Beyond the Modula NeoD CC (split from original thread)

[JonMarsh]

Thanks, Thomas, but my taper jig is a much more attractive shade of blue; I'm sure that will improve the quality of the cuts obtained, regardless of the price of the respective saws! :rofl:

Actually, this project will probably require a combination of techniques- Tom's taper jig fixture plus some things I did with the M8ta.

Keep in mind a lot of documentation and the web site will probably be up in about three weeks- including a lot of piece part documentation, and intermediate step documentation.

The basic construction for the front panels in the first version (premium gas) requires four layers of 3/4" MDF; a 1" X 12" X 4' LBL panel (laminated Bamboo lumber) and a 1/4" MDF panel cut to the same size. Two sections are built up 1-/2" thick by gluing together 3/4" MDF a foot wide to start. These are the "middle" and "back" panels of the front panel. The "front" part of the front panel is layered up from the LBL panel and the 1/4" MDF/HDF panel. There's work done on all three sections before bringing them together, including back chamfer on the rear of the "front" part of the front panel. Particularly for the front panel I'll be explaining/describing a detailed methodology for construction and fabrication. Naturally, you're free to innovate... just be thoughtful and careful. :W

BTW, enclosure side and back walls are 1" thick, and unless you have ready access to 1" MDF (I don't, except in full size sheets which are too big for me to handle in my work space by myself and with my previous back surgery), so these are layered up from MDF 2' X 4' thinner panels, like the front panel. Bottom and top are 1-1/4 thick.

All will be revealed soon.... :W

[JonMarsh]

I'm also excited about this project, and give a hearty thanks-in-advance for the level of effort and documentation you're putting in to this.

I think your cautionary note is one that could well be factored out and made a sticky. Even looking at the 'budget' version, materials look like they'll run at least $1k per pair, possibly more depending on final xover. This is in line with the NeoDCC, but still may be a bit surprising to some.

For my part, I'm looking at this project as something of a stake-in-the-ground. I've been experimenting with my own projects for ~15 years or more, but have never produced a 'finished' product. Time to put up or shut up, and the fact that it's ambitious is of course part of the attraction. Plus, it's a excuse to finally get a vacuum press :T Work won't probably start in earnest until the fall, but there's lots of shop prep needed before then anyway.




Can I interpret this as an indication that the BT3K will be sufficient for the facets? The fact that you used one for the M8 project was (I think - memory is getting fuzzy) a part of what drew me to the saw in the first place.

Yes, I believe that the BT3K or BT3100 would be sufficient, with the right fixtures and methods. At this time I have a "Frankensaw" which is part BT3100 and part Craftsman 21829. Blades are a mix of Forrest and Rigid and Freud. Specific blades will be recommended for specific cutting operations, that is the tech specs, not the brand necessarily.

I expect the crossovers will run around $300 each or more- most likely more, due to the size of components needed for a ~300 Hz lower midrange crossover and zobel. So even a pair in the low cost configuration, crossover parts and drivers are looking most likely at $300 for drivers plus $600 for crossovers let's say $900 to $1,000. That actually compares pretty favorably with the SDX-1 NeoD CC, which is about the same.

Additionally, for this larger cabinet, internal wiring is an additional expense due to length; the SDX-1 NeoD CC used Cardas Cross speaker cable, as did the M8ta; I'll be using that on the woofers and midrange, but braided Kimber 4TC cable for it's lower inductance on tweeter. Naturally, this is something anyone can substitute with what they're comfortable with price or performance wise.

The cabinet cost will be another matter; I haven't tried to estimate that yet, but will before I'm done, for now it's in the noise floor. I've chosen a moderate cost controlled approach to veneering, mixing paper backed veneers and nicer figured and birds eye solid veneers in Maple- the latter are a bit more spendy, but under control, and of course that part is "optional". More signifiant is probably tools and clamps. I have more routers and clamps an any normal or sane person living in a two bedroom condo/townhome would be expected to have. But I've bought more for doing this project. Fortunately Woodcraft was having a sale on 24" Bessy clamps, and if you shop around you can find some good prices on tools these days. Picked up a Bosch Colt variable speed router to facilitate some of the work, with some of the accessory mounts. I also have a DeWalt 621 setup with a Jaspar jig, two Hitachi 12V semi-dedicated to certain tasks, a Bosch router on a Freud Table, a Porter cable trim router dedicated to my Veneer tooling fixture, and a Ryobi router on one of my router table plates for the Frankensaw for dado type work. I know that's seriously wacked, but it's also productive. One other very desirable tool is a biscuit joiner; I'll be using my DeWalt on this project again. Though I'm really just a wires and sparks kind of guy, I don't treat the wood working side lightly. :W


I think you have the right idea, doing your research and prepping your shop to get ready to work. All the info you need to determine if this is a path you'd like to try should be up in a few weeks. I'm also planning on taking these to ThomasWs in Denver during RMAF barring any work disaster or other unplanned events. Usually we get together with some of the folks we're in touch with over the internet in past years during RMAF, so there might be a chance for others to hear these at Thomas's and offer somewhat less biased opinions than Thomas or I might have. I'll be bringing out a complete small system to demo them with, running full range, no hookup to Thomas's IB subs. :W

[CraigJ]

Jon,

Looking extremely good, and can't wait to see your results. In the meantime, is there a link to show us what a finished diy aluminum plate, like what we will be using with the RS100, looks like? I believe that I understand what's to be done, but a picture will most likely say 1000 words.

Again, thanks you and am looking forward to hearing my first Deulund style crossover. Actually, I may get to solder my first crossover yet. :rofl:

BTW, have you had a chance to try out the "new and improved" BT3100?

Craig

[JonMarsh]

I have done some of my first prep work on panels and ripping- just planning on putting a new blade in it this afternoon, a Forrest Woodworker II with a 0.125 kerf and 15 degree ATB teeth for the heavy ripping work. It has longer rails than my BT3100 originally came with; now I have two router plate pieces, one with and one without the router, and I think it may be handy for cutting some of the front panel pieces to have two linked sliding miter tables. When I said "Frankensaw", I meant "Frankensaw". :W Otherwise all the hardware behaves pretty much the same; has a nicer motor switch; motor appears to be mostly the same; I have a spare new BT3100 motor I picked up on eBay a couple years ago just in case, which appears to be compatible. One of the fences is now setup permanently as a router fence. This is the way I like to work- set stuff up carefully, once, then don't tear it down. When you have limited time like I do, it's productive that way! :T

I'll see if I can't get a piece drawing of the adapter plate done this weekend and post it- it's NBD, just 1/4" aluminum plate cut to the outside dimensions and corner radius of the C79-6, and inside dimensions cut to the 3" hole cutout the RS100 needs; I figure I'll just tap the aluminum for 6-32 screws to mount; with six of them, things should fit fine, do a thin film RTV skinned gasket. I'll probably black epoxy paint the piece after fabrication. With cutting wax and an old carbide blade you can cut aluminum like that on the power saw, just take the feed VERY easy. The best blade to use is a multi-tooth thick kerf carbide with a negative hook angle of 5-7 degrees and a triple angle face grind alternated with a straight grind; but most folks won't probably go out and buy a $100 blade to mount a $25 driver. Of course, in manufacturing environment, that's the sort of thing you do. Used to make custom heatsinks that way from extrusions back in the 70's when prototyping power amplifiers.

I am probably going to build 4 cabinets right off the bat, so I can be mounting and testing drivers in one set, and veneering and finishing the other set.

[CraigJ]

So, do you want me out Sunday the 21st, or Monday the 22nd?

[JonMarsh]

Well, it depends, Craig- if you came out Sunday, we could do like we do for work on special events, have a team "Kick off" dinner, review the planned agenda, create a joint mission statement, you know, build up to launch a stretch goal! :W

Otherwise, I'd suggest Monday, so you can have this last weekend with your family, because after ET gets done working you to the bone, you'll have wished you signed up for a two week labor stint in the Bespin mines instead! :W When you see them next, you'll never be the same.... :E :twisted:

[bigg]

RS100/aluminum plate

Hi Jon,

You are going to use the RS100 for the midrange? I have never heard of the rear masking problem before (newbie). Is this a problem with that specific driver or is it caused by the deep front baffle? Your knowledge and energy in speaker design humbles a lot of us.

On a another note. Have you ever tried the Supra Classic wire from Madisound? I used it in my EROS 2s and I love it. No measurements, just my ears. OTOH, I didn't use anything else in them to campare to. Thank you for your time.

[---k---]

Hi Jon,

You are going to use the RS100 for the midrange? I have never heard of the rear masking problem before (newbie). Is this a problem with that specific driver or is it caused by the deep front baffle? Your knowledge and energy in speaker design humbles a lot of us.

It is caused by the shallow depth of the driver and the thickness of the baffle. You'll notice in almost all of the Mission Accomplished projects, the back of the baffle needs to be rounded over to let the driver breath and prevent rear masking. The smaller the driver and thicker the baffle, the more important this is. I think Jon documents it well in the Modula MTM thread.

John Krutke had a post a few months back discussing the NHT closeout speakers. They didn't have the back of the baffle rounded over. His measurements showed problems in the frequency response due to this.

Jon's super front baffle makes this even more of a problem, which is why he is discussing the aluminum plate route.

[bigg]

SDX-2 Ardent

Thanks ---K---,
That makes perfect sense! I'll make sure to check that for all my projects. :thanku:

[JonMarsh]

VERY much on the money, Ryan. Thanks for saving me the words. Just a reminder, this is what an RS100 looks like...



With a thick front baffle, this driver would be hopelessly rear masked; even a 3/4" panel with significant chamfer is less than ideal for this little guy, which is otherwise a very nice part for the money. This is a common situation with 4" to 5" drivers if they don't use neodymium magnets. There was a very nice Seas 5" woofer in the 80's that I used to build 2.5 way near field monitors for some of the recording studios in Boulder, with a very large magnet and short window openings in the frame; I custom fabricated and had anodized at a local vendor a complete aluminum front panel assembly for those. It was the only way around the issue, and it worked so much better than an attempt at a MDF panel sonically and structurally, it was also a different way of getting a very stiff launch panel for the drivers. I still had some of those panels left that I didn't toss until about ten years ago.

The C79-6 also has a shallow rear window, so the rear masking characteristics and complexity of baffle design are also important. It also has a very, very stiff frame. A desirable characteristic in a midrange driver where the resonance frequency of magnet and basket may fall within the acoustic transfer range of the driver.


Bigg, there are two versions being designed and built for this project.

The "premium grade", with dual Seas ER18RNX midwoofers, Accuton C79-6 midrange, and Scanspeak D3004/662001 tweeter, and the "budget value version", with dual Dayton RS180-8 midwoofers, Dayton RS100-4 as the midrange, and Seas 27TBFCG as the tweeter.

Using the same baffle mount and cutout for the RS100 as the C79 will keep open the option of upgrading the system configuration, though obviously at substantial expense. It would be possible to come up with a midrange driver crossover only upgrade path, if there's interest.

Electrical parameters for crossover components will be given (inductor DCR, for example), and a suggested BOM based on Evil Twin's recommendations for performance, and then anyone has the option to substitute lower cost components. Obviously the crossovers will be different electrically.

I've never tried the Supra classic bulk cable; it looks like a very good quality
zip cord with good metallurgical characteristics and a low corrosion insulation. I personally favor wire constructions with multiple sizes of high quality OFC conductors at golden mean ratios with braided geometries that reduce inductance. A minimal example of that sort of construction is the Kimber 4TC and 8TC; a more extreme example is Cardas Golden Reference Speaker Cable (5 AWG with 0.031 uH/ft loop back inductance ) or the newer Cardas Clear or Clear Beyond, with about 1/3 the loop back inductance of the Reference series. Very, very pricey, though. I think I'll remain happy with my Golden Reference for a while...

For those who believe all cables sound the same, this last part of the discussion is a moot point, and can be ignored. :W

[bigg]

cables

Hi Jon,
Where do you get the Kimber 4TC, 8TC and the Cardas Golden Reference cables for your crosovers? You obviously know more about this than I do. So I'll go by your recommendations. ;x(

[JonMarsh]

I don't use Cardas golden reference in the cabinet wiring- those are my external speaker cables.

Kimber 4TC and 8TC I've bought in the past from Welborne Laboratory; Cardas Crosslink SP Bulk is available from DIYCable.com; it's 10 AWG equivalent. I've used Kimber 8TC to make "semi-budget" DIY speaker cables with low resistance- that was documented in one thread years ago. They work pretty nicely, unless you A/B them on a good speaker/system against Cardas Golden Reference or Neutral Reference. The NeoD CC's M8ta's, Modula MTM, were built with Cardas Crosslink as internal wiring, as that is my usual preference.

Note that I will be evaluating some other combinations, and may wind up using Cardas SE-15 singly or in parallel on the tweeter connections; SE-15 is AWG 15.5, with loop back inductance of ~0.065 uH/foot. It's a coax construction cable with Cardas's usual golden ratio constant Q copper Litz construction. May use this on both mid and tweeter, probably doubled up for the tweeter. It depends on results of listening tests that ET will run.



