But geometrically on a wider horn the mid drivers can get closer to the throat. The center of the mid, if the mid's edge is up to the throat is closer (as measured along the axis of the horn). I've always seemed to need to get the mid port as close as I can. (wanting to cross it as HIGH as possible).
Patrick,
Good to see you found the reason for your gross simulation errors, I was beginning to think you had inhaled a bit to many fiberglass fumes
.1) Yes, the far larger narrow dispersion horn will have more on axis output.
Some folks, like yourself, don't want a four foot deep cabinet.
2) Yes, given the same output requirements, the HF on the far larger horn could be crossed lower, as could a far larger wide dispersion horn.
3)A narrow angle horn does not allow the mids to play any higher, and anything from 50 x 50 to a 100 x 100 on a 28" or so square exit will allow a typical 1" exit driver to play loud enough to cause hearing damage in the 800 Hz range in the typical home listening environment.
An 8" cone driver can easily reach 800 Hz in an offset horn of the same dimension.
Looks like you have some extra HF drivers arriving soon
.Art
just as a quick reality check: in my implementation of the unity concept, i had to pad the bms 1inch compression driver down about 12dB to equal the 4x 6inch mids. Crossover is 1khz, mid entry points are about 10cm away from the compression driver exit, 45 by 45 degree angle, 2500 square cm mouth. these are really crazy loud for home cinema, making the power output of the behringer A500, which i use to power the subs, the limiting variable of the system by far.
Hey MaVo
Out of curiosity:
Which compression ratio do you use for the 6"ers (i.e. the midrange port area?)
/Thomas
two ports per driver, both 8mm diameter inside the horn and maybe 3-4cm diameter on the driver facing side, making it a round conical column of air. i think i could have made them bigger on the horn inside without affecting the sound quality of the compression driver, but this works too, at least up to home cinema reference level. (never listened louder.)
Edit: cone area 100cm² per driver. wood 15mm thick.
Edit: cone area 100cm² per driver. wood 15mm thick.
Last edited:
Hi Patrick Bateman,
Would it be worth considering lowering the Vas limit in Hornresp, or is this just a "one-off" compression driver design exercise on your part?
Kind regards,
David
This means that you're using a compression ratio of 25/1; or does the conical shape somehow compensate for the relatively small exit holes?
Harman has a great paper on loudspeaker ports, and one of the things noted in the paper was that bowtie shaped ports performed very well.
So basically a port that's shaped like this:
An externally hosted image should be here but it was not working when we last tested it.
works better than a port that's shaped like this:
An externally hosted image should be here but it was not working when we last tested it.
Now a horn is not a port, of course. But this is still air we're talking about, and if you think about the air path in the horn, it's basically a giant version of that bowtie shape above.
IE:
big volume of air ---> necks down to small volume of air ---> expands to giant volume of air
big - small - big
Also, they found that the 'bowtie' shape shouldn't be symmetric. It should neck DOWN faster than it necks back UP. And if you think about it, that's what the frustum is doing, as it feeds the conical horn.
http://www.aes.org/tmpFiles/elib/20130415/11094.pdf
"When designing a port for maximum acoustical output,
both the inlet and the exit fluid dynamics should be
balanced. The geometry for best exit flow is different from
that for inlet flow. Inlet flow is best with a very large taper
(NFR close to 1.0). For exit flow a very slow taper is best
(NFR closer to 0). This points to an NFR of 0.5 as the optimum."
I stumbled across this paper while trying to reverse engineer (what else) another Danley box: http://www.diyaudio.com/forums/subwoofers/226235-featherweight-title-fight.html#post3296900
Last edited:
Hi JLH,
The Vas minimum limit will be lowered to 0.001 litres (1 cubic centimetre) in the next release. I'm not sure though, how appropriate it really is to be describing a compression driver in terms of Thiele-Small parameters
.Unfortunately, changing all the input parameter fields on the main screen to three decimal places is not a practical proposition at this stage.
Kind regards,
David
Any explanation as to why the "1/4 wave reflection notch" is up higher in frequency than expected?
Maybe because the reflected sound has a wavelength much bigger than the 1" horn throat? Will the reflection occur at a certain area as related to wavelength?
If the throat is tiny, how far up the horn does a pressure wave travel before it reflects back towards the mouth?
The update has now been released.
http://www.diyaudio.com/forums/subwoofers/119854-hornresp-334.html#post3457517
Kind regards,
David
Increased flare angle
A question about the increased flare angle on the last ~1/3 of the Synergy horn:
How large is the angle increase, it does not this seem to be a fixed angle ratio? - What I mean is: some designs employ a 50% increase, others much less. Is it more a trial and error thing?
/Thomas
A question about the increased flare angle on the last ~1/3 of the Synergy horn:
How large is the angle increase, it does not this seem to be a fixed angle ratio? - What I mean is: some designs employ a 50% increase, others much less. Is it more a trial and error thing?
/Thomas
Last edited:
Trial and testing, available mouth space, horn depth, all play into the design, but as Don Keele wrote back in his AES paper "What's so sacred about exponential horns", approximately the last third of the conical section is roughly doubled.
So a 25 degree horn goes to around 50, a 60 to around 120, but as the main angle increases, the "break" angle does not need to go as wide, a 90 would not need to go to 180.
The additional angle does end up making the conical more exponential at low frequencies, increasing output over a conical.
