I have two line array projects in progress.
One with 25 Vifa TC9FD 18-08 in an alu enclosure.
The alu enclosure is at the metal workshop right now, where they make the holes etc.
The other project really fills the bill as "stupid cheap line array".
24 Aura Sound AS3-75-16FR @ $1.5 a pop.
The freight of 100 pcs to Denmark was actually more expensive than the drivers!
24 pcs per side was chosen as the resulting impedanse can be very flexible:
12 parallel: 2.6 Ohms
8 parallel: 6 Ohms
6 Paralel: 10.7 Ohm
4 parallel: 24 Ohms
3 parallel: 42.7 Ohms
2 Parrallel: 96 Ohms
all in series: 384 Ohms.
Did anybody say Tube OTL amp??
The attached picture is the enclosures for the Aura Sound drivers. 16 mm mdf. The drivers will be mounted from the inside of the front plate. That is actually one of the problems with the Aura Sound drivers. They are difficult to mount! The best way is to mount them inside the enclosure, but you need to make the holes very precise!
One with 25 Vifa TC9FD 18-08 in an alu enclosure.
The alu enclosure is at the metal workshop right now, where they make the holes etc.
The other project really fills the bill as "stupid cheap line array".
24 Aura Sound AS3-75-16FR @ $1.5 a pop.
The freight of 100 pcs to Denmark was actually more expensive than the drivers!
24 pcs per side was chosen as the resulting impedanse can be very flexible:
12 parallel: 2.6 Ohms
8 parallel: 6 Ohms
6 Paralel: 10.7 Ohm
4 parallel: 24 Ohms
3 parallel: 42.7 Ohms
2 Parrallel: 96 Ohms
all in series: 384 Ohms.
Did anybody say Tube OTL amp??
The attached picture is the enclosures for the Aura Sound drivers. 16 mm mdf. The drivers will be mounted from the inside of the front plate. That is actually one of the problems with the Aura Sound drivers. They are difficult to mount! The best way is to mount them inside the enclosure, but you need to make the holes very precise!
Attachments
Last edited:
I've been trying to figure out how to mount my drivers for some time now. With CNC, mounting the tweeters flush and the NS3-16Ohms up to the mounting plate is easy (though pincushion is a pain), but trying to flush-mount the ring is a big problem. I thought of ~0.2" MDF (or other material) CNC'd to fit perfectly over the front of the drivers, clamping the mounting flange, but that would be a HUGE undertaking, and the CNC'ing would be very expensive. It really kills the purpose of this project, and somehow, I doubt that it's worth it. It would look cool though...
I think I'm just going to stick to a big chamfered hole and surface mount the drivers from the front. Especially for the NS3's, that thin flange shouldn't have any real negative effects. The NS525 I'd imagine plays too low a frequency to be greatly effected by the front mount instead of flush mount either.
Greg
Looks great Koldby
I'm in the process of breaking in my NS3-16s and have 35 of them done. They will fulfill the 400Hz to 6KHz portion of my 3-way line array and I'll use 21 of them with a center-to-center spacing of 82.5 mm.
Looking at the frames of them, I'm going rear mount into 12mm plywood and will angle cut the holes with a Roto Zip. 76mm mounting hole with the circle cutter but will angle the tool so it "flares" the hole to 79mm at the front. Finish off by rounding the holes with a orbital sander to smooth. Since the NS3's will share the space with a woofer line, I'll seal them off with a 90mm deep X 82.5mm wide sealed/stuffed box to protect them. A second bezel will be cut to slide over the "mid box" and make the bezel 25mm thick to prevent vibration.
This will take a large amount of time to do, nothing zippy about a Roto Zip but it does allow me to cut the holes at beveled angles. Making them inset into the wood is a pain but it will protect them and mounting a protection grill very simple since it can be surface mounted.
Good luck putting them together without breaking the bezel and look forward to your measurements and listening impressions. Meanwhile, I just wait for a warm day as winter does not seem to be helping at this point.
