@Camplo, this is an ancient, but still relevant paper on experiments with Tractrix horns.
Part I primarily deals with the mathematics.
Part I primarily deals with the mathematics.
Last edited:
In regards to the polar matching. A 15" at 630hz(potential xover) has a wide polar.....The Tractrix 350 at 630hz is going to have a slightly smaller spread....This is what we want right? The polars to be as close as possible? With a larger horn the polars at 630hz become increasing smaller....if I understand the philosophy correct, that is a step in the wrong direction.
Exactly. Yoyve got it right. But the disadvantage of the wide dispersion of the horn at the xo frequency is poorer loading often requiring eq and resulting increased excursion. Wouldn't work for pa, but for home it works. Keep in mind im using a 1" exit compression driver at high levels, your 2" should survive just fine.
Loading isn't an issue at home, I think coverage is.
According to 'deep horn afficionados' it is exactly the (nearly) perfect impedance matching that give these horns their distinct character.
The trade-off is limited directivity and considerable beaming, which often leads to using multiple horns.
According to 'deep horn afficionados' it is exactly the (nearly) perfect impedance matching that give these horns their distinct character.
The trade-off is limited directivity and considerable beaming, which often leads to using multiple horns.
So it does if I understand the math right, but 'we' apparently are missing something based on this Altec tech bulletin from way back when, which states that ka = 2 = half space, so in my math challenged, simple minded way chose long ago to interpret its formula as ka = 2*radius/SoS.
It goes on to state that a 15" ka = 1 = 317 Hz and since it halves/octave, ka = 2 = 634 Hz, so a major discrepancy.
All that said, using 900 Hz, k = ~3, which happens to be the half space directivity index [DI] with a Q = 2, so ka = DI, k = Q?
Regardless, no one has ever corrected me and FWIW, Dr. Geddes among others well versed in the 'art' of speaker design have [repeatedly in some cases] posted that half space [ka or DI?] for a 15" is ~900 Hz, which is of course ~frame size whereas these Altec charts are based on effective piston diameter, which BTW don't jive with a 15" frame's 12.75" - 12.875" effective dia./ Sd specs used when converting to T/S, so had to have been originally published in the 50s when drivers had smaller surrounds.
GM
OK, shows you how much I retain nowadays once posted.
'Sounds' like it needs [significant] damping, especially if you want it to be as 'tight' as a sealed alignment plus probably needs significant time/phase delay.
GM
Correct, you'll have to include the baffle area also since it's < ~900 Hz.
Hmm, a 350 Hz tractrix only loads down to ~350*[2]^0.5 = ~495 Hz, so basically has no LF power handling for a driver rated with a 500 Hz/2nd order minimum XO.
Of course for HIFI/HT apps you may not need it, but consider Altec's 500 Hz 511 expo horn that loads down to ~340 Hz with the then premium prosound 802 series compression driver rated 30 W @ 500 Hz/2nd order, which many [hard core] Altec fans don't like at all unless XO'd at > 700 Hz due to sounding harsh compared to the same driver on its smaller scaled 800 Hz 811 horn.
In short, even using the well proven half octave/2 slope orders doesn't guarantee a great performance and no doubt in my mind it's one of the reasons why the pioneers started out with 1st order overlaps [and then some in a few cases] for 2nd order XOs.
GM
Depends, see my 511 horn expose........
One of the major things I learned from the pioneers is that if the horn is smaller than whatever it mates to, then at least in the horizontal plane it needs some form of baffle to complete it from both a loading and polar response POV, though for super tweeters it need only be as acoustically large as the mids/HF horn's flat polar response.
GM
I gotta stop reading post before work....gonna make me late
GM heres where I sourced that formula http://www.gedlee.com/Papers/directivity.pdf
I much rather the Ka 2 be 900 btw, it would mean that the polar is that much wider for the 15" at 630hz, which per my design philosophy is better.
GM heres where I sourced that formula http://www.gedlee.com/Papers/directivity.pdf
I much rather the Ka 2 be 900 btw, it would mean that the polar is that much wider for the 15" at 630hz, which per my design philosophy is better.
I wish to know what this formula is? Whered did you get 350hz? Besides the name I mean? In the writing it says minimum freq is 520. When I use "~SoS/perimeter of mouth flange" I get 331hz...and "~SoS/4/D" gives me 260hz....I got those formulas from you but I may be using them wrong
I don’t post here much anymore as work has encompassed and fed my passion for DIY Audio but I’ll add this.......I’ve never given any concern for response above 15khz since testing the theory of full range operation 30 years ago. There’s simply nothing of use there in a closed space and better off without it as it will always be extremely uneven due to the immense amounts of reflected surfaces. It causes extreme ear fatigue pretty quickly. If you can get a 2” to play cleanly and flat to 12khz or so, sans the measurement game, you’ve won and would be better off in the real world listening environment UNLESS your space is extensively and professionally acoustically designed and treated accordingly.
