BSC discussion - split from Usher two way thread

[ch83575]

I know that I am going against the grain for this forum, but I personaly think that we should not fault a disign for having little or no BSC. I have been playing the level of BSC in my design (which uses the 9930 that PE has on sale) and I am not convinced one way or the other about its necessity. I started looking at BSC more criticaly after seeing a pair of Linn speakers taken apart with crossover exposed... no BSC at all! And these are speakers I respect very much (my refrence speakers are Katans). I realize that they are a speaker that are not for everybody (probably because of the lack of BSC) but they also have virtues that I do not hear very often in comperably priced speakers (also probably due to the lack of BSC).

I dont mean to start a BSC war here, I just wanted to point out that Dr. D'Appolito is fully aware of the issue and chose to handle it the way that he did. Also, I would not be suprized if the 1.8mH did provide a dB or 2 of BSC and the quoted SPL is just optimistic. Or perhapse the driver is more sensitive than we think, the manufacturer quotes it at 87dB, and Zaph's SPL is calculated with his TS params, not measured.

Anyways, I think this kit looks like a real winner, even if we are just looking at what you get for the price!

-Chad

[sreten]

Hi,

All I'm saying is things like this should be spelt out in the design.
PEs frequency response graphs IMO leave a lot to be desired.
I'm also saying it won't sound anything like Zaphs designs.
The limited attenuation of the tweeter also indicates no BSC.

I really do not like misleading frequency response graphs.




/sreten.

[ch83575]

Originally posted by sreten
I'm also saying it won't sound anything like Zaphs designs.

For sure! I cant decide how much BSC I like in my design (I am not even concidering full correction, but I think there is something strangely compelling about none) but I would never imply that it isnt a major design decision and drasticaly changes the sound of the speaker system. This design obviously has either very little or no BSC and would sound very different from Zaph's. I was actually suprized when I switched my tweeter from the H1212 to the 9930 by how much less efficient the Usher is. So, it doens't have very much padding, but it doesn't have much efficency either.

I also totaly agree about PEs design documentation. I would guess that this is just like a Stereophile FR graph where the near-field is merged to the far without any baffle effects added in. Even if this design did have a dB or 2 of BSC the graph still wouldn't look anything like that in free-space. They are obviously trying to sell their kits, but they are selling them to a different crowd than Wilson Audio is!

I really wish I had the $ to build this kit as-is and Zaph's L18 design. It would be like polar oposites I expect. And I am sure there are those that would prefer one and those that would like the other. I am pretty sure which I would like better .

-Chad

[Zaph]

Wooooo BSC war!

Most of my designs have near-full BSC. It's a design decision based on the fact that I don't design for compromised speaker setup. In other words, I don't design for tiny rooms, placement up against a wall, or in a bookshelf.

With Joe's 1.8mH inductor, there is some BSC there, but not as much as I might do. This isn't a design flaw, it's just choice. Joe is wise in attempting to cover a few more placement possibilities than I might. As a result, in some larger rooms, his Usher system might sound a small bit leaner in the low end. Rest assured that it probably still sounds great however because the Ushers are great drivers and Joe is a good designer.

BTW, beware of the "calculated" SPL on my T/S parameter sheets. They are only calculated, and as such they are not accurate. The more accurate level is on the frequency response curve.

[ch83575]

Zaph, you get discussed a lot on these boards but I never actually think about you sitting and reading what we are saying about you and your work (note to self: try not to bad-mouth Zaph as much ).

Have you ever tested the 9930? What do you think of the other design choices Joe made with this design? Is the 9930 totaly up to the 2000 crossover? Because it looks the 8945A could go a little higher without the major dip/peak getting too close to the crossover point. It just seems like the 9930 would be hard pressed to have lower distortion or sound less strained than the woofer with that crossover point, and the woofers breakup seems rather benign in terms of distortion. What are other peoples feelings on the 2k crossover?

Oh... and just to play devils-advocate Linn would say that close wall placement is not a "compromise". All rooms have walls regardless of size, not everybody has enough room to place their speakers 2-3ft from the walls.

-Chad

[TerryO]

Originally posted by Zaph
Wooooo BSC war!

Most of my designs have near-full BSC. It's a design decision based on the fact that I don't design for compromised speaker setup. In other words, I don't design for tiny rooms, placement up against a wall, or in a bookshelf.

Zaph,

I'd be interested in what you consider near-full BSC. Some of the locals here in the Northwest have had discussions on that very topic. I realize that there is a difference between designing for a particular application or a specific room and a more generic approach that should work in "most" situations.
I tend to believe that 3dB works in my somewhat small room, Dan Wiggins' designs tend toward a more generic 4dB, with cabin or room gain doing the rest. Others have opted for a full 6dB.
What say you?

Best Regards,
TerryO

[Zaph]

Originally posted by TerryO
I'd be interested in what you consider near-full BSC.

Near full BSC could be anything from 4 to 6 dB.

A lot of factors contribute to what I select for BSC, including low end response shaping based on the selected tuning, midbass and bass distortion levels, and more, on top of the normal room size and speaker placement options.

This may seem out of character coming from "Mr. Objectivity" but in addition to the considerations listed above, I often let my ears decide what amount of BSC sounds best for a given design. The right amount can be hard to model or predict due to so many variables, so a good listening test helps. I've never done any less than 4dB for a speaker that wasn't designed to be placed against the wall or used very nearfield.

[Zaph]

Originally posted by ch83575
Have you ever tested the 9930? What do you think of the other design choices Joe made with this design? Is the 9930 totaly up to the 2000 crossover? Because it looks the 8945A could go a little higher without the major dip/peak getting too close to the crossover point. It just seems like the 9930 would be hard pressed to have lower distortion or sound less strained than the woofer with that crossover point, and the woofers breakup seems rather benign in terms of distortion. What are other peoples feelings on the 2k crossover?

Oh... and just to play devils-advocate Linn would say that close wall placement is not a "compromise". All rooms have walls regardless of size, not everybody has enough room to place their speakers 2-3ft from the walls.

I've never tested any Usher tweeters, but I hope to change that someday. (unless we count the Dayton RS28, which is an Usher product)

I do call close wall placement a compromise, at least when done with normal rectangular shaped boxes. It moves the first rear wall reflection higher in frequency where it's more audible, and the response curve aberrations caused by the early reflection have less "room diffusion" to smooth out the percieved sound. Bottom line, up against a wall sounds worse than a few feet out. I don't know, I suppose it could just be called a placement limitation rather than a compromise.

In-wall speakers have the potential to be less of a compromise than a box speaker placed close to the wall, though of course they have their own set of compromises.

[ucla88]

Another reason the baffle step issue is problematic is that the on axis spl drop/rise is 6 dB, but the power response drop/rise is only 3 dB

See JohnK's discussion here

http://www.musicanddesign.com/PowerMatching.html

Depending how far away you listen to your speakers, and what the room is like, and what your speaker is in relation to the boundry...

So there is no clear cut answer. I think you have to target between 3-6dB. Closer to 6 if your speakers are out from the walls and you're "close" to the speakers. Closer to 3dB if you're well beyond the critical distance and or the speakers are close to the bounary/walls.

[ch83575]

Originally posted by Zaph


I do call close wall placement a compromise, at least when done with normal rectangular shaped boxes. It moves the first rear wall reflection higher in frequency where it's more audible, and the response curve aberrations caused by the early reflection have less "room diffusion" to smooth out the percieved sound. Bottom line, up against a wall sounds worse than a few feet out. I don't know, I suppose it could just be called a placement limitation rather than a compromise.

I guess you are right. If you choose to make a speaker sound ITS best in any room as opposed to sounding the best it possibly can, but only in some rooms that can acomodate the proper setup... well that is the definition of a compermise isn't it. It seems to me more and more that this hoby is nothing but one compermise after another and balancing what you will and will not compermise with is the real art of speaker designing .

I also feel that the choice of BSC can greatly effect the pace and timing of the speaker. I know there are those (Zaph included) that dont put much value in such things, but there are a lot of people (more in England than US) that think of it as one of the most important factors in music reproduction. I feel from my own listening experiments that the more BSC you introduce the less the transients line up with the bass. This causes the speaker to have what I can only describe as poor timing. Linn has no BSC and the best timing I have heard, but I am not absolutly enthraled with their voicing and sonic realism with orchestral music. I don't have anything but my ears to back this up with though, so I dont expect to be given much heed. Perhapse there is some other brave soul out there who can help me find a reason for what I am hearing that could be tested and illustrated with a chart or graph.

-Chad