no, I would not use the same resistor for mid and tweeter, those sections would be completely separate
I have already posted Don's crossover, see back, I do not think it uses equal sections.
I'm sold,
If 14 of them can handle a crossover at 160 Hz without issues, then 21 of them should not have any issues crossed at 400 Hz. My plan is to run 7 parallel/3 series in their own 0.4 cubic foot (12 liter) lightly stuffed sealed enclosure to protect them from the twelve 5" woofers pressure. I've heard it is best to divide each parallel string into it's own enclosure before running them in series? Basically, divide the mid box into 3 pieces?
My woofers are 3 segments of 4 parallel/3 series so with the 7P/3S configuration so I can play around with shading if I feel like it. Thanks for the update and advice on arraying the Aurasound 3" FRs.
If 14 of them can handle a crossover at 160 Hz without issues, then 21 of them should not have any issues crossed at 400 Hz. My plan is to run 7 parallel/3 series in their own 0.4 cubic foot (12 liter) lightly stuffed sealed enclosure to protect them from the twelve 5" woofers pressure. I've heard it is best to divide each parallel string into it's own enclosure before running them in series? Basically, divide the mid box into 3 pieces?
My woofers are 3 segments of 4 parallel/3 series so with the 7P/3S configuration so I can play around with shading if I feel like it. Thanks for the update and advice on arraying the Aurasound 3" FRs.
Yeah, I don't remember now on the sections but was thinking it was by groups of 6, with one group unattenuated. Will have to double check that.
Edit: Earlier I had said groups of 4, but the schematic you show has groups of parallel drivers to total over 40 drivers. Maybe that was for the tweeter. Anyway, whatever scheme you come up with as long as it's shaded the proper amount, that's what matters.
Nice find on the drivers, BTW.
Edit: Earlier I had said groups of 4, but the schematic you show has groups of parallel drivers to total over 40 drivers. Maybe that was for the tweeter. Anyway, whatever scheme you come up with as long as it's shaded the proper amount, that's what matters.
Nice find on the drivers, BTW.
Hi Pano, I know that shading is important. An if you ask Don, surely he will tell you its important. But I am not so sure how very important it is.
People are listening to straight line arrays, some are using power tapering, some are not. Which ones are right?
You see, the whole concept of array is already an compromise. Sometimes we have to make right compromises, sometimes not. When it comes to the shading in Keele's CBT, I am not so sure its absolutely necassary. See, the curved shape is already doing great job! Please compare the two pictures attached (I hope it works, I coppied them from Dons lecture).
As you can see, there is not that much difference in directivity and beamwidth. Its much smooter in shaded, but its not that different.
Attachments
Now to continue the discussion, for me there are cons and pros for Legendre Shading. Offcourse Don will tell you its essential, he is probably perfectionist. But to me the added complexity is a drawback. I have heard his CBT array at RMAF this year, I talked to him. My concern with shading is that those poor little fullrangers at the bottom of array, those 0 dB ones are working hardest, you can see them moving!, yet those ones on the top are barely working. And all that added complexity in the wireing of the crossover.
And the benefit? I admit, nearly ideal directivity index. Well, nothing is ideal, and believe me, the listening room will brink way more irregularities than Legendre shading or not. Having all sharing the same power gives me way more room for power handling, I guess.
My opinion.
And the benefit? I admit, nearly ideal directivity index. Well, nothing is ideal, and believe me, the listening room will brink way more irregularities than Legendre shading or not. Having all sharing the same power gives me way more room for power handling, I guess.
My opinion.
Thanks for the charts,
Your RTA display with the NS3 full ranges showed an increase in output from 1.5KHz to 4KHz from what I'm assuming to be the rise shown on the Aurasounds frequency graphs. According to Don's charts, you should see a drop off in that range even though the NS3 will rise--but not enough to make up for the drop. In reality, that does not show up so it is theory VS reality yet again?
My tweeter line does show that downward tilt--they "should" put out around 106dB at one watt considering there are 48 of them at 90 to 93dB each from 4,500Hz to 20KHz. I am assuming about a 3dB loss per octave as it goes up, the Apex Jr. tweeters have a 3dB rise from 4,300Hz to 8,000Hz which negates the natural loss to around 10KHz. It could use a bump at 10KHz+ for a bit more "air" but not too objectionable. The woofer array is around 96dB so it is obvious I'm losing tweeter output since they should be cooking my hair and don't.
