Yes, you are right! That's why I was asking for a concept and how or in which locations it's used. That is the main reason on where to put the crossover points and the choice of drivers. The Beyma AMTs are kind of excempt from that because the WG which has to be built for them can adjust to that. Using them without a WG is possible but doesn't grant a uniform dispersion.
That's actually wrong. Stacking horns increases the acoustical impedance and does indeed extend the cutoff frequency significantly, ie the Martin W-bins go ~10-13 Hz lower in a stack. It depends on the construction and principle, for example, tapped horns usually don't profit much from stacking in regards of extension but in spl per unit. If TH are stacked in a too big cluster, they can even lose performance because the horns low extension can create a phase shift with the direct radiated sound and cause a cancellation. Same on horn reflex or horn hybrid systems.
Yes sorry I should have typed 100 dB sensitivity WITH 100 w RMS power handling.
I should clarify. The type of events I'm thinking to use this system for are small edm events. So these will more often than not be indoors. Especially nowadays becuase here in the Netherlands summer doesn't seem to exist anymore. I won't be surprised if almost all events move indoors in the future. Because giving events outside is hopeless. This year it is at least. So I should probably aim for medium directivity?
I would say the mid/tweeter crossover could be quite a bit higher indeed to create a nice directivity pattern that matches with the tweeter. Of course I'll have to see what the directivity patern will be in the finished waveguide.
What Art Welter mentioned about the line length and what I mentioned is that your proposed 1.3m high cabinet will not approximate a line source and have a cylindrical wave front.
Have you seen the DIY PM60 / PM90 design that uses a co-axial BMS compression driver with a pair of horn loaded 12" drivers? The woofer rear chambers are ported, and the bass horn is bent around the HF horn.
The ~ 32 kg PM90 uses an RCF HF950 horn, BMS4594HE coaxial compression driver and a pair of 12" neodymium woofers. The finished cabinet exhibits good directivity control down to 300 hz. Some people have used an RCF woofer, some use the B&C 12NDL76, depending on market availability. The woofer chosen needs to both fit ( magnets almost touch the compression driver ) and have enough motor strength and the correct parameters.
The PM60 variant uses the same compression driver, but the HF horn is changed to an 18 Sound XT1464, the cabinet is slightly different in width compared to the PM90.
EDIT: apparently RCF discontinued retail sales of the RCF HF950 horn around 2019, so it would be almost impossible to obtain.
Have you seen the DIY PM60 / PM90 design that uses a co-axial BMS compression driver with a pair of horn loaded 12" drivers? The woofer rear chambers are ported, and the bass horn is bent around the HF horn.
The ~ 32 kg PM90 uses an RCF HF950 horn, BMS4594HE coaxial compression driver and a pair of 12" neodymium woofers. The finished cabinet exhibits good directivity control down to 300 hz. Some people have used an RCF woofer, some use the B&C 12NDL76, depending on market availability. The woofer chosen needs to both fit ( magnets almost touch the compression driver ) and have enough motor strength and the correct parameters.
The PM60 variant uses the same compression driver, but the HF horn is changed to an 18 Sound XT1464, the cabinet is slightly different in width compared to the PM90.
EDIT: apparently RCF discontinued retail sales of the RCF HF950 horn around 2019, so it would be almost impossible to obtain.
Last edited:
Building a pair of PM90s right now. Incredible output for a 1m high box! There are a few videos of them on YT.
I've literally built tons of FLH bass cabinets and compared hundreds in outdoor stacks and flying configurations, Martin being one manufacturer that I have had particular interest in since back in the late 1970's.
A stack of bass horns does increase the acoustical impedance, but your assertion that their cutoff frequency extends "significantly", much less ~10-13 Hz is simply not backed with actual measurements or simulations.
The array's low end increases in relative level to the upper response, and smooths response compared to a single, but the cut off frequency does not change more than a few Hz even with eight cabinets.
As an example, this simulation of a typical single undersized mouth front loaded bass horn designed for use in multiples has a low corner frequency of 37.81Hz.
