More waveguide madness

[Dennis H]

As noted on the Mad board, Earl Geddes has designed a commercial speaker with a 15"
woofer and a 15" waveguide tweeter crossed at 900Hz. He has a whitepaper on the page
that talks about a "Refractive Sound Plug" to eliminate HOM (high order modes), i.e.
reflection and refraction inside the waveguide.

Over the last several years Dr. Geddes has worked on ways to further reduce the
HOM in his waveguides. This has resulted in the Refractive Sound Plug or RSP (™ and
Patent Pending). This plug is made of very low density open cell polyurethane foam. The
idea is to absorb the HOM as they travel down the waveguide. Since the HOM travel a
longer path in the device than the axial wave, the HOM will experience more sound
attenuation that the axial wave. Further, since all waveguides have some reflection at the
mouth (a large radius will reduce this, but not eliminate it) the RSP also helps to
eliminate this standing wave. In essence the RSP allows only the desired direct wave to
propagate freely and attenuates all other waves. The result is a device that has
significantly diminished all forms of internal diffraction, resonance and HOM. A device
that sounds as natural as any piston source yet possesses characteristics unobtainable by
any flat source.

The pic is too small to see what's going on. Is the whole waveguide stuffed with mostly
transparent foam, like some of the old JBLs used for grills, or are the walls lined with
more absorbant foam with a hole at the throat?

Edit: well the pic sure isn't very small when displayed inline. It's pretty clear the
thing is stuffed. So WTF? Must be grill-type foam?

http://www.gedlee.com/Summa.htm

http://www.gedlee.com/images/enclosure.jpg"]http://www.gedlee.com/images/enclosure.jpg

[ThomasW]

It was being demonstrated at the RMAF. The room was so small and the speaker so ugly I couldn't force myself to sit down and listen...... :roll:

As I walked by the door opened. They had the covers or whatever it is you see in the pic removed so you could see the dome and the woofer. Otherwise yes it looked like this...

[Dennis H]

Thanks for the smaller pic, Thomas. So, does anyone have any insight about what kind of foam he's using in the waveguide?

[ThomasW]

The matrix of the foam in the Gedlee design has much smaller holes than that used on the old JBL designs.

As fussy as Earl is he probably had something custom made, either that or he bought it from McMaster-Carr ...:wink:

[dwk]

I think Earl must get the award for most iconoclastic setup at RMAF. He was driving his speakers with gear purchased retail for a grand total of $220. A Toshiba DVD player, generic Japanese receiver, and Home Depot cables.

Yes, the speakers are a bit unattractive, but to compensate they're also big The science behind the 'RSP' is pretty simple, although I'm not sure how it's 'refractive' as that typically implies separation by differences in speed. In this case, it's simply a longer path length. Apparently the driver output is attenuated ~10dB by the RSP, so this isn't really a candidate for the waveguide-loaded dome tweets that have been discussed around here.

Bizarrely enough, in the elevator heading to Earls exhibit I ran into a name I recognized from various lists etc. I don't remember the details of his horns (I don't think they're OS waveguides) but he used PetSmart fish-tank filter material to stuff his horns and reports that it makes a BIG subjective difference in the sound of the compression driver.
Depending on the bandwidth of the waveguide, I'm still not convinced that lining the walls of the horn might not provide a similar benefit by attenuating reflections but leaving the axial wave untouched, although it's entirely possible that it would interfere with the expansion and cause unwanted artifacts.

As for the sound Earls speakers, I'm not entirely sure what to think. In some respects, my reaction was very similar to when I heard the Orion the first time - initially rather unexciting, but after a bit of time it was apparent that it did allow listening 'into' the recording. Tonal balance really seemed a bit thin though, and I think they may have been missing some upper treble energy. It didn't do anything to dissuade me that the waveguide idea has merit, though.

[Patrick Bateman]

How That Foam Works

Thanks for the smaller pic, Thomas. So, does anyone have any insight about what kind of foam he's using in the waveguide?

Geddes has invented many things*, and I think the refractive foam plug is his most elegant and effective innovation. He's explained how it works on Audioasylum, and it is quite simple. The foam simply attenuates high order modes.

Here is how this works. Picture a soundwave that is reflected at the mouth of the waveguide. As the soundwave is reflected at the mouth, and then again at the throat, it will travel through the foam a minimum of THREE times. Simply put, the primary wave will be attenuated minimally, while the high order modes will be attenuated to a much lower level.

Subjectively, I personally find that this invention removes most of the stridency and fatigue that is so commonly associated with high efficiency designs. It literally gives you the best of both worlds. You get a design with directivity and high efficiency, but you DON'T have to cope with the fatigue and unease that are endemic to traditional horns. How great is that? There aren't too many inventions in audio that let you have your cake and eat it too.

I can confidently say that I will NEVER use a compression driver without this foam again.

Patrick B

P.S. You can buy the foam over the counter in any city, but it's not cheap. It is VERY specialized, and difficult to manufacture. But it's worth every penny.



* As I recall, Geddes began exploring waveguides fifteen years ago, also invented the bandpass alignment, and has patents on amplifiers and noise control devices on his resume.

[dwk]

The matrix of the foam in the Gedlee design has much smaller holes than that used on the old JBL designs.

As fussy as Earl is he probably had something custom made, either that or he bought it from McMaster-Carr ...:wink:

From discussions on audioasylum, it's 30ppi polyurethane foam. I did a quick search at the time, and it seemed to back up the idea that it's not particularly common stuff. Earl has also said it's not exactly cheap as well, but I didn't press for details.

I picked up some of the above-mentioned fish-tank-filter material, and it looks to be similar, although that might well be deceiving. Off-white rather than black as well, which would probably look rather cheezy when deployed. Somewhat pricey ($10 for a little box of 4 pcs 8.5x3x0.75"). The box says 'Fluval 4Plus'.

[Hennie]

It is an open cell filter foam. Filter foams are specified in PPI whereas most others are specified I.T.O mass density.

Here are examples, have a look if Dow Performance Foams has something similar.

http://www.sondor.co.za/mat/mat009.asp

http://www.sondor.co.za/mat/mat010.asp

Some filter foams may be polyester and not polyurethane.

Some acoustic foams may also be suitable. They are also specified in PPI.

[Patrick Bateman]

From discussions on audioasylum, it's 30ppi polyurethane foam. I did a quick search at the time, and it seemed to back up the idea that it's not particularly common stuff. Earl has also said it's not exactly cheap as well, but I didn't press for details.

I picked up some of the above-mentioned fish-tank-filter material, and it looks to be similar, although that might well be deceiving. Off-white rather than black as well, which would probably look rather cheezy when deployed. Somewhat pricey ($10 for a little box of 4 pcs 8.5x3x0.75"). The box says 'Fluval 4Plus'.

Doug,

Did you measure it's effect on frequency response? I have, and was pleasantly surprised by the results.

John

[dwk]

Doug,

Did you measure it's effect on frequency response? I have, and was pleasantly surprised by the results.

John

No - I only picked it up last week after RMAF, but haven't done anything with it. I'm pretty much in permanent experimental mode, and don't have any compression drivers set up at the moment. I have a couple of the cheapie Seleniums, but I hacked up the lenses I was using to experiment with 'horn'-loading normal dome tweets.

I'm hoping to get to it someday, though.

[Patrick Bateman]

measurement of my compression drivers with the foam

No - I only picked it up last week after RMAF, but haven't done anything with it. I'm pretty much in permanent experimental mode, and don't have any compression drivers set up at the moment. I have a couple of the cheapie Seleniums, but I hacked up the lenses I was using to experiment with 'horn'-loading normal dome tweets.

I'm hoping to get to it someday, though.

Doug,

It's interesting to see just how linear the measurements are. Before I'd measured the foam's effect, I'd anticipated that the attenuation would be erratic. But my measurements show a broad and even attenuation that affects the highs more than the lows.

I'm willing to post my measurements, but please don't make any assumptions about what Geddes has done. His waveguides do not suffer from the same compromises as mine. The dimensions that I use are SEVERELY compromised, and I'm positive that his measure much much better. In fact, he has measurements posted online. Without further ado, here are the measurements of my JBL 2470s with and without the fluval foam.



I had to kludge together two graphs in photoshop, so please excuse the quality. The DB levels are NOT equivalent, I didn't bother to measure the voltage on the two graphs. You can see from the general shape of the two curves that it's quite well behaved. Because of this, it's easy to come up with a passive circuit that can cope with the curve's shape.

[Dennis H]

As predicted you can get the stuff in bulk from McMaster-Carr. 4'x10' rolls up to 1" thick, available in 30ppi and 60 PPI. Page 605, scroll down to "Polyurethane Foam Air Filter Rolls." They also have 4'x6' pads up to 2" thick in 20, 45, 65, 80, and 100 PPI on page 606.

http://www.mcmaster.com/

[Patrick Bateman]

That's Me!

In this case, it's simply a longer path length. Apparently the driver output is attenuated ~10dB by the RSP, so this isn't really a candidate for the waveguide-loaded dome tweets that have been discussed around here.

Stay tuned. I'm going to try the 30ppi foam with a planar tweeter. I'll have measurements up soon. [ john ]

Bizarrely enough, in the elevator heading to Earls exhibit I ran into a name I recognized from various lists etc.

That was me I'm posting under a pseudonym now. I don't need any static at work for my hobbies! [ john ]

I don't remember the details of his horns (I don't think they're OS waveguides) but he used PetSmart fish-tank filter material to stuff his horns and reports that it makes a BIG subjective difference in the sound of the compression driver.

Depending on the bandwidth of the waveguide, I'm still not convinced that lining the walls of the horn might not provide a similar benefit by attenuating reflections but leaving the axial wave untouched, although it's entirely possible that it would interfere with the expansion and cause unwanted artifacts.

If I were to reduce the amount of foam, I would remove the middle part. I'm guessing most high order modes occur at the throat and at the mouth. I don't have measurements to prove this though, and my JBLs have sensitivity to burn (107db @ 1 watt!)[ john ]

[AJINFLA]

While Dr. Geddes work is something that I have followed closely and has had influence on my design decisions, I still can't see how his speaker won't suffer from the same spectral imbalance that all monopole bass designs will have in a normal room.
I'd still like to hear them of course. That would only be fair. But I have (still) never heard a box speaker that didn't sound like a box. X material and 700lb concrete enclosures included. How can it sound like anything other than it is?
Perhaps he could repeat his listening tests vs the Orions. I'm fairly certain I would know the statistical outcome there .

Cheers,

AJ

[dwk]

That was me

Ahhhhh, I see. You even signed your previous post and I didn't pick it up. D'oh!

Are you going to try a planar or a ribbon? I'm certainly interested in the results. No progress this weekend as it rained all Sunday (leading to snow overnight) and my garage/workshop roof leaks like a sieve....

[dwk]

While Dr. Geddes work is something that I have followed closely and has had influence on my design decisions, I still can't see how his speaker won't suffer from the same spectral imbalance that all monopole bass designs will have in a normal room.

If I understand his position correctly, he would probably say that this is a problem with your room, not with the speaker. His speaker has a very smothly increasing power response, which shouldn't suffer from percieved spectral imbalance (in a post to the bass list he indicated that his subjective study indicated that smoothness of power response had higher correlation to subjective approval than did the flatness of the power response). He'd then point you at his HT book which tells you how to construct a proper room

I do think it's an interesting question, though - if I had more time/money I would like to try dipole bass modules with my still-not-finished planar waveguides. Of course, it's not really possible to transition from a monopole to a dipole without either an on-axis response aberration or a power response aberration either, so it would appear to be a pick-your-poison situation.

[Patrick Bateman]

Ahhhhh, I see. You even signed your previous post and I didn't pick it up. D'oh!

Are you going to try a planar or a ribbon? I'm certainly interested in the results. No progress this weekend as it rained all Sunday (leading to snow overnight) and my garage/workshop roof leaks like a sieve....

If i had a ribbon, I would use one. But all I have in my garage are four planars. I have a Monsson flat panel, a BG Neo3pdr, a Dayton Pt-2, and a handmade planar panel I found on e-bay.

I found that the neo3 mated to a tractrix horn extended the response down to under 1000hz. Even more surprising was that the response was flatter than a compression driver. I think this is due to the horn loading inherent in the neo3pdr faceplate. I'll post graphs sometime this week.

[ralphs99]

Hi Patrick,
Your experiments with lining your 2370's with foam are very interesting. I would be careful about adding anything near the throat and I don't think there is any point in doing so. The throat and mouth of a horn are high pressure low velocity regions. Conversely, the middle part of the horn is where velocity is highest. Acoustic foam works by reducing particle velocity and dissipating the energy as heat, and will therefore be most useful in the middle of the horn.
Interestingly TAD make a horn with a small section of foam half way to the mouth.
Even more interesting is US patent 4893695 which appears to describe the horn-full-of foam idea in detail. Even without the pictures it still makes for interesting reading. I can post the text if anyone is interested. Or maybe someone else has access to the full patent and can post it.

Cheers,
Ralph.

[Dennis H]

Even more interesting is US patent 4893695 which appears to describe the horn-full-of foam idea in detail. Even without the pictures it still makes for interesting reading.

Thanks for that, Ralph. That patent appears to be a variation on what dwk was talking about, lining the horn with foam, rather than plugging the horn with it. It actually looks like a very good way to match the radiation pattern of a dome tweeter to a dipole mid as discussed in other threads. You'd just cut a straight sided hole in the baffle and insert a foam or felt plug with a tapered hole to form the 'waveguide.' Here's a pic of the most basic version in the patent.

[JonMarsh]

While Dr. Geddes work is something that I have followed closely and has had influence on my design decisions, I still can't see how his speaker won't suffer from the same spectral imbalance that all monopole bass designs will have in a normal room.
I'd still like to hear them of course. That would only be fair. But I have (still) never heard a box speaker that didn't sound like a box. X material and 700lb concrete enclosures included. How can it sound like anything other than it is?
Perhaps he could repeat his listening tests vs the Orions. I'm fairly certain I would know the statistical outcome there .

Cheers,

AJ

Amen, brother.

The only way I know to ameliorate the monopole bass isues is the kind of setup and positioning hassles that Avalon and Wilson routinelly used. I have gotten very good results with approach years ago (in the 70s, actually), but not many people can practially put the speakers in the middle of the room and invest in a lot of room treatment. Dipoles are much more forgiving to get to a certain basic level of performance, though wringing the last bit of imaging they're capable of DOES usually require going to the near middle of the room trick.

~Jon

[Patrick Bateman]

Natural Bass Without A Dipole?

Jon,

Doug hit the nail on the head. Geddes is looking at the speaker and the room as a complete system. So the treatment of the room itself goes hand in hand with the bass alignment. This room treatment is described in his Home Theater book. It's very simple and effective, but it is not the traditional treatment used by Wilson. As I understand it, the use of traditional room treatment with a constant directivity device would have unintended negative consequences. Again, it's described in much more detail in the book.

I hope it doesn't sound like I'm against dipoles. I'm not, in fact I've heard dipole bass sound VERY natural. The tricky part is finding a midrange with the same radiation pattern and power response.

:PB:

Amen, brother.

The only way I know to ameliorate the monopole bass isues is the kind of setup and positioning hassles that Avalon and Wilson routinelly used. I have gotten very good results with approach years ago (in the 70s, actually), but not many people can practially put the speakers in the middle of the room and invest in a lot of room treatment. Dipoles are much more forgiving to get to a certain basic level of performance, though wringing the last bit of imaging they're capable of DOES usually require going to the near middle of the room trick.

~Jon

[ThomasW]

Here's a quote I lifted from a post Earl made on the MAD board. I thought is was interesting since we've been discussing the construction of his speakers.

The Summa's could easily be built into a wall owing to the fact that the front baffle is independent from the rear. Thus you would simply buy the set without the rear enclosure and mount it into the wall. What I always say to people in this situation is just hide the speakers - like I do There is no way that an interior decorator (aka a wife) could not come up with a design whereby the speakers were behind a wall made of fabric instead of gypsum. I have done this in my last three homes. The only constraint that needs to be placed on your wifes ideas is that the materioal of the wall needs to be acoustically transparent. This is easilt tested by simply blowing through it. If you can easily blow through it its OK, if not its too tight.