18 Hz Horn Sub (Simulation)

[Hoerninger]

In this forum a low frequency horn sub has been mentioned, I have simulated it with a JBL 2226 H. (Hornresp by David McBean)

It goes down to 18 Hz (-10 dB) and is very flat until 22 Hz (-3 dB). It is simulated for use in a corner but placing in front of a wall can be compensated by slightly increasing the bass by tone control.

It is huge, 2134 liter netto, about 2600 liter brutto. Living in a house for your own it will perhaps give you a corner place of 1 square meter with a height of 2,70 m.

The horn length is 5,8 m and the horn mouth is 12200 cm*cm, a bit more than one square meter.

Efficiency is about 108 dB/ W /m, max SPL with 600 W at 100 Hz will be about 137 dB.

Omitting the 200 liter (!) rear chamber volume will produce irregularities in SPL. Increasing throat chamber volume (in front of the cone, same area) will make the high frequency roll off steeper - due to cone excursion it can't be made smaller than 1000 cubic centimeter (1 liter).

The pictures show you in detail:
1. Input parameters (JBL 2226 H)
2. Overview (Schematic diagramm)
3. SPL responce (corner/ 0,5 x Pi)
4. SPL response (wall/1,0 x Pi)
5. Electrical impedance
___________
Peter

PS: The shown SPLs are 8 dB too low.

[Hoerninger]

6. Diaphragma displacement with 1 W /8 Ohm
7. maximum SPL

[Steve Schell]

Hi Hoerninger,

At a glance, your response curve simulations look a lot like the measured response of my large horn subs. The main difference is that my subs tend to roll off faster on the top end, usually beginning to fall by 80Hz. These subs use a 15" driver, have a 16 cubic foot back chamber and a 15Hz. exponential flare (50" doubling of cross sectional area) of 18' to 20' path length. The measured response has an 108dB average sensitivity at one watt from 20 to 80Hz. measured at the box exit.

According to a published equation by Frank Massa (RCA, 1930s), the horn subs I have built should be able to achieve one acoustic watt of output at 20Hz. with .080" of cone motion from a 15" driver. The Altec 515-16G driver I have been using most of the time is rated for .150" peak to peak cone motion, so this is well within it's capabilities.

I settled on a 16 cubic foot (!) back chamber mostly because reducing the volume from this figure began moving the lowest response peak upwards in frequency. I'm not sure how much reactance annulling is taking place, but this back chamber does serve to deliver the most SPL at low frequencies from the horn while keeping the energy off the back of the driver from interfering with the horn's response in the room.

Horn loading can be an amazing thing when exploited fully... over 105dB in room at 20Hz. from a single 515!

[yggdrasil]

What is your thoughts and experience with time alignment of the large bass horns?

The driver in a 5.8 meter horn is approximately 18 millisseconds further away from the listener than a direct radiator.

[Hoerninger]

What is your thoughts and experience with time alignment ...

... some thoughts:
My consideration starts with phase shift. 5,8 m corresponds to 180 degrees at ca. 30 Hz. 180 degree phase shift can simply be compensated by polarity switch, but 30 Hz as upper frequency limit is very low. I would prefer ca. 80 Hz as Steve does. (Then there is no need for a 15 inch driver for the mains, may be a 10 inch driver can be used.)

If I would dare to build such a horn sub I would seriously think over electronic time delay for the mains. Because of the low deviding frequency there will be an electronic dividing network anyway.

... some experience:
With Paragon XXL there is a difference in path lenght of about 1,4 m corresponding to 122 Hz and 180 degree phase shift. Although the deviding frequency is with 500 Hz much higher, I have never observed inhomogenities in sound stage. We even used it for support of the chorus with male and female voices and the choir leader was very satisfied.
(But I admit I have never tested it out like a hifi system at home.)

My thoughts and experiences are contradictory. In reality I would start with a 6 dB/oct deviding network with some Rs and Cs at the amp inputs and listen what will happen.
___________
Peter

[yggdrasil]

Thanks Peter.

Some background....

I grew up with horns and love the explosive nature. I miss that in my primary system today. On the other side, I don't want to give up the low frequency response of my 4-ways with the 2245 at the bottom end.

Now an opportunity to establish a third large system has come up... A small construction flaw at our cabin has led water into the south wall, and we will have to redo it. When doing so, we want to add a new rest room for me and my wife. In a weak moment she gave in and gave permission for something approximately the size of my 4-ways. They are a little smaller than the 4345 (not as wide, but a bit deeper).

I realize that 2 sub horns will be considerably larger, but still in the same "class" .

Last week I got a 2-way 250Hz electronic filter finished and installed (DIY). I got a lot more headroom from my amps. First impression was that I got more tired with the new filter installed, but then I realized that the listening level was up around 10dB. These filters have the possibility to plug in all-pass sections with both +180/+360 and -180/-360 degrees sections at selectable frequency. Then -180 degrees at 30Hz on all the upper drivers should do the trick.

Anyway the project is at least one year away from realization. Have some parts waiting for a project already. Have to build a chassis for the electronic filter, then probably a Zen v9 for the upper drivers on my 4-way and a more powerful amp for the 2245's.

So many projects, so little time.... So much fun.