Does curving the array limit the amount of downward tilt as the frequency rises? Personally, I know the loss is there but not as severe as Don's charts show. It is there so I'll add an array of NS3 mids to take advantage of their rising response in that region and to limit beaming when crossing them at 4,300Hz.
Your RTA display with the NS3 full ranges showed an increase in output from 1.5KHz to 4KHz from what I'm assuming to be the rise shown on the Aurasounds frequency graphs. According to Don's charts, you should see a drop off in that range even though the NS3 will rise--but not enough to make up for the drop. In reality, that does not show up so it is theory VS reality yet again?
My tweeter line does show that downward tilt--they "should" put out around 106dB at one watt considering there are 48 of them at 90 to 93dB each from 4,500Hz to 20KHz. I am assuming about a 3dB loss per octave as it goes up, the Apex Jr. tweeters have a 3dB rise from 4,300Hz to 8,000Hz which negates the natural loss to around 10KHz. It could use a bump at 10KHz+ for a bit more "air" but not too objectionable. The woofer array is around 96dB so it is obvious I'm losing tweeter output since they should be cooking my hair and don't.
Does curving the array limit the amount of downward tilt as the frequency rises? Personally, I know the loss is there but not as severe as Don's charts show. It is there so I'll add an array of NS3 mids to take advantage of their rising response in that region and to limit beaming when crossing them at 4,300Hz.
Yes, but a valid one, I think. It's an interesting question. Keele's charts certainly show improvements with the shading, but it isn't night and day.
I also saw those poor little drivers down at the bottom working real hard and smelled the hot resistors!
Considering the killer price you got on the drivers, you might end up spending as much on the shading resistors. Of course I really want you to try it and report your findings on the differences, but I'm not buying the resistors nor doing all the wiring, am I?
i thing that much more interesting overlay is this
this is 1ft infront of the cbt array, again, no eq, in the middle of array, directly in on axis, (blue) than 20 degrees of axis (magenta), and 45 degree (green) of axis...
i thing it nicely shows cbt behaviour
(i took more angles, 5, 10, 30...but the measurement just fits righ in beween shown)
this is 1ft infront of the cbt array, again, no eq, in the middle of array, directly in on axis, (blue) than 20 degrees of axis (magenta), and 45 degree (green) of axis...
i thing it nicely shows cbt behaviour
(i took more angles, 5, 10, 30...but the measurement just fits righ in beween shown)
Attachments
Last edited:
since the 3" auras rolled of after 10k, I had to put tweeters on those
I added 18 tweeters, wired as 16 ohm to match midranges
crossover is single cap only
attached are two pictures for those who are interested
also the fr response with tweeters (no sub attached, just an array)
measures well, sounds good
I added 18 tweeters, wired as 16 ohm to match midranges
crossover is single cap only
attached are two pictures for those who are interested
also the fr response with tweeters (no sub attached, just an array)
measures well, sounds good
Attachments
now, when I was adjusting the subwoofer level, something strange happened...
(this is old ugly sub which I use as stand mostly, but this time I had to put electronics off the sub, since it was shaking the cd player quite a bit
its a closed box of unknown volume, pretty big, with two 15" woofers
yes, there are two woofers, one on each side
the woofers are BOSS unknown model)
anyway, long story short, I tested the subwoofers configured as dipol or bipol
simply by switching the polarity on one woofer
subwoofer sounded much better in di-polar configuration, the bass was cleaner, deeper...
so I wanted to verify what I was hearing, so I did more detailed measurement
what measurement showed was that not only there was no cancelation or loss in di-polar woofers arrangement, there was gain ~5-6 dB!
