You guys have seen a bunch of the Unity horn projects I've built.
I tend to use cheap drivers, and those cheap drivers frequently have low BL.
Does high BL smooth out the response?
Some folks have been publishing hornresp sims of midbass horns, and I noticed that when they're using high BL prosound drivers, the response curve in the passband is epically smooth, compared to the sims I get for my cheap $10 and $20 midbasses.
I don't have many high BL drivers in my collection, but here's a Unity horn I made with a B&C 8" midbass:
Thread is here: https://www.diyaudio.com/community/threads/bend-it-like-bateman.378855/
For comparison's sake, here's a Unity horn I made last month using four Parts Express ND64-16s, which are low BL:
Thread is here: https://www.diyaudio.com/community/threads/nehalem.396920/page-2
I tend to use cheap drivers, and those cheap drivers frequently have low BL.
Does high BL smooth out the response?
Some folks have been publishing hornresp sims of midbass horns, and I noticed that when they're using high BL prosound drivers, the response curve in the passband is epically smooth, compared to the sims I get for my cheap $10 and $20 midbasses.
I don't have many high BL drivers in my collection, but here's a Unity horn I made with a B&C 8" midbass:
Thread is here: https://www.diyaudio.com/community/threads/bend-it-like-bateman.378855/
For comparison's sake, here's a Unity horn I made last month using four Parts Express ND64-16s, which are low BL:
Thread is here: https://www.diyaudio.com/community/threads/nehalem.396920/page-2
If it's not clear, my thought process is that the high BL of a prosound motor may have stronger control over cone motion, reducing/minimizing resonances. Note in the ND64 horn how there's a monster of a peak at 2600hz.
Maybe this gives some answers: https://www.speakerplans.com/index.php?id=faq6
The cone needs to be considered as well, stiffness as well as shape.
The cone needs to be considered as well, stiffness as well as shape.
Apart from that I do use some compression drivers with insane magnets which are anything but smooth.
8NDL51 has an EBP of 322Hz and the Dayton ND64-16 has an EBP of 449Hz
So the cheap Dayton actually seems to have a superior EBP for horn loading. I'm guessing the fundamental issue might just be a resonance caused by the aluminum cone.
Many times those resonances aren't so apparent when the drivers are used as direct radiators, but you put the driver on a horn and it's like the horn becomes and x-ray for resonance
I was able to EQ it away to an extent, but I'm not super keen on using 8-10dB of EQ in the octave where our ears are most sensitive AND where the xover point is located:
When you are going for compression, you also have to consider the throat of your horn and how it matches the driver. This is a highly sensitive part of a driver/horn combination and a lot can and will go wrong here. By the nature of the construction resonances are excited much easier and as you call it "x-rayed".
Last edited:
Maybe it's just that the cones on pro drivers are just thicker, more rigid, and therefore are more pistonic in the passband?
Definitely goes a long ways in the right direction based on my own 60 Hz bass horn experiments and Tom's on-line musings WRT (sub) bass horns in general back on the Basslist BB?* mirroring my experience, i.e. it's the driver's ability to remain pistonic over its entire BW, which requires both cone rigidity/damping + sufficient effective motor strength to 'plow' through the varying acoustic pressures of compression horn loading same as required for any true sealed/vented (infra) sub woofer (< ~60 Hz).
* Which IIRC 'triggered' the much misunderstood/maligned in abject ignorance 'woofer speed' doc among other seminal musings over time
* Which IIRC 'triggered' the much misunderstood/maligned in abject ignorance 'woofer speed' doc among other seminal musings over time
Rigid might be a bad thing here. My Dayton ND64-16s have a peak of close to 10dB in the octave from 1500 to 3000Hz, while the B&C 8NDL51 has a peak that's about half as loud.
The Dayton is aluminum, the B&C is treated paper.
The spec sheet from Dayton shows it's smoother than my measurements show, so I'm trying to figure out why:
B&C's 8NDL51 measurement on a baffle
My measurement on a Unity horn
Dayton's ND64-16 measurement on a baffle
Subjectively, those prosound midbasses (in general) seem to perform pretty nicely through the midrange. My old Summas with the 14TBX100 had a great midrange, and my Yamaha DXR12 with a midbass that looks a lot like the Eminence Alpha is also quite nice. Of course both speakers use them as a direct radiator so it's a bit of an "apples and oranges" situation.
If anyone's wondering why both spec sheets show a rising response, and my measurements show a falling response, I believe that's because the horn loading takes the output radiated by the midbass and spreads it across a wider beamwidth. IE, same amount of energy but wider beam means less energy on axis. Also, compression chambers act as an acoustic low pass filter.
There is a way to think about this based on the driver's impedance curve. If the cone mode is well coupled to the voicecoil, which isn't always the case, then it will show up in the impedance curve. Let's say the impedance curve has a 1ohm bump at 10ohms due to a cone mode. That means it is like a parallel resonator in series with a 10ohm resistor. With voltage drive this is equivalent to a 1ohm Zo parallel resonator shunted by a 10ohm resistor. To be critically damped we would need a 1ohm shunt resistor, so we are far from critically damping the resonance. Most cone modes produce less of a bump than this, so it's unlikely that voltage drive is meaningfully affecting the cone resonance. Add to that the fact that the speaker's impedance is partially imaginary and the cone mode is probably only loosely coupled to the voicecoil and it's even worse than that.
So I think it's more likely that high quality drivers just have more optimized cone designs which are better behaved in themselves.
A way to test this would be to take a horn with a high quality driver and use it with a series resistor. If the high Bl is meaningfully reducing cone modes then there should be a significant worsening of performance.
EDIT: Then again fitting a horn to a driver will introduce air column resonances so may add more modes that could be affected by Bl. But in this case the above still applies. Add a series resistor, does the dispersion pattern worsen?
So I think it's more likely that high quality drivers just have more optimized cone designs which are better behaved in themselves.
A way to test this would be to take a horn with a high quality driver and use it with a series resistor. If the high Bl is meaningfully reducing cone modes then there should be a significant worsening of performance.
EDIT: Then again fitting a horn to a driver will introduce air column resonances so may add more modes that could be affected by Bl. But in this case the above still applies. Add a series resistor, does the dispersion pattern worsen?
Last edited: