What's problematic about a high bookshelf crossover? - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 7Likes
  • 1 Post By lovinthehd
  • 1 Post By Taking Notes!
  • 1 Post By Bill Fitzmaurice
  • 1 Post By Bill Fitzmaurice
  • 2 Post By PrimeTime
  • 1 Post By Bill Fitzmaurice
 
Thread Tools
post #1 of 16 Old 01-05-2016, 02:32 PM - Thread Starter
AVS Forum Special Member
 
Sean Spamilton's Avatar
 
Join Date: Jul 2012
Location: Toronto, ON
Posts: 1,345
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 444 Post(s)
Liked: 546
What's problematic about a high bookshelf crossover?

Hey all.

I'm just curious about why exactly a bookshelf would have a very high crossover. It seems to me that most have a crossover of around 1800 - 3000hz, however it seems bookshelf speakers from Fluance use 6000+hz.
http://fluance.com/product/Signature....eng-169.html#

Their previous lines have been reviewed by Audioholics before, and although get fairly good reviews - most reviewers do take issue with the crossover being that high (though they don't elaborate).

It's funny, as they don't use such a high crossover for any of their other speakers - and the towers from their new line look quite interesting (90hz sensitivity / 8ohm / 3 way with two 8" woofers).
http://fluance.com/product/Signature...s.eng-168.html

What are the potential pratfalls exactly of using such an arrangement? I've contacted Fluance about it and received this in reply:

"Hi Sean,

We can see your point, but I can tell you that this new line has been in design for quite some time and a professional sound engineer had been employed for the entire design process.

The settings were found to provide the best synergistic effect from all the components and speakers within the Signature Line
."

I've actually had a set of earlier Fluance years ago - and found that their towers were really unbelievable for their price-point, but even back then their surrounds were the weak point.

With a 30 day free trial & free shipping both ways - there's not too much to lose, but I'd like to be better educated about the odd crossover design.

Cheers!
Sean.


Sean Spamilton is offline  
Sponsored Links
Advertisement
 
post #2 of 16 Old 01-05-2016, 02:36 PM
 
lovinthehd's Avatar
 
Join Date: Dec 2007
Location: OROR
Posts: 16,231
Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 4588 Post(s)
Liked: 4795
Depends on the drivers being used....
darthray likes this.
lovinthehd is offline  
post #3 of 16 Old 01-05-2016, 04:24 PM
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 15,491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 5770
A 6kHz crossover is fine if the midbass has an adequate dispersion angle that high, which means it would have to be a four incher or smaller. I run a 6khz crossover myself, with four inch midbasses.
Bill Fitzmaurice is offline  
Sponsored Links
Advertisement
 
post #4 of 16 Old 01-05-2016, 04:27 PM - Thread Starter
AVS Forum Special Member
 
Sean Spamilton's Avatar
 
Join Date: Jul 2012
Location: Toronto, ON
Posts: 1,345
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 444 Post(s)
Liked: 546
Quote:
Originally Posted by Bill Fitzmaurice View Post
A 6kHz crossover is fine if the midbass has an adequate dispersion angle that high, which means it would have to be a four incher or smaller. I run a 6khz crossover myself, with four inch midbasses.
Thanks Bill (and LovintheHD) . I won't question the mathematics of it, but is 4" the ceiling? In the case of those bookshelves they use a 5" driver...
Sean Spamilton is offline  
post #5 of 16 Old 01-05-2016, 04:42 PM
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 15,491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 5770
Quote:
Originally Posted by Sean Spamilton View Post
Thanks Bill (and LovintheHD) . I won't question the mathematics of it, but is 4" the ceiling? In the case of those bookshelves they use a 5" driver...
The general rule of thumb is not to run a driver much higher than where the actual cone diameter is more than one wavelength. A five inch driver has about a four inch cone, where one wavelength is 3400Hz, so a 6kHz crossover is pushing it.
Bill Fitzmaurice is offline  
post #6 of 16 Old 01-05-2016, 05:32 PM - Thread Starter
AVS Forum Special Member
 
Sean Spamilton's Avatar
 
Join Date: Jul 2012
Location: Toronto, ON
Posts: 1,345
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 444 Post(s)
Liked: 546
Quote:
Originally Posted by Bill Fitzmaurice View Post
The general rule of thumb is not to run a driver much higher than where the actual cone diameter is more than one wavelength. A five inch driver has about a four inch cone, where one wavelength is 3400Hz, so a 6kHz crossover is pushing it.
Hmm.... I'm trying to read up about it. Post 4 from this article about DIY speaker design states:

Quote:
A dome tweeter radiates over a wider angle near the crossover as the dome is smaller. If a crossover was set above a few kHz in this case, and we based it solely on the on-axis frequency response, there would be less total sound from the woofer just below the crossover. This could contribute to tonal problems and may cause the tweeter to stand out rather than blending in.

All the same, there are good reasons to use a woofer into this upper region where the design specifically calls for it, usually where a tweeter with controlled directional qualities is used.


It also states:

Quote:
You will firstly need to choose a crossover frequency that gets the best out of the drivers. If we choose too high a frequency, we risk poor sound from the woofer, and poor matching to the tweeter.
http://www.diyaudio.com/forums/multi...asurement.html

So basically the danger with a too high crossover (unless the driver is extremely directional) is that the tweeter stands out too much, and the woofer will have a dip in the frequency response below the crossover?

What I find odd is that assuming they are using the same driver (and it looks like it) for the midrange in the towers and as the woofers in the center, they have it crossed over at 2300. What could be the reasoning for (assuming the same tweeter) crossing it over at two wildly different frequencies?

Last edited by Sean Spamilton; 01-05-2016 at 05:35 PM.
Sean Spamilton is offline  
post #7 of 16 Old 01-05-2016, 08:21 PM
AVS Forum Special Member
 
Red99's Avatar
 
Join Date: Sep 2014
Location: Pacific Northwest
Posts: 1,289
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 776 Post(s)
Liked: 418
Because one is a 3 way and the other is a 2 way would be my guess.


Red99 is offline  
post #8 of 16 Old 01-06-2016, 06:38 AM
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 15,491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 5770
Quote:
Originally Posted by Sean Spamilton View Post
So basically the danger with a too high crossover (unless the driver is extremely directional) is that the tweeter stands out too much, and the woofer will have a dip in the frequency response below the crossover?
The issue will be in the off-axis response at the upper end of the midbass pass band. It won't be seen in an axial response chart.

Quote:
What I find odd is that assuming they are using the same driver (and it looks like it) for the midrange in the towers and as the woofers in the center, they have it crossed over at 2300. What could be the reasoning for (assuming the same tweeter) crossing it over at two wildly different frequencies?
An MTM center needs to cross over no higher than where the midbasses are 1 wavelength apart center to center. 2300Hz would allow them to be spaced about six inches apart CTC, and as that's clearly not the case it doesn't explain it. They could, and should have, spaced the midbasses much closer than they did, calling into question the statement a professional sound engineer had been employed for the entire design process. Needless to say said engineer wasn't Joe D'Appolito, or myself for that matter. When you see the midbasses so far apart it's a safe assumption that the basic design came out of the marketing department, where what it looks like is of far more concern than what it sounds like.
I can understand why the center crossover is as low as it is. While it's not as low as it should be, it's probably as low as that tweeter can go. Why one would cross that same tweeter and midbass at 6kHz in the bookshelf makes no sense to me.
Bill Fitzmaurice is offline  
post #9 of 16 Old 01-06-2016, 08:06 AM
Member
 
Taking Notes!'s Avatar
 
Join Date: Aug 2009
Posts: 167
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 92 Post(s)
Liked: 83
The nice thing about a 6k crossover is it is on the high side of the midrange our ears are most sensitive to.
It takes great skill to design a crossover that is audibly transparent in the midrange and things become a little more forgiving as you go up in frequency.

So in that sense a 6k crossover is great!

