diyAudio (
-   PA Systems (
-   -   How much db a 90db driver can gain using a Horn? (

rhythmsandy 11th December 2012 07:29 PM

How much db a 90db driver can gain using a Horn?
im about to build a horn speaker for a theatre but would like to know how much spl will a horn gain using a 91db 6.5 inch midbass driver?

JMFahey 11th December 2012 09:20 PM

Short answer: replace it with a 12" 96/97dB speaker ;)
Even in a shoebox it will be a better option.
And nothing stops you from building a *good* 12" speaker cabinet, of course :)

bear 11th December 2012 09:34 PM

It doesn't quite work that way.

The horn loading is most effective lowest in the range. At higher frequencies the horn tends to go away and the driver beams, more or less like a direct radiator...

So if you did the freq response of a driver that was flat in a horn without the horn it would likely look more or less like a rising response... generally speaking...

Otoh, if you can create some compression, there may well be some gain mid band... you can simulate this stuff using hornresp program and see what does what.

One of the things that a horn does well in a PA/SR situation is to provide directivity.

Directivity is good to get the sound where you want it and not on the walls, ceilings and back into mics...

You might want to look at the paraline thread, that's a good fast, cheap way to make a very high performance PA speaker, imo. You will still need some 12" PA speakers to round it out, but you'll need them anyhow before ur done.

Ain't no free lunch, sorry to say.


rhythmsandy 12th December 2012 02:34 AM

im thinking about very fast drivers which has super low inductance as low as 0.08mH which in general can overcome beaming. I have used Hornresp software to simulate for which just with 1w i got 105db sensitivity and there are very few +- 1db ups and downs apart from that it looks damn good.

But i have a question that I want to use good quality drivers like vifa or scanspeak do you think it will work in the horn?

I could still go with pa drivers 12 inch ones but never liked its sound quality.

btw what mid drivers that I max big screen LCR uses?

rhythmsandy 12th December 2012 03:40 AM

can i use array of 20 6.5 inch drivers? to reach target spl?

bear 12th December 2012 03:49 AM

You need to define your application clearly.

I have no clue what you really are trying to accomplish from your OP... so there is no way to give a good answer.


chris661 14th December 2012 09:50 PM


Originally Posted by rhythmsandy (
im thinking about very fast drivers which has super low inductance as low as 0.08mH which in general can overcome beaming.

Who told you that'd stop beaming?


Kaufmancd 14th December 2012 10:13 PM


Originally Posted by rhythmsandy (
can i use array of 20 6.5 inch drivers? to reach target spl?

A series/parallel line array of 9 drivers will maintain the original driver's impedence. That is, if the 6.5" driver is an 8 ohm then three parallel banks of three series connected drivers will give the amp an 8 ohm load.

With that said, the sensitivity gain for a 9 driver line will add about 9 dB. so your equivalent array will have a 99 dB sensitivity. You'll need a good sized amp to drive them, but you should be able to get plenty loud with them.

Another advantage of the line array is the vertical beaming, meaning more of your sound energy will reach farther into the room. A vertical line array doesn't hurt the horizontal dispersion at all.

It's a great way to build if you've got a bunch of the 6.5" drivers.

JMFahey 15th December 2012 11:52 AM

The problem with simulations is that they sometimes assume certain things which actually don't happen in the real world.
Couldn't download Hornresp ; Mr Mc Bean's site seems to have some problem, but anyway reading that it's based on Olson's horn model, gives me a clue.
If that's so, yes, Olson's model is excellent, but it assumes you have a planar, perfect audio wave at the throat, with perfect frequency response.
Simply slamming a regular speaker at a horn's throat provides none of that.

So the horn will do its magic, no doubt about that, but only for the frequencies for which that speaker can more or less approach those conditions.
1) you don't have a flat cone, but a concave conical one.
Supposing it were a perfectly rigid piston (which it is not) "sound from the edges would reach the horn mouth earlier than sound from the center".
Now, at low frequencies, you can assume that the path length difference is small, compared to the full wavelength, so it works reasonably well.
But at higher frequencies this does not hold true.
Just measure the cone depth and calculate the frequency at which it indicates 1/4 wavelength, and there you have the maximum frequency that horn will reasonably help.
2) the cone is not perfectly rigid at all.
At low frequencies you can consider it so, but not above them.
So your effective piston diameter will become smaller, while the horn throat diameter stays the same.
Coupling decreases dramatically.
Sound comes straight out of the (now smaller) cone, without caring about the horn.
Which is exactly what bear said.
So in practice, a horn coupled to a simple speaker (not a horn driver which is a very different thing, with phase plugs and such), will help it at lower frequencies, nowhere else.
And since horn size is usually limited (unless it's a fixed installation and you have unlimited space), there is also a low end limit too.
So in practice, and that's the experience of all of us who built and used them, they usually help in mid bass situations.
And as bear also said, the main advantage lies in providing directivity, which helps both in open air events, and in very reverberant closed rooms (think basketball stadiums and such).
So what can you expect from horn loading a small driver in a reasonable sized horn?
Bsically, same sound as before, but with somewhat louder midbass.
Will it match an efficient regular PA cabinet?
Doubt so.
Will it provide 105dB/1W/1M ?
Sorry, but no way.
Not even in a narrow band. Sorry.
But of course, if you want to try it, go ahead.
This is DIY Audio. after all :)

PS: just in case it's not clear: I'm sure the program is fine. It's just that a simple speaker does not properly fulfill the driver role it expects.
As a side note: what I read about it brings warm memories.
When I started studying Engineering (1969), I also programmed in Fortran, punched the program on paper tape (much cheaper than cards) , and run the program in an IBM 1620 such as this one:

DrBoar 29th December 2012 08:35 PM

A horn is an aid not a miracle device (unless you buy it in TV-Shop) Hornloading a 90 db device will increase output and directivity but a pro device with a magnet the size of a bowling ball and 110dB/watt is something else alltogether. Horn modeling has its limitations I have measured output linear up to 1200 Hz with drivers that should have a mass rolloff above 150 Hz, so some of the asumptions in models with rigid cones and so on does deviate a lot from IRL.

All times are GMT. The time now is 02:56 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio