Why does driver arraying have to be difficult? You can use my Near Field Line Array White Paper for this. See this link:
http://www.audioroundtable.com/misc/nflawp.pdf
The good stuff is on Page 17.
I quote:
"While ultimately the SPL of the speaker will be measured, in the development process you can compute the system sensitivity (or efficiency) impact of a line array. First, assume that an individual driver in the line has a known SPL value. Next assume that the drivers in the line have overlapping acoustical radiation patterns and are spaced within a wavelength center-to-center from each other as stated in Table II. Thus, the acoustical improvement (efficiency gain) at 1 watts, 1 meter distance is given by:
Efficiency Gain = 10*log (Number of Drivers Driven)
while the sensitivity gain or loss at 2.83v, 1 meter is:
Sensitivity Gain/Loss = 10*log (Nominal Driver Impedance/Nominal Array Impedance)
If the nominal array impedance is less than the individual driver impedance, the array sensitivity increases or is a gain. If the array impedance is greater than the individual driver impedance, then the sensitivity decreases or becomes a loss.
Hence, for the overall system
System Efficiency = SPL + Efficiency Gain
System Sensitivity = SPL + Efficiency Gain + Sensitivity Gain/Loss
I give an example on Page 18.
Jim
http://www.audioroundtable.com/misc/nflawp.pdf
The good stuff is on Page 17.
I quote:
"While ultimately the SPL of the speaker will be measured, in the development process you can compute the system sensitivity (or efficiency) impact of a line array. First, assume that an individual driver in the line has a known SPL value. Next assume that the drivers in the line have overlapping acoustical radiation patterns and are spaced within a wavelength center-to-center from each other as stated in Table II. Thus, the acoustical improvement (efficiency gain) at 1 watts, 1 meter distance is given by:
Efficiency Gain = 10*log (Number of Drivers Driven)
while the sensitivity gain or loss at 2.83v, 1 meter is:
Sensitivity Gain/Loss = 10*log (Nominal Driver Impedance/Nominal Array Impedance)
If the nominal array impedance is less than the individual driver impedance, the array sensitivity increases or is a gain. If the array impedance is greater than the individual driver impedance, then the sensitivity decreases or becomes a loss.
Hence, for the overall system
System Efficiency = SPL + Efficiency Gain
System Sensitivity = SPL + Efficiency Gain + Sensitivity Gain/Loss
I give an example on Page 18.
Jim
Correct.
You maintain the single driver 4 Ohms so no gain or loss there.
You double the Sd from 1 driver to two drivers which gives you +3dB,
then double it again from two drivers to four drivers which gives you another +3dB, so a total gain of 6dB.
All the best
Derek.
Excellent, thanks Jim, that seems to support the statements from Derek and Kindhornman very well.
Gain with use of 4 drivers with same impedance = 10* log(4) = 6.02 dB.
I had seen and used your method to figure out the theoretical gain in my own specific case.
A couple random questions. I've been spending most of my free time trying to learn as much as possible about anything array related. A couple things I've read that I'm not sure whether was someone's speculation, imagination or there is actual relevance.......
1. When series parallel wiring drivers that there will be phase issues between series sets that will be audible. I've heard to keep series sets in separate cabinet space. I've also read that if possible it's best to series OR parallel wire the whole string of drivers but with arrays with 20+ drivers I've yet to encounter or hear of any amp that would handle this job.
2. I believe in one of the links here of another DIY'er that had a positive array experience claimed that soldering all the connections would or could introduce impedance issues and suggested crimp connectors. I haven't used those in the last 20 years even in car audio.
Any input would be great. Thanks,
Wes
1. When series parallel wiring drivers that there will be phase issues between series sets that will be audible. I've heard to keep series sets in separate cabinet space. I've also read that if possible it's best to series OR parallel wire the whole string of drivers but with arrays with 20+ drivers I've yet to encounter or hear of any amp that would handle this job.
2. I believe in one of the links here of another DIY'er that had a positive array experience claimed that soldering all the connections would or could introduce impedance issues and suggested crimp connectors. I haven't used those in the last 20 years even in car audio.
Any input would be great. Thanks,
Wes
That's a hard question to answer, as there will be lots of opinions either way. 
I have read the same claims, and claims that say the opposite as well. As I haven't done any tests on this subject I remain neutral until I've seen some convincing proof. It would have to be a bit more than subjective info to sway me either way.
This picture will show a bit of what I did though:
The crimp connectors wasn't done for the reason you stated above. I wanted a serviceable construction, being able to correct problems if they should arise. Though I've read several times that soldered connections aren't always the best choice. I've measured differences in wiring. Another hot debate. So it would make sense there can be differences in soldered connections as well.
About the sharing of cabinet space I've also read lots of claims. One that stuck in my head was something Lynn Olsen wrote on his big Ariel thread. It made me close each chamber for at least the first area behind each driver. Making sure that any connection between spaces is a damped passage. I don't have one big open space as seen above, but every chamber is still connected internally, though at different spots, not one or two holes down the entire line. Again, no scientific proof so no definitive standpoint from my side.
At one point every time I read the forum I wanted to adjust my plans, even when the build had started. One day I decided to trust in what I had and just go for that. There are lots of claims about everything. As many claims to dispute it too. In the end, you'll have to make up your own mind.
Even if I didn't answer the question directly, you've seen my solution. I hope it helps you make up your own mind. People's claims can be very persuasive. If you put enough believe in a claim you will hear the evidence. Even if it isn't valid. It's harder to unlearn such believes. It made me measure every step of the way. Even though I do think I have good ears, once you have fooled yourself you might start to think a bit different about these subjects.
I have read the same claims, and claims that say the opposite as well. As I haven't done any tests on this subject I remain neutral until I've seen some convincing proof. It would have to be a bit more than subjective info to sway me either way.
This picture will show a bit of what I did though:
The crimp connectors wasn't done for the reason you stated above. I wanted a serviceable construction, being able to correct problems if they should arise. Though I've read several times that soldered connections aren't always the best choice. I've measured differences in wiring. Another hot debate. So it would make sense there can be differences in soldered connections as well.
About the sharing of cabinet space I've also read lots of claims. One that stuck in my head was something Lynn Olsen wrote on his big Ariel thread. It made me close each chamber for at least the first area behind each driver. Making sure that any connection between spaces is a damped passage. I don't have one big open space as seen above, but every chamber is still connected internally, though at different spots, not one or two holes down the entire line. Again, no scientific proof so no definitive standpoint from my side.
At one point every time I read the forum I wanted to adjust my plans, even when the build had started. One day I decided to trust in what I had and just go for that. There are lots of claims about everything. As many claims to dispute it too. In the end, you'll have to make up your own mind.
Even if I didn't answer the question directly, you've seen my solution. I hope it helps you make up your own mind. People's claims can be very persuasive. If you put enough believe in a claim you will hear the evidence. Even if it isn't valid. It's harder to unlearn such believes. It made me measure every step of the way. Even though I do think I have good ears, once you have fooled yourself you might start to think a bit different about these subjects.
Yes he does, as well as many others. I bet there are threads on here talking about this subject though. These kind of discussions never end. As with many subjects they remain open ended for discussion and belief. I 've picked my poison. Not planning on a change any time soon. We've been there before on this thread too: Solder vs crimped
Part of the question about the crimped vs solder story is missing in this. There are differences in solder type and there are major differences in crimped connections. While I worked in Aerospace most but not all connections were done with crimped connection, on the circuit level obviously you use solder but for connectors and such it would be a crimped connection. But this is where it gets important. the crimped connections that you will make with a typical hand held crimping tool is not the same as what is used at a professional level. The difference being that a professional crimped connection is expected to be gas tight, and I fail to see how any of the cheap hand held crimping tools can meet that standard. A long life crimped connection will only survive with low impedance and little degradation if it is originally gas tight. If not then you will find that you are better off with the soldered connection, but make sure you clean the connection after soldered just as it is recommended that a circuit board is cleaned of flux after assembly. There is no perfect solution but the crimped connection that we make with consumer crimping tools have no real advantage over a soldered connection besides the option of using male/ female connectors that we can easily disconnect. For a permanent connection I personally would just use good old fashioned lead silver solder and forget about it.
Kindhornman,
That's pretty much my take on it. I suppose whatever works for your situation is what dictates your method of choice. I highly doubt anyone has conclusively determined either method to have an audible advantage. If some of the well respected designs use crimp then I suppose it's fine for me though it goes against what I've always thought was best. I guess I will flip a coin or something.
Wes
That's pretty much my take on it. I suppose whatever works for your situation is what dictates your method of choice. I highly doubt anyone has conclusively determined either method to have an audible advantage. If some of the well respected designs use crimp then I suppose it's fine for me though it goes against what I've always thought was best. I guess I will flip a coin or something.
Wes
Good thing I crimped the connectors on the cable as best as I could and soldered it up once crimped. Basically to seal it off. But I didn't want to solder the wires to the drivers. Wait, I didn't...
As said, many options and opinions. It's how well it holds up over the years that will be important down the road.
Yes, it's hard enough to get it right if you want to keep the drivers as close as possible. That doesn't leave a whole lot of room between the drivers. Where the connectors are I have MLV rubber (as seen in the picture) as the divider. 2 drivers are mounted on each side of that rubber "wall" with connectors facing each other. It made wiring things up a little easier. Plus the connectors would otherwise be mounted on the same side everywhere. (alternating the driver orientation solves that).
The things we worry about (lol)
As said, many options and opinions. It's how well it holds up over the years that will be important down the road.
Yes, it's hard enough to get it right if you want to keep the drivers as close as possible. That doesn't leave a whole lot of room between the drivers. Where the connectors are I have MLV rubber (as seen in the picture) as the divider. 2 drivers are mounted on each side of that rubber "wall" with connectors facing each other. It made wiring things up a little easier. Plus the connectors would otherwise be mounted on the same side everywhere. (alternating the driver orientation solves that).
