diyAudio (
-   Multi-Way (
-   -   Midbass horn disappointment (

Corvus corax 8th December 2012 04:09 AM

Midbass horn disappointment
I have been trying a number of things recently looking for a good match for my TAD 4002/ Stereolab horn system. Tired of the somewhat hooty upper bass presentation offered by my big 15 inch TL system, I decided to build the 100hz midbass horn designed by John Inlow, which looked very promising.

I took the annotated drawings posted on his site and proceeded to fill in the missing bits of information by drawing the system out full size to get my panel widths/etc., and put the whole thing together with 1/2 inch baltic birch, which apparently is not enough for the task. The resulting horns are extremely stiff, and relatively dead thanks to some 3/4 inch red oak bracing placed every 6 inches or so, added after John offered that the horns needed stiffening to perform as designed. I also cut back the 5 inches of extra width I had added to the last panels in an effort to extend the range of the horn.

Problem is: At the moment, this midbass horn has no midbass response to speak of. No slam, no punch, no… bass, really. Sounds great when the harpsichord is rocking, but the kick drums need not apply, nor the lower 2 strings of the bass guitar, either.

The woofer being used is the B&C 12PE32. The driver has an fs of 51 hz, a recommended sealed box volume of 1.1 cu ft, which will give a tuning of 85 hz. The chamber in this horn is 380 cu in, which becomes 190 cu in once the driver is installed. My best guestimate of the resulting chamber resonance is something on the order of 400 hz or so… Is this too high for a midbass horn? I can fabricate a simple baffle and measure the chamber resonance sans horn, but where should it be tuned on a 100 hz horn flare?

norman bates 8th December 2012 10:17 AM


I'd been told a few things on bass horns.
That driver should work fine.

But to me the bass horn seems a little small and a little shallow, even if the simulation says otherwise. 21" x 21" doesn't seem big enough.

Do I know better than horn response ? no, but there are a few rules out there that should not be forgotten. You can find formulas to find out the area of a mouth needed.

A 100hz horn (that I'd like to try someday), for enough mouth (sitting on concrete ideally), you'd need around 5ft2, or I think 27"x27". And ideally the horn length is 1/4 wavelength of 1/2 an octave lower than the mouth cutoff. So 100hz mouth would like 1/4 wavelength of 75hz, or 3.76' long (42").

And I imagine the knee is sharp, so there may seem like nothing below 100hz (no bass support). -3db @100hz with 18db rolloff would seem like no bass to my ears too. That'd be like crossing to a subwoofer but without a subwoofer connected.

And the rear chamber, that I can't help you with, but small is usually good, we arn't making a woofer box.

His 80hz horn looks better to me (26.5" x 26.5") but still a little short horn length to me at 34" deep.
I noticed that his simulation for the 80hz horn is 1pi, that would be on a concrete floor AND up against a concrete wall.
Maybe his simulation of the 100hz horn is in 1pi also................ Wiki - projects by fanatics, for fanatics

Take a klipsch Lascala.
a 24" x 24" mouth, with what, under 36" of length ?
It peaks at 250 and 140hz (meaning the mouth is too small causing ripples).
It has 120hz mouth and 100hz taper rate.
Below 120hz, it is acting like a woofer in a box (horn loading not helping).
It's -10db in klipsch's concrete perfect golden ratio room with an unknown amount of room gain to make that number.


DrBoar 8th December 2012 01:29 PM

With such a extremly low Q driver the horn will drop of well above 100 Hz, try an "inferior" driver something lika a plain vanilla Emnience driver with a Q of 0.5 or so.
Or try putting a 10-20 Ohm resistor in series with your driver to increase the Qes, and try if you like this better.

The Beyma I used (Doctor Boar; DIY loudspeakers and HiFi) is similar to your driver with regard to Q and Mms.

djn 8th December 2012 01:52 PM

2 Attachment(s)
Are you running subs??? Slam comes from the combination of a strong mid AND sub signals. Here are the HR and real plots for that horn. I can see how you would think there is no slam or punch.

I had the same disappointment when I first got my Edgar 80hz horns. I could hear the low-mid music, but it had no slam. I put in a couple subs and BINGO!!!

Corvus corax 9th December 2012 01:35 AM

I went ahed and dusted off the old Totem lightning sub (tiny, but reasonably capable in my small listening room), and the system has picked up quite a bit in the enjoyment department. I had originally built these horns longer, with an expanded mouth, but it was suggested that I shorten the horn back to the original profile, and the original mouth size- it was 24X24, but is now 20X20. It sounds a little more controlled now, but I guess the fast rolloff is what was getting me.

djn 9th December 2012 04:27 AM

I can see that Corvus. I just finished my speakers tonight and even though I installed two 4" x 10" long ports, I can still plug them with plumbers plugs and bring up the low end with subs. It will all depend on which one sounds better.

Corvus corax 9th December 2012 01:03 PM

Running the numbers using the formulas published by Edgar in his show horn article gave me completely different results with regards to mouth size, throat size, and length. I think perhaps, John was looking to optimize the impulse response of the system, which he mentions specifically several times on his site. THe main thing I was looking for with this component of my system was a good match to my Stereo Lab 250 hz horns, and I think this horn does just that on the top end- I can run the midbass without crossover and get a good level match with the main horn- the sensitivities seem to line up- on paper, anyway…

djn 9th December 2012 02:13 PM

If it mates well with the mid horn, that is great....half the battle really. The other half is the bottom end. Below is a pic of those horns in John's house. Notice the HUGE Tapped Horn sub behind his horns.

That is an interesting choice for driver. Normally a horn driver has a Qts of .6 or so. The B&C is .18 I wonder what John was thinking about when he chose that driver.

What crossover are you using? The MiniDSP might be a very good choice.

This is all very interesting to me as I was all set and ready to build John's 80hz horn when a pair of cabs fell in my lap. I like John's work.

pooge 9th December 2012 02:39 PM


Originally Posted by djn (Post 3277307)
That is an interesting choice for driver. Normally a horn driver has a Qts of .6 or so. The B&C is .18 I wonder what John was thinking about when he chose that driver.

I'm a bit perplexed by this statement. The formula 2*fs/Qes is often cited as a figure of merit for a horn driver's high end. In this case, Qes (which is close to Qt, should be as low as possible to extend the high frequency response an adequate range to get up to, say 500Hz. See Edgar\'s show horn article. Wondering if you are confusing a horn driver with a sealed box driver. Two different animals.

djn 9th December 2012 03:52 PM

I have now place to point for reference, it is just something I\'ve read in posts and talked to others about. As a matter of fact Bruce told me that when I asked him about what driver to use in his 80hz straight horn.

All times are GMT. The time now is 07:35 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2