Horn Speakers - Is it me or.......

[kyle_neuron]

It seems to me that the internal geometry (like a tall pyramid) would result in the mid and low frequency ranges having local pressure maxima at the apex, where the compression driver's diaphragm is located. Does that happen; and if so does is it of any consequence; and if so does that constrain your compression driver choices?

I can maybe impart some ‘user experience’ here on that topic before Tom gets in with the real science

A few years back I went out to the Brazilian Danley dealer with a friend from the States to investigate a recurring issue with them losing HF drivers in their bigger Jericho and Synergy horns. It was also a good excuse for a working holiday but that’s another matter… anyway, at 100% import tax, losing an Italian premium HF diaphragm is a costly affair, even more so when there are four per box. I flew sixteen spares in my luggage to stock them up… The shows these guys do run 24 hours for a solid week, full blast (and I mean face meltingly loud) in the sun, wind and dust.

Anyway, one such theory was the same you suggest here. We wondered if the sheer amount of rolling 180 BPM+ low end (psychedelic trance, don’t ask!) coming off the six 15” drivers in their Jericho could be putting tons of extra stress on the diaphragms. So we tested it, with a scope attached to the otherwise unconnected and unamplified diaphragm terminals, hammering the crap out of the rest of the box. We even put the other Jericho speaker and eight subs against the mouth of the horn, and tried that. There was evidence of slight movement, but certainly nothing that would suggest the diaphragms would be forced to slam into the phase plug even when running hard.

I’m guessing the loading down the throat end is sufficiently spread out on the bigger boxes to help alleviate that, and the acoustic or physical filtering approaches the design is built around also work ‘backwards’.

Remember there are a ton of cone midrange with their own taps, and those likely spread out some of the pressure too. It’s something that could maybe benefit from a swanky COMSOL model to verify further, but the field testing seemed to answer the question for me.

Either way we resolved their issues with some adjusted DSP settings, a much more rigorous maintenance & cleaning policy, some extra scrim or mesh across the gaps, and a general guide to bring more rig for the gig - putting out delay lines at 40 metres, for example.

View attachment 168973Is that stuff for real ? i was looking at compression drivers to learn what they are about a bit more . Lots of manufacturers seem to list the power in the " testing tube "
but this is on a 40X20 horn ? how can that little driver still push 105db at 300hz with only 1W ??

Do most compression drivers have this kind of compromise ? this one seems to drop quite fast before 10khz
but goes very low for its size , most of the smaller ones drop real fast under 1KHZ but most go up to 30khz iirc

I am totally impressed by this graph .lol

It is for real, but typically the drivers don't sound so hot at 300 Hz. Not the ripple on the green trace, a sign of breakup modes. Having a driver which can do 300 Hz in the lab test condition is a good indicator that you'll get decent results to 600 Hz, for example.

A plane wave tube is a common test because it's a known construct with an analytical solution; you can more readily expand the PWT results of different drivers to a horn of your own, than test data taken on different horns with different coverage patterns and acoustic loading.

That particular BMS is the midrange version I believe. There's a variant which is the full coaxial, using two diaphragms. They're good, but there's a narrow resonance in the 'vanilla' versions which I personally find unpleasant. It can't be EQ'd out either. The later HE versions cost more cash, but massively reduce that issue.

If you're looking at high end 'pro'ring radiators, then may I also suggest the B&C DCX464 and Faital Pro HF1440? The former has exceptional mid-range performance, where the latter has the smoothest HF detail of this type - in my subjective experience. I think the Celestion Axi2050 is already known here, even if it's not really the same 'thing'. All of them benefit from a very optimised horn.

 

[JeanKazamer]

It is for real, but typically the drivers don't sound so hot at 300 Hz.

Good info mate, appreciated . This one graph was mounted on a horn though .. Kinda deep down pressure there !

I have seen the DCX464 in some reviews and tubes and it seems to be a great driver . The waterfall plots are quite good also.

Items about: Voice Coil Test Bench

audioxpress.com

I am using their testbench section , quickly reviewing most of their tests to try and understand how different driver designs and compromises performs . Quite interesting for a newbie such as me, getting loads of into in no time from all those and fun to read

Also they seem to be using horns to test most( or all ) of the compression drivers

And finding interesting drivers like this tweeter :

Test Bench: Wavecor TW030WA12 30-mm Cloth Waveguide Loaded Tweeter

This Test Bench is about the new Wavecor TW030WA12, a new 30-mm cloth waveguide loaded tweeter intended for home audio applications. The TW030WA12 features a 30-mm wide surround precision-coated cloth diaphragm optimized for high-frequency cut-off above 20 kHz. Its most differentiating feature...

audioxpress.com

 

[Teeter]

Loooong story short, I prefer my Klipsch "Horn'" bookshelf speakers over former JBL. lol

 

[ernestcarl]

Here's something we don't normally see:

More detail of the installation: https://www.fulcrum-acoustic.com/case-studies/crusader-stadium-at-university-of-mary-hardin-baylor/

I noticed there were some small areas where lobing(?) or mid/HF drop was more pronounced, but not very much considering the rather large area covered. I wonder how much suppression of the sound to the surrounding areas was achieved...

The configuration of the horn itself also looks unique -- you can see the back with the compression drivers and some discussion of the design and possible use cases below:

 

[dasdoing]

I noticed there were some small areas where lobing(?) or mid/HF drop was more pronounced, but not very much considering the rather large area covered. I wonder how much suppression of the sound to the surrounding areas was achieved...

PA is not easy.
on the left side (bottom) you have this cave like structure at the back causing issues.
at the backside you have this eco caused probably by the wall on the right.
and on the right the wall you get some boost from the wall.

overall it probably sounds better than 99% of stadiums

 

[dasdoing]

after some weeks with combo speaker with a compression driver. there is something to the transients that a dome doen't seam be able to reproduce. things suddenly sound real, and domes suddenly sound "like speakers". it's surprising that there is no real explaintion yet on the effect out there. it might be the controled dispersion (more direct sound = less time distorsion). or could it be that domes take longer for the soundwaves to rise to peak?
now what sounds worse is the range above 10kHz (smearing caused by the phase plug). arguebly the less important range, but if you pay atention there you might dislike horn sound.

at the end I can confirm what I read several times: once you like them, you would miss them

 

[ernestcarl]

after some weeks with combo speaker with a compression driver. there is something to the transients that a dome doen't seam be able to reproduce. things suddenly sound real, and domes suddenly sound "like speakers". it's surprising that there is no real explaintion yet on the effect out there. it might be the controled dispersion (more direct sound = less time distorsion). or could it be that domes take longer for the soundwaves to rise to peak?
now what sounds worse is the range above 10kHz (smearing caused by the phase plug). arguebly the less important range, but if you pay atention there you might dislike horn sound.

at the end I can confirm what I read several times: once you like them, you would miss them

Certainly a lot of the differences also comes from the directivity and dispersion characteristics of the drivers... but there definitely something I feel lacking in the other waveguided domes I've listened to despite the more uneven (probably more flawed) response of my own smaller compression & coaxial waveguided S8 monitors, for instance. But the further away the distance I position these speakers... and even after applying quite a substantial shelving manually (-2.4 to -2.7 dB), I still find myself preferring the HF clarity of the S8 over my KH120s and JBL LSR305s. Could be just personal preference for their small horn sound "coloration", perhaps... either way I haven't regretted getting them and have no thoughts of replacing these for a long time -- until the internal DSP or amp breaks down eventually, anyways.

 

[kongwee]

Well my DSP studio monitor is this:

 

[dasdoing]

Certainly a lot of the differences also comes from the directivity and dispersion characteristics of the drivers... but there definitely something I feel lacking in the other waveguided domes I've listened to despite the more uneven (probably more flawed) response of my own smaller compression & coaxial waveguided S8 monitors, for instance. But the further away the distance I position these speakers... and even after applying quite a substantial shelving manually (-2.4 to -2.7 dB), I still find myself preferring the HF clarity of the S8 over my KH120s and JBL LSR305s. Could be just personal preference for their small horn sound "coloration", perhaps... either way I haven't regretted getting them and have no thoughts of replacing these for a long time -- until the internal DSP or amp breaks down eventually, anyways.

I thought you were eye-balling a Fulcrum (natural evolution to your S8?). A RX699 with a sub perhaps

 

[ernestcarl]

I thought you were eye-balling a Fulcrum (natural evolution to your S8?). A RX699 with a sub perhaps

I believe it was the double 8" & coax reference I was mostly interested in... but, the single 12-inch sub really is the current "weak link" in my setup needing augmenting (starts to compress above ~103 dB) rather than the speakers. Well neighbors started to complain even with that modest sub... so doesn't seem prudent getting an even higher SPL capable replacement now. I wonder if their separate townhouse unit acts as an "bass trap" extension to my listening room system. Was never intentional since I really had no control with the previous construction work -- otherwise, I would have made it more soundproof -- seems the effect may be both a blessing and a curse? Hmmmn

 

[dasdoing]

I believe it was the double 8" & coax reference I was mostly interested in

Saw that one...thought it was a little overkill for a room lol. a shame they have no double 6" or even 5".