However, the point Bill is making is that a 5" cone driver asked to reproduce sound up to 6k will get "beamy". The sounds at those frequency will be more directed and you will not have room reflections in balance with those of the tweeter or lower frequencies from the woofer.

How objectionable it will be is hard to say. Much depends on your room/setup, and there may be some specifics on how Fluance addressed this issue which help to mitigate this issue.

That is the theoretical analysis.
We look forward to your empirical evaluation, if you choose to follow through.

PS - speaker designs always involve compromises. At $200/pair the compromises are more pronounced. I would not be too quick to discount this speaker as poor, but ideally would like to see off-axis frequency response graphs.
Sean Spamilton likes this.

Cheers,
Kurt (aka KEW)
Taking Notes! is offline  
post #10 of 16 Old 01-06-2016, 01:51 PM - Thread Starter
AVS Forum Special Member
 
Sean Spamilton's Avatar
 
Join Date: Jul 2012
Location: Toronto, ON
Posts: 1,345
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 444 Post(s)
Liked: 546
Quote:
Originally Posted by Bill Fitzmaurice View Post
The issue will be in the off-axis response at the upper end of the midbass pass band. It won't be seen in an axial response chart.

An MTM center needs to cross over no higher than where the midbasses are 1 wavelength apart center to center. 2300Hz would allow them to be spaced about six inches apart CTC, and as that's clearly not the case it doesn't explain it. They could, and should have, spaced the midbasses much closer than they did, calling into question the statement a professional sound engineer had been employed for the entire design process. Needless to say said engineer wasn't Joe D'Appolito, or myself for that matter. When you see the midbasses so far apart it's a safe assumption that the basic design came out of the marketing department, where what it looks like is of far more concern than what it sounds like.
I can understand why the center crossover is as low as it is. While it's not as low as it should be, it's probably as low as that tweeter can go. Why one would cross that same tweeter and midbass at 6kHz in the bookshelf makes no sense to me.
Quote:
Originally Posted by Taking Notes! View Post
The nice thing about a 6k crossover is it is on the high side of the midrange our ears are most sensitive to.
It takes great skill to design a crossover that is audibly transparent in the midrange and things become a little more forgiving as you go up in frequency.

So in that sense a 6k crossover is great!

However, the point Bill is making is that a 5" cone driver asked to reproduce sound up to 6k will get "beamy". The sounds at those frequency will be more directed and you will not have room reflections in balance with those of the tweeter or lower frequencies from the woofer.

How objectionable it will be is hard to say. Much depends on your room/setup, and there may be some specifics on how Fluance addressed this issue which help to mitigate this issue.

That is the theoretical analysis.
We look forward to your empirical evaluation, if you choose to follow through.

PS - speaker designs always involve compromises. At $200/pair the compromises are more pronounced. I would not be too quick to discount this speaker as poor, but ideally would like to see off-axis frequency response graphs.
Thanks for the fantastic information gentleman. While I would wholeheartedly expect $200 speakers to have to make compromises, if there are flagrant violations of speaker design 101 incorporated into them it's certainly something I was hoping to be made more aware of. I know that some fantastic engineers have managed to make magic out of seemingly little the last 10 years or so - (Dr. Hsu, Andrew Jones etc, Paul Barton to an extent); I suspect that a designer of that caliber wasn't employed here.

If I might be so bold, Bill, how exactly have you done the calculations here:

Quote:
An MTM center needs to cross over no higher than where the midbasses are 1 wavelength apart center to center. 2300Hz would allow them to be spaced about six inches apart CTC, and as that's clearly not the case it doesn't explain it.
I googled 'how long is a wavelength' and was treated to answers FAR outside my level of understanding. Is there a simplified formula? I.e. X/Y should be =< the size of the driver? and is a wavelength always the same measurement? I'd like to be able to look at an MTM arrangement going forward and be able to use your explanation to effect.

