I found some small folded horn enclosure units with a 40mm x 20mm driver. The SPL is rated at 83 @ .5m. (8 ohm, 120 to 14k, 2 watts) Neat little "Toys" and the price is right..
Would I expect a decent volume level if I were to parallel/series 16 of these to achieve 8 ohms in a panel/tower mount?
Would I expect a decent volume level if I were to parallel/series 16 of these to achieve 8 ohms in a panel/tower mount?
Last edited:
Add 3dB for each extra driver, if they are all aimed directly at you.
Maybe a vertical line array on a parabolic curve to focus them.
At 1m the spl drops to about 77dB. At 3m (normal listening distance) it drops to about 67dB spl. All on-axis.
Other things equal, you lose 6dB spl for every doubling of distance.
Take into consideration the directionality of the horns, though. They should also be mounted as close together vertically as possible.
Don Keele's CBT (Constant Beamwidth Transducer) Page
Last edited:
Think I'd mount them sideways, up against each other with the tabs horizontal.
The line would be about 4' 8" high. Set the height so the middle of the line is at ear level. A curved mounting surface is better than flat.
Here's a pic.. (Hope it works!)
An externally hosted image should be here but it was not working when we last tested it.
I'd mount them all the same way, with the driver end close to the baffle's edge.
Make the two speakers as left and right mirror images.
The curvature depends on the listening distance.
Sounds like a plan! Thanks everyone for your input!
I envision a "Holder" for the units to be as minimal as possible. Perhaps 1" x 2" lumber or 3/4" PVC. I'll decide when I have one in my hand. I'll doll it up somehow..
As for the PVC, I figure I could leave a small amount of room between the units so I could bow the array slightly and find a sweet spot? Just some more brain splooge..
Another winter project under way! (With hopes the Wife doesn't shoot me when 32 of these things arrive in the mail!)
I envision a "Holder" for the units to be as minimal as possible. Perhaps 1" x 2" lumber or 3/4" PVC. I'll decide when I have one in my hand. I'll doll it up somehow..
As for the PVC, I figure I could leave a small amount of room between the units so I could bow the array slightly and find a sweet spot? Just some more brain splooge..
Another winter project under way! (With hopes the Wife doesn't shoot me when 32 of these things arrive in the mail!)
Re the end sensitivity, it depends on how you wire them, but if it's in a series - parallel array of 4 packs of 4 drivers, then you will have a net sensitivity gain of 0 and an efficiency gain of 12dB up to the point that the individual outputs no longer sum coherently through the different phase-angles & you need to curve the array to focus at a specific point to avoid this at the listening position.
If advertised on-axis SPL is 83dB 0.5m for 1 driver, then as noted at 1m it will be 77dB, at 2m 71dB etc. Assuming they're in the array above, then at 1m output will be approx. 89dB, ignoring HF losses etc. through the array effect. If mounted in an untapered array, output will drop 3dB per doubling of distance within the Fresnel zone, unlike a point source where it will decrease -6dB per doubling of distance. As you move further back (into the Fraunhofer zone) the array becomes a 'point source' once again & its behaviour is consistent with that, SPLs dropping at -6dB per doubling of distance.
I'd be wary inasmuch as I'd buy one of these to start off with & see if you like the sound. If you do, well & good. If you don't, using 32 of them isn't likely to change that, unless the only reason you didn't like them was distortion through excessive drive level, which using multiple units would reduce.
If advertised on-axis SPL is 83dB 0.5m for 1 driver, then as noted at 1m it will be 77dB, at 2m 71dB etc. Assuming they're in the array above, then at 1m output will be approx. 89dB, ignoring HF losses etc. through the array effect. If mounted in an untapered array, output will drop 3dB per doubling of distance within the Fresnel zone, unlike a point source where it will decrease -6dB per doubling of distance. As you move further back (into the Fraunhofer zone) the array becomes a 'point source' once again & its behaviour is consistent with that, SPLs dropping at -6dB per doubling of distance.
I'd be wary inasmuch as I'd buy one of these to start off with & see if you like the sound. If you do, well & good. If you don't, using 32 of them isn't likely to change that, unless the only reason you didn't like them was distortion through excessive drive level, which using multiple units would reduce.
Last edited:
Thanks, Scottmoose..
Yes, I plan to configure them to end up at 8 ohms. Might try 4 ohms too!
It's not a bank breaker. At $1.43 each, I can have some fun! These units are completely tuned and in cabinets. I can't buy the drivers alone that cheap..
I have a handful of those PAM8403 chip amps laying around so I can make use of them if all else fails..
Yes, I plan to configure them to end up at 8 ohms. Might try 4 ohms too!
It's not a bank breaker. At $1.43 each, I can have some fun! These units are completely tuned and in cabinets. I can't buy the drivers alone that cheap..
I have a handful of those PAM8403 chip amps laying around so I can make use of them if all else fails..
Thanks to Scottmoose for posting the correct answer to the original question of "what is the overall sensitivity?". Essentially the answer is the log (base10) of the number of drivers in the array plus the sensitivity gain (or loss) of the wiring configuration. In this case 16 drivers yields an efficiency gain of log (16) or 12 dB while the impedance configuration (4 parallel strings of 4 drivers in series) yields an 8 ohms impedance. Scott also states the sensitivity at 1 meter distance vs. the 0.5 meter spec--a 6 dB difference.
Those of you interested in a write-up on this subject can read pages 17-19 of my Near Field Line Array White Paper. These pages cover how to handle system efficiency improvements and arraying issues. See:
http://www.audioroundtable.com/misc/nflawp.pdf
Jim
Those of you interested in a write-up on this subject can read pages 17-19 of my Near Field Line Array White Paper. These pages cover how to handle system efficiency improvements and arraying issues. See:
http://www.audioroundtable.com/misc/nflawp.pdf
Jim
Last edited:
Thanks, Jim. I appreciate the knowledge! (Hope I understand it right)..
Figuring an 8 ohm load at the amp..
I parallel 2 units to achieve 4 ohms. I do this with all 16 units. I then parallel 2 sets of the units, which will give me a 2 ohm load. I then series the 4 combinations, which will give me an 8 ohm load at the amp..
The fact that I'm connecting the 4 combinations in series, there is no gain. The paralleled units, in groups of 4, will gain 3db per group, as in paralleling two drivers in a cabinet, with a net gain of 12db overall..
The power is being divided equally into each 2 ohm load (4 units) in the series and will react as 2 paralleled drivers, being the available power is being divided 4 times in each 2 ohm load..
Is the 'ol boy seein' this right?
Figuring an 8 ohm load at the amp..
I parallel 2 units to achieve 4 ohms. I do this with all 16 units. I then parallel 2 sets of the units, which will give me a 2 ohm load. I then series the 4 combinations, which will give me an 8 ohm load at the amp..
The fact that I'm connecting the 4 combinations in series, there is no gain. The paralleled units, in groups of 4, will gain 3db per group, as in paralleling two drivers in a cabinet, with a net gain of 12db overall..
The power is being divided equally into each 2 ohm load (4 units) in the series and will react as 2 paralleled drivers, being the available power is being divided 4 times in each 2 ohm load..
Is the 'ol boy seein' this right?
Last edited:
OGKG,
You are getting into your combos. Lots of way to connect drivers. Remember that more drivers connected in series will raise the overall array impedance (above the nominal impedance of a driver) and show sensitivity loss in the array. More drivers connected in parallel will lower the array impedance and contribute to array sensitivity gain. Just figure out the impedance of the array for your connection and use the equations I have in the white paper. Then add the impedance sensitivity contribution to the array efficiency improvement to get the overall sensitivity.
Jim
You are getting into your combos. Lots of way to connect drivers. Remember that more drivers connected in series will raise the overall array impedance (above the nominal impedance of a driver) and show sensitivity loss in the array. More drivers connected in parallel will lower the array impedance and contribute to array sensitivity gain. Just figure out the impedance of the array for your connection and use the equations I have in the white paper. Then add the impedance sensitivity contribution to the array efficiency improvement to get the overall sensitivity.
Jim
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.