The difference in the low corner going to four cabinets is hardly visible on the chart, with 8 units the cutoff has dropped by only 1.59Hz to 36.22.
As Ivan Beaver wrote: "it "appears" as if they are going low-but in reality it is just the part of the response that is already being reproduced is just smoother and louder."
Ivan has measured a lot more horns than I have before he took the job as the chief engineer measuring and installing Danley Sound Labs front loaded and tapped horns more than a decade ago.
If you expect multiple cabinets to extend the cutoff frequency significantly, you will be quite disappointed when you measure their response.
Art
Even if it’s just smoother and louder I’ll still take it. -3dB relative to midband DOES go down. Local peaks in the excursion profile also tend to be reduced. I’ll take that too.
I cannot fathom putting super expensive home hi-fi dome tweeters in a pro PA rig.
You’ll blow them all out $$$$$$$$ in the first or second howl of feedback.
I would consider the eminence Textreme driver in a compression horn.
You’ll blow them all out $$$$$$$$ in the first or second howl of feedback.
I would consider the eminence Textreme driver in a compression horn.
Well, that's complicated.
Yes, I totally agree, the depth of a sigle bin got almost the same cut off frequency as a stac,. I never disputed that and I agree the usable lower frequency of a single w-bin is closer (much closer) to a stack. But the context here is to bring the reality to someone who applies (and expects) HiFi-standards. Okay, it's a simulation. But if we consider the 100 Hz peak to be lower (well, almost flattened), let's say by -3 dB, then the f6 (not the usualy -3dB for a flat response) is at 60 Hz. I totally agree, the horn is perfectly usable to ~45 Hz - if you discount for the spl loss, low-cut that's needed and the spl loss. The stack gets a f6 of ~30 Hz according to your simulation, and that's well above what I've claimed it would gain. I DO know horns are rated at the cut-off frequency and/or the maxspl but the OP is someone who comes from HIFI and that's why I (maybe lazyly?) avoided applying PA standards. You can use it at lower frequencies but you still lose 6dB (or more) in spl and/or (more) maxspl. So please disregard I've tried to tell him how it is in HiFi-terms.
Yes, it's usable even below the -6 dB, yes, I know the stack doesn't improve the usable fr but the level of it instead nor that I explained why it's not a LA but I saved tons of explainations to direct it into the right direction without sidetracking it with the (admittedly appropable) explaination of directivity, spl gain, stacking gains or even dispersion in the bass (dual side stack, center stack, toothrow, cardioid, etc). I also mentioned that the top is not a real LA very early on and that the dispersion diversity is a real problem. It's not that I don't understand each issue but that it's a way to try to put it to the OP to easily understand it.
Regarding the BMS4594HE or other variations of the BMS coax driver, it does not perform well on a diffraction horn and there are almost none non-DH horns available on which it would perform well. Aside from that, the BMS comes not even close to the resolution of the Beyma AMT, which seems to be main goal of the OP. Plus, it's even more expensive. The AMT does not give the 'in your face' sound like the BMS but I think the OP does not want that at all.
Yes, I totally agree, the depth of a sigle bin got almost the same cut off frequency as a stac,. I never disputed that and I agree the usable lower frequency of a single w-bin is closer (much closer) to a stack. But the context here is to bring the reality to someone who applies (and expects) HiFi-standards. Okay, it's a simulation. But if we consider the 100 Hz peak to be lower (well, almost flattened), let's say by -3 dB, then the f6 (not the usualy -3dB for a flat response) is at 60 Hz. I totally agree, the horn is perfectly usable to ~45 Hz - if you discount for the spl loss, low-cut that's needed and the spl loss. The stack gets a f6 of ~30 Hz according to your simulation, and that's well above what I've claimed it would gain. I DO know horns are rated at the cut-off frequency and/or the maxspl but the OP is someone who comes from HIFI and that's why I (maybe lazyly?) avoided applying PA standards. You can use it at lower frequencies but you still lose 6dB (or more) in spl and/or (more) maxspl. So please disregard I've tried to tell him how it is in HiFi-terms.