Thanks again!
Sean.
Sean Spamilton is offline  
post #11 of 16 Old 01-06-2016, 02:33 PM
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 15,491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 5770
Quote:
Originally Posted by Sean Spamilton View Post
I googled 'how long is a wavelength' and was treated to answers FAR outside my level of understanding. Is there a simplified formula?
Wavelength is the speed of sound divided by frequency. The speed of sound varies with everything from the altitude to the temperature and weather. I use 1130 feet per second, or 13500 inches per second, as a reasonable average. That makes 1 wavelength at 1kHz 1130/1000 for 1.13 feet or 13500/1000 for 13.5 inches.

1 wavelength center to center spacing is critical with a horizontal MTM, otherwise above that frequency the two drivers operate as independent sources, leading to comb filtering on the horizontal plane. With a vertical MTM spacing isn't critical. Combing will still occur, but on the vertical plane, where you're far less likely to hear it, unless your LP is very close to the speaker. This is why MTM is seldom used for nearfield monitors, which typically are listened to at distances of less than six feet.
Sean Spamilton likes this.
Bill Fitzmaurice is offline  
post #12 of 16 Old 01-06-2016, 03:17 PM - Thread Starter
AVS Forum Special Member
 
Sean Spamilton's Avatar
 
Join Date: Jul 2012
Location: Toronto, ON
Posts: 1,345
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 444 Post(s)
Liked: 546
Quote:
Originally Posted by Bill Fitzmaurice View Post
Wavelength is the speed of sound divided by frequency. The speed of sound varies with everything from the altitude to the temperature and weather. I use 1130 feet per second, or 13500 inches per second, as a reasonable average. That makes 1 wavelength at 1kHz 1130/1000 for 1.13 feet or 13500/1000 for 13.5 inches.

1 wavelength center to center spacing is critical with a horizontal MTM, otherwise above that frequency the two drivers operate as independent sources, leading to comb filtering on the horizontal plane. With a vertical MTM spacing isn't critical. Combing will still occur, but on the vertical plane, where you're far less likely to hear it, unless your LP is very close to the speaker. This is why MTM is seldom used for nearfield monitors, which typically are listened to at distances of less than six feet.
That's perfect. So looking at various crossovers for instance: (spdS / given crossover frequency)
2500hz should be no more than 0.452 feet ~5 1/2 inches apart center to center
1800hz should be no more than 0.623 or 7 1/2 inches apart center to center
1500hz should be no more than 9 inches apart etc etc...

How would that calculation change (or would you alter the calculation) if you moved the tweeter up above the center-line of the driver?

From Audioholics:
Quote:
Driver directionality is primarily a function of the relationship of the radiated wavelength to the effective piston diameter of the driver. Other factors can also affect directionality, such as cone shape, cone stiffness, breakup modes, among others, but the first determinant of directionality is driver size.) Bringing the dual midranges in much closer to each other minimizes the distance between their acoustic centers, for an effectively smaller radiating dimension, which maximizes their horizontal dispersion.
Can the same formula be applied? (I would assume so).

Cheers!
Sean.
Sean Spamilton is offline  
post #13 of 16 Old 01-06-2016, 06:59 PM
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 15,491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 5770
Quote:
Originally Posted by Sean Spamilton View Post
How would that calculation change (or would you alter the calculation) if you moved the tweeter up above the center-line of the driver?
If you mean placing the midbasses together with the tweeter above them the calc doesn't change. But doing so allows you to actually place the midbasses close enough together. With the tweeter between them it's almost impossible.
Sean Spamilton likes this.
Bill Fitzmaurice is offline  
post #14 of 16 Old 11-02-2016, 03:17 PM
AVS Forum Special Member
 
bluewizard's Avatar
 
Join Date: Jul 2008
Posts: 4,746
Mentioned: 38 Post(s)
Tagged: 0 Thread(s)
Quoted: 2319 Post(s)
Liked: 1531
As someone else said, the crossovers depend on the drivers chosen and the overall design.

Back in the old days when speakers looked like monkey coffins, they made true 3-way speakers, meaning Woofer, Mid-Range, and Tweeter. In a system like this, for a variety of reasons, the crossovers were likely 800hz and 5khz. As a starting point, most 3-way designs assume roughly 3 octaves in the midrange. So starting at 800hz, then next octave up is 1600, and the one after that is 3200hz, and the last octave would be at 6400hz. That's pretty high well outside the range of most normal instruments. So, it is typically dropped down into the range of 4500hz to 6000hz, with 5000hz to 5500hz being common.