Well neighbors started to complain

like I said in the other topic I think the modes are the worst in terms of propergation in a building. after all it's them that are the strongest right at the wall. ever thought of running a loud sweep while meassuring it at your neighbor's place?

 

[ernestcarl]

ver thought of running a loud sweep while meassuring it at your neighbor's place?

Would be interesting to test... but I'm not too chummy with any of them. We sort of acknowledge each other and say hello occasionally, but don't really "talk". LOL... I'm probably not what you would call the most personable guy around the area.

 

[Kvalsvoll]

after some weeks with combo speaker with a compression driver. there is something to the transients that a dome doen't seam be able to reproduce. things suddenly sound real, and domes suddenly sound "like speakers". it's surprising that there is no real explaintion yet on the effect out there. it might be the controled dispersion (more direct sound = less time distorsion). or could it be that domes take longer for the soundwaves to rise to peak?
now what sounds worse is the range above 10kHz (smearing caused by the phase plug). arguebly the less important range, but if you pay atention there you might dislike horn sound.

at the end I can confirm what I read several times: once you like them, you would miss them

Dome tweeters can sound quite good, but will never be like the best HF solutions, for clarity and dynamics. Horn is one part of that, then the driver itself is also important.

It is easy to underestimate capacity requirements for high frequencies, because when you look at the energy spectrum for music, it falls off towards higher frequencies, but when you look at the high-passed waveforms the peak level can be comparable to that of the lower frequency range.

 

[dc655321]

It is easy to underestimate capacity requirements for high frequencies, because when you look at the energy spectrum for music, it falls off towards higher frequencies, but when you look at the high-passed waveforms the peak level can be comparable to that of the lower frequency range.

Can you explain in more depth what you mean by this?

 

[Newman]

I think he is saying that the energy spectrum for HF is lower, not because the voltages/SPLs are lower, but because the duration of the voltages are shorter, ie high loudness levels in HF tend to be of short duration. So tweeters still need to be able to deliver high peaks without crapping out.

But that's just my guess, let's see hear from him.

 

[dc655321]

I think he is saying that the energy spectrum for HF is lower, not because the voltages/SPLs are lower, but because the duration of the voltages are shorter, ie high loudness levels in HF tend to be of short duration. So tweeters still need to be able to deliver high peaks without crapping out.

But that's just my guess, let's see hear from him.

You may be right.
It would depend on the HPF itself of course (among other things), but I would think that Parseval may have something to say about this (in general).

 

[LTig]

I think he is saying that the energy spectrum for HF is lower, not because the voltages/SPLs are lower, but because the duration of the voltages are shorter, ie high loudness levels in HF tend to be of short duration. So tweeters still need to be able to deliver high peaks without crapping out.

But that's just my guess, let's see hear from him.

I think it's a good guess. I always wondered why Genelec 2-way speakers had identical power amps for woofer and tweeter. Then there was a discussion at ASR about how much power active speakers might need for the different drivers. I used Audacity to highpass a music sample and was very astonished that the high passed signal sometimes had higher peaks than the full range signal.

So I think that the average power of the tweeter amp (the area under the curve) is smaller than the average power of the woofer amp, but the peak power is the same. So you need a high power amp for the tweeter as well but it may need less cooling.

 

[dasdoing]

https://meyersound.com/news/m-noise-test-signal/ said:

It was discovered that the crest factors of music and pink noise were similar at frequencies below 500 Hz, but with music exhibiting a steadily rising crest factor at higher frequencies.

Crest factor - Wikipedia

en.wikipedia.org

So while music averages pink-noise-ish, the peaks in HF are still pretty loud

 

[dasdoing]

Would be interesting to test... but I'm not too chummy with any of them. We sort of acknowledge each other and say hello occasionally, but don't really "talk". LOL... I'm probably not what you would call the most personable guy around the area.

a good excuse to have more contact with the neighbours lol.

no, I understand that in some regions this is kind of a tabu. I still think talking with the neighbors is very healthy for coexistence and mental health

 

[dasdoing]

let's compare Erin's Instantaneous-Compression™ from two top dome-speaker to a to CD speaker, shall we:

So there is undoubtly some loss in the transients at reference-level-ish even with "state-of-the-art" dome tweeters.
and it seams that when you are forced to match woofer(s) to CDs you get better results in the mids, too