Yes, it's usable even below the -6 dB, yes, I know the stack doesn't improve the usable fr but the level of it instead nor that I explained why it's not a LA but I saved tons of explainations to direct it into the right direction without sidetracking it with the (admittedly appropable) explaination of directivity, spl gain, stacking gains or even dispersion in the bass (dual side stack, center stack, toothrow, cardioid, etc). I also mentioned that the top is not a real LA very early on and that the dispersion diversity is a real problem. It's not that I don't understand each issue but that it's a way to try to put it to the OP to easily understand it.
Regarding the BMS4594HE or other variations of the BMS coax driver, it does not perform well on a diffraction horn and there are almost none non-DH horns available on which it would perform well. Aside from that, the BMS comes not even close to the resolution of the Beyma AMT, which seems to be main goal of the OP. Plus, it's even more expensive. The AMT does not give the 'in your face' sound like the BMS but I think the OP does not want that at all.
When you wrote that FLH cutoff frequency (Fc) drops ~10-13 Hz when used in multiples you disputed me saying it won't- you wrote "That's actually wrong".
Now you totally agree that the Fc does not drop ~10-13 Hz 🙂
Glad we could clear that up..
The reality is the EQ required for a single FLH and multiple FLH are quite different, but Fc is not much different.
There are a lot of professional PA sound engineers that would disagree with that statement.
That said, the divide between diffraction horn advocates and haters is not limited to the PA world.
Cheers,
Art
Last edited:
No, you don't (based on what you have written so far).
Your goal is very, very far away from your present knowledge and experience.
Have you ever designed (measured and simulated in software) and actually build a simple two-way hi-fi loudspeaker?
No? I thought so...
And you are intending to burn your $100,000 (or, even worse - someone else's) on line array... of which you know nothing about!? I am sorry, but I have to ask - are you serious?
Last edited:
You are absolutely right! 😁 Yes I have build speakers before, three way actually. Just have a peek at my Avatar. Just of more simple design. Yes I did some basic modeling and yes of course I have measured them, setup crossovers etc. They have very even off axis FR actually. And they sound well, better than 90% of what I've heared in the high end store where I worked some time.
This type off system is just new territory for me and I'm unfamiliar with the physics at play in LAs and horns. Which is why I'm here to learn. I just put my idea out here and will learn from your comments as I go. I'm currently also indulging myself in some of theory behind LAs now online and I'm learing to use Akabak.
As far as I could find out it seems that adding a slight curve to the cabinet helps evening out the polar behavior of LAs? This was anyway my intend. I made up the maximum venue/area size I think I would probably take on and put it in Fusion to calculate the requered angles. The vertical dispersion would have to be 24 degrees to go from a short persone standing in front to a tall person standing in the back.
To reach the width off the venue at the back (for which the top tweeter would be responsible) the dispersion would have to be 40 degrees. Yet to reach the whole front line it would have to be 150 degrees so as wide as possible. Which is a problem. So idealy the top tweeter would have a waveguide tuned for more directivity and the bottom tweeter to be almost direct radiating and having the wave guide gently transition between the two WG geomitries over the length of the cabinet. L Acoustics deploys something like this. However to properly match the dispersion of the tweeter and midrange at the crossover this would mean either each layer of drivers would require a different XO frequency or the midrange would also have to be matched by wave guides. Both of which would make the whole a lot more complex but it's something to think about.
I'm now trying to learn about the phase behavior of LAs. But just reading about it I find it hard to understand so I'm hoping to get a better understanding by vizualising it in simulations and see what happens when I change things.
As I said before I gonna work at a AV company, do a lot of research and maybe some courses to learn everything I need and I will not take on this project before I'm confident in my abilitys. That would be stupid and a very effective way to get myself in great trouble! 😄 But why not start my learning curve and the design of this system right now? I'm 21 you know...
Attachments
Never too late to start..
I was a bit younger when when I started building vertical arrays:
Almost a half century ago..
