Qs about baffle diffraction

[ethan]

In doing my homework for building my speakers I have come upon Roman Bs article (http://www.rjbaudio.com/Audiofiles/t...fraction.html)
about baffle diffraction and had a question. I am looking at a 3 way design and I understand the baffle diffraction as he explained it, and the question is this.

If the baffle used in my speaks is the 9" (just used as an example) like in Roman's article, I will have to worry about baffle diffraction in the upper end of the midrange as well correct? From what I can tell with a 9" baffle, diffraction will begin to occur at right about 1500hz, correct? Unless the mid is crossed lower, which seems rather unlikely.

TIA
Ethan

[Pete Schumacher ®]

Provided Link: Baffle diffraction simulator


> In doing my homework for building my
> speakers I have come upon Roman Bs article
> (http://www.rjbaudio.com/Audiofiles/t...fraction.html)
> about baffle diffraction and had a question.
> I am looking at a 3 way design and I
> understand the baffle diffraction as he
> explained it, and the question is this.

> If the baffle used in my speaks is the
> 9" (just used as an example) like in
> Roman's article, I will have to worry about
> baffle diffraction in the upper end of the
> midrange as well correct? From what I can
> tell with a 9" baffle, diffraction will
> begin to occur at right about 1500hz,
> correct? Unless the mid is crossed lower,
> which seems rather unlikely.

> TIA
> Ethan

Baffle diffraction occurs across the spectrum. You'll notice a drop off in bass response, but you will also get some "ripple" in response above that which is mitigated by rounding the edges.

Download the baffle diffraction simulator to see what the effects will be for your design.

[hongrn]

I use John Murphy's formula, and diffraction for a 9" baffle starts at 503 Hz. For a full 6 dB baffle step compensation, assuming a 8 ohm system, the resistor is 8 ohms, and the inductor is 1.8 mH. May be the experts can chime in to see if I'm correct.

Hong

[cjd]

> I use John Murphy's formula, and diffraction
> for a 9" baffle starts at 503 Hz. For a
> full 6 dB baffle step compensation, assuming
> a 8 ohm system, the resistor is 8 ohms, and
> the inductor is 1.8 mH. May be the experts
> can chime in to see if I'm correct.

Confuse not baffle diffraction and baffle step.

You speak of baffle step.

Diffraction is, as someone else mentioned, something that occurs across the frequency spectrum. I pick up an interesting 1800Hz dip on my 14" wide (with 3/4" chamfer) 3-ways - felt alleviates this tremendously.

C

[hongrn]

I stand corrected. Read the question too fast. Thanks Chris.

Hong

[dlr]

Provided Link: Dave's Speaker Pages


The other response are all good. If you are interested, I have a couple of sections at my web site specifically on diffraction as well as an article in the June 2005 issue of audioXpress.

You will have a significant influence with a 9" baffle, both on the tweeter and the midrange. The woofer should be operating exclusively in 4-pi space (below the baffle step).

Diffraction of the tweeter response due to the midrange or woofer will also occur. This can be as severe as that of the baffle edges.

Just to clarify the post by cjd, baffle step is part of diffraction. Step is not just related, it is one of the aspects of diffraction.

dlr

[madrok]

> Confuse not baffle diffraction and baffle
> step.

The smooth baffle step shown on John Murphy's site applies to a spherical baffle. It turns into a spiky slide with a rectangular cabinet face. Diffraction occurs all across the spectrum. And sound waves wrapping around the cabinet edges causes the step. But it's more pronounced in certain frequency ranges than others.

[curt_c]

Below is the modeled diffraction plot of a woofer on an 8" x 20" baffle. Note that there is a significant ripple dip between 450 Hz and 3K. Not only is the woofer affected, but a midrange or tweeter on this baffle would be exposed to substantial response variations due to diffraction effects.

I used to assume that only the woofer and midrange was affected significantly by diffraction effects, but I have since found for flat mid/tweeter summation, baffle diffraction must be addressed.

C

[cjd]

Well, yes. Ultimately, it comes down to the same physics.

However, means of dealing with them differ as well as the impact (and obvious-ness to the listener). Baffle width will tell you the fundamental/harmonic points by nature of the distance(s) of course, and this relates to baffle step and diffraction alike.

I think I'm tired though, as my mind wandered off and forgot where it was going just now. Re-reading this post doesn't help. :P 12+ hours at work, got home in time to sleep and get up for work today...

C

[ethan]

Ahh... I see! Thanks much for the input and that link! There is so much to consider when designing... wow.

> Baffle diffraction occurs across the
> spectrum. You'll notice a drop off in bass
> response, but you will also get some
> "ripple" in response above that
> which is mitigated by rounding the edges.

> Download the baffle diffraction simulator to
> see what the effects will be for your
> design.

[madrok]

I don't see the tweeter being effected by that at all, as long as you have a xover above 2kHz.

There IS a relative 2dB peak by 800 Hz. And a 3dB dip by 250Hz. THAT's an issue. I guess it depends on how picky we are.

[dlr]

> I don't see the tweeter being effected by
> that at all, as long as you have a xover
> above 2kHz.

You can't judge the midrange or tweeter by that graph. It's for a woofer only with its directionality included. The midrange and tweeter will be far worse.

dlr