Geeze finally I have gotten to the point I can post my plan and see if any of you gurus have some input that may point me in a better direction....if anyone notices something wrong please point it out.
I built a pair of nano neos, and a custom version of a ntn for a center channel and are very impressed. So I got it in my head if I build a wwtww with the same drivers it would probably do what I want for 2 channel without a sub when I have time to sit down and actually listen to the music, rather than watch movies or just have the music in the background (like it always is)....
So here is my plan of attack:
I will build a wwtww based on the nano neo/ntn build. I do not have a golden ear or even close, I do know what sounds good to me but I do not know if what I like is really how it is supposed to sound (aka flat response). I have built the nano neos so I can voice my project off the nanos. I will use Jeff Bagby's software and work off the speaker building bible.
Here is what I have done so far:
I SPL traced the info off of PE's website for the Dayton nd105-4 and the nd20fa-6 (the two drivers used in the nano neos/ntn's). I used this information to enter into response modeler and create simulated in box measurements.
Here is my theory: The Nano's are a proven design and the accuracy of simulated measurements are somewhat questionable, but that is what I have. If I model the nano neos and then model the wwtww I can just do my best to match the curves, right??
The box will be 7"w x 11"d x 50"t this high to get the tweeters to ear level.. (internal airspace 41") just a snidge over 1.1 cu ft with a 3 inch port tuned to 39 hz 1/3 of the way up the airspace box area. This will be a 2.5 way with the outer woofers crossed over to provide baffle step.
So the meat and potatos:
I have attached screenshots of MY simulated version of the nano neos (with the newest crossover off of Chris's album, for comparison to my version of wwtww.
I modeled crossover parts, and I think I am going to start with this:
Tweet: 4.3 uf cap with .25 mh coil (12 db) padded to taste
Inside woofers: .8 mh coil with 3.3 uf cap (12db) zobel: 10 uf cap and 10ohm res
Outside woofs: 2.0 mh coil (6db)
Now this is the crossover for the nano neos used on the inside woofers with the same coil that is on the woofers for the ntn version.
I have attached MY simulated version of the nano neos (1st attachment) and the wwtww is the other two attachments.
Now it was harder to curve fit the two than I thought it would be...are these two curves pretty close to sound similar?
Anyone have any comments?? Does this look like a good start?
Fire away with anything that you may notice- I probably need more help than I realize.....
I have found personally (and others have said it here as well) that what I simulate in PCD matches VERY closely with what I measure. That being the case, my opinion is that what you have modeled in both simulations would be very difficult for me to listen to. There is a 10 dB delta from 100 Hz to 20KHz with the emphasis on the top registers.
Thinking about this a bit more.... I can't believe that the Nano Neos or NTN's have the frequency response curves you simulated. Too many people really like how they sound and I'm pretty sure that wouldn't be the case if they truly had that FR. Not to mention Chris knows what he's doing. So my guess is this:
The ND105-4 frequency response is measure at (as published by PE) 1 Watt/1 meter. Since it is a 4 ohm nominal driver that most likely means 1.00 Volt/1 meter.
The ND20FA-6 frequency response (as published by PE) says 2.83 Volt/1 meter.
That would equate to somewhere around 1.5 dB error in the tweeter's SPL in the simulation. If you just SPL traced the curves and imported them into Response Modeler, then PCD, that could account for the rising response. I have made that same error before
Hope this helps or at least moves you in the right direction
Thinking about this a bit more.... I can't believe that the Nano Neos or NTN's have the frequency response curves you simulated. Too many people really like how they sound and I'm pretty sure that wouldn't be the case if they truly had that FR. Not to mention Chris knows what he's doing. So my guess is this:
The ND105-4 frequency response is measure at (as published by PE) 1 Watt/1 meter. Since it is a 4 ohm nominal driver that most likely means 1.00 Volt/1 meter.
The ND20FA-6 frequency response (as published by PE) says 2.83 Volt/1 meter.
That would equate to somewhere around 1.5 dB error in the tweeter's SPL in the simulation. If you just SPL traced the curves and imported them into Response Modeler, then PCD, that could account for the rising response. I have made that same error before
Hope this helps or at least moves you in the right direction
I was pretty sure my ugly response curves were not "reality" between the continual rise and the big droop in the crossover region.... Would it be advisable to try and fix my files(at least for spl), or try and curve match (my theory) the known good nano neo curve to what I am trying to build??
Hopefully some of the more expert members will jump in here with some great advise, but if it were me.... I would take the raw .frd file you traced in SPL Trace for the tweeter and open it in SPL View. Under the "Tools" menu you can select "Adjust Magnitude". I would adjust the magnitude by -1.5 dB. Then start over by importing that into Response Modeler, adding the correct diffraction and baffle step loss for your baffle dimensions, and finally extract minimum phase. Then I would import the Response Modeler tweeked ND105-4 .frd and .zma and the new ND20FA-6 .frd and .zma files into PCD.