The Nexo ID84 is a more current example of good way of doing what you'd like to do:
www.youtube.com/@devinlsheets_alphasound
Dave Rat has probably also covered much of what you need to know in the sound reinforcement buisiness, and got started in the DIY approach to sound in a similar era as me:
https://www.youtube.com/@DaveRat
His Powersoft M-Force SuperSubs are an example that DIY is not dead.
Hello there, I've been learning a bit about horns and did some first simple calculations to get an idea for the required horn dimensions. For the tweeters I've landed on a oblate spheroidal waveguide as is said to have low distortion and good, even directivity while also offering good loading. Kind of the best of both conical and exponential horns. I've made a program in excel where I can visualize a horn geometry based on given requirments. But now I'm wondering. All of the formulas that this is based on are for cillindrical waveguides yet the waveguide that I need will be rectangular. To keep the same performance from the waveguide. Do I simply stay with the section view geometry of the round horn and stretch it out? Or do I need to recalculate it to area?
The horns for the mid bass will probably be hyperbolic with T around 0,5~0,6.
The horns for the mid bass will probably be hyperbolic with T around 0,5~0,6.
Attachments
That array on the Nexo speaker is a very cool concept actually. May or may not look into doing something like that too. Might be a very cool way too control the directivity of the midranges. Perhaps the midrange would have to be pretty far appart to achieve effective diractivity. Something to try out in akabak.
I have by the way landed on the Beyma 6MCF200Nd as midrange driver. It has a very nice 97 dB sensitivity with 200 W AES power handling. And even though it's also a sealed back midrange. It is actually perfectly flat from 400 Hz. So you get to use the full 97 dB sensitivity instead of only being left with 96 or 93 from the B&C sealed back mids. It also has very nice dispersion up to 2 kHz. There aren't any measurments out there on any of the main websites but I did find a thread about it where it measured very well and the owner praised it as the best he had heared and that it matched perfectly to the TPL-200, which makes sense and it's probably designed by Beyma to do so. The midwoofer will most probably a Beyma too because some of their models are just the most efficient and powerfull. They often also have low Qts which is great for the mid bass horn.
Though when I was playing around with a exponential horn graph. I noticed that the bigger the throat, the shorter the horn could be. Which seemed counterintuitive to me. Shouldn't the throat be smaller to achieve better loading with a given size horn?
I have by the way landed on the Beyma 6MCF200Nd as midrange driver. It has a very nice 97 dB sensitivity with 200 W AES power handling. And even though it's also a sealed back midrange. It is actually perfectly flat from 400 Hz. So you get to use the full 97 dB sensitivity instead of only being left with 96 or 93 from the B&C sealed back mids. It also has very nice dispersion up to 2 kHz. There aren't any measurments out there on any of the main websites but I did find a thread about it where it measured very well and the owner praised it as the best he had heared and that it matched perfectly to the TPL-200, which makes sense and it's probably designed by Beyma to do so. The midwoofer will most probably a Beyma too because some of their models are just the most efficient and powerfull. They often also have low Qts which is great for the mid bass horn.
Though when I was playing around with a exponential horn graph. I noticed that the bigger the throat, the shorter the horn could be. Which seemed counterintuitive to me. Shouldn't the throat be smaller to achieve better loading with a given size horn?
Last edited:
The shorter the horn the narrower its BW, so for a given flare frequency/factor you can 'slice' a section out of it to make a shorter horn like the pioneers did such as the Altec A7 (110 Hz 'slice' out of a 55 Hz flare with an upper cut off around 338 Hz, the originally intended driver's upper mass corner (Fhm = 2*Fs/Qts where all box design theory peters out) and similar.
Only a quick listen so far. Certainly loud and bright. The PHL seems to have smoother breakup behavior.
I doubt there will be a big difference when eq'd and low passed at a reasonable frequency.
I'll try for some measurements this weekend.
I doubt there will be a big difference when eq'd and low passed at a reasonable frequency.
I'll try for some measurements this weekend.