduction available if your requirement is for full bandwith coverage. The only other type of speaker that can do this remarkable feat is the full range drivers and they suffer reduced treble. Many choose this because they like the sound!
2nd order and higher are not phase coherent and therefore suffer from what I call time smear. In the xover region acoustic energy arrives at the ear from two drivers in slightly different timeframes causing the music to smear in the time domain. You can not do anything about it! I prefer my music unsmeared.
Either I am right or you are! There are many good examples of first order speakes out there so I think my position is rather safe!
Terry
But what would you say to the argument that time smear is inaudible and insignificant?
If it was so significant why dont high end manufacturers bother with them yet theire speakers are still regarded by many as some of the best. Eg. B&W, wilson audio etc. Even studios rely on these speakers.
In general I agree with your position, certainly on the point about intended driver usage and appropriate selection.
There are crossovers higher than first order that are also transient-perfect such as subtractive delay, but implemented passively that one can only work for a 2-way system. It suffers from the same issue as does a first order system (or any non-coincident one for that matter), the response degrades off-axis, more so I think for the subtractive delay.
In fact, it is now a rather easy task to create a multi-way system that has flat phase on the design axis. Programs such as SoundEasy implement the filters on a PC and even very high order crossovers can have flat response and flat phase. It does require that they be Linkwitz-Riley to have flat frequency response as well, but that's academic in this situation.
The one aspect that may make the first order the better option and that has no workaround is the power response. I tend to think that the power response for first order may be the best one for a multi-way system. But that's debatable just as are so many other considerations.
Suggested reading : Crossover Analyses
dlr
Last edited:
- Status
- Not open for further replies.