I'm in the process of breaking in my NS3-16s and have 35 of them done. They will fulfill the 400Hz to 6KHz portion of my 3-way line array and I'll use 21 of them with a center-to-center spacing of 82.5 mm.
Looking at the frames of them, I'm going rear mount into 12mm plywood and will angle cut the holes with a Roto Zip. 76mm mounting hole with the circle cutter but will angle the tool so it "flares" the hole to 79mm at the front. Finish off by rounding the holes with a orbital sander to smooth. Since the NS3's will share the space with a woofer line, I'll seal them off with a 90mm deep X 82.5mm wide sealed/stuffed box to protect them. A second bezel will be cut to slide over the "mid box" and make the bezel 25mm thick to prevent vibration.
This will take a large amount of time to do, nothing zippy about a Roto Zip but it does allow me to cut the holes at beveled angles. Making them inset into the wood is a pain but it will protect them and mounting a protection grill very simple since it can be surface mounted.
Good luck putting them together without breaking the bezel and look forward to your measurements and listening impressions. Meanwhile, I just wait for a warm day as winter does not seem to be helping at this point.
Now with all the 24 drivers in one of the enclousers.
I am very ancious to hear this line source. I measured the Aura Sounds and the Vifas as single drivers on a baffle and they were suprisingly similar. So I will use the Aura Sound as they are and with a DCX 2496 to equalize them. The phase behavior of such a system is very very hard to mimic with a multiway system.
Koldby
I am very ancious to hear this line source. I measured the Aura Sounds and the Vifas as single drivers on a baffle and they were suprisingly similar. So I will use the Aura Sound as they are and with a DCX 2496 to equalize them. The phase behavior of such a system is very very hard to mimic with a multiway system.
Koldby
Attachments
Agreed. Very nice job. I'm glad I haven't started working on my array yet, cause I may copy that rear mount. Is there anything special with the routing? Any specific angles/curves that I should work for? I'm sure a double fluted edge is not especially great. Keep in mind that I'm not going to have those drivers playing much above 7k.
Greg
Greg
sony neo 5.25
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
So now the system is up and running.
Amazing how good a sound from even an uncorrected version is.
First I made measurements with Holm Impulse and EQéd them in DCX 2496.
Later I downloaded a trial version of Audiolens speaker/room correction software
and measured them in that program and made a correction file form that.
That is amazing!
I have EQéd them to 20 Hz and that is a little too much if you play in a big room and play loud. So I will be using them with a sub.
They NEED brake in time and that is not over yet, so you have to wait for the final result.
It is my plan to use these as my HT system and the Vifa line source as my music system.
Attached is 4 files.
Two with the Holm Impuls measurements of a single Aura Sound NS3 and one with Aura Sound and Vifa on to of each other.
The other two attachments are from Audiolens.
One is the uncorrected NS3 linesource inroom response @ about 4 meters distance and the second is the simulated response after audiolens has corrected them.
Koldby
Amazing how good a sound from even an uncorrected version is.
First I made measurements with Holm Impulse and EQéd them in DCX 2496.
Later I downloaded a trial version of Audiolens speaker/room correction software
and measured them in that program and made a correction file form that.
That is amazing!
I have EQéd them to 20 Hz and that is a little too much if you play in a big room and play loud. So I will be using them with a sub.
They NEED brake in time and that is not over yet, so you have to wait for the final result.
It is my plan to use these as my HT system and the Vifa line source as my music system.
Attached is 4 files.
Two with the Holm Impuls measurements of a single Aura Sound NS3 and one with Aura Sound and Vifa on to of each other.
The other two attachments are from Audiolens.
One is the uncorrected NS3 linesource inroom response @ about 4 meters distance and the second is the simulated response after audiolens has corrected them.
Koldby
Attachments
Last edited:
Thanks again, Koldby
I've broken in 44 of mine in with only 6 left to go. The weather might warm enough to start making sawdust later next week so I'm excited to get the project running in the next few weeks. Then again, it might snow but I'm thinking positive for a few weeks of warm enough weather to get started.
Mine will cross at 400Hz and 6KHz to other arrays and it looks smooth enough
according to your charts to proceed with those points. I'm taking a wild guess
that 21 of them will give me 95 to 98dB at one watt of sensitivity so they would blend in with my woofer and tweeter lines. Am I close or really off with my estimate?
Looking forward to your thoughts and results after a few days of break in, it is looking good now as a speaker and very good considering the price. The Auras look really good compared to the 3.5 inch Vifas--very close.
Thank you for your efforts and sharing with the DIY community.
I've broken in 44 of mine in with only 6 left to go. The weather might warm enough to start making sawdust later next week so I'm excited to get the project running in the next few weeks. Then again, it might snow but I'm thinking positive for a few weeks of warm enough weather to get started.
Mine will cross at 400Hz and 6KHz to other arrays and it looks smooth enough
according to your charts to proceed with those points. I'm taking a wild guess
that 21 of them will give me 95 to 98dB at one watt of sensitivity so they would blend in with my woofer and tweeter lines. Am I close or really off with my estimate?
Looking forward to your thoughts and results after a few days of break in, it is looking good now as a speaker and very good considering the price. The Auras look really good compared to the 3.5 inch Vifas--very close.
Thank you for your efforts and sharing with the DIY community.
18,
I managed to find a spreadsheet that someone put out. Not sure if it's accurate, but there's so much math that I assume the guy who made it is smart... Ok, I'm pretty sure it's about right. 3 drivers in series, 7 parallel sets, 16 ohms input w/ 82 db per driver, final impedance 4 ohms.... 101.24 dB.
Considering the price, yes. I've set up my Vifas with an open baffle on a 1/4" piece of birch plywood in my cubicle at work. I'm continually impressed with them (except for bass, since I only have 1 pair). They're definitely quality drivers, but for $1.50 each, those Auras rock. I think I'm going to finally cement a design soon.
Actually, what I'm really excited about is the fact that 16x5.25s and 24x3s both yield 8 ohms and 101 db and 98 db respectively. That's just perfect for 3db worth of BSC. The tweets are probably going to need some padding though, and I'll need bass reinforcement to run passive... this is a good thing for those with room correction receivers, though not my original goal.
Greg
Thanks Greg,
I stacked up 48 tweeters, the Apex Jr. ones and they almost touch each other running the line. The SPL calculators claim something crazy like 106dB at one watt so I wired them for around 11 ohms and have a plate of power resistors to pad them down. Lesson #1 is tweeters don't couple like woofers do so when I add the mid panels, I'll rewire the tweeters for around 5 ohms to get them as loud as possible. If they are too loud, I have plenty of padding resistors laying around.
The woofers will couple together, I have 12 of them at 85dB each so expect around 96dB or so from them. Anything above that is gravy in my book.
One guy tested out the Apex Jr. tweeters, ran them 7S/7P for around 6.5 ohms and he ended up with around 95 to 96dB at one watt. They make it out to about 16K then fall off quickly, close enough for me. Figure running them at 4.5 ohms should give me around 96 to 97dB so I'll wire the Auras to either 12.8 ohms (20 of them 4S/5P) or around 7 ohms (21 wired 3S/7P) as my final tuning.
My SPL meter will tell me how to wire the mids to match the woofers output so I'll just make the panel hold 21 of the NS3-16 full ranges. If I need a drop in output, I'll run 20 of them at 12.8 ohms with the 21st one sealed off and not connected at the bottom of the line. Always good to keep my options open.
The woofers "match" the tweeters in output so the mid range line is the final question. I'm hoping I don't need to delve too deep into the dark art of passive crossovers but am prepared if need be.
