Mid-bass FLH design considerations???

[NEO Dan]

I've got the itch to take a crack at my first mid-bass horn and after reading a few threads I've got a few general questions. Personally the system I have in mind for myself is a single 15" or 12" FLH mating to a large 4722HF ScreenArray wave guide in the ~700Hz neighborhood. CD should be either the BMS 4555 or FaitalPRO HF144.

Is 1Π really reasonable, I see a few Hornresp sims are done like this?
What is a reasonable compression ratio when driver excursion is low, say within ±3-5mm WRT 12", 15", 18" drivers?
Is taking the time to be very accurate with simulation worthwhile?
Is it reasonable to have a flat bottom like so?:


Any general guidance to put me on the right track would be greatly appreciated.

[m R g S r]

did you get my pm's?

[NEO Dan]

U got PM's

[NEO Dan]

Am I in the wrong forum, maybe horns are filed under exotic?

[revintage]

No problem with even two flat sides. I suggest bottom and the side outwards to the sidewall.

Don´t mind excursion at all!

1,0 *Pi will mostly do.

You will have very hard to reach 700Hz with 12" or 15".

Also take a few hours to learn Hornresp, it is an invaluable tool.

Probably noone answers as this thread is so close to the two other previously published.

Just doing this out of compassion, can´t help you anymore.

[NEO Dan]

Thanks for the reply Lars,
I'd like to have a thread of general mid-bass horn design info. It would be great if anyone with basic skills could have the guidelines for design.

I'm interested in info like what happens when your horn is loaded 2 Pi in the mids and 1 Pi in the lows? How do you manage the transition in the design?

I've got lots of seat time on Hornresp and the pencil sketch above is a stylized CAD rendering.

[John Sheerin]

How would you make the radiation angle change with frequency? Obviously the pattern will narrow with a non-cd horn as frequency increases, but that will be accounted for if you look at the on-axis response in hornresp.

1*pi is reasonable if your horn mouth is against a floor and wall. If it's just on the floor, it's probably not reasonable - depends on how far it is from the wall.

Compression drivers have compression ratios around 10. I've built bass horns with compression ratios that high as well, but if your cone is not very stiff and you drive it to high levels, you can damage the cone. 4-5 is a more typical limit for midbass horns with large drivers with light cones.

You might have a tough time matching up of the directivities of the midbass horn and the midrange horn if you do what you're proposing (700hz crossover).

As far as 'gathering all information on x' in a thread, I'd suggest not bothering. There's already tons of stuff posted here if you search and read enough. Instead start designing. As you encounter issues, ask specific questions. You're more likely to get answers that way. Then redesign as appropriate.

[NEO Dan]

Thanks for the reply John,
Although I don't know what an ideal polar map looks like I figured on using the Hornresp to look at what should be happening with the design.


Maybe someone has a good looking polar of a 90 degree WG they could throw up as a example?

I also was thinking that 1Pi would start where I couple to the rear wall, so whenever I'm within the distance of X there should be a transition.


At any rate a collection of do's and dont's with explanations and related Hornresp screens and other illustrations would be a great resource for someone getting started.

[John Sheerin]

You can find the beamwidth data for the HF waveguide mentioned in the JBL data sheet. You just want to match up to that around the xover point. The hornresp polar map you show isn't very useful since it's not normalized to the frequency response (which I don't think is possible in hornresp). However you can look at the beamwidth plot or look at the polar pattern at a discrete frequency which will also tell you the DI which you can match up to the JBL data around the crossover region.

[NEO Dan]

Thanks John,
I had not looked at beam width:

Back to the drawing board