Geddes on Waveguides

[gedlee]

Must have missed these.

There is an excellent paper by Roberto Magalotti in JAES on doing exactly this - a PWT for measurements. I can only paraphrase and say that the density of stuffing needs to vary in a very precise way so as not to reflect any sound. Not an easy thing to do but Roberto describes how. He is now very high up in pro audio engineering companies. He deserves it.

 

[Patrick Bateman]

I can't comment on specific designs, particularly ones that I am not familiar with, but what you suggest sounds more like hand-waving to me.

I think Sheeple is referring to my musings on the JBL M2 waveguide, over in the thread I started on that speaker.

In a nutshell, the waveguide has a diffraction slot whose length varies based on angle. This has the effect of widening the beamwidth on the horizontal and vertical axis, at the expense of the diagonal axis.

This is similar to how a vertical diffraction slot works. Except a vertical diffraction slot widens the beamwidth on the horizontal axis alone.

IE, if we have a waveguide of a depth that is fixed, a diffraction slot can offer beamwidth that's wider than would be otherwise possible.

I made these musings before A.T.H. had the ability to simulate diffraction slots, and once that feature was added, the software proved that my theory was correct.

Of course, there's no free lunch, diffraction slots come with drawbacks.

 

[gedlee]

In a nutshell, the waveguide has a diffraction slot whose length varies based on angle.

"angle" here is around or off axis?

 

[mabat]

[...] In a nutshell, the waveguide has a diffraction slot whose length varies based on angle. This has the effect of widening the beamwidth on the horizontal and vertical axis, at the expense of the diagonal axis. ...

I still hold a view that this is mainly because the included angles of the walls are simply different as a result - the nominal coverage angle is wider horizontally than diagonally - and that the throat itself has very little to do with it per se. What the throat does, however, is it spreads/blurs the higher order stuff in space and time. I believe it's actually pretty the opposite of a diffraction slot. The M2 waveguide looses pattern control at the high frequencies as any other horn with that throat diameter but it does so pretty gradually. I believe that's because of the throat varying its length around the horn. I can simulate the same thing on a circular wavguide (as shown in my thread).

 

[AllenB]

When you try to constrain path lengths to set the wavefront shape to fit a square horn, and instead of a phase plug you use the simple flat method of changing the throat shape, it tends toward being a diamond.

 

[GM]

I'd like to ask about a plane wave tube

PWT design, use: http://www.angelofarina.it/Public/Standing-Wave/palnewt.pdf

 

[Omholt]

Both the JBL M2 and 2384 get's narrower in the highs. Especially horizontally. Vertically they do slightly better. Overall, M2 doesn't really have a particular uniform directivity. 2384 is better but I haven't seen an independent proper measurement of it.

In comparison, a horizontal polar of a diffraction slot horn below. Take not that this is measured indoor without any gating. The directivity is much wider than what it actually is, which is common with indoor measurements.

 

[300Z]

I haven't seen an independent proper measurement of it.

Erin did a Klippel test a little while ago.

IMHO, the narrowing directivity in the top end of the M2 isn't really that much of an issue since there isn't much content up that high anyway.

 

[Omholt]

300Z:
I was referring to the JBL 2384. Read the context.
The measurements I showed of the M2 were Erin's!

 

[gedlee]

The directivity is much wider than what it actually is, which is common with indoor measurements.

Why would that be? Indoor measurements can be just as accurate as outdoor ones.

 

[Omholt]

Where's the quote button on the latest message?
Gedless wrote: "Why would that be? Indoor measurements can be just as accurate as outdoor ones."

I'm not really sure, but I'm seeing this when I compare my indoor measurements with anechoic ones. And I hear others are experiencing the same thing.

Below is an indoor measurement of the Klipsch K402 (no gating FIY). Same thing here. It's much wider than the approximate 90° directivity Klipsch has published.

 

[mabat]

I think what Earl might have missed is your mention that "this is measured indoor without any gating.".
Is that really what you mean by an indoor measurement?

 

[Omholt]

Whether I gate or not doesn't make a difference of the beamwidth.

 

[mabat]

That's true but it will affect the measurement which won't be correct.

 

[gedlee]

You are correct Marcel, it is not correct. I did catch the gating comment, but wanted to see if Omholt realized this was the issue.