small horn + waveguide PA?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
All right.....whats the scoop on the Karlson impedance match invention?
I am going to have to leave it up to you who have knowledge.

Can you increase the effective acoustic diameter by scalloping the edge of a horn? It would take me too much time I think to try and search and digest everything when the knowledge lurks somewhere in the minds of diyAudio members.

I like the design direction of bjornos bass bazookas coupled with a horn. Im on the verge of ordering more 2 inch drivers to pursue the unity as per the suggestion. If the Karlson concept could reduce the mid horn diameter at the expense of more length, that would be perfect. In an ideal portable PA system, the horns could fit just inside the bass bazookas although as the design is, they are too narrow in diameter. Small point. With some minimal legs, the bass towers could double as stands also for the horn. It would be really cool...

Isnt there an iron law about size vs efficiency? I wonder how bjornos bass tubes compare. In fact what is the type of speaker with the greatest efficiency per volume everyone should know that I guess? hmmmm electrostatic flat panels maybe...being theoretically two dimensional....

Max
 

Attachments

  • karlston slot.jpg
    karlston slot.jpg
    30.8 KB · Views: 193
Yep, it's proportional to the cutoff frequency (Fb) cubed times box volume (net Vb), divided by the electrical quality factor (Qes). The proportionality factor is different for different designs, sealed/reflex/whatever. This is the best efficiency possible within the laws of physics, naturally there are ways to get less efficiency for the same box which is what most actual speaker alignments wind up being as the trade-off to get a wide band-width (BW).

GM
 
the high frequency Karlson unit was the 7" long slotted tube used in his 1965 2-way X15 system - - I'm not sure how good of a midrange one could make but Robert Reams (Neural Audio) made a double slotted "Clothespin" tube which I believe was driven by an EV12L


Karlson's 1961 letter to Carl Martin Poppe explaining the asymmetric projector
pattern
http://home.earthlink.net/~buddhaboy2/Let61.pdf
 
SCIENCE!
Thomas Dolby

I guess I begin the long process....
"we believe the cause of death was multiple iterations"

Hornresp! what a fantastic program! He blinded me...with SCIENCE

I expected the size of bjorno's much appreciated Unity design using 4 peerless 2 inch to have a mouth of 18 inches probably. But I started to play with his template in Hornresp and it appears that the diam is more like 7 inches, which is what the concept calls for, size wise. I started to play with parameters, and it appears that the gain is due to the 6 db half space radiation and 6 db from using four drivers, while the horn serves to modify only a few db in the 3k hz and up region. Actually, it confuses me a little that if I start with full space radiation as the baseline, then add baffle area, it seems to reduce the high frequency as opposed to boosting..... Maybe its due to how the measurement is specified in the field.....

So I could be entirely wrong in all of the above thinking, but if its about right, it makes me wonder about the value of the horn,with its driver mounting difficulties when the drivers themselves should have the same spl at the low end, and full range on up....
Thats to say that simply mounting the four drivers on a flat baffle at the horn mouth and using the horn as a labyrinth, the low freq spl looks the same, and the high end might be smoother..?

Now using a much longer horn, and the same mouth area, its possible to get about a 10 db boost but with periodic peaks of about 6 db. I wonder, in this application, how bad that would sound. Are the peaks due to reflections and phase interaction? Is it worth persuing a sub optimum mouth size horn in order to achieve a usable response down to 200 Hz?

Max
 
uh oh. My mistake. Maybe I should have said offset driver location. Im not at all familiar with these things yet. Its just the bjorno placed a picture of a Unity indicating what could be possible and this threw me off...I guess the thing with the unity is that the mid drivers can be placed to load the horn at the right flare and yet maintain phase with the compression driver at the throat....or something like that, and HR doesnt model two different simultaneous driver locations. In that case I suppose you cant tell at all what mid band response you would end up without the rear driver in the mix....

thank you,
max

Max
 
Last edited:
Hi there MaxAxe: I would like to prpose a diffrent solution: a vented enclosure using 2 drivers, and for apperance ( audance reaction ) a horn section coupled to the enclosure vent. Some sound will exit the horn, near your ear and the enclosure will be housed in a form of a back pack with the drivers sound exiting from each side. This concept is the result of an experimental enclosure I constructed a couple years ago, and the parts are still gathering saw dust in the shop. If you are interested, you can contact me at ddroke@msn.com ...regards, Michael
 
I have done some subjective experimentation. I rigged up a 2 ft long horn, conical with an exponential mouth region, 8 inches diameter, and placed it over one of my diminutive 2 inch Peerless 830970. I compared it with a 6.5 inch Pioneer car speaker with integral tweeter in a large enclosure, my reference monitors, as well as a kind of nice bookshelf with a ribbon tweeter.

The horn was surprisingly good sounding and more "real" when playing woodwinds and brass, except the largest brass. It suffered with orchestra sounds. But it put out much more bass than I expected. It was louder and superior to the bookshelf for this application, and a trade-off with the car speaker. In some ways it sounded more realistic, as if there was a real instrument in the room, than the monitors. I will have to find my SPL meter and freq generator.

I do yearn for a bit more power handling on the low end as well as low end extension, which leads me back to the original question, how can you extend the freq response of an undersized horn. I cant use a backpack because Im always switching between my instruments, and Im usually in sort of classy resort lounges and whatnot. Better would be the the stand alone bass unit like bjorno's bazooka.

Im finding the visual parameters are stringent. The conical flare of the two foot horn gives it a slight witches hat look or something. Its big.

Does anyone have an idea of how to incorporate four of the peerless? Hornresp suggests a very small throat diameter with no speaker offset forward along the horn.

thanks,

Max
 
GM

"Anyway, no way to use a Unity concept to hide your actual instrument in, so not sure why you're interested in it"

I cant hide the actual instrument in any design, it has to go beside it, attached with outriggers.

Of the Unity someone says;

"arguably the first major evolution in horn technology in the last 60 years. Ultra clear reproduction through the high-mid band with extremely linear frequency response..... it minimizes, if not eliminates, the major phasing and lobing problems typically associated with your standard multiple horn 3-way box. By mechanically time-aligning the mid and High components it makes for a simpler to use system..."

I could make an interesting small PA with bjorno,s towers that would double as speaker stands...they look neat. That would not be for the instrument. but I thought about incorporating multiple peerless units housed in little brass domes...but I fear its beyond my scope. These are time consuming projects, what to say of the advanced engineering required. I would like to have the spirit of embracing the ideas that come up. In the end, we shall see what actual system I build.


Max
 
Agreed WRT the Unity's SQ, but even the 'small' 300 Hz Unity is quite heavy, so would require at least a unipod with a swivel base to support it and the 80 Hz Unity is so large, heavy that even ideally balanced may still be a distracting 'handful'. Then there's the issue of coverage angle..........

A M.I. horn has a much wider one than any of the Unitys, so a special one would have to be made and it would have a very large mouth which means more weight.

Bottom line, I may be missing the obvious, but to me, your goals seem completely at odds with the physics of the situation.

GM
 
GM

I wouldnt doubt that its me who is missing the obvious. Just to clarify a bit, and then you can set me aright; The primary goal was a trombone sized horn covering as much range as possible. I would attach this with some outriggers to my musical instrument.

I think this is feasible, and after my initial tests, I can see that a horn with 8 inch diam x 24 inch + or - size is adequate even without bass reinforcement.
The question becomes which driver is best suited, and I would like suggestions. Im considering the low end to be around 200 Hz. The low end could be extended with a free standing bass unit, or simply allow the main PA to fill in, as I can send a direct line from the synth to the mixer. So Im confident the goal can be achieved. This is coming soon, because I really need something.

A secondary idea presents itself, as I think Ive mentioned, which is to build a set of Unity horns and use those neat looking towers of bjorno to hold them up. Does that seem unrealistic? I would have to engineer legs, and no doubt its a significant project. We normally lift 50 lb speakers onto stands 6 feet high. Size wise, Im looking at dimensions of 16.5 x 16.5 x 14 for a unity going to 300 Hz. The Danley SH 100 is 20.5 in x 20.5 x 9. 45 lb and covers 70 Hz to 20k at 110 Deg, 123 SPL. (Is that really a horn?) This is a performance increase compared to say bose 802's which are similar in size at 13 x 20.5 x 12.5 and 31 lb. We often use those. Im not confident In results unless someone had a design worked out.

You mention dispersion. Would you be surprised that no musicians I know perform in stereo? I dont like it myself, but thats how it is. So with two horns, L and R angled out, you can get coverage, and in many rooms, limited coverage is more ideal. I used to pan our instruments, but in the mangled reverberant spaces you play, its futile.

What are your suggestions?

Thank you,
Max
 
Last edited:
Hi there Maxi: Good to hear you are experimenting, as DIY'ers normally do. Hope you will keep the community appraised of your progress, results and observations of the finished project. Yes Tom Danley's products are based on his patetened invention the Tapped Horn. You can read further about them by calling up Danley Sound Labs site and/or looking at the collaborative TH site on this forum. Note that a DIY plan for the Danley TH Spud is on this forum. ...regards, Michael
 
I do yearn for a bit more power handling on the low end as well as low end extension, which leads me back to the original question, how can you extend the freq response of an undersized horn.

Does anyone have an idea of how to incorporate four of the peerless? Hornresp suggests a very small throat diameter with no speaker offset forward along the horn.

You either add a bigger driver or more drivers which means a higher compression ratio (CR) for a given throat area which in turn means adding a filter chamber and phase plug, i.e. create a compression driver. For all but the most avid and/or knowledgeable, such a project is beyond the scope of DIYers. Ditto for designing/building a Unity concept horn.

GM
 
I think this is feasible, and after my initial tests, I can see that a horn with 8 inch diam x 24 inch + or - size is adequate even without bass reinforcement.

The question becomes which driver is best suited, and I would like suggestions.

The Danley SH 100 is 20.5 in x 20.5 x 9. 45 lb and covers 70 Hz to 20k at 110 Deg, 123 SPL. (Is that really a horn?)

Size wise, Im looking at dimensions of 16.5 x 16.5 x 14 for a unity going to 300 Hz.

Hmm, an 8" mouth in free space loads down to ~432 Hz and a 24" path-length combined with an 8" bell loads down to ~157 Hz, so from this we see that if sized/shaped like a bass trombone its F3 is going to be somewhere around 244 Hz if I did the math right and yet it plays down to ~55 Hz albeit at a very low SPL because of a human's ability to maintain a high back pressure against the horn's.

I assume you started with a ~1.6" diameter throat though and liked what you heard, so bumping up to a big, heavy 2" exit compression driver should get you down to the ~432 Hz mouth frequency at high SPL, but you're still going to need a large, efficient bass bin such as an old Altec 825 VOT cab to mate to it and get down to ~55 Hz Fb.

Of course you can try a Unity style conic horn with four Peerless 2" drivers positioned one on each wall, though FWIW I'd be surprised if they can handle the power abuse they're likely to 'feel' for any length of time. FWIW, my one experiment with dual 4" wide BW (~80-14 kHz) drivers on a 60 deg Unity concept horn didn't work as well as I thought it might, but 90 deg showed some promise. Unfortunately, 'life' got in the way and I haven't been able to follow up on it.

I was trying for a wider HF BW than you though, using a 1:1 CR, so may not be an issue if you roll off the ~5 kHz-up BW with whatever slope order it takes to keep the mids/lower treble clear sounding.

WRT the SH-100, this isn't a Unity concept horn, but a coaxial driver in a small vented cab with a nicely contoured baffle and as such its directivity won't be anywhere near 110 deg over its entire pass-band.

No, not surprised it's all mono and I thought we were dealing with a single horn, hence the concern over too narrow a dispersion. A M.I. horn has such a narrow angle expansion that they are near omni over much of their BW.

GM
 
d to the g,
Right, I like those horns, rotating or not.

GM,
Thanks for your input. I wonder how you derived 244 hz for the f3 on a trombone. Sorry you didnt get to pursue your Unity.

A single horn for the instrument, yes. Im not at all worried about dispersion with that. Two for a small PA, call it project B, or project #463 (a lot of projects) if you prefer. Then dispersion is an issue.

Anybody have a suggestion for a compression driver good for 200 HZ, moderate power of not more than say 40 watts? Im not very familiar with them.

No chance of approximating the required throat geometry for 4 peerless 2 inchers?

I wonder how well Hornresp models the geometry of the throat? Of course there is no phase plug...but you can input throat area and volume, and then conical sections.

Im not totally new to building, its just been awhile. We used a line array, 3 way system with rotating elements and a stuffed closed 15 inch box for the bass. It was the sweetest small system I have used. We didnt need monitors. I built a TL D-Apollito (?) layout with pretty nice components. The TL didnt really work out as well as expected, but it still sounds good. The gradual slope rolloff for the tweeters as he designed the crossover put a lot of stress on them. Its a sad thing to fry a nice tweeter.

I have experimented with a quite crude Karlson tube, and by jove, it works! Not as much low end as the horn, but more highs, and a more open sound. I wonder if the length of the slit should be equal to 1/4 wave, or what, and also the effect of the throat diameter. Its an interesting invention, and just a little enticing since its bio-mimetic. Nature has a good way with optimization....

Back to the original question, Interestingly, a rather long vent can coax more extension as modelled in Hornresp. At least you can trade horn length, and cut-off slope....

SCIENCE!

High regards to all readers

Max
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.