Depends, see my 511 horn expose........
One of the major things I learned from the pioneers is that if the horn is smaller than whatever it mates to, then at least in the horizontal plane it needs some form of baffle to complete it from both a loading and polar response POV, though for super tweeters it need only be as acoustically large as the mids/HF horn's flat polar response.
GM
Agreed, smaller horns and specifically shallow waveguides 'need' a baffle.
This isn't necessary if the mouth rolls back.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
And I need to quit posting in the wee-wee hours, but then I'd rarely post and limit it to simple, short, repetitive stuff I can mostly copy/paste.
So for now, I'm sticking with Altec's understanding of polar response plotting.
It's a 1/2 octave spread calculation.
In scanning, I misunderstood what you wrote here, though the point I was trying to make is still valid, but will try to elaborate more as time permits since Tractrix is a special case and frankly, question their min. XO points unless modern drivers have at least a 4x greater usable excursion compared to the [vintage] Altec and similar drivers I'm familiar with: Is it possible to cover the whole spectrum, high spl, low distortion with a 2-way?
GM
Completely agree, although I may not have tested it to the extent of some others here, find that for music reproduction:
If you get good response to over 12khz it's better to just be happy with that instead of chasing rainbows.
My biggest problem with higher frequencies is that there's so much noise there, psu's, lamps, turbos on car engines, etc etc. Right now sitting in a place where the monitor is really loud, and the sensors for the fire alarm system must have some heavy noise on the bus because they are LOUD!
Even if one piece of electronics does not sound noisy, it can interfere with other equipment and make noise there instead.
It's a nightmare! Thankfully I do not hear so high up anymore, still wondering why the monitor is bugging me since it makes noise at 19khz and little else. Maybe I'll have to review my listening test scheme...
At any rate, electronics noise is really annoying!
And then you get mirror reflections from hard surfaces, so it can be really hard to figure out what makes the sound.
Sorry about the rant.
I need to study more, but I think the tractrix 200 model, might be the go to, it changes the polar scheme but raising the xover on the 15"might of been a good idea anyway.
I've considered a ported box instead, seems that excursion is lowered with sacrifice of group delay among other things, but for this design, the lowered excursion might out weight the benefits of a sealed enclosure. It seems with dual AE 15m's I can just touch loudness goals at the limits of the excursion. I'm hoping that I have completely over shot real life expectations, in the loudness department, which was the goal in the first place, leaving me with even less excursion at reasonable volumes, application dependent. My current 12's are 200rms, with a claimed "114 dB Music; 117 dB Peak"...thats measured at who knows what distance, I assume 1 meter. Dual 15m's in a ported enclosure can do 115db @ 280watts putting the excursion right at its limits at 56hz (according to modeling), but who in their right mind wants to hear 56hz at 121db....think I just sold myself on that woofer. With eq I don't think I'd have to give the speakers more than 300watts ever...
Thanks again for suggestions, I still am trying to figure out the best horn tractix size (350 vs 200)
Heres the parts list.
JBL 2451
JBL D16R2451SL Diaphragm
Acoustic Elegance 15m
CTS1200 for woofers
Tractrix 350 or 200?
Still searching for another amp and active eq solutions
LAB GRUPPEN IPD 1200 DSP Amplifier is on the watch list to power the CD
Almost there
I've considered a ported box instead, seems that excursion is lowered with sacrifice of group delay among other things, but for this design, the lowered excursion might out weight the benefits of a sealed enclosure. It seems with dual AE 15m's I can just touch loudness goals at the limits of the excursion. I'm hoping that I have completely over shot real life expectations, in the loudness department, which was the goal in the first place, leaving me with even less excursion at reasonable volumes, application dependent. My current 12's are 200rms, with a claimed "114 dB Music; 117 dB Peak"...thats measured at who knows what distance, I assume 1 meter. Dual 15m's in a ported enclosure can do 115db @ 280watts putting the excursion right at its limits at 56hz (according to modeling), but who in their right mind wants to hear 56hz at 121db....think I just sold myself on that woofer. With eq I don't think I'd have to give the speakers more than 300watts ever...
Thanks again for suggestions, I still am trying to figure out the best horn tractix size (350 vs 200)
Heres the parts list.
JBL 2451
JBL D16R2451SL Diaphragm
Acoustic Elegance 15m
CTS1200 for woofers
Tractrix 350 or 200?
Still searching for another amp and active eq solutions
LAB GRUPPEN IPD 1200 DSP Amplifier is on the watch list to power the CD
Almost there
@ 30hzs that is...So the 12's theoretically would be down 3 or more at by 30hz, pretty sure the f3 on those is 32hz. The power for the 15m's will be continuous throughout their usable range, I'm hoping it is a significant increase over the current subs. Interesting enough, there is a huge bubble of output possible in between 30-40hz for the dual 15m set up, within xmax limits., thanks to port tuning.