This paper always pisses me off. I had consulted with Charley at Peavey and after that he "invented" this device. It was a direct copy of my OSWG with a very slight modification at the throat (made because his drafting software could only do circles). Not a single mention of me or my work in their entire set of publications.
My first attempt will look something like this. Little bigger, 23" x 22" x 36" intrnal dimensions. THe sides will bend back. Internally it will be a TL, with vent on the bottom. It will house an Altec 416-8a. The horn will be a Iwata 290 with either Altec 288 or JBL 745 or 850. I am interested in opinino s here for the comparison speaker which will probably be based on AE Td12 variant and waveguide of some sort.
http://www.diyaudio.com/forums/group-buys/166312-waveguides-horns-143.html#post3505136
http://www.diyaudio.com/forums/group-buys/166312-waveguides-horns-143.html#post3505136
I'd often wondered if the benefits of an OS waveguide could be achieved with a simple spline between the throat and the mouth. In other words, I know that there's one line that's mathematically correct. But I can draw a spline in Xara in two seconds. When I made an OS waveguide using the formula and a spreadsheet, my waveguide took about 30 hours to assemble. (I built it up, layer by layer, using pieces of 3mm cardboard.)
"Perfect is the enemy of good" Voltaire
I'm quite certain that they're a very good sounding system, relative to the "typical" systems used in audiophilia. Great drivers, optimized electronics and enclosure, and a high performance horn, by what we've seen. That doesn't mean there's nothing that can be done better, or is already being done better by someone else.
My own system is hardly pertinent to design criticism of a system I've never seen nor heard. If you must know, it's a hybrid offset bipole, utilizing underhung neo motor 15"s in a low diffraction vented cabinet and my oblate spheroid waveguide build, with a neo compression driver whose diaphragm I've modified for smoother top end. That's the front section. The rear is a little simpler, sealed JBL 12" and a waveguide loaded soft dome tweeter.
None of that is relevant here- why are you making this about the people/their systems and not the specific design criticisms?
System
Perhaps he's just curious what everyone has. Or wants to have...
Right now, mine is LCR in front using three JBL 4430's. Switching to the Summa would definitely eliminate some of the peaks that tend to be irritating. (Based on my audition of the Summa's) The nice, easy, undistorted dynamic range sure is nice compared to the smaller "acoustic suspension" boxes that previously served in the front.
Bruce
Perhaps he's just curious what everyone has. Or wants to have...
Right now, mine is LCR in front using three JBL 4430's. Switching to the Summa would definitely eliminate some of the peaks that tend to be irritating. (Based on my audition of the Summa's) The nice, easy, undistorted dynamic range sure is nice compared to the smaller "acoustic suspension" boxes that previously served in the front.
Bruce
The thing is with CNC why use the "almost correct" formula? I do not know how much difference it makes, but I suspect it us not much.
Actually the OS contour IS a spline between the throat and the mouth. What peavey did was a circle and then a straight line, which is not a spline. In my software I can get a perfect OS contour with three points and three angle at each point - a basic three point spline.
I totally understand how you feel, Earl. I've put a lot of effort into some loudspeaker designs too, and so then when they are copied or work is heavily based on my ideas, it's nice when I get some credit out of the deal. Some people do, but some don't. The worst offenders try to take credit, almost quoting their source but wording it like it was their own idea. Just kinda chaps me sometimes.
So for the record, when I designed my waveguides, I want you to know I remembered what you said about the catenary shape, and how your OS horns are a rotated catenary. I remembered what you said about other catenaries, prolate spheriod and elliptic cylinder surfaces too. It was your influence that drove me towards the catenary profile for my own waveguide/horns.
I always knew I liked radial horns better than CD horns with sharp edges, so I gravitated towards devices that had some similarities with waveguides already. But without your influence, I would have basically made a conical horn that transitioned the compression driver exit to the horn's wall angle with an arbitrary radius. I probably would have chosen a curve much like Hughes did. But after talking with you, I studied the catenary and afterwards, strongly agreed with you that it was the most desirable flare profile shape.
Likewise, I've always given a tip of the hat to Klipsch for their influence on me with the cornerhorn, to David Smith, Don Keele and John Eargle at JBL for their influence on me with DI-matched two-ways, you and Todd Welti on multisubs and others mentioned in numerous posts on my Pi Speakers forum. I think it's really important to give credit where credit is due.
The whole impetus for making the Summa was to use all the great things about the 4430's that I had used for decades but eliminate some of the harshness that I found with their horn. I was very content with the results, but still consider the 4430's to be state-of-the-art for their day. In fact when I first did this I had no intention of making them for other people, just myself. But I got so many requests after people heard them that I decided to give it a try.
If it makes you feel any better, I work in software, and it seems like this is how a lot of guys become millionaires in this arena:
step 1) Some dude copies an open source program, makes a few tweaks to it, and then claims it's a new invention
step 2) Investment capital comes along and hands the dude a few million dollars to promote the software
step 3) After a few years the software company is gobbled up by another more established software company. (Big software companies often grow their business by acquiring smaller software companies.)
Bottom line - the original inventor of the software rarely makes a penny off of this process. (Since it was originally open source.)
OTOH, I shouldn't complain too loudly, considering that writing software pays my rent 🙁
4430's
I often wonder if smoothing the throat with file, sandpaper and little filler material would help to some degree. The junction between the two halves of the horn/waveguide is not very smooth.
I often wonder if smoothing the throat with file, sandpaper and little filler material would help to some degree. The junction between the two halves of the horn/waveguide is not very smooth.