The things we worry about (lol)
Last edited:
No kidding. There's a fine line between attention to detail when planning your project and too much attention to detail that prevents you from ever making any progress. I've spent many a night losing sleep over something I haven't even begun to build. Thank you all for what you contribute to this community as I would most likely never consider this project without the knowledge I've gained here.
Wes
Sample Music? Test Music?
Wesayso,
Just a question about your array and what sort of music you use to test the system? I found some recommendations on the internet and tried one of them out yesterday with interesting results.
Steely Dan – Jack Of Speed
Daft Punk – Get Lucky
Crooked Still – Little Sadie
https://www.cambridgeaudio.com/blog/test-your-speakers-cambridge-audio-engineer
http://www.diyaudio.com/forums/mult...s-work-tested-simple-terms-8.html#post4628944
How do you find the quality of You Tube videos? Are there free high res music downloads that you use to test, and maybe tell us how it sounds? I think there may be some useful information there.
Maybe you could try this video or if you have the high res even better.
Ave Maria - Schubert (Michael Lucarelli, Classical guitar ...
classical guitar you tube ave maria
3:58
https://www.youtube.com/watch?v=LNUJA9N14eM
My home theatre system (sub + 2 speakers at the moment) just made it sound incredible. Very moving. Why is guitar like this?
Wesayso,
Just a question about your array and what sort of music you use to test the system? I found some recommendations on the internet and tried one of them out yesterday with interesting results.
Steely Dan – Jack Of Speed
Daft Punk – Get Lucky
Crooked Still – Little Sadie
https://www.cambridgeaudio.com/blog/test-your-speakers-cambridge-audio-engineer
http://www.diyaudio.com/forums/mult...s-work-tested-simple-terms-8.html#post4628944
How do you find the quality of You Tube videos? Are there free high res music downloads that you use to test, and maybe tell us how it sounds? I think there may be some useful information there.
Maybe you could try this video or if you have the high res even better.
Ave Maria - Schubert (Michael Lucarelli, Classical guitar ...
classical guitar you tube ave maria
3:58
https://www.youtube.com/watch?v=LNUJA9N14eM
My home theatre system (sub + 2 speakers at the moment) just made it sound incredible. Very moving. Why is guitar like this?
Last edited:
Youtube is usually crap, even at 1080 resolution.
Here's a link to some nice classical recordings, in hi-res. They have different format to play with, like DSD files. Stereo and 5.1 surround files too.
2L High Resolution Music .:. free TEST BENCH
Here's a link to some nice classical recordings, in hi-res. They have different format to play with, like DSD files. Stereo and 5.1 surround files too.
2L High Resolution Music .:. free TEST BENCH
Thanks! Any high res files should be OK? Popular music? How will know it's high res?
So, I did some tests on the tiny Tymphany PMT drivers, and yes, 4 of them net 6 db over one driver if two pairs are connected in parallel each and then in series, i.e., get the same impedance as one driver.
But, I just can't figure out why the formula doesn't reflect this:
SPL db/2.83V/1m = 20 x log10 (Bl x Sd/Re/Mms) + 88.5. Quadrupling the area in this equation while maintaining the same Re gets you +12 db.
But, I just can't figure out why the formula doesn't reflect this:
SPL db/2.83V/1m = 20 x log10 (Bl x Sd/Re/Mms) + 88.5. Quadrupling the area in this equation while maintaining the same Re gets you +12 db.
ra7,
Not sure where you got that formula from but it seems that one of those factors should be a denominator and everything shouldn't be multiplied. Increasing the moving mass by four should decrease the output it seems rather than that being a multiplier. Something doesn't appear correct in that equation.
Not sure where you got that formula from but it seems that one of those factors should be a denominator and everything shouldn't be multiplied. Increasing the moving mass by four should decrease the output it seems rather than that being a multiplier. Something doesn't appear correct in that equation.
I'd have to agree with Perceval that youtube isn't cutting it.
I've heard some high res material that did sound quite convincing, but I've also heard CD's that were at least as good as their high res counterpart.
To me it seems more a question of final mastering differences.
For example, I have Steely Dan's album Gaucho, which is one of the reference albums for me, on CD and as a high res counterpart. I can't say the high res does any better than the CD for me.
At one point I thought High Res really was better, until I discovered it changed the timing of my ambient channels. Once I took care of those differences I had a difficult time telling them apart.
There are some songs I still like better on high res though. Maybe they just did a better job on the mastering.
I had seen your list, even though my collection of tracks differs it is similar in diversity of content. I just try a lot of different material and some songs are keepers to be called my personal reference tracks. It can be that I get someone over to visit/listen and play one of their favourite songs that I add it to my list. I try to keep it as diverse as possible.
One thing I did notice, the more I get it right (for me) in tonal balance and measurements (used as a control tool), the more everything I play just seems to work.
Spend some time to get more direct vs reflected (room) sound. That's where the biggest leap comes from i.m.h.o. Try and minimise early reflections, even if it's just temporary. You'll notice the difference.
Last edited: