Geddes on Waveguides

[gedlee]

Originally Posted by angeloitacare

Dr Geddes might give his comment of this " wisdom " :
….

There is never any "wisdom" in Romy's rants.

[gedlee]

Originally Posted by soongsc

CD horn is quite real based on simulation for axisymmetric condition. It just doesn't get there with a 90deg wall angles.

From where I sit thats not true.

[rcw]

The point I would make is that the concept of constant directivity is not the same thing as the mechanism used to achieve it.
It is possible to achieve c.d. and minimise diffraction and higher mode production.
This is amply demonstrated in the data refered to and posted in these pages, for both compression drivers and dome radiators.
rcw.

[Patrick Bateman]

Originally Posted by tinitus

What do you think of this BMS driver

I suppose its kind of special having 1.5" throat, and only 1.75" voice coil

http://www.bmspro.info/Specification...al_info.0.html

Or is it a misunderstanding because of the ringradiator diaphragm
They seem to have removed much of the previous information about their diaphragms
Now they just call it annular diaphragm design

Wow! That is outrageous extension for a compression driver with a 1.5" throat. There aren't many 1" models with clean output about 15khz.* I'm using a BMS 4540ND in my car, and the only compression driver that equals it above 10khz is over $1000. I've measured it extensively and it has output beyond 20khz.

I really like the BMS ring radiators.

* yes, I know there are compression drivers that go to 20khz. The difference is that they get the output up high via diaphragm resonances, which sounds raspy to my ears

[Patrick Bateman]

Originally Posted by Wayne Parham

Earl - I was going back through the archives to find the photos of you and Duke from GPAF 2005. Wanted to post one on the HOM measurement thread. Thought I'd post here too, sort of memory lane.

• Geddes and AudioKinesis at GPAF 2005

I remember thinking the woofer and tweeter were about 20" apart, rough guess. Is that about right?

I'm going to bringing my Accord, with OS waveguides, horns and tapped horns, to the Audio Fest on Mercer Island tomorrow.

http://www.audiokarma.org%2Fforums%2...%3Fp%3D3057229

If I can squeeze them into the car, I'll be bringing my Summas up too. Should be interesting. I've managed to cram them into my car before, but it's tricky...

Info on the car is here:

http://www.diyaudio.com/forums/showt...hreadid=148006

If anyone is in the area, would be great to discuss waveguides, horns, etc... Should be a fun show.

[gedlee]

Originally Posted by rcw

The point I would make is that the concept of constant directivity is not the same thing as the mechanism used to achieve it.
It is possible to achieve c.d. and minimise diffraction and higher mode production.
This is amply demonstrated in the data refered to and posted in these pages, for both compression drivers and dome radiators.
rcw.

Actually I have not seen this for dome radiators, some simulations but no actual measurements of a real device.

The bottom line however is that there aren't different ways to achieve CD. You can easily do the inverse problem to calculate what is required from the source to achieve ideal CD in the far field and there aren't two solutions. There are different approximations that one can make that lead to slightly different configurations, but there are not that many options. Basically, I don't see how a dome can achieve CD in any configuration. I know you claim that it can be done using this or that technique, but stated quite simply, I haven't seen it practice yet. And until thats been done its all hypothetical.

[gedlee]

Originally Posted by Patrick Bateman

The difference is that they get the output up high via diaphragm resonances, which sounds raspy to my ears

Wouldn't this have to be true of all compression drivers? No diaphragm is going to 20 kHz and above with NO resonances. How well are they controlled is the issue. And I suspect that most of the problem above 10 kHz is acoustic - its very hard to get those frequencies out of the device. They tend to rattle arround inside - they don't want to come out to play. Thank goodness that they have so little effect on the total sound quality.

[Patrick Bateman]

Originally Posted by gedlee

Wouldn't this have to be true of all compression drivers? No diaphragm is going to 20 kHz and above with NO resonances. How well are they controlled is the issue. And I suspect that most of the problem above 10 kHz is acoustic - its very hard to get those frequencies out of the device. They tend to rattle arround inside - they don't want to come out to play. Thank goodness that they have so little effect on the total sound quality.

Because the BMS is a ring radiator, the peak is pushed above audibility in many of their compression drivers. My only experience is with the 4540ND - but I'm eager to try the other models. My Summas have the B&C, I've listened to Unity horns with Tad, and in my car I'm running BMS 4540ND.

Above 10khz, the BMS leaves the B&C in the dust, and gives the Tad a run for it's money.

Of course, below 2000hz, the B&C and the TAD eat the BMS alive, and below 2000hz is very important.

So it's a tradeoff of course. The BMS 4540ND has a delicacy and "airiness" in the top octave which reminds me of a ribbon tweeter.

This is a measurement that Brandon (augerpro) did of a B&C on an OS waveguide. You can see there's a peak at 16hz, which is the domes breakup, and then the output drops off a cliff.


My measurements of a BMS 4540ND on an OS waveguide. This is from my car. So the response is a lot more ragged than you'd get in an anechoic measurement. Nonetheless, there's no peak at all, and response continues beyond 20khz. If the peak is there, it's above 20khz.


TimG at audio heritage did this measurement of a 4540ND on an ENG1-90 waveguide, which is clearly inspired by an OS waveguide. Same results : extension beyond 20khz.

[soongsc]

Originally Posted by gedlee

From where I sit thats not true.

I've recently been running some sims on larger configurations. They get very CD without the on axis dip. I will show measurement data when the time is right. But I already have a pretty good feeling what is causing the on-axis dip in your designs.

[gedlee]

Originally Posted by soongsc

I already have a pretty good feeling what is causing the on-axis dip in your designs.

I already KNOW what causes the dip and I've posted it here a number of times. Its the mouth diffraction which adds out of phase from the direct sound at precisely one frequency when precisely on axis. An elliptical mouth will make this go away.

We've been all through this before, where have you been?