You rhetorical question has done it's job.
Fundamental character floors like directivity / dispersion / distortion. Can fix the fundamental with DSP but not necessarily the harmonics.
I think I had a eureka moment regarding horns for CDs. They must be the same width as the woofer beneath to allow directivity matching. And that directivity match will happen at a specific frequency. This is one of the factors that determines crossover frequency. And then there are a load of trade offs when looking at smaller woofer with smaller horns. A smaller horn will have better dispersion at higher freqs but then the smaller associated woofer will have to either move more or be high passed higher or I have to accept higher HD.
I think my first design choice is to go with sealed subs with qtc~0.6. About 4 cu ft each. Two RSS390HEs running off the RMX4050HD in stereo at 4R will be enough for the low end. I was ok with my xls10s and these can move far more air. Sealed subs are easy to eq, no port resonance or noise and more flexibility on crossover frequency. Should also be a better match for the room gain. This then allows more choice for the 2 ways above them. Penalty is more compression and higher HD vs vented. Not against vented but do I need that extra output / lower compression?
It also appears the RSS390he is really meant for sealed. Although people do make LLTs with them.
Hoping my understanding is on the right track...?
Fundamental character floors like directivity / dispersion / distortion. Can fix the fundamental with DSP but not necessarily the harmonics.
I think I had a eureka moment regarding horns for CDs. They must be the same width as the woofer beneath to allow directivity matching. And that directivity match will happen at a specific frequency. This is one of the factors that determines crossover frequency. And then there are a load of trade offs when looking at smaller woofer with smaller horns. A smaller horn will have better dispersion at higher freqs but then the smaller associated woofer will have to either move more or be high passed higher or I have to accept higher HD.
I think my first design choice is to go with sealed subs with qtc~0.6. About 4 cu ft each. Two RSS390HEs running off the RMX4050HD in stereo at 4R will be enough for the low end. I was ok with my xls10s and these can move far more air. Sealed subs are easy to eq, no port resonance or noise and more flexibility on crossover frequency. Should also be a better match for the room gain. This then allows more choice for the 2 ways above them. Penalty is more compression and higher HD vs vented. Not against vented but do I need that extra output / lower compression?
It also appears the RSS390he is really meant for sealed. Although people do make LLTs with them.
Hoping my understanding is on the right track...?
I see your point. No didn't try a notch filter but this next attempt at speaker design I most likely will do this.
This is beginning to sound good.. but a few details.
Good start, but it is far from as simple as this.
Not the case. It depends on the device.
There is no 'better', it depends on the situation.
You're right. There is no "better" just more appropriate.
I am currently leaning towards using a cd in a horn with a large sealed midrange. Don't want the complications of a port if I can help it. Want to get driver integration right.
What size? Not sure yet as that depends what dispersion profile I would like / need. Too much would cause other issues like excessive reflections. (Other trade offs will be fàctors but for now dispersion is the focus)
However, through my reading/hunting around, it looks like 15" speakers in general are serious drivers but the 10s and 12s look a bit weak in comparison... Even from same manufacturer.
Used to have 2 x 8" drivers per side doing the mids. A 10" would be a step back in radiating area. I like the idea of the low HD of a large midrange ticking over.
Very different approach to before. It was all about the FR last time and having heavy cabinets.
There is some comedy visual value in having a 15" subwoofer and a 15" midrange each side 🙂
I am currently leaning towards using a cd in a horn with a large sealed midrange. Don't want the complications of a port if I can help it. Want to get driver integration right.
What size? Not sure yet as that depends what dispersion profile I would like / need. Too much would cause other issues like excessive reflections. (Other trade offs will be fàctors but for now dispersion is the focus)
However, through my reading/hunting around, it looks like 15" speakers in general are serious drivers but the 10s and 12s look a bit weak in comparison... Even from same manufacturer.