I stacked up 48 tweeters, the Apex Jr. ones and they almost touch each other running the line. The SPL calculators claim something crazy like 106dB at one watt so I wired them for around 11 ohms and have a plate of power resistors to pad them down. Lesson #1 is tweeters don't couple like woofers do so when I add the mid panels, I'll rewire the tweeters for around 5 ohms to get them as loud as possible. If they are too loud, I have plenty of padding resistors laying around.
The woofers will couple together, I have 12 of them at 85dB each so expect around 96dB or so from them. Anything above that is gravy in my book.
One guy tested out the Apex Jr. tweeters, ran them 7S/7P for around 6.5 ohms and he ended up with around 95 to 96dB at one watt. They make it out to about 16K then fall off quickly, close enough for me. Figure running them at 4.5 ohms should give me around 96 to 97dB so I'll wire the Auras to either 12.8 ohms (20 of them 4S/5P) or around 7 ohms (21 wired 3S/7P) as my final tuning.
My SPL meter will tell me how to wire the mids to match the woofers output so I'll just make the panel hold 21 of the NS3-16 full ranges. If I need a drop in output, I'll run 20 of them at 12.8 ohms with the 21st one sealed off and not connected at the bottom of the line. Always good to keep my options open.
The woofers "match" the tweeters in output so the mid range line is the final question. I'm hoping I don't need to delve too deep into the dark art of passive crossovers but am prepared if need be.
Last edited:
So now the system is up and running.
Amazing how good a sound from even an uncorrected version is.
First I made measurements with Holm Impulse and EQéd them in DCX 2496.
Later I downloaded a trial version of Audiolens speaker/room correction software
and measured them in that program and made a correction file form that.
That is amazing!
I have EQéd them to 20 Hz and that is a little too much if you play in a big room and play loud. So I will be using them with a sub.
They NEED brake in time and that is not over yet, so you have to wait for the final result.
It is my plan to use these as my HT system and the Vifa line source as my music system.
Attached is 4 files.
Two with the Holm Impuls measurements of a single Aura Sound NS3 and one with Aura Sound and Vifa on to of each other.
The other two attachments are from Audiolens.
One is the uncorrected NS3 linesource inroom response @ about 4 meters distance and the second is the simulated response after audiolens has corrected them.
Koldby
Very nice!
JSS
So now the system is up and running.
Amazing how good a sound from even an uncorrected version is.
First I made measurements with Holm Impulse and EQéd them in DCX 2496.
Later I downloaded a trial version of Audiolens speaker/room correction software
and measured them in that program and made a correction file form that.
That is amazing!
I have EQéd them to 20 Hz and that is a little too much if you play in a big room and play loud. So I will be using them with a sub.
They NEED brake in time and that is not over yet, so you have to wait for the final result.
It is my plan to use these as my HT system and the Vifa line source as my music system.
Attached is 4 files.
Two with the Holm Impuls measurements of a single Aura Sound NS3 and one with Aura Sound and Vifa on to of each other.
The other two attachments are from Audiolens.
One is the uncorrected NS3 linesource inroom response @ about 4 meters distance and the second is the simulated response after audiolens has corrected them.
Koldby
Promissing results! I guess the brown Holm Impuls measurement is the Vifa? I have great hopes after seeing this! I love the way you mounted the aura's. Very clean look! I'm glad I did not start yet. Something to think about
.Woohoo!
Finally have all 50 of the NS3-16's broken in and remembered I have 8 spare drivers to play with. Figure I'll make two pair of mini-lines out of scrap wood I have laying around with 4 in each little box. This way I can get used to cutting angled holes with scrap and not screw up the huge bezels required. Use the pair of mini arrays and test them as 4, run the two boxes on top of each other for an 8 speaker mini array and see how the frequency/SPL changes.
I figure at some point, the SPL at a specific frequency will not go up when I make the switch from 4 to 8 drivers. As the frequency falls, the output should increase and at the +3dB point, exactly one octave above that point should tell me the SPL of the line.
In theory, I should be able to calculate what 21 of them will do from the 4 to 8 mini array frequency testing. On paper--if you hold your mouth right.
At the minimum, I'll at least get experience cutting beveled or flared holes with the Roto-Zip and see if my wood working skills are up to snuff with scrap wood. I'll keep the thread posted once I figure out what SPL does at what
frequency with 4 of them and 8 of them running at the same power level. At least I can see the frequency curve change from 4 to 8 to gain a better understanding of how these things interact with each other.
More to come--depending on weather.
Finally have all 50 of the NS3-16's broken in and remembered I have 8 spare drivers to play with. Figure I'll make two pair of mini-lines out of scrap wood I have laying around with 4 in each little box. This way I can get used to cutting angled holes with scrap and not screw up the huge bezels required. Use the pair of mini arrays and test them as 4, run the two boxes on top of each other for an 8 speaker mini array and see how the frequency/SPL changes.
I figure at some point, the SPL at a specific frequency will not go up when I make the switch from 4 to 8 drivers. As the frequency falls, the output should increase and at the +3dB point, exactly one octave above that point should tell me the SPL of the line.
In theory, I should be able to calculate what 21 of them will do from the 4 to 8 mini array frequency testing. On paper--if you hold your mouth right.
At the minimum, I'll at least get experience cutting beveled or flared holes with the Roto-Zip and see if my wood working skills are up to snuff with scrap wood. I'll keep the thread posted once I figure out what SPL does at what
frequency with 4 of them and 8 of them running at the same power level. At least I can see the frequency curve change from 4 to 8 to gain a better understanding of how these things interact with each other.
More to come--depending on weather.
Thanks to all for the nice comments
The red (brown) is the Vifa and the second curve on that sheet is THD. The large peak on the NS3 curve is not the characteristic of the driver, but my wife opening a door!!! ;-)
18hurts :
looking foreward to your results.
Koldby
I guess I ougth to have commented the graphs.
The red (brown) is the Vifa and the second curve on that sheet is THD. The large peak on the NS3 curve is not the characteristic of the driver, but my wife opening a door!!! ;-)
18hurts :
looking foreward to your results.
Koldby
All my NS3-16's are broken in with at least 6 hours on them,
Dug around, have plenty of scrap wood, mismatched input terminals and old power cord wire to build two 4 mini-line array sealed speakers for testing. Only need 42 of them for the lines so the other 8 will be to experiment with.
I only care about 400Hz and up but would like to test them down low for educational reasons. My plan is to plot them with test tones from around 80Hz to 16KHz and see the difference in SPL when changing from 4 to 8 drivers.
The weather and life is not cooperating very much, but I'll do the hit and run style of building when it warms up enough in the garage to make sawdust.
For full range testing, I am planning about 1 liter per driver considering the Vas is 1.12 liters and will line the small boxes with old blue jean denim to soak up some of the internal bounce. The denim will keep the poly fill from being sucked into the vented pole pieces of the NS3-16 FRs.
Would 1 liter be enough to get reasonable results at 80 to 200Hz testing or should I increase the size of the box to give them more room to breath?
Dug around, have plenty of scrap wood, mismatched input terminals and old power cord wire to build two 4 mini-line array sealed speakers for testing. Only need 42 of them for the lines so the other 8 will be to experiment with.
I only care about 400Hz and up but would like to test them down low for educational reasons. My plan is to plot them with test tones from around 80Hz to 16KHz and see the difference in SPL when changing from 4 to 8 drivers.
The weather and life is not cooperating very much, but I'll do the hit and run style of building when it warms up enough in the garage to make sawdust.
For full range testing, I am planning about 1 liter per driver considering the Vas is 1.12 liters and will line the small boxes with old blue jean denim to soak up some of the internal bounce. The denim will keep the poly fill from being sucked into the vented pole pieces of the NS3-16 FRs.
Would 1 liter be enough to get reasonable results at 80 to 200Hz testing or should I increase the size of the box to give them more room to breath?