If my thread has turned bad, it has not for me. I have been given so much to consider and read about. There are a few high sensitivity designs it's been suggested I build, and I'm considering them. I considered the JBL M2s, but I added it up over 5 grand. And I'd need another fund for the Crown amplifiers.
Someone privately offered me their design very kindly. If that happens, my search will be over, as I will build them.
The monkey coffins are amazing I am sure. But I am after high sensitivity.
If the thread is trivial or useless to you, it is not to me. I am learning. I've been given choices, and am thankful to everyone helping and contributing. I am old now, and on a fixed income. I promise you I will NOT spend my money foolishly. That is why I am on here, learning.
Also, is there a good way to search online for builds with specific drivers? I type in too much info, and it doesn't work so well.
Someone privately offered me their design very kindly. If that happens, my search will be over, as I will build them.
The monkey coffins are amazing I am sure. But I am after high sensitivity.
If the thread is trivial or useless to you, it is not to me. I am learning. I've been given choices, and am thankful to everyone helping and contributing. I am old now, and on a fixed income. I promise you I will NOT spend my money foolishly. That is why I am on here, learning.
Also, is there a good way to search online for builds with specific drivers? I type in too much info, and it doesn't work so well.
Make sure you understand what numbers you're looking at. The sensitivity of most loudspeakers is limited by the sensitivity of the woofer. Many manufacturers and designers therefore just quote the woofer sensitivity of the woofer into 2pi (half space) as the overall sensitivity of the whole loudspeaker.
Some woofer examples (for 2pi):
- The Calpamos woofer has a power efficiency of 99 dB-SPL (1W, 1m). The voltage sensitivity is almost 101 dB-SPL (2.83Vrms, 1m).
- The Monkey Coffin woofer has a power efficiency of 97 dB-SPL (1W, 1m). The voltage sensitivity is about 98 dB-SPL (2.83Vrms, 1m).
- The ScanSpeak woofer of the B1371 has a power efficiency of (almost) 88 dB-SPL (1W, 1m). The voltage sensitivity is 92 dB-SPL (2.83Vrms, 1m).
- The Calpamos sensitivity is specified as 93 dB-SPL (2.83Vrms, 1m).
- The Monkey Coffin sensitivity is specified as 90 dB-SPL (2.83Vrms, 1m).
- The B1371 sensitivity is specified as 93 dB-SPL (2.83Vrms, 1m).
Here's a recipe to estimate the true sensitivity of a loudspeaker system:
- Don't trust the marketing department!
- Determine the voltage sensitivity of the woofer at 2.83 Vrms / 1 m (see here).
- Subtract 6 dB for the baffle step.
- Subtract 1 or 2 dB for mechanical and electrical losses.
Last edited:
I remember member sreten used to link to DIY collections, and if this is where the thread wants to go, here are a couple of those...
Zaph|Audio
Paul Carmody's DIY Speaker Pages
Zaph|Audio
Paul Carmody's DIY Speaker Pages
High efficiency is NOT a virtue. It is a tradeoff between sensitivity on the one hand and both resonance problems and f-3dB/box size on the other. I wonder why people tend to forget this. Maybe because resonances can sound dynamic like and low frequency reproduction in a conventional living room is a pain in the @#$!
Excellent explanation btw, Matthias!
Excellent explanation btw, Matthias!
Last edited:
True! While I am not exactly sure what you mean by resonance problems, basic physics have told us a long time ago that there is always a compromise between efficiency, box size and low-frequency extension. See attachment.
Attachments
Fixing cone resonances either by changing the stiffness/mass ratio to push resonances up or increasing the damping to lower the magnitude will invariably lead to adding mass to the cone. High efficiency drivers do tend to have quite some cone breakups just because the tradeoff was chosen in favor of the sensitivity.
We don't have to start a discussion about the resonances themselves and/or if we can notice this or that level of resonant behavior here, I just wanted to point out the tradeoff thing.
"Here's a recipe to estimate the true sensitivity of a loudspeaker system:
Also can anyone help me understand why dave123 used two AMT's side by side? I've never seen this done.
- Don't trust the marketing department!
- Determine the voltage sensitivity of the woofer at 2.83 Vrms / 1 m (see here).
- Subtract 6 dB for the baffle step.
- Subtract 1 or 2 dB for mechanical and electrical losses."
Also can anyone help me understand why dave123 used two AMT's side by side? I've never seen this done.
"Subtract 6 dB for the baffle step.
