Jean Michel on LeCleac'h horns

[soongsc]

Originally posted by mige0
Thanks tend for your work, ans also that you brought your page back!





Below the according sound field / pressure distribution calculated by CARA (for the W.M. Hall measuements):




most striking for such a simplified simu ?no?



and the corresponding wave front propagation:





QED


Michael

Bear in mind that this is illustrative and not an engineering sim.

[mige0]

Originally posted by Salas
Personal calls between mige0 and gedlee removed. Although expressed in a civil manner, still personal calls. Avoid.

The most interesting part – deleted !
What a pity...

Originally posted by panomaniac
Interesting. Had not thought about using CARA to simulate a waveguide. What did you do, build a triangle room and put a point source loudspeaker in it?
Walls are made of what?

Originally posted by panomaniac
That's why I asked what he made the wall of. In Cara you can specify. Maybe put a door there? But I think CARA sees that as no reflections.

But whatever, it at least illustrates a point, even if exaggerated. If the CARA sim is close, it does show no refections until the wave comes back.

Originally posted by soongsc

Bear in mind that this is illustrative and not an engineering sim.

Allow me to come back to the tricks I used to stretch the capabilities of CARA to a topic it isn't intended for (in its current version at least).
I'd like to focus first on what we can gain in understanding what's happening with horns first and outline the limitations of CARA visualisation on our very topic afterwards.


Michael

[mige0]

What we have seen in my last posting is that the islands seen in W.M. Hall's measurements clearly reflect the sound field inside the horn.
I took some liberty here to stretch the term "sound field" to characterise pressure distribution inside the horn speaker whereas it usually is meant to describe pressure distribution in the listening room .

This pressure distribution is nothing else than the SPL level (for a specific frequency) at any point in space – meaning – we easily can localise hot spots.

One thing to stay clear about is that the SPL distribution shown is for a specific frequency *only* - meaning – it is calculated to reflect / visualise W.M. Hall's measurements at 800Hz.

The islands usually would change their shape and place inside the horn if the frequency is altered.

So – at a certain point in space there can be a hot spot for one frequency – translating to a peak in the frequency response for that very frequency taken at exactly that point in space – whereas for other frequencies the sound field looks totally different – making up for the ragged FR's we are all familiar with.

Going back to my Grand Canyon example – people at different seats would have different impression of the "sound quality" of Zeus' lightning.

From a horn design point of view it is desirable to keep the islands in place as what is cooked inside the horn is spilled out into the room and we would like to be able to correct for that raggedness for a as wide as possible room angle.


Ok, all above is more or less a side note to CD characteristics – or better put – how deviations from CD translate from the underlying mechanisms.



#############

One additional thing I wanted to come back in detail are the phase gradients also displayed in W.M. Hall's measurements.
The ones Earl and W.M. Hall got confused with the term "wave front"


Below the CARA simu of a sinus stimulus showing how the mouth reflection interferes with the original wave (at the very frequency of interest)



We see a "movie" taken at subsequent phase angles of the stimulus source. If there were no mouth reflection we simply would see a smooth sinus wave "moving" from right to left.

Now – if we freeze this movie at 0 (180) 90 (270) deg and draw the lines along its zero crossing we would get the equi-phase gradients in space shown by W.M. Hall.

These phase gradients tell us that in addition to the hot spots seen earlier we get phase shifts for specific frequencies at specific points in space.



##############


To sum up the hot spot island and the phase gradients seen in W.M. Hall's measurenets - this is what's diffraction > reflection > delay > interference (and thus HOM included) is all about.

Nothing new under the sun !


The point to pay attention to is that the "wave *front*" isn't affected by this interference mechanism.


There *is* a influence on the shape of the "wave front" by the mechanism of diffraction though (not visualised by my CARA sims), but it is a relative simple straight forward effect (plus some reflection part) involved in "bending the wave front around the corner" as presented and outlined by Jean-Michel and David McBean on several occasions.



Michael

[soongsc]

Whenever there is impedance mismatch, there will be reflection. If the impedance transition is gradual, then the reflection will be less concentrated.

[Pano]

Yes, I see those reflections all the time on video and computer graphics cables. Maybe I even hear them on SPDIF cabels.

But if the impecdance transistion is gradual, what does that do to the reflections? Are they actaully lower in amplitude, or just spread out and diffused?


Thanks for the sims, Michael. Would you mind posting your CARA room and box files so we can play? He email. Thanks!

[Ed LaFontaine]

But if the impecdance transistion is gradual, what does that do to the reflections? Are they actaully lower in amplitude, or just spread out and diffused?

Both, The reflections occur over a wider range of frequencies and are lower in amplitude.

http://www.tolvan.com/edge/help.htm

Run some simulations in "The Edge". Compare the diffraction off a corner with that from a large radius edge.

[mige0]

Originally posted by panomaniac
Interesting. Had not thought about using CARA to simulate a waveguide. What did you do, build a triangle room and put a point source loudspeaker in it?
Walls are made of what?

Yes – basically its a triangle room – side walls are non-absorptive - front wall is semi-absorptive to account for the mouth reflection (100% absorptive needed for *no* mouth reflection).
In fact it makes no difference how much reflection you allow – the pattern stays stable - its just the magnitude that's changing – and this isn't calibrated anyway.

Originally posted by tinitus
I dont understand
Are you saying that sound wave hits the room air like if its was like a brick wall
Obviously it will "hit" the air and loose support from waveguide, spread out, and loose intensity
Maybe a very small part will bounce back, but I suppose not like if it hit a wall
Or it will be hit by the next coming wave

Tinitus – its only the horn itself and the SPL distribution within the horn that is modelled here – no room at all.
Keep in mind though that what is cooked inside the horn is spilled out into the room.

Originally posted by panomaniac
That's why I asked what he made the wall of. In Cara you can specify. Maybe put a door there? But I think CARA sees that as no reflections.

But whatever, it at least illustrates a point, even if exaggerated. If the CARA sim is close, it does show no refections until the wave comes back.

No door – its called an "open window" in CARA (at least in German version). You can set whatever absorption you like (over full frequency bandwidth) in the so called "material editor".

Originally posted by soongsc

Bear in mind that this is illustrative and not an engineering sim.

Spot on.
As good as CARA is in providing a "in depth" understanding of the mechanisms involved, the limited it is when it comes to accurate numbers – it simply isn't made for that.
The beauty of CARA is that it puts some shade of light on aspects usually not in the focus of horn / wave guide design.

One exception though is the area at the top end of any horn or wave guide where throat dimension becomes comparable to wave lengths.

Here its seems CARA can capture what's going on whereas BEM can't.
One other advantage of CARA is the low resources it needs (compared to BEM).
Unless you have access to "Deep Blue" BEM simply can't be done at realistic time consumption at higher mesh resolution and top end frequency.

This is an interesting area as horns seem to suffer in this department and some good simus might help a lot.

The basic limitations when stretching CARA's capabilities to horn simulation I already have outlined when I introduced my "bunch of pretty pictures" in:


http://www.diyaudio.com/forums/showt...31#post1847931
http://www.diyaudio.com/forums/showt...04#post1848104
http://www.diyaudio.com/forums/showt...39#post1848339


One additional thing to notice is that we focus here on the SPL distribution on a single plane – hence bottom and ceiling of the horn are made to be fully absorptive to not interfere.

Another thing to notice is that CARA does not account for edge diffraction.
Kind of exception is the "diffraction" at the driver / throat joint as this is quite well accounted for with the drivers directivity.
The other exception is the mouth diffraction which is pretty accurate modelled (as far as the sound field *inside* the horn is concerned) with a simple reflection.

It is most interesting that my simus and that "ancient" measurements from W.M. Hall are so accurately congruent.
Given that CARA does not account for diffraction along the horn contour (a no brainer along a conical horn anyway) this tells us that by far the most errors in horns stem from reflection.

Reflection from the mouth sure - but even more so form the reflections along the horn contour due to finite throat dimension when we enter the top end. My thoughts about this are outlined here:

http://www.diyaudio.com/forums/showt...98#post1854398

The point here is that we could have *any* contour - be it conical, Lecleach, OS or whatever - it does not matter – in this frequency range the contour has practically no control over the interference created – building up for a very special directivity behaviour at the top end.

Michael


Sorry for getting slightly OT, Jean-Michael

[wxa666]

Originally posted by mige0
What we have seen in my last posting is that the islands seen in W.M. Hall's measurements clearly reflect the sound field inside the horn.
I took some liberty here to stretch the term "sound field" to characterise pressure distribution inside ...

Hm, There are several horn simulation packages around. Some are very elaborate, use tremendous computing power. But still have to be approved by measurement. Too in this thread the terms "impulse", "frequency" and the like had to be explained, because it has all been mixed up with eagerness, streching the terms deliberately as been said in the citation.

What does this debate aim for? The LeCleac'h contour could be simulated thoroughly. That has been done with the result that it lacks defined directivity. Second to that a lumped element model of the intended (quite old fashioned) driver has proven to be useless. In that this model predicts obviously wrong results. Has a LeCleac'h contour yet been examinated for some irrelevant details as where to it throughs all that precious sound - directivity?

I think the use of a "streched" CARA to support a shallow misconception of what an impulse is may be seen as, so, what do You think? But is there anything left to be said about the LeCleac'h contour I didn't get yet?

Thank You

[thend]

But if the impecdance transistion is gradual, what does that do to the reflections?

With horn, there are gradual impedance transition according to wave length, so frequency.
More sound is directive (treble) more there are an adaptative impedance, indeed the wall's horn have less effect on directives frequencys (less reflection)...

That's why, like we can see on my page "horn reflection", further down in frequency there are more reflection.
But the horn allows a gradual larger wave front (acoustic transformer), and when the wave front is large, more the impedance transition gives bass frequency, it depends on the size of horn, and more the wave front is large more there are directive (less reflection).

So, for the impedance/reactance point of view, the fonction of horn is to allow a gradual larger front in good conditions, otherwise we have more reflection.

This is my way of thinking.

[jzagaja]

Tangband offers interesting flat units that resembles Balanced Mode Radiators: W3-1797S and W4-1757SB. Question is can we use modal transducer in a waveguide? I think we can if one chose right kind of pressure feed and loading that does not destroy disc balancing. Then we can use these inexpensive drivers with possible simple phase plug and full range with transmission line.