I recommend to experiment with di-polar subwoofers, it seems to me that when woofers are not fighting against each other, creating big ups and downs in preasure inside the box, they move more air in the room
that's what my measument shows anyway, and I am sure many will not believe
I have no special theory about it, just that in di-polar arrangement the membranes move in unisono, so they move more freely, when in bi-polar configuration one is preventing the other to move
(this is old ugly sub which I use as stand mostly, but this time I had to put electronics off the sub, since it was shaking the cd player quite a bit
its a closed box of unknown volume, pretty big, with two 15" woofers
yes, there are two woofers, one on each side
the woofers are BOSS unknown model)
anyway, long story short, I tested the subwoofers configured as dipol or bipol
simply by switching the polarity on one woofer
subwoofer sounded much better in di-polar configuration, the bass was cleaner, deeper...
so I wanted to verify what I was hearing, so I did more detailed measurement
what measurement showed was that not only there was no cancelation or loss in di-polar woofers arrangement, there was gain ~5-6 dB!
I recommend to experiment with di-polar subwoofers, it seems to me that when woofers are not fighting against each other, creating big ups and downs in preasure inside the box, they move more air in the room
that's what my measument shows anyway, and I am sure many will not believe
I have no special theory about it, just that in di-polar arrangement the membranes move in unisono, so they move more freely, when in bi-polar configuration one is preventing the other to move
Attachments
Thank you for the RTA readings,
I'm still breaking in my box of NS3-16's, 31 of them completed and will start building the new bezels in two weeks. Such is the slow progress in the winter as I have an affinity to breaking in speakers when the temps drop below freezing.
After listening to my 2-way line arrays, I've come to the conclusion that in the upper treble range, they hit a "wall" and the output of the line hit's it max at the higher frequencies and any additional drivers adds sound output to lower frequencies.
My wild guess is by using 21 drivers, the output should increase below 2KHz but not above it, maybe smooth the response even flatter? I'm crossing the NS3's at 400 and 6,000 Hz to woofer and tweeter lines so on paper--it is looking better all the time.
Thanks again for the results--I'll have mine in about a month and hope I can get close to your readings.
I'm still breaking in my box of NS3-16's, 31 of them completed and will start building the new bezels in two weeks. Such is the slow progress in the winter as I have an affinity to breaking in speakers when the temps drop below freezing.
After listening to my 2-way line arrays, I've come to the conclusion that in the upper treble range, they hit a "wall" and the output of the line hit's it max at the higher frequencies and any additional drivers adds sound output to lower frequencies.
My wild guess is by using 21 drivers, the output should increase below 2KHz but not above it, maybe smooth the response even flatter? I'm crossing the NS3's at 400 and 6,000 Hz to woofer and tweeter lines so on paper--it is looking better all the time.
Thanks again for the results--I'll have mine in about a month and hope I can get close to your readings.
You may have already passed over the idea of shading, but I thought I'd bring up the idea of using multi-tap transformers instead of resistor networks.
For example, on my shelf I have a number of Tannoy THP-60 and THP-30 transformers out of ceiling speakers. They have taps for -3, -6, and -9dB, which correspond to Keele's shading steps, IIRC. It would raise the array efficiency somewhat, and might even be less expensive depending on on the resistors it replaces.
For example, on my shelf I have a number of Tannoy THP-60 and THP-30 transformers out of ceiling speakers. They have taps for -3, -6, and -9dB, which correspond to Keele's shading steps, IIRC. It would raise the array efficiency somewhat, and might even be less expensive depending on on the resistors it replaces.
Hi, adason! Tell me, in your opinion: what's the principial difference between Keele's radial arrays with electrical shading and J-arrays, wide used in professional sound reinforcement, where the question of having of equal levels on each vertical direction and distance is solved by different number of drivers in each vertical direction. Sorry for bad english! Thanks!
Hi guys, sorry for the delay in reply...I was not expecting any posts in this old thread.
Bill, thanks for good suggestion, as a matter of fact, I do have similar transformers...the ones from 70 volt distrubution boxes...I might experiment with those, but for now the arrays are sounding superbly.
rogozhin, I know nothing about J-arrays, I only gathered info about Don's CBTs
Bill, thanks for good suggestion, as a matter of fact, I do have similar transformers...the ones from 70 volt distrubution boxes...I might experiment with those, but for now the arrays are sounding superbly.
rogozhin, I know nothing about J-arrays, I only gathered info about Don's CBTs
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.