I have a feeling you might be able to significantly improve the 4430s with some foam in the throats- 30ppi reticulated foam is used in the throats of my waveguides
An externally hosted image should be here but it was not working when we last tested it.
going back to the phase plug. The 2426 on the 2344 can sound a little harsh or spitty, but I bet running foam from 1-2cm proud of the diffraction slot, all the way back to the phase plug (you'd have to remove the bugscreen and fill the roughness that results at the driver/horn junction) would improve them mightily. This is based upon Geddes foam work, where he fills the whole darn thing with foam.
Speaker Foam - Filter foam for speaker covers, noise reduction
I readily hear the benefit with vs. without my small (Yellow in my pic) plugs, and my waveguides don't use the diffraction slot. Forgive the mess- it doesn't look quite so bad behind the speakers, normally 🙂
Foam
I picked up some foam made to fill gutters to keep them from clogging at Home Depot the other day. It looks quite a lot like open cell reticulated foam. $6 for around a 6' length. Cheap experiment! Have not had time to cut any neat little plugs as of yet. Maybe this weekend.
It still bothers me that the throats are so rough. Exactly at the point they need to be smooth.
I picked up some foam made to fill gutters to keep them from clogging at Home Depot the other day. It looks quite a lot like open cell reticulated foam. $6 for around a 6' length. Cheap experiment! Have not had time to cut any neat little plugs as of yet. Maybe this weekend.
It still bothers me that the throats are so rough. Exactly at the point they need to be smooth.
That was my path too. I had made constant directivity cornerhorns, influenced from Klipsch obviously, but using constant directivity horns. I liked the overall character of the speaker very much, but its tweeter was too "spitty" and harsh, so I soon switched to radials having no sharp edges. I liked them much better.
Likewise, when I saw the JBL 4430, I liked the approach, seeing it as a way to make a speaker that had uniform directivity but that didn't need to be placed in corners. But again, I prefered radial horns to any of the CD horns of the day. A good radial horn has nearly constant directivity in the horizontal plane, and its response is very smooth. So I could make some really good speakers using this combination, using the DI-matched two-way concept, but using smoother horns. They sounded great, and I still like those era/model speakers, even to this day.
I do think a good waveguide beats the radials, especially in imaging. A good waveguide is silky smooth and has spooky good imaging. But I do tend to caution people about the waveguide they choose, because some of them are no better than the old CD horns from the 1970s. A good waveguide measures smooth, and doesn't need a lot of equalization in the crossover beyond mass-rolloff compensation.
It does- but most of what I saw was very large cells- and thus larger plastic "webbing" for the structure, more along the lines of 10 than 30, and might not be as effective. Definitely worth a shot. Smoothing them with sandpaper is probably a good idea. I'd want to either remove the driver, or suspend the speaker between two chairs facing down, as done with setting up Linn LP12s so you can work from below, so that the dust falls downward (and naturally wipe it out with a damp paper towl afterwards).
That was the opposite of my intention, the obvious answer being included
It seems to me to be a very rationale choice to learn from more experienced bulders/listeners than to just begin building horns. Of course each has its own merits, but already i have confirmed a secondary woofer choice in some of the comments made, the JBL 2235. Dr. Geddes mentioning the JBL 4430(third time in the thread) pretty well establishes this speaker as a decent, if not very good, refernce point. This speaker can be easily built with second hand components or we can build off its strengths and address its weaknesses. I thank you for your response and system description. I and I believe others, would enjoy a picture of it, if possible. Step away and consider the talent that has assembled in this thread and made fairly clear statements in terms of personal choice. What potential builder would not be swayed?
Fair enough- it came across (to me) as "who are you to judge, what's so great that you've got that beats an M2?"
My speaker, front and back- note that the back section now has the horn right up against the frame of the 12". Driver integration in the rear is superb.
And for kicks, here is a highly modified JBL CD5525 using what appears to be a DDS ENG90 waveguide with a square flange (which was nicely routed to mate with the frame of the 12" JBL 2206h). The HF is jbl 2426h, as used in the 4430 (well, that's the 2425h, IIRC, but it's effectively the same driver with cosmetic/mounting differences)
These were used in disneyland parades, and the cabinet is fiberglassed- making it extremely strong, heavy, durable. The woofer got a light weatherproofing coating but as a JBL 4" coiled 12", it's got the construction where a light coating won't make much difference.
I'll be filling the front section (note that the cab walls are a couple inches proud of the baffle) with reticulated foam, with higher density foam towards the sides, to ensure that the front panel isn't a resonator.
My speaker, front and back- note that the back section now has the horn right up against the frame of the 12". Driver integration in the rear is superb.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
And for kicks, here is a highly modified JBL CD5525 using what appears to be a DDS ENG90 waveguide with a square flange (which was nicely routed to mate with the frame of the 12" JBL 2206h). The HF is jbl 2426h, as used in the 4430 (well, that's the 2425h, IIRC, but it's effectively the same driver with cosmetic/mounting differences)
An externally hosted image should be here but it was not working when we last tested it.
These were used in disneyland parades, and the cabinet is fiberglassed- making it extremely strong, heavy, durable. The woofer got a light weatherproofing coating but as a JBL 4" coiled 12", it's got the construction where a light coating won't make much difference.
I'll be filling the front section (note that the cab walls are a couple inches proud of the baffle) with reticulated foam, with higher density foam towards the sides, to ensure that the front panel isn't a resonator.
Last edited:
Thanks, I couldn't figure out how to make that work without an amp connected. So, I suppose that means the raw HF response looks a bit like this:
Attachments
- Status
- Not open for further replies.