Subtract 1 or 2 dB for mechanical and electrical losses."
This looks very pessimistic or worst case scenario.
Just use two/side of these woofers in parallel, calculate with 6dB baffle step loss (worst case scenario) + all other losses (that latter is negligible because of active xo) and you have about 99dB/2.83V/1m sensitive woofer section with F3 between 30-40Hz. But it requires large box.
FaitalPRO | LF Loudspeakers | 18FH500 (8Ω)
Subtract 1 or 2 dB for mechanical and electrical losses."
This looks very pessimistic or worst case scenario.
Just use two/side of these woofers in parallel, calculate with 6dB baffle step loss (worst case scenario) + all other losses (that latter is negligible because of active xo) and you have about 99dB/2.83V/1m sensitive woofer section with F3 between 30-40Hz. But it requires large box.
FaitalPRO | LF Loudspeakers | 18FH500 (8Ω)
Last edited:
Well, it's simple. Take a JBL 2226H and put it in a 200 liter enclosure, port frequency at 40 Hz:
It does 98dB at 2.83V in half space (meaning it's mounted in the floor of a very large room or open space) at 40Hz.
Put it in a box of 100 liter and you lose about 5dB:
Dave could. I can tell you get a lot of comb filter effects just a slight little bit off axis if you drive them with one signal. He might have lowpassed one of them at 3kHz...
Attachments
Unfortunately it's not pessimistic or worst case. It's just simple physics. Take a look at the data out there and you'll see that physics are a reality (the Calpamos and the Monkey Coffing are good examples). The only "way out" would be to argue that the baffle step is less than 6 dB if the speakers are installed near a wall or a corner of the room. However, I prefer some flexibility of where I put my speakers.
That's why I sayd "worst case". Because 6dB baffle step loss and 1-2dB electrical loss (due passive xo) is the worst case. But since the op wants to use active xo, there would be no electrical loss.
It does not appear correct to assume FaitalPRO specifies 1W/1m sensitivity referenced to Re value of a unit, rather to 8 ohm as the driver itself already is. You can calculate the TSP sensitivity at 1W to confirm that. The FR graphs in datasheet are 2.83V already.
As with almost all pro sound drivers, the dB/1W value is based on this formula: 1W = 2.83V into 8 Ohm load. So a nominal 8 Ohm driver are measured at 2.83V and not really at 1W because the actual impedance curve is rarely 8 Ohm. So the frequency response of a nominal 8 Ohm FaitalPRO driver are measured at 2.83V into half space in a virtually endless large box, check that.
C'mon, we're already writing 2021!
C'mon, we're already writing 2021!
Last edited:
I believe you are right. For the 12PR320 I get this:
Datasheet says sensitivity (1W/1m) = 97 dB-SPL
Calculated from TSP:
- 95.3 dB-SPL (1W/1m)
- 97 dB-SPL (2.83Vrms/1m)
Yet another trick by the marketing dept.?
The importance of half space vs. full 4pi space is both important and hard to interpret, as the listening room plays quite a role. At wavelengths of 3m and more the acoustic sum of a driver at -say- 0,5m above floor level is higher because of the contribution of that same floor. The same applies to the front wall, which can be adding SPL too. The couch against the back wall will again bring more sound pressure at the listening position. As long as the phase differences between original sound and reflection(s) are smaller than 120 degrees, boundaries will contribute to the total SPL at the listening position. This all is highly dependent of the geometries and wavelengths involved. So…
Another option I have not seen in this thread: coaxial compression drivers.
Would the existing Kappalite’s be able to go up to match some of the better BMS coaxial compression drivers? Definitely high efficiency.
You’d get the physical format of a big two way with one waveguide, but compression driver efficiency down low to meet the big woofers.
I’m not familiar with them, but BMS may offer some that go lower than 500 Hz?
Would the existing Kappalite’s be able to go up to match some of the better BMS coaxial compression drivers? Definitely high efficiency.
You’d get the physical format of a big two way with one waveguide, but compression driver efficiency down low to meet the big woofers.
I’m not familiar with them, but BMS may offer some that go lower than 500 Hz?
Yes. Those SONY speakers are excellent. I have heard them. A firned has a pair. One of the best kept secrets in audio.
Why are you assuming that you need to do a baffle step correction all the time. There are plenty of systems out there that don't use or need it depending on driver and baffle size.
The losses in the low pass filter can be minimized by using low DCR inductors
You are always going loose some sensitivity but not necessarily 8db
Rob 🙂