How Does One Determine The Frequency By Which Diffraction Loss Will Occur?

[ReDress]

Hey guys,

I was reading a part of the white paper for the Grimm Audio LS1 and stumbled upon a part in which they began to talk about how they desired something like a constant directivity style of response above 300hz, and then have it transition to Omni - leading to their use of wide baffles

But I was curious, how does one determine the frequency by which the diffraction loss will begin to occur, relative to the baffle size; essentially, what size will lead to what frequency (and any other such information upon this topic would be greatly appreciated as well)

(Side question: The benefits of small baffles are "better" imaging, yet I do not understand the benefits of large baffles (besides pushing the diffraction loss further down - how might that be beneficial when compared to the traditional small baffle?)

But I digress, your guys' help and responses are all greatly appreciated

Much love/Many thanks in advance, everyone
(Hopefully my question makes sense to you guys...)

 

[AllenB]

The frequency by which the diffraction loss

Think of wavelengths comparable to the baffle, eg driver to edge 1/4wl before sound is diffracted (but not lost).

The benefits of small baffles are "better" imaging

I've never understood this and I don't believe it. I suspect people say this because there is more room energy. It would have to be frequency selective energy.

The best imaging I've heard was from the largest speakers I've built.

the benefits of large baffles (besides pushing the diffraction loss further down - how might that be beneficial when compared to the traditional small baffle?)

If one is going to allow low mids to spread further and try to compensate by turning them up it may not sound the same. Of course many of us do it all the time. There is a band above, maybe 500-700Hz where things should be kept cleaner (wrt diffraction).

 

[Michael Bean]

Hey guys,

I was reading a part of the white paper for the Grimm Audio LS1 and stumbled upon a part in which they began to talk about how they desired something like a constant directivity style of response above 300hz, and then have it transition to Omni - leading to their use of wide baffles

But I was curious, how does one determine the frequency by which the diffraction loss will begin to occur, relative to the baffle size; essentially, what size will lead to what frequency (and any other such information upon this topic would be greatly appreciated as well)

(Side question: The benefits of small baffles are "better" imaging, yet I do not understand the benefits of large baffles (besides pushing the diffraction loss further down - how might that be beneficial when compared to the traditional small baffle?)

But I digress, your guys' help and responses are all greatly appreciated

Much love/Many thanks in advance, everyone
(Hopefully my question makes sense to you guys...)

I use this : Tolvan Data

Mike

 

[planet10]

A couple articles on BSC start here: Baffle Diffraction Step

The 2nd one has a 1st order approximation of where it occurs.

dave

 

[CharlieLaub]

According to THIS page from True Audio, the following relationship(s) can be approximated from Olsons' work:

f(3) = 115/W(B)
(where W(B) is the baffle width in meters)

or

f(3) = 380/W(B)
(where W(B) is the baffle width in feet)

 

[ReDress]

Thanks for the responses thus far everyone - they've been interesting and helpful

As an extension of my question - BSC requires a padding down of x-amount of db (Is it usually safe to say around 6db for narrow baffles?), as such, how does one determine the amount padding (in terms of db) required based on the baffle size? Such as large baffles (I presume they need less padding down due to their size, but again, how does one figure out the amount of reduction necessary based on size?)

Your guys' continued help and comments are greatly appreciated and welcome.

 

[Michael Bean]

Thanks for the responses thus far everyone - they've been interesting and helpful

As an extension of my question - BSC requires a padding down of x-amount of db (Is it usually safe to say around 6db for narrow baffles?), as such, how does one determine the amount padding (in terms of db) required based on the baffle size? Such as large baffles (I presume they need less padding down due to their size, but again, how does one figure out the amount of reduction necessary based on size?)

Your guys' continued help and comments are greatly appreciated and welcome.

Did you look at the link I provided for Tolvan's "The Edge" program? It will show you the frequency response aberrations caused by diffraction, and recommend passive and active filters to fix it. It works very well. The size of the baffle doesn't affect the magnitude of the diffraction, only the frequency(s) at which it occurs.

Mike

 

[ReDress]

Can't download any software on this computer (so my apologies in advance for not having directly taken a definitive look - I'm just a lad, and as such, parents wouldn't be too pleased with me downloading software onto the main comp and don't really have many other resources to use/download onto)

 

[planet10]

BSC requires a padding down of x-amount of db (Is it usually safe to say around 6db for narrow baffles?), as such, how does one determine the amount padding (in terms of db) required based on the baffle size?

Baffle size only gives you an approximation of where. The amount is dependent on many things. Room. Speaker placement in the room. Taste. If the speaker has gain in the bass (a good example being many back loaded horns where the output from the horn is up in level compared to the direct radiation higher up).

dave