Baffle Difraction &

[johninCR]

You're talking about OB and line arrays which significantly affect the acoustical loading of the drivers, baffle step, imaging, and sound stage. If you want better imaging then you need to look into power tapering the lines.

As far as reaching 250hz as a dipole bass roll off starting point only a minimal construction will get you there and the "visual width rule" is simply wrong. Swept back "wings" add more distance for the rear wave to travel to your ears using the same amount of wood so actually the opposite is true.

 

[ThorstenL]

Konnichiwa,

if I'm interpreting correctly, the wider baffle apparently gives the soundstage more depth and solidity (but less width).

Yes, even more so with soffit mounted, controlled directivity studio monitors (I mean real monitors, not Yamaha NS-10 or BBC LS-3/5).

What narrow baffles do is to make sure that where the drivers have a wide dispersion a lot of sound is radiated to the side walls. This will invariably lead to an overly wide Image (even with recording having a nartrow soundscape), due to reflections.

Sayonara

 

[ThorstenL]

Konnichiwa,

Swept back "wings" add more distance for the rear wave to travel to your ears using the same amount of wood so actually the opposite is true.

Actually assuming the Speaker is in a 2Pi envoironment the rear wave never reaches the listener, so what does that make your view?

Sayonara

 

[Paul W]

Test Mules

It is interesting to spend time with Paul V's (FRD consortium) BDS spreadsheet. Ignoring the dipole backwave for the moment, wider baffles including large-radius roundovers provide much smoother midrange and treble response.

For example, my current baffles are "visually" 19 inches wide with 45 degree wings. Current plans (developed with the help of BDS) call for new baffles 25" wide, including 6" radius roundovers. Response simulations are much smoother with the new baffle design. My "dipoles" begin a transition to monopole at about 100-150Hz and are 100% monopole above 800Hz, so YMMV.
Paul

 

[johninCR]

That doesn't do anything to my view. I understand that the narrower the front baffle, the higher the frequency where half space radiation starts. This point is not determined by visual baffle width. It is determined by the front baffle dimensions and shape. I guess our difference in view point is because we are talking about 2 different things with regard to effective baffle width on an OB speaker. I'm thinking about dipole bass roll off, while you are talking about baffle step. KYW, you are only leading to confusion when you talk about -6db @ 50hz with a 24" wide baffle and change the topic to half space radiation.

Regarding the 20" minimum baffle width, since a singer's voice and few instruments propagate their sound from a 20" or wider flat surface, how can that be the optimum way to reproduce the sound with our speakers? It may get closer to how the mixing engineer hears it, but that isn't necessarily optimum.

I would also contend that the narrower baffle would actually decrease the magnitude of side wall reflections that reach the listener because more of the frequency range is allowed to wrap around the sides of the baffle instead of unnaturally being forced into half space.

 

[Paul W]

Yup...two related, but different, subjects are being discussed. Directionally, my response was to Ralph's original question.

Reading the BDS documentation, you can see that overall baffle width (visual width) does impact diffraction...and only with sharp edges is it a transition "point".

 

[johninCR]

Thanks Paul W,

My response wasn't directed at your post and agree with your points about roundovers. I'd think that a trapezoid shape combined with the big radius roundovers would make it even smoother. With OB's and a flat baffle, are baffle edge diffraction and baffle step still important to consider at all? I can visualize how they would be considerations if you have wings, but with a flat baffle I'm not so sure.

BTW, how are you making your big radius round overs? I saw someone use very closely spaced saw cuts not completely through used on MDF used to make nice smooth 4" radius corners. I'm thinking about a similar approach to make some horn flares. Have you come up with something even easier.

 

[ThorstenL]

Konnichiwa,

I'm thinking about dipole bass roll off, while you are talking about baffle step.

But they are two sides of the same coin.

KYW, you are only leading to confusion when you talk about -6db @ 50hz with a 24" wide baffle and change the topic to half space radiation.

I di not change the topic to halve space, I merely wanted to help you think straight. The -6db@ 50Hz are results from both mathematical analysis and measurements.

Regarding the 20" minimum baffle width, since a singer's voice and few instruments propagate their sound from a 20" or wider flat surface, how can that be the optimum way to reproduce the sound with our speakers?

First, no-one said it was the optimum, it is the MINIMUM, secondly, we are talking re-production, not production. The sound has already been produced and recorded.

I would also contend that the narrower baffle would actually decrease the magnitude of side wall reflections that reach the listener because more of the frequency range is allowed to wrap around the sides of the baffle instead of unnaturally being forced into half space.

Oh, so basically RADIATING MORE sound towards the sides REDUCES sidewall reflections. I am sure there is some logic underpinning that, but it escapes.

I think on the subject of acoustics you are by far to biased by a view how things "should be" in your view to really get what is going on. Just read the sig.

Sayonara

 

[johninCR]

Konnichiwa,
But they are two sides of the same coin.

No, 2 different coins that are related.

I di not change the topic to halve space, I merely wanted to help you think straight. The -6db@ 50Hz are results from both mathematical analysis and measurements.

I am thinking straight. You aren't getting -6db@50hz because your 24" baffle is radiating into half space at that frequency. It's not.

First, no-one said it was the optimum, it is the MINIMUM, secondly, we are talking re-production, not production. The sound has already been produced and recorded.

I used the word MINIMUM in my post.

Oh, so basically RADIATING MORE sound towards the sides REDUCES sidewall reflections. I am sure there is some logic underpinning that, but it escapes.

It escapes because you are the one who needs to take another look at your sig. With a narrower baffle a higher percentage of the spectrum is wrapping around the baffle and hitting the side walls at perpendicular angles and even at angles behind the plane of the baffle. These reflections will have to bounce off of walls more times and travel more distance to reach your ears, so they will be lower in magnitude when they reach your ears. With the wider front baffle, a higher percentage of the frequency spectrum retains the half space radiation pattern and have a more direct route of reflection to get to your ears, so they are higher in magnitude.

I think on the subject of acoustics you are by far to biased by a view how things "should be" in your view to really get what is going on. Just read the sig.

Someone who is obviously intelligent and knowledgeable but has a closed mind has no place putting something like that in their sig. Just because you'll never admit when YOU "don't get it" doesn't give you the right to talk down to other people. I'm just here to learn. I don't pretend to know everything or anywhere close. When I see something I think is wrong I try to discuss it until I understand why I'm wrong. Simple geometry proves your "visual width rule" is wrong as related to wings and bass extension for an OB speaker and now I've shown that the reflected energy of your wide baffle speakers is stronger when it reaches your ears. I doubt that you will admit either and you're likely to not respond or will try to twist the topic again.

Note that I'm not saying that wide baffle speakers don't sound better than narrow ones. That may very well be the case. If so, I'd like to learn why since that's not how real world sounds propagate.

 

[Paul W]

On a finite baffle, BED clearly comes into play on the top end where most of us use monopole drivers. As KYW mentions, real-world dynamic drivers (with magnets, frames, etc) only approximate dipoles in the midrange...so the actual in-room result can be a very-well-mixed bag!

I just try to model as much as I can and then leap from there.


Here is a source for large radius quarter and half-rounds. I have not ordered from them yet, so personally cannot speak to either quality or service. (the plywood parts are rumored to be "inconsistent"...MDF quarters may be much better).

anderson roundovers

I have (in hand) 3" radius MDF quarter rounds from Rockler...these are good quality and Rockler service was excellent.

 

[ThorstenL]

Konnichiwa,

I am thinking straight. You aren't getting -6db@50hz because your 24" baffle is radiating into half space at that frequency. It's not.

I never suggested there was any link. I suggested that you consider, for clarity, the case of a dipole in halve space, in other words on the floor.

You seem adamant on insiting on your views, to which you are welcome and unwilling to consider things as they are. That is perhaps sad, but seems unchangable.

I shall not try to change your views any more, I provided you with enough clues, if you still don't get it, you don't.

Sayonara

 

[johninCR]

Konnichiwa,

I never suggested there was any link. I suggested that you consider, for clarity, the case of a dipole in halve space, in other words on the floor.

You seem adamant on insiting on your views, to which you are welcome and unwilling to consider things as they are. That is perhaps sad, but seems unchangable.

I shall not try to change your views any more, I provided you with enough clues, if you still don't get it, you don't.

Sayonara

Exactly as I thought, no response.

 

[wintermute]

Did I miss something?

OK has all of the discussion about baffle width and imaging been to do with OB, or has it been general (ie applicable to sealed or other alignments)??

I just went back a checked, and the line array that was in the first post was clearly (from the linked picture) not an open baffle, and this was the speaker which Ralph was specifically asking about the imaging and soundstage....

He then went on to say he was thinking of making an Open dipole (I asume) line array and asked about wings......

I was always under the impression that OB needed wider baffles than non OB designs, should I forget all of this discussion with relation to non OB designs????? or is it equally applicable??

Tony.

 

[cc00541]

KYW:

Actually assuming the Speaker is in a 2Pi envoironment the rear wave never reaches the listener, so what does that make your view?

Seems like everyone is talking open baffles in a 4pi environment here. Suddenly you start talking about 2pi and soffet mounted baffles. Stay on topic if you please!

And your statement about dipole bass roll off and BSC being two sides of the same coin is just plain bogus. As you should know, One is due to the back wave cancellation from the driver, and the other is due to the loss of support of the front baffle and the transition of 2pi to 4pi environment. Clearly these are two different phenomena and NOT the same coin. JohninCR has it right.

I never suggested there was any link. I suggested that you consider, for clarity, the case of a dipole in halve space, in other words on the floor.

Again a bogus statement. A dipole in 2pi space would be an infinite baffle, not a dipole. A dipole by definition can only exist in 4pi space, otherwise it becomes a monopole.

And I as well do not understand your signature. –Oh wait! I get it! You are referring to yourself as the ‘latter’. Ah, it makes perfect sense now.

C

 

[cc00541]

Tony:

The narrow baffle is an aid to imaging as the time delay between the direct and reflected sound from the baffle edges is very small.

This quote is from my original reply, and applies to monopole speakers in general, although I'm sure there are those who will dispute this, so please take it as my personal experience only, not a statement of fact.

If the speaker is a 2 way or better, poor phase tracking through the crossover region can also have a significant impact on imaging and localization. Imaging is also affected by speaker placement in the room and the room acoustics. As I'm sure you know, speakers improperly placed, can have as much or more detrimental impact in the resultant sound than the other causes I've mentioned.

C

 

[ThorstenL]

Konnichiwa,

Seems like everyone is talking open baffles in a 4pi environment here. Suddenly you start talking about 2pi and soffet mounted baffles. Stay on topic if you please!

I AM on topic.

Soffit Mounted Speakers (as is done to most serious studio monitors) ensures a constant acoustic envoironment is maintained over the widest possible frequency range. This together with well controlled directivity tends to ensure among other things a power response that throughout the critical frequency ranges remains identical to the on axis frerquency response and such arrangements (in fact infinite baffles) offer the most accurate imaging (accurate as in matching the recorded event).

I suggested to consider a dipole in a 2PI envoironment purely as a didactic aid to JohninCR as he kept going on about when the rear wave from a dipole reaches the listener. If we locate the listener and dipole speaker on an "infinite plain" the rear wave never reaches the listener and what happens then? Thta was the situation I posited to JhninCR to help the thinking through of the situation.

As said, both Items are strictly on topic.

Soffit mounted speakers offer the widest "baffle" area, ensuring most accurate imaging and by implication, vestigal baffles (very narrow baffles) do the opposite.

And your statement about dipole bass roll off and BSC being two sides of the same coin is just plain bogus. As you should know, One is due to the back wave cancellation from the driver, and the other is due to the loss of support of the front baffle and the transition of 2pi to 4pi environment.

In other words, both are due to the fact that the sound "wraps" around the baffle as certain geometric relationships between wavelength and baffle size come into play.

The key difference is that the BDS does not show up in the power response and that it levels out at -6db while the dipole loss does show up in the power response and continues.

But the fundamental causes and implications are identical.

A dipole in 2pi space would be an infinite baffle, not a dipole. A dipole by definition can only exist in 4pi space, otherwise it becomes a monopole.

Would it now? Good, you have just asserted that an open baffle located on a floor ios a monopole. Now would you kindly supply the experimental proof for you theory shattering assertation?

Now I'll really let this rest. If you insist in not wanting to understand, who am I to force you to contend with facts to contrary.

Sayonara

 

[Grandma´s_SUB]

Hello!

1. Why did linkwitz said,that,at the h-baffle effective diameter is deph of the »box«.For my understanding D at the h-baffle should be lenght+width of te »box«.Point is to separate +and-(dipole) wawe.So the so distance and area between wawes must be kept long and big as possible
-In short:Is D of H-baffle w+d

2. A lot of people wire 4 or more drivers(at dipole,of course) to use them as subwoofer…
-Hello!

1. Why did linkwitz said,that,at the h-baffle effective diameter is deph of the »box«.For my understanding D at the h-baffle should be lenght+width of te »box«.Point is to separate +and-(dipole) wawe.So the so distance and area between wawes must be kept long and big as possible
-In short:Is D of H-baffle w+d

2. A lot of people wire 4 or more drivers(at dipole,of course) to use them as subwoofer…
-reasons:
*long wawe lenght cancelling---low bass response below Fqeual---add more drivers to rise up Sensitivity
*High cone excursion below resonance---add more drivers to lower it
Here things become complicated…Why?
When we add drivers SPL rises(3db per a doubling+3db if they are wired paralell,per a doubling)..I REALLY dont see reason,why cone excursion falls…Logicly should be that cone excursion rises..Why?If you need more SPL you need to move more air… So if SPL rises,cone excursion rises
-In short:WHY and WHERE(paralell,series, self amped) cone excursion falls?

3. So if cone excursion falls about 4 X we can drive 4drivers with 400W(if one is 100W), but this is NOT logical:SPL RISES,CONE EXCURSION FALLS!

Dean