That's not typically how they design 3-way speakers today. Today, many Tower speakers are 2-way speakers with an additional Low-Bass driver. So, they would be Low-Bass, Mid-Bass, and Tweeter. Just to illustrate the Low-bass/Mid-bass might cross at 300hz, and the Mid-bass/Tweeter might cross in the range of 2500hz to 3500hz. These new style speakers could be true 3-way or they may be 2.5-way. Meaning the Low-Bass and Mid-Bass share a section of the very low bass range.

However, there are still more traditional 3-way design being made.

So, the intend and the design philosophy will determine the crossover, in combination with the specific drivers chosen.

With small diameter bass drivers you have to make a trade off. If you want deep bass, then you lose midrange. If you want midrange, then you give up low bass to get it. In a 2.5-way or 3-way Low/Mid-Bass design the Low-Bass driver can be specifically chosen to go deep, while the Mid-Bass can be specifically chosen to reach up higher into the midrange. Thus making the overall design easier.

With in a certain perspective, the old school 3-way - Woofer, Mid-Range, Tweeter - designs were the easiest, because each driver is well within it working range. In new school 3-way - Low-Bass, Mid-Bass, Tweeter - matching components is more difficult, but still workable. The hardest design, relative to picking the drivers is a 2-way design. It is hard to find the right compromise between low bass and good midrange. But when a good 2-way design works, it usually works very well.

Steve/bluewizard

Last edited by bluewizard; 11-04-2016 at 03:53 PM.
bluewizard is offline  
post #15 of 16 Old 11-02-2016, 04:45 PM
AVS Forum Special Member
 
PrimeTime's Avatar
 
Join Date: Mar 2003
Location: Lower California
Posts: 3,247
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 800 Post(s)
Liked: 493
Quote:
Originally Posted by Taking Notes! View Post
However, the point Bill is making is that a 5" cone driver asked to reproduce sound up to 6k will get "beamy". The sounds at those frequency will be more directed and you will not have room reflections in balance with those of the tweeter or lower frequencies from the woofer.
Query: Isn't keeping the sound off your walls a good thing?

The whole "room reflections in balance" thing is quite controversial. This hi-tech Feng Shui unfortunately has flowed from the JBL folks to justify imposing Public Address Figures Of Merit to interior spaces. IMHO, it is of little merit. If you want to listen to your walls, get a Bose 901.

Bill F's point of the "one wavelength" rule of thumb is partially correct. It comes from the physics of cone movement vs. the speed of sound propagating along the cone surface from the voice coil. Actual observations show that this pistonic range of motion (where the entire cone moves together, in phase) is usually limited to the region below where the cone is roughly 1/2 wavelength in diameter (it can be a little higher depending on cone material). You can usually see it with the microphone placed inches from the speaker under test; the frequency response will show a "kink" where the initial cone breakup (non-pistonic operation) begins.

Linear operation in the so-called "breakup" region, aside from the beaming phenomenon, is not necessarily bad. It depends on a complex combination of other factors (cone mass, voice coil mass, suspension, suspension treatment) which determine whether the total output of the speaker is properly damped.
Jon Lane and BufordTJustice like this.
PrimeTime is offline  
post #16 of 16 Old 11-02-2016, 06:41 PM
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 15,491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 5770
Quote:
Originally Posted by PrimeTime View Post
Query: Isn't keeping the sound off your walls a good thing?
It's not a bad thing, but you can overdo it. Having a midbass with far less off-axis dispersion as the tweeter in the crossover region isn't such a good idea either. Dispersion is enough of an issue that the recommended practice for determining how high to run a driver is typically no higher than where response at 30 degrees off-axis is down no more than 6dB from axial response.
Taking Notes! likes this.
Bill Fitzmaurice is offline  
Sponsored Links
Advertisement
 
Reply Speakers

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off