At that point I would work on your very own design by simulating in PCD. I would frequently post up your screen shots from PCD. I am pretty sure there are a good number of helpful members here that can make awsome suggestions and help you realize your goal.
Thinking about it some more, I would adjust the woofers up to 2.83 V/m which would be + 3 dB instead of adjusting the tweeters down. Then everything would be corrected to the same level.
I will have to "adjust" the SPL of those drive units and see how it goes. Thanks for the input, I am glad there is a tool for doing this.
"At that point I would work on your very own design by simulating in PCD. I would frequently post up your screen shots from PCD. I am pretty sure there are a good number of helpful members here that can make awsome suggestions and help you realize your goal."
I actually did work this design in pcd, it just happened to come in very near to an existing crossover, for the 1 woofer existing design, taking the bsc out of the full range woofers. The inductor on the .5 woofers was just a guess.
Good catch on the voltage discrepancy, Craig. I'm sorry I don't have much sage advice on the design itself, as I'm currently wrestling with my own issues with the ND-series woofers. Maybe I'll have more input in a week or so.
Come Get Down And Eat Best Food, Sharp. Cee? Sharp.
I will have to "adjust" the SPL of those drive units and see how it goes. Thanks for the input, I am glad there is a tool for doing this.
"At that point I would work on your very own design by simulating in PCD. I would frequently post up your screen shots from PCD. I am pretty sure there are a good number of helpful members here that can make awsome suggestions and help you realize your goal."
I actually did work this design in pcd, it just happened to come in very near to an existing crossover, for the 1 woofer existing design, taking the bsc out of the full range woofers. The inductor on the .5 woofers was just a guess.
Cliff
Hi Cliff.
Attached is what I THINK the Nano Neos look like. You can see the uncut ND20 (about 91-92dB?) @ 2.83v, and with about a 9-10dB cut to play with the woofer. A single ND105-4 in MY box, cut, and crossed about 3.5k or so. Looks to end up around 81dB with about 3-4dB of BSC.
As I told you, I've fiddled with the tweeter's Darren K. CLIO.txt file AND the SPLTraced ND105 file, to get a response that matched my "in-box" measurements taken last October. THOSE showed a hot spot in the 2k-3k range (in the original 6el XO) which I've toned down on all the current XO designs.
It SEEMS like you're running the current Nano Neos XO, but I don't think that will work real well on your "inside" woofs since they'll be 8n nominal and that low pass was meant for 4n.
Fiddling around with your idea, I'm looking at a 6n series resistor on the high pass ("out front" - instead of the L-pad) and a coil change to 0.20mH.
For the insde woofers I'm looking at something closer to the ntn LP:
1.6mH series coil, 1.5uF shunt cap, Zobel with 5uF and 20 ohms.
For the outside pair, same Zobel with a 4mH low DCR coil. You can most likely "cascade" this 2nd pair in PCD, but I don't think the software I'm using can do that (or I just don't know how).
I see about a +5dB sensitivity gain, with a full +6dBSC on top of that.
The ND105-4 frequency response is measure at (as published by PE) 1 Watt/1 meter. Since it is a 4 ohm nominal driver that most likely means 1.00 Volt/1 meter.
The ND20FA-6 frequency response (as published by PE) says 2.83 Volt/1 meter.
On the spec sheet, the 4n woofer is about 82-1/2dB at 1 watt, which should be TWO volts (not 1v). At 2.83v (which is 2w into 4n nominal), there'd be a +3dB increase. On the PDF plot, to ME, it looks to be more like 81dB. At ANY rate, the plot (which I traced) is certainly NOT at 2.83v, it MUST have been measured at 1w (2v). So, I added 3dB to normalize the sensitivity to 2.83v for my sim (which, I'm guessing, yielded about 84dB over the range where I used it, below 3.5kHz).
Found it. It was listed under the Product Details:
"Dayton Audio ND20FA-6 3/4" Neodymium Dome Tweeter
Don't let the low price fool you! This versatile tweeter is a huge step above the average "budget" tweeter. Conceived with line array thoughts in mind, it is finally possible to achieve the inter-driver spacing needed to minimize comb filtering out to nearly 20 kHz. But its uses don't stop at line arrays-- it has the sensitivity, smoothness, and power handling to individually do a two-way with most small woofers. Or, use it in an A-pillar in your car as a great inexpensive component tweeter. Give them a try for your next budget or micro speaker-- you will be pleasantly surprised! Specifications: *Power handling: 15 watts RMS/30 watts max *VCdia: 3/4" *Impedance: 6 ohms *Re: 5.2 ohms *Frequency range: 3,500 - 25,000 Hz *Fs: 2,005 Hz *SPL: 91.5 dB 2.83V/1m, 90 dB 1W/1m *Dimensions: A: 1-3/4", B: 1-5/16", C: 3/4"."
wwtww with nd105 design thread input please!!
![$vboptions[bbtitle]](images/techTalkHeader-leftGray2.jpg)
![$vboptions[bbtitle]](images/techTalkHeaderRepeatGray3.jpg)
Reply With Quote
Originally Posted by PWR RYD
