Hi,
After auditioning the Blumenhofer Grand Goia at an audio show, I've been researching these kind of speakers. Speakers with some kind of horn/waveguide on the bass section.
I hope somebody can answer some questions I have about these.
What is the effect of these kind of 'horns'? Are they there for directivity control, improve efficiency or both? Or something else?
In some of these designs, a part of the driver is covered by the mouth of the horn. I don't question there is a good reason behind it, but intuitively, it doesn't make sense to me.
Don't you lose Sd, efficiency, SQ like this?
What are the rules/formula(s) to determine the size of the mouth of the horn, or what portion of the driver to cover?
After auditioning the Blumenhofer Grand Goia at an audio show, I've been researching these kind of speakers. Speakers with some kind of horn/waveguide on the bass section.
I hope somebody can answer some questions I have about these.
What is the effect of these kind of 'horns'? Are they there for directivity control, improve efficiency or both? Or something else?
In some of these designs, a part of the driver is covered by the mouth of the horn. I don't question there is a good reason behind it, but intuitively, it doesn't make sense to me.
Don't you lose Sd, efficiency, SQ like this?
What are the rules/formula(s) to determine the size of the mouth of the horn, or what portion of the driver to cover?
Attachments
Do you mean like picture 1 where the bass drivers are covered at the horn throat, or like picture 3 where the tweeter horn mouth casts a shadow over the bass drivers?
These are nice examples of front loaded horns with vented chambers.
I've been reading about these recently and want to try and build one (80Hz - 600Hz).
I'm a novice and slowly learning HornResp. I had the same questions and this program helped me understand more than reading, changing the various parameters of horn size, driver, chamber and then viewing the expected results.
They are for directivity control and improve efficiency to frequencies above the horn cutoff.
The port vent is tuned to compliment (match) the horn loaded portion, extending the drivers range.
Partially covering the driver is fine when crossover point is low enough. If crossover is higher (above 1kHz or so) there would most likely be phase plugs in front of the drivers.
As for rules, don't believe there are hard rules. Size of horn mouth, depth of horn, choice of driver all factor into your target frequency range.
I've been reading about these recently and want to try and build one (80Hz - 600Hz).
I'm a novice and slowly learning HornResp. I had the same questions and this program helped me understand more than reading, changing the various parameters of horn size, driver, chamber and then viewing the expected results.
They are for directivity control and improve efficiency to frequencies above the horn cutoff.
The port vent is tuned to compliment (match) the horn loaded portion, extending the drivers range.
Partially covering the driver is fine when crossover point is low enough. If crossover is higher (above 1kHz or so) there would most likely be phase plugs in front of the drivers.
As for rules, don't believe there are hard rules. Size of horn mouth, depth of horn, choice of driver all factor into your target frequency range.
Having a smaller throat opening than the cone is referred to as compression, as in a compression driver. You can have a compression ratio of as little as 1.5:1 to higher than 10:1
A direct radiator relies on cone velocity to move air, whereas when you provide compression, you can call on the motor strength to create pressure. Imagine it as similar to the difference between you throwing a light ball, and a shot put.. Horn drivers usually have larger magnets for this reason.
Hornresp by David McBean in the Subwoofers subforum is able to do simulations of this.
A direct radiator relies on cone velocity to move air, whereas when you provide compression, you can call on the motor strength to create pressure. Imagine it as similar to the difference between you throwing a light ball, and a shot put.. Horn drivers usually have larger magnets for this reason.
Hornresp by David McBean in the Subwoofers subforum is able to do simulations of this.
The concept of a cone driver operating in a FLH with around 2:1 to 3:1 compression ratio will usually yield hifi worthy results with a healthy efficiency gain if the driver is capable of sufficient midrange extension. There are a few drivers I prefer for this application, which are the B&C 8PE21, 12PE32 and 12MH32. These are IMO the best sounding, most reasonably priced and easily available drivers for the job.
The expo-reflex design can be tricky to get flat response from. Balancing out the FR between both parts of the desogn to yield the best efficiency and linearity without ending up with annoying resonances from the reflex compartment isn't easy. Very often over-dampening the back reflex enclosure is necessary to make it all sound clean and you're trading efficiency for linearity. If you get it right, it can sound better than some higher efficiency direct radiating designs which more often than not sound rough and fatiquing, even at low levels.
I wish there was more reliable simulation available on expo-reflex cabs. I'm sure hornresp can do a decent job, but there are so many small factors which usually add up to large variables that can't be factored into the sims.
The expo-reflex design can be tricky to get flat response from. Balancing out the FR between both parts of the desogn to yield the best efficiency and linearity without ending up with annoying resonances from the reflex compartment isn't easy. Very often over-dampening the back reflex enclosure is necessary to make it all sound clean and you're trading efficiency for linearity. If you get it right, it can sound better than some higher efficiency direct radiating designs which more often than not sound rough and fatiquing, even at low levels.
I wish there was more reliable simulation available on expo-reflex cabs. I'm sure hornresp can do a decent job, but there are so many small factors which usually add up to large variables that can't be factored into the sims.
Hi, I have not heard such system, but here is some morning coffee reasoning about it.
Thinking layman terms it works as acoustic bandpass filter, more sensitivity on the pass band. Mouth size relates to the low end of the bandwidth and throat size to the high end. Having smaller mouth / throat frequency goes up, and bigger it goes down. Unless the horn is very big, meters, there is no boost to very low frequencies and perhaps separate even bigger system for low bass is needed. One could EQ down the bandpass boost to level it out with the low bass, if there is no separate low bass system. Increased sensitivity means reduction in excursion, which affects distortion from that particular driver. Also directivity control which affects how sound radiates into the room. Diffraction at the mouth would be significant with all implementations on the opening post, not sure how audible this is, past 1ms delayed secondary sound source at the perimeter for the whole bandwidth I predict. Non-linear distortion exchanged to linear, no idea which one is more audible/worse/better.
From this I would reason that benefits for home use are: directivity control 🙂 Practical stuff: If there is no separate bass system then prepare to buy a DSP if you like to hear the lows, which usually needs to be louder, not quieter, than the mid bass. I suspect it could sound very good but see them mostly as marketing leverage, pattern control for home listening levels is possible with cardioid system as well, or open baffle, except sensitivity goes down. For big room / small amplifier, why not. Having built few ~1meter sized boxes without too complicated shapes I'd say they would be quite pain to build, which would be no go at least for me 🙂 I preserve rights to change the opinion if I ever hear one and it sounds significantly better than what I've heard before. I like big speakers and understand importance of dynamics and have some idea what it takes. I'm not sure if something like this would be best compromise at least for typical living room use, which is the perspective I'm looking from.
Thinking layman terms it works as acoustic bandpass filter, more sensitivity on the pass band. Mouth size relates to the low end of the bandwidth and throat size to the high end. Having smaller mouth / throat frequency goes up, and bigger it goes down. Unless the horn is very big, meters, there is no boost to very low frequencies and perhaps separate even bigger system for low bass is needed. One could EQ down the bandpass boost to level it out with the low bass, if there is no separate low bass system. Increased sensitivity means reduction in excursion, which affects distortion from that particular driver. Also directivity control which affects how sound radiates into the room. Diffraction at the mouth would be significant with all implementations on the opening post, not sure how audible this is, past 1ms delayed secondary sound source at the perimeter for the whole bandwidth I predict. Non-linear distortion exchanged to linear, no idea which one is more audible/worse/better.
From this I would reason that benefits for home use are: directivity control 🙂 Practical stuff: If there is no separate bass system then prepare to buy a DSP if you like to hear the lows, which usually needs to be louder, not quieter, than the mid bass. I suspect it could sound very good but see them mostly as marketing leverage, pattern control for home listening levels is possible with cardioid system as well, or open baffle, except sensitivity goes down. For big room / small amplifier, why not. Having built few ~1meter sized boxes without too complicated shapes I'd say they would be quite pain to build, which would be no go at least for me 🙂 I preserve rights to change the opinion if I ever hear one and it sounds significantly better than what I've heard before. I like big speakers and understand importance of dynamics and have some idea what it takes. I'm not sure if something like this would be best compromise at least for typical living room use, which is the perspective I'm looking from.
Last edited:
@tmuikku: Regarding audibility of linear distortion vs nolinear distortion, I found an interesting article here 🙂 :
http://www.gedlee.com/Papers/AES06Gedlee_ll.pdf
http://www.gedlee.com/Papers/AES06Gedlee_ll.pdf
This concept was also used in the Altec VOT, apparently using woofers specially designed for high compression ratios. The main reason being efficiency.
In the linked article, there's this:
"The bad news about this is that very low bass extension was not possible, because with a bass horn, the lowest possible frequency was limited by the horn’s exit width. Once the wavelength becomes wider than the horn exit, the response precipitously drops."
That would mean if the exit of the horn is say 1m, frequency would drop rapidly at 343hz. Sounds very high to me, any thoughts?
https://audioxpress.com/article/altec-lansing-s-voice-of-the-theatre-speakers-what-you-need-to-know
In the linked article, there's this:
"The bad news about this is that very low bass extension was not possible, because with a bass horn, the lowest possible frequency was limited by the horn’s exit width. Once the wavelength becomes wider than the horn exit, the response precipitously drops."
That would mean if the exit of the horn is say 1m, frequency would drop rapidly at 343hz. Sounds very high to me, any thoughts?
https://audioxpress.com/article/altec-lansing-s-voice-of-the-theatre-speakers-what-you-need-to-know
Attachments
Efficiency is always good but never more necessary than when it was difficult to find higher powered amps.
Moving to direct radiators took away the benefit of directivity.
We have the luxury to pick and choose.
Moving to direct radiators took away the benefit of directivity.
We have the luxury to pick and choose.
I'm sure it's possible to make some kind of 3d printed 'sponge' or Swiss cheese-like matrix (or stacked pebbles or something) that controls the rate of expansion of the air volume, while making the directivity behave like it's a direct radiator mounted on a flat baffle.
Yup.
The altec horn loads to around there then transfers to being a direct radiator…..
Jbl cabs that are similar shaped, their own graphs show the same thing.
Compression ratios = djk (rip) said a community compression driver (vhf 100 ?, 4:1 compression ratio) had a smoother sound vs the 10:1 typical ratios.
Perhaps that is the audible result exchanging linear for non/linear distortion….
The altec horn loads to around there then transfers to being a direct radiator…..
Jbl cabs that are similar shaped, their own graphs show the same thing.
Compression ratios = djk (rip) said a community compression driver (vhf 100 ?, 4:1 compression ratio) had a smoother sound vs the 10:1 typical ratios.
Perhaps that is the audible result exchanging linear for non/linear distortion….
At some point it might be a case of balancing 2 different sources of non-linear distortion. The cone displacement may be higher when less compression is used, but the air itself also distorts at the throat.
Jbl 4560.
Rolls off below 200hz.
Maybe 6-7db down at 90hz vs 200hz.
Mouth isn't big. Nor is the horn too deep.
Rolls off below 200hz.
Maybe 6-7db down at 90hz vs 200hz.
Mouth isn't big. Nor is the horn too deep.
The transition point from horn to direct radiator is what decides how linear/well it will perform. Many designs use a high Q box tuning to boost efficiency down low, which completely ruins the sound. I have built a few of these in my days. The best results I got was with 2 x 12MH32, giving a range of 40 - 600 hz +/- 3 dB with a 2.5:1 compression ratio and 130 liters rear compartment volume. Efficiency was about 104 dB/W. Wish I still had the cabinet drawing. It was a great sounding, punchy and accurate design that blended very well with various larger HF horns.
What, in simple terms, happens at the transition point from horn to direct radiator?
The boxes in the first post and, from what I can gather from info on the Altec VOT designs, are reflex boxes with a horn added. I don't know if the port is the only reason these have decent bottom end, but all of those do a good ~50hz
The boxes in the first post and, from what I can gather from info on the Altec VOT designs, are reflex boxes with a horn added. I don't know if the port is the only reason these have decent bottom end, but all of those do a good ~50hz
Profiguy - can you share an example of a design mentioned where the high Q box completely ruins the sound ? Curious to review bad examples.
CoolJazz - I was told the following by an experienced builder regarding FLH with vented chamber (more in context to midbass):
"In a standard bass reflex with the drivers used tuning would be in the 45-50Hz region, for maximally flat response to the lowest possible frequency. By tuning it higher, along with a smaller than optimal rear chamber, the port output is higher, at the expense of low frequency extension, so that the port output matches that of the front wave. This also increases the effective driver resonant frequency, enhancing the horn sensitivity at the lower end of its range"
HornResp allows you to simulate this balancing act described above and mentioned by others here. Take a ported cabinet response. Then take a front loaded horn response. The basic design concept is combing the two responses to work as one unit, the ported portion for the lower part of the response and the horn portion for the higher part of the response. Its tricky to get the two to combine in one nice looking response. I find it challenging as there is many factors to consider, once you get the simulation close, you can play with parameters and adjust. HornResp also allows you to view response together or separately to see what each portion is contributing.
Norman Bates - after reading DJK's recommendation on another forum about the favorable lower compression ratio of the Community M200, I definitely agree with his ears/opinion after acquiring a pair. It was a massive improvement to my system at home. DJK's recommended horn, BRH90 was also part the upgrade. There is something to be said about the smooth relaxed sound of a lower compression ratio. Still dynamic and detailed but in a natural desirable way. No fatigue, no honk, no harshness, very low distortion. I use it from 600Hz to 3.5kHz with minimal eq.
CoolJazz - I was told the following by an experienced builder regarding FLH with vented chamber (more in context to midbass):
"In a standard bass reflex with the drivers used tuning would be in the 45-50Hz region, for maximally flat response to the lowest possible frequency. By tuning it higher, along with a smaller than optimal rear chamber, the port output is higher, at the expense of low frequency extension, so that the port output matches that of the front wave. This also increases the effective driver resonant frequency, enhancing the horn sensitivity at the lower end of its range"
HornResp allows you to simulate this balancing act described above and mentioned by others here. Take a ported cabinet response. Then take a front loaded horn response. The basic design concept is combing the two responses to work as one unit, the ported portion for the lower part of the response and the horn portion for the higher part of the response. Its tricky to get the two to combine in one nice looking response. I find it challenging as there is many factors to consider, once you get the simulation close, you can play with parameters and adjust. HornResp also allows you to view response together or separately to see what each portion is contributing.
Norman Bates - after reading DJK's recommendation on another forum about the favorable lower compression ratio of the Community M200, I definitely agree with his ears/opinion after acquiring a pair. It was a massive improvement to my system at home. DJK's recommended horn, BRH90 was also part the upgrade. There is something to be said about the smooth relaxed sound of a lower compression ratio. Still dynamic and detailed but in a natural desirable way. No fatigue, no honk, no harshness, very low distortion. I use it from 600Hz to 3.5kHz with minimal eq.
nice.
Djk
"The VHF100 is so clean it doesn't sound like it has any HF, although it goes higher and flatter than Altec and JBL 1" drivers. Just goes to show how much distortion we are used to accepting.... The Avantgard Trio uses the Beyma 380M with the M200A (customized), the VHF100 is about 4X the cost of the Beyma (in Europe)."
I see the m200 has about 1.8 : 1 compression ratio.
I can't find the vhf100 compression ratio.............
Some say that compression drivers sound "hard".
I doubt the community ones sound that way.
Djk
"The VHF100 is so clean it doesn't sound like it has any HF, although it goes higher and flatter than Altec and JBL 1" drivers. Just goes to show how much distortion we are used to accepting.... The Avantgard Trio uses the Beyma 380M with the M200A (customized), the VHF100 is about 4X the cost of the Beyma (in Europe)."
I see the m200 has about 1.8 : 1 compression ratio.
I can't find the vhf100 compression ratio.............
Some say that compression drivers sound "hard".
I doubt the community ones sound that way.