Tapped Horn Directivity

[weltersys]

There have been various claims regarding the directivity of tapped horns, but since I had not seen any tests done comparing them to bass reflex or standard horns, I did some testing myself.
As I had suspected, TH in themselves have little more directivity than other types of cabinets below around 60 Hz, but do become more narrow in dispersion above.

The LF directivity others have noted when TH are used in multiples comes primarily from the increased frontal area acting as a plane, transitioning from 1/2 space to 1/4 space. This same effect happens to some extent with any type of speaker.

The BR is basically omnidirectional within about 1 dB to 60 Hz, then looses about 3 dB at 125, and around 7 dB at 160 Hz.

The TH is basically omnidirectional within about 2 dB to 60 Hz, then looses about 12 dB at 125, and around 9 dB at 160 Hz.

The C horn is basically omnidirectional within about 3 dB to 60 Hz, then looses about 7 dB at 125, and above.

The TH with the extender shows the advantage of a large frontal area, it has directivity all the way down to below 30 Hz. At 60 Hz, it has around 7 dB difference, and 125 Hz, a bit over 14 dB. A huge directionality increase from a sheet of plywood.

The BR cabinet and the TH both used the same 18” speakers, the C horn a 15”.
The C horn is a spiral horn about 16.5 meters long, the TH is also a spiral around 3 meters long.

Tests were done with the microphone 10 meters from the front of the cabinet, the cabinet was rotated around the center of the cabinet front so the distance from the exit to the microphone was always 10 meters.
Tests were made on axis, 45, 90,135, and 180 degrees off axis, 180 off axis the cabinet is pointing away from the microphone.
In the case of the TH with the horn extender, the extra width of the cabinet would make the length about 1/3 of a meter longer when at 180 degrees than the other cabinets.
Cabinet dimensions are height, width, depth in inches:
BR 36 x 24 x 17.5
C Horn 26.5 x 22.5 x 30
TH 45 x 26.5 x 22.5
TH with extender 56.25 x 53 x 37.5
The extender is a simple 90 degree “wave-guide”, 45 degree each side, parallel top and bottom. A dummy cabinet was placed above the TH cabinet to cover the “hole”.

As it turns out, buildings and fences, even though more than 40 feet from the cabinet and test microphone had some fairly significant effects on the test, in most cases the 135 degree results were lower than the 180 degree, odd to say the least. A pump house (10 x 8 x 8 feet), located about 10 meters from the test microphone makes the response look “lumpy”, the cabinets all have quite smooth response when tested at 2 or 3 meters.

The surroundings make the tests flawed, but the flaws apply to all the cabinets.

Art

[Djim]

Very nice findings and good side info, thanx...

I guess a plane (non expanding) extension will work also but maybe less below a certain freq.

[weltersys]

Originally Posted by Djim

Very nice findings and good side info, thanx...

I guess a plane (non expanding) extension will work also but maybe less below a certain freq.

One other finding in the course of over an alphabet's worth of different mouth shapes and trying various corner ramps was that adding corner ramps just changed some minor frequency dips and peaks in frequency.

I ended up using no corner ramps at all in the Keystone TH design.

Art

[Djim]

Lol, Art it looks like this corner "thingy" is going to hunt me forever around here...

Anyways, they can have a more effective change when the corner is situated on the right spot. That’s why findings are so different. But sure they can give changes (even its minor) in the higher regions of the band pass. But still, I prefer to see them more as corner corrective tools instead of anything else...

Btw can you produce 360 degree graphs of your directivity findings?

[boydon_lepasci]

Hey Art, not to be a choosy beggar, but is there a writeup of the "Keystone" forthcoming? Your comments regarding reflectors and mouth geometry hint at volumes of background info.

[weltersys]

Originally Posted by Djim

Lol, Art it looks like this corner "thingy" is going to hunt me forever around here...

Anyways, they can have a more effective change when the corner is situated on the right spot. That’s why findings are so different. But sure they can give changes (even its minor) in the higher regions of the band pass. But still, I prefer to see them more as corner corrective tools instead of anything else...

Btw can you produce 360 degree graphs of your directivity findings?

The Smaart graphs I posted covered 180 degrees, in retrospect rotating the speakers the opposite direction as well (for 360 degrees) would have given some more useful data points, since the test environment was not symmetrical. I suspect readings at 225 dergrees, pointing more away from the backyard fence would show a greater reduction than the 135 and 180 graphs.

That said, with the flawed test environment, I don't think I'll bother trying for more directivity tests.

I am still a bit amazed how much effect the distant surroundings had on the findings, looks like when the test microphone is 10 meters from the sub, 30 meters or more to buildings may be needed for test results to be valid.

[weltersys]

Originally Posted by boydon_lepasci

Hey Art, not to be a choosy beggar, but is there a writeup of the "Keystone" forthcoming? Your comments regarding reflectors and mouth geometry hint at volumes of background info.

I would not know where to start that book, not including the directivity tests, several weeks of sims with Hornresp, I conducted tests on at least 60 actual (plywood, sawdust, and lots of clamps) versions of the Keystone sub.

There are still a few more to do...

The funny thing (ha ha) is the design basically ended up being one I drew on the back of an envelope before downloading Hornresp.

[JLH]

Originally Posted by weltersys

The funny thing (ha ha) is the design basically ended up being one I drew on the back of an envelope before downloading Hornresp.

I've had this very thing happen to me before. I let my analytical mind lead me away from my initial concept only to end right back where I started. Its the darnest thing. LOL.

[jbell]

I finally had a somewhat calm warm day to do a bit of testing, so I could add in directivity tests for the SS15.

The interesting numbers are what happens to directivity as you lay it on it's side. (the 50hz reading is not a typo, I did it a couple times, just to make sure....)

This is single cabinet, sine waves, 10 meter test, over 100' from any buildings, etc... (not calibrated to 28v or anything, just turned up enough to test)

[weltersys]

Originally Posted by jbell

I finally had a somewhat calm warm day to do a bit of testing, so I could add in directivity tests for the SS15.

The interesting numbers are what happens to directivity as you lay it on it's side. (the 50hz reading is not a typo, I did it a couple times, just to make sure....)

This is single cabinet, sine waves, 10 meter test, over 100' from any buildings, etc... (not calibrated to 28v or anything, just turned up enough to test)

Jim,

Looks like the SS15 couples to the ground plane a bit better on it’s side at 50 Hz, but the 180 degree off axis result (-6 dB) is the same in either upright or on it’s side.

I find it interesting that your measured 50 Hz SS15 directivity is almost twice the Keystone’s, while the 75 and 100 Hz are more similar.
Also of interest, at 100 Hz upright, your 90 and 180 degree levels are the same.

I would have expected the larger front panel of the Keystone to provide more directivity than the SS15.

It would be interesting to look at the SS15 phase response at 50 Hz on axis and 180 degrees off, might give some insight to why it has so much directivity that low.

Unfortunately, I can’t test 100 feet away from buildings on my property.
The buildings affected my results quite a bit, so it is impossible to say if the differences between our 50 Hz measurements are due to local surroundings, or the cabinets.

Thanks for the test results !

Art