AVS Forum banner
1 - 17 of 17 Posts

·
Registered
Joined
·
3,575 Posts
Discussion Starter · #1 ·
The starting point


We all understand that very few can have the ideal centre speaker, a clone of the L/R mains - like those with an AT screen, or coax speakers that can lay on their side, etc.


So given that we are going with a horizontal centre compromise, choosing a W-T/M-W dramatically reduces the lobing potential a M-T-M has.


OK, we have a 3-way centre speaker in a W-T/M-W configuration crossed over to a sub at 80Hz...

The question


- what would be the ideal crossover point for the woofers to the mid (as there will still be lobing potential in the crossover range)


- what would be the ideal crossover point for the mid to tweeter (is there an architecturally ideal one, or is this purely based on the mid and tweeter drivers chosen)
 

·
Registered
Joined
·
345 Posts
Is this a DIY speaker or are you comparing existing 3 way center channels? I think your idea of it being dependent on the speakers is right. Every crossover which is made right is made to the speakers based on their measurements. Basing it off of manufacturer's data will get close but sometimes the data isn't very consistent. Also I know a few designs which use identical drivers but completely different crossovers and cabinets to achieve different goals.
 

·
Registered
Joined
·
894 Posts
My two cents...you do not want bass to muddle your center...cross it over at 80 or higher if you can get away with it without localization of your sub...but for sure 80.


Even if your center is capable of more...center is for dialogue first..everything else second.. Just my .02 take it for what it is worth..which is probably less than .02
 

·
Registered
Joined
·
3,575 Posts
Discussion Starter · #4 ·

Quote:
Originally Posted by psycholis  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23200812


Is this a DIY speaker or are you comparing existing 3 way center channels?

No, it is to understand the science better to be better able to evaluate existing designs and/or make suggestions for a custom project.


Certain frequencies are more susceptible to lobing, presumably if the W-to-M crossover is below this the design would have less (none?) issues.


For the two identical drivers (whether MTM or WTMW) there is a relationship between 1/2 wavelength and centre to centre of the drivers. How far below that frequency is the highest frequency you wnat those drivers to play?



There are very knowledgeable speaker designers on this forum. I am hoping some of them can better explain the science in this area to contribute to all of our education. That way we can make more informed decisions about our compromise centre speakers (which most of us need to use).


That's one of the main reasons we are all here isn't it.
 

·
Registered
Joined
·
345 Posts
I don't have the answers you're looking for but I know what the sort of default answer would be. The easiest solution would be to have 3 identical speakers across the front. Regardless of what configuration or setup a center channel has, it will still be different from a main speaker. I guess the question then becomes - is that difference noticable to be distracting during multi-channel or movie listening? I don't own any multi-channel music but my MTM center hasn't ever bothered me during dialogue sections of movies.


In a way I want the answer to your questions too because I'm currently brainstorming a DIY design that uses a D'appolito configuration in a tower. A few manufacturers have them (Polk, Aperion, EMP, Salk) and I kind of like how it looks compared to a single midrange.
 

·
Registered
Joined
·
638 Posts

Quote:
Originally Posted by rick240  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23200492

The starting point


We all understand that very few can have the ideal centre speaker, a clone of the L/R mains - like those with an AT screen, or coax speakers that can lay on their side, etc.


So given that we are going with a horizontal centre compromise, choosing a W-T/M-W dramatically reduces the lobing potential a M-T-M has.


OK, we have a 3-way centre speaker in a W-T/M-W configuration crossed over to a sub at 80Hz...

The question


- what would be the ideal crossover point for the woofers to the mid (as there will still be lobing potential in the crossover range)


- what would be the ideal crossover point for the mid to tweeter (is there an architecturally ideal one, or is this purely based on the mid and tweeter drivers chosen)

I have no experience in crossover built and design. But i would say the best crossover points depends on the drivers used. Alot depends on the off axis dispersion of the drivers.

I agree, there must be knowledge people here that can answer but it might be too trouble and complicated to explained.
 

·
Banned
Joined
·
15,491 Posts

Quote:
Originally Posted by rick240  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23200492



OK, we have a 3-way centre speaker in a W-T/M-W configuration crossed over to a sub at 80Hz...

- what would be the ideal crossover point for the woofers to the mid (as there will still be lobing potential in the crossover range)

- what would be the ideal crossover point for the mid to tweeter (is there an architecturally ideal one, or is this purely based on the mid and tweeter drivers chosen)
In the case of a horizontal speaker you have to consider not only the off-axis response but also the driver center-to-center distance, to prevent comb-filtering. That distance must be no more than 1 wavelength at the crossover frequency, while the two woofers must also be no more than 1 wavelength apart CTC within their passband. At the same time the drivers should be crossed over so that their off-axis response never dips by more than -6dB at 30 degrees. Achieving all of these requirements isn't easy, and very few centers actually do so.
 

·
Registered
Joined
·
3,575 Posts
Discussion Starter · #8 ·

Quote:
Originally Posted by Bill Fitzmaurice  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23203191


In the case of a horizontal speaker you have to consider not only the off-axis response but also the driver center-to-center distance, to prevent comb-filtering. That distance must be no more than 1 wavelength at the crossover frequency, while the two woofers must also be no more than 1 wavelength apart CTC within their passband.

So, since the two woofers are within their passband at the crossover to the midrange their minimum CTC distance likely dictates the highest possible W->M crossover point (as it will be a greater distance than the CTC of woofer to mid).


Am I understanding this all right?
Quote:
At the same time the drivers should be crossed over so that their off-axis response never dips by more than -6dB at 30 degrees.

What sort of things would impact this?
Quote:
Achieving all of these requirements isn't easy, and very few centers actually do so.

No kidding - a lot to achieve.




PS. Thanks for the lesson.
 

·
Registered
Joined
·
638 Posts

Quote:
Originally Posted by Bill Fitzmaurice  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23203191


In the case of a horizontal speaker you have to consider not only the off-axis response but also the driver center-to-center distance, to prevent comb-filtering. That distance must be no more than 1 wavelength at the crossover frequency, while the two woofers must also be no more than 1 wavelength apart CTC within their passband. At the same time the drivers should be crossed over so that their off-axis response never dips by more than -6dB at 30 degrees. Achieving all of these requirements isn't easy, and very few centers actually do so.

Thank you Bill for taking the time.
 

·
Registered
Joined
·
638 Posts

Quote:
Originally Posted by rick240  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23203363


So, since the two woofers are within their passband at the crossover to the midrange their minimum CTC distance likely dictates the highest possible W->M crossover point (as it will be a greater distance than the CTC of woofer to mid).


Am I understanding this all right?

What sort of things would impact this?

No kidding - a lot to achieve.




PS. Thanks for the lesson.

I did a couple online calculation of different xover point and even at much higher ( 700 hz) , it seems that the wave length is long enoug to not cause any com filtering. No sure if is right or wrong.

For example, 700hz have a wave length of 1.6 feet. That distance should be long enough to have acomodate a tweeter/ mid and two 7 in woofer. In order words, the distance between woofer CTC should still fall within 1.6 feet.

Not sure my calculation is right or not.
 

·
Registered
Joined
·
16,930 Posts
I can't answer specifically.


But the problem with standard MTMs is that the midrange frequencies, particularly, - which, since dialog is the most prevalent content, are obviously essential to good center channel performance - are the most affected by the lobing/comb filtering that occurs with the midwoofer to midwoofer distances encountered in most MTMs. Putting all these frequencies in a single dedicated midrange driver sidesteps the issue. There still may be issues with the lower frequencies in the woofers, but they'd be less noticeable in that lower frequency range than the more important midrange frequencies.


As far as issues around the crossover point between the midrange driver and the adjacent woofers, you have to rely upon the manufacturer to have at least provided a reasonably designed speaker.


The crossover between the mid and tweeter is dictated by the design and drivers used and is not going to suffer from any issues that can really be directly attributed to the distances between the drivers and the crossover used.


As sort of an aside, do not dismiss 2.5-way MTMs, where only one midwoofer operates through the critical midrange frequencies. This also sidesteps the midrange lobing/comb filtering issue.


edit: I know you know all of this, probably as well as if not better than I do. I guess I am just trying to say that once you've addressed the main issue with MTMs by going to a dedicated midrange driver, the other stuff is not so much of a concern. Do you have any specific models in mind?
 

·
Registered
Joined
·
638 Posts

Quote:
Originally Posted by sivadselim  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23204657


I can't answer specifically.


But the problem with standard MTMs is that the midrange frequencies, particularly, - which, since dialog is the most prevalent content, are obviously essential to good center channel performance - are the most affected by the lobing/comb filtering that occurs with the midwoofer to midwoofer distances encountered in most MTMs. Putting all these frequencies in a single dedicated midrange driver sidesteps the issue. There still may be issues with the lower frequencies in the woofers, but they'd be less noticeable in that lower frequency range than the more important midrange frequencies.


As far as issues around the crossover point between the midrange driver and the adjacent woofers, you have to rely upon the manufacturer to have at least provided a reasonably designed speaker.


The crossover between the mid and tweeter is dictated by the design and drivers used and is not going to suffer from any issues that can really be directly attributed to the distances between the drivers and the crossover used.


As sort of an aside, do not dismiss 2.5-way MTMs, where only one midwoofer operates through the critical midrange frequencies. This also sidesteps the midrange lobing/comb filtering issue.


edit: I know you know all of this, probably as well as if not better than I do. I guess I am just trying to say that once you've addressed the main issue with MTMs by going to a dedicated midrange driver, the other stuff is not so much of a concern. Do you have any specific models in mind?

Yes, Rick is refering to a wt/mw set up.

According to Bill, it seems that the distance between tweeter center to midrange still matters and is best the distance to be shorter then the crossover wave length. I dont have any experience with crossover design but i guess that could relate to constructive or instructive comb filtering?
 

·
Registered
Joined
·
3,575 Posts
Discussion Starter · #13 ·

Quote:
Originally Posted by RicardoJoa  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23204753


it seems that the distance between tweeter center to midrange still matters and is best the distance to be shorter then the crossover wave length.

These shouldn't be an issue as they are vertical - they would of course be subject to vertical comb filtering though.
 

·
Registered
Joined
·
3,575 Posts
Discussion Starter · #14 ·

Quote:
Originally Posted by sivadselim  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23204657


I can't answer specifically.


But the problem with standard MTMs is that the midrange frequencies, particularly, - which, since dialog is the most prevalent content, are obviously essential to good center channel performance - are the most affected by the lobing/comb filtering that occurs with the midwoofer to midwoofer distances encountered in most MTMs. Putting all these frequencies in a single dedicated midrange driver sidesteps the issue. There still may be issues with the lower frequencies in the woofers, but they'd be less noticeable in that lower frequency range than the more important midrange frequencies.

This is what I gathered from reading threads over the past couple of years.


What I wasn't clear on was the CTC guidelines for best performance. It seems that
 

·
Banned
Joined
·
15,491 Posts

Quote:
Originally Posted by sivadselim  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23204657


The crossover between the mid and tweeter is dictated by the design and drivers used and is not going to suffer from any issues that can really be directly attributed to the distances between the drivers and the crossover used.
Both the distances between the drivers and the driver off-axis responses are critical in determining the correct crossover frequency. If there is a common failing with centers it's not adhering to the basic rules where these factors are concerned. Some centers are so poorly designed that one can only assume that whoever designed them had no clue what they were doing, and must have believed that 'wavelength' is a term only used by surfers.
 

·
Registered
Joined
·
3,575 Posts
Discussion Starter · #16 ·

Quote:
Originally Posted by Bill Fitzmaurice  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23205164


and the driver off-axis responses are critical in determining the correct crossover frequency

Is this part of the published specs for a driver?

Edit: of course it is, just not always labelled in a legend - d'oh!


So presumably one wants to aim for the section of FR where both the on-axis and 30 degree lines are coincident - right?


Does it impact both crossovers (W-M and M-T)?


What considerations lead to a "correct" crossover choice?


- Thanks
 

·
Registered
Joined
·
16,930 Posts

Quote:
Originally Posted by Bill Fitzmaurice  /t/1468135/what-are-ideal-crossovers-for-a-w-t-m-w-centre-speaker#post_23205164


Both the distances between the drivers and the driver off-axis responses are critical in determining the correct crossover frequency. If there is a common failing with centers it's not adhering to the basic rules where these factors are concerned. Some centers are so poorly designed that one can only assume that whoever designed them had no clue what they were doing, and must have believed that 'wavelength' is a term only used by surfers.
Right. I was specifically addressing an over/under mid/tweeter arrangement. And I was trying (albeit clumsily) to say that, generally, all the variables involved in a speaker's design are already considered (or should be) when choosing driver distances, crossover parameters, etc.. Yes, it CAN all be screwed up, royally. But there are not invariable and inevitable issues such as, for example, the lobing/comb filtering that occur with a horizontal MTM design.
 
1 - 17 of 17 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top