Used to have 2 x 8" drivers per side doing the mids. A 10" would be a step back in radiating area. I like the idea of the low HD of a large midrange ticking over.
Very different approach to before. It was all about the FR last time and having heavy cabinets.
There is some comedy visual value in having a 15" subwoofer and a 15" midrange each side 🙂
Following this thread with interest.
https://www.diyaudio.com/community/...-a-nice-ob-dipole-woofer.421955/#post-7893703
https://www.diyaudio.com/community/...-a-nice-ob-dipole-woofer.421955/#post-7893703
Interesting! Folks generally describe mine as anything but comical......... decades old, poor quality pic:
Attachments
I have a new contender for the 15 inch midrange. The Faital pro 15pr300. It appears to have slightly "better" (looks like it can be crossed over higher before it becomes too directional) dispersion than most other 15s. It also looks quite well behaved off axis. I.e. the off axis response follows the on axis response. I think that means that the sweet spot will be wider. It also means that any reflections will be truer to the on axis sound. Off axis behaviour has to be right by design and like harmonics cannot be dealt with via DSP.
One question I do have is, is it essential that the CD horn is the same width as the midrange? I know I thought it is above but now I'm wondering if that's true. If the dispersion of horn matches according to datasheet irrespective of size then is that ok?
Have an interesting set of requirements to get the waf. The woofers must be able to flex and the whole system must look a bit crazy. CD horn + 15 inch idea was met with much enthusiasm.
Quite enjoying all this learning. It's something new 🙂
One question I do have is, is it essential that the CD horn is the same width as the midrange? I know I thought it is above but now I'm wondering if that's true. If the dispersion of horn matches according to datasheet irrespective of size then is that ok?
Have an interesting set of requirements to get the waf. The woofers must be able to flex and the whole system must look a bit crazy. CD horn + 15 inch idea was met with much enthusiasm.
Quite enjoying all this learning. It's something new 🙂
I'm performing an experiment to test this requirement. I've bought a 15" nominal 14" baffle hole driver to match up with a 12.7" wide CD horn. H-14EA horn has nominal 60 deg wide dispersion. I'm trying to copy a Peavey SP2(2004) which has a 1508-KADT woofer and a 19" wide horn. I decided matching the horn inlet with one from the same brand, might be more important than having a full width horn from paudio or somesuch. Inlet angles are unspecified for the cheaper horns on the market. Main problem with the Peavey, they have pawn shop value, are visible through the window from the street, and one pair has been stolen already.
One variable not discussed much, time alignment of the diaphragm of the woofer and the CD. The SP2 has the CD back from the woofer about 10". SP2 spec sheet says the time misalignment high to low is about 0.48 millisec. I might be able to do better than that with the CD horn sitting in notches on top of the woofer cabinet. Will it matter?
At 1kHz, it's wider than 90 degrees, ~90 at 1400Hz, -10dB@2kHz, somewhere around 60 degree.
No. Different horn profiles require different widths for the same dispersion at a given frequency.
If matching the horn's horizontal dispersion to the woofer's through the acoustical crossover region is the goal, size is irrelevant.
That said, the vertical center to center distance may be a concern, minimizing it allows a smoother seated to standing pattern.
Time/phase alignment does matter.
12" is ~1.13ms, .48ms is only about 6.45".
If the SP2 uses a series "protection" capacitor on the HF driver, that introduces a 90 degree phase shift which is equivalent to 1/4 wavelength somewhere in the response.
Too little information to draw conclusions from the information you have provided.
What about a coaxial tweet/mid and a larger dia midbass, Dayton's reinforcing the lows of course
Coaxial mid/tweet is a possibility. I do currently have a couple of 500w plate amps kicking about and another DAC.
My original idea before starting this thread was to run a CD coaxial as the top end. Then I started reading about their drawbacks like having a moving cone behind the CD and how the CDs are not always that great in the first place. And then some people reckoned that a superior result could be had by using a separate midbass and horn. There's probably more fun and more degrees of freedom with a separate midbass and horn.
A coaxial does solve the vertical spacing between drivers. A 15" midbass would require a very low crossover frequency to keep the distance between drivers within a quarter of a wavelength. Or I have to accept some lobing.
Again compromises. Is a 15" coaxial from someone like b&c a real possibility...? Or is the moving cone behind the CD a real issue? Having a separate cone for midbass and a smaller coax worth the extra cost? Or with fewer drivers I can spend more on each one.
Questions, questions, considerations and consequences. Never put this kind a thought into my speakers before. Maybe that's where I have been going wrong
My original idea before starting this thread was to run a CD coaxial as the top end. Then I started reading about their drawbacks like having a moving cone behind the CD and how the CDs are not always that great in the first place. And then some people reckoned that a superior result could be had by using a separate midbass and horn. There's probably more fun and more degrees of freedom with a separate midbass and horn.
A coaxial does solve the vertical spacing between drivers. A 15" midbass would require a very low crossover frequency to keep the distance between drivers within a quarter of a wavelength. Or I have to accept some lobing.
Again compromises. Is a 15" coaxial from someone like b&c a real possibility...? Or is the moving cone behind the CD a real issue? Having a separate cone for midbass and a smaller coax worth the extra cost? Or with fewer drivers I can spend more on each one.
Questions, questions, considerations and consequences. Never put this kind a thought into my speakers before. Maybe that's where I have been going wrong
Last edited:
I've been thinking about a 3 way high efficiency system of my own so will offer a couple of thoughts. Bare in mind that AllenB, GM, and Weltersys each have forgotten more than I'll ever know so it's great they're providing input. A 15", or 2, on the bottom end makes sense. I feel an 8" pro midrange, or midbass, covering approximately 250 to 2500, provides a convenient way of covering the vital mid band. Then you have the choice of any 1, 1.4, or 2" CD, with horn, to handle from 2500 up. Others can give better info on which CD and horn.
Or ... use a Beyma 12AX30ND 12" coaxial with a 15", or 2, for the bottom.
Or ... use a Beyma 12AX30ND 12" coaxial with a 15", or 2, for the bottom.
So some decisions and purchases have been made. I like to try and make the final decisions myself. If you live by the sword... etc. I can only blame myself 🙂
For the horns have decided to go with RCF HF950s with Ciare PR614s on the way for comparison using Beyma CP755/TI compression drivers. Not quite large format... Nothing extreme in the specs as far as I could tell.
The midbass (midrange) ended up being the sb acoustics Nero 15MWN700D mentioned earlier in this thread. I liked the low inductance and the EBP of 86 and that they go down to 70Hz in a sealed cabinet according to Sims.
By my reckoning I have a decent directivity match at 1 kHz.
Listened to this lot free air using a crude crossover. They go loud but strangely effortless. Probably cause they are PA speakers. Sounded ok even though the setup is completely compromised.
It is indeed something different. The main goal.
For the horns have decided to go with RCF HF950s with Ciare PR614s on the way for comparison using Beyma CP755/TI compression drivers. Not quite large format... Nothing extreme in the specs as far as I could tell.
The midbass (midrange) ended up being the sb acoustics Nero 15MWN700D mentioned earlier in this thread. I liked the low inductance and the EBP of 86 and that they go down to 70Hz in a sealed cabinet according to Sims.
By my reckoning I have a decent directivity match at 1 kHz.
Listened to this lot free air using a crude crossover. They go loud but strangely effortless. Probably cause they are PA speakers. Sounded ok even though the setup is completely compromised.
It is indeed something different. The main goal.
FYI/FWIW, etc.
(250*2500)^0.5 = 790.57 Hz mean
13543/pi/790.57 = 5.45"/13.84 cm eff dia
(pi*13.84^2)/4 = 150.44 cm^2 = between 7 - 8" dia.