YMMV; we don't discuss cables on HT Guide Mission Possible. :W There are other areas of the design you may want to focus your funds on first, depending on your means and goals.

Personally, I wish all wiring and capacitors sounded the same- excellent, and with no impact on musical transparency. They don't, though; and it's up to everyone to decide for themselves how important that is and how they want to deal with it, you know? In most cases, if someone is agonizing over a $25 tweeter versus a $45 tweeter, they're not going to be using a Duelund or Clarity MR cap in the tweeter crossover, or even a Jantzen Z-Superior (used in the NeoD CC) or Jantzen Z-Silver cap. While it's relatively easy to convince someone to get the better more expensive tweeter or midrange driver, in some cases you may be better off with a lower cost driver with good CSD and the better crossover components. Making these tradeoffs to greatest effect is not easy.

[dwk]

I think you have the right idea, doing your research and prepping your shop to get ready to work. All the info you need to determine if this is a path you'd like to try should be up in a few weeks.

Well, my shop has been in chaos for far too long, so it needs attention. From my experience in other areas this type of task basically never gets done unless you define a concrete goal. I guess I have the 'goal' part.

It's actually pretty eerie, since what you describe with your BT3K setup sounds very close to where I've been headed. I already have 2 units in quasi-franken mode on a single stand, but I haven't really done much work to figure out the correct overall arrangement. Still, the basic idea of having multiple stations to reduce the amount of setup/tear-down and hence having to re-set measurements/dimensions is part of the attraction.

Heck, I'd even thought about the idea of 2 SMT's working in tandem as a big crosscut sled. I think it's a viable idea, although it'll require some alignment tricks to get them perfectly parallel.

I'm also planning on taking these to ThomasWs in Denver during RMAF barring any work disaster or other unplanned events. Usually we get together with some of the folks we're in touch with over the internet in past years during RMAF, so there might be a chance for others to hear these at Thomas's and offer somewhat less biased opinions than Thomas or I might have. I'll be bringing out a complete small system to demo them with, running full range, no hookup to Thomas's IB subs. :W

Yeah, I missed out on the similar event for the NeoD CC, so I'm hoping to make this one. Unfortunately, we're already booked for our vacation (hiking in Bryce/Zion) leaving on Oct 3rd which is the Saturday of RMAF. Hopefully you can front-load the visit. On the upside, I work about 2 blocks from the RMAF venue, and since I know ThomasW is pretty close to there, there may be some flexibility.

[JonMarsh]

Well, my shop has been in chaos for far too long, so it needs attention. From my experience in other areas this type of task basically never gets done unless you define a concrete goal. I guess I have the 'goal' part.

It's actually pretty eerie, since what you describe with your BT3K setup sounds very close to where I've been headed. I already have 2 units in quasi-franken mode on a single stand, but I haven't really done much work to figure out the correct overall arrangement. Still, the basic idea of having multiple stations to reduce the amount of setup/tear-down and hence having to re-set measurements/dimensions is part of the attraction.

Heck, I'd even thought about the idea of 2 SMT's working in tandem as a big crosscut sled. I think it's a viable idea, although it'll require some alignment tricks to get them perfectly parallel.



Yeah, I missed out on the similar event for the NeoD CC, so I'm hoping to make this one. Unfortunately, we're already booked for our vacation (hiking in Bryce/Zion) leaving on Oct 3rd which is the Saturday of RMAF. Hopefully you can front-load the visit. On the upside, I work about 2 blocks from the RMAF venue, and since I know ThomasW is pretty close to there, there may be some flexibility.

Nice vacation plans!!

This is exactly the kind of information that's helpful to have in advance, as "front loading" the visit is something already up for discussion- actually, I would likely be out a few days earlier that week before RMAF. As usual, the main issue will be seeing how much time I can get off from work, and if I'm not on business travel, I may be able to extend my stay, since I often work from a home office, depending on ongoing activities at work.

[dwk]

Using the same baffle mount and cutout for the RS100 as the C79 will keep open the option of upgrading the system configuration, though obviously at substantial expense. It would be possible to come up with a midrange driver crossover only upgrade path, if there's interest.

I'd been wondering about the viability of mix-n-match between versions. The difference in sensitivity between the midrange units probably make this close to a redesign, rather than a drop-in, unfortunately. For me personally, it's the Illuminators that are the gut-check part rather then the C79's; an extra $600 for better midrange drivers is somehow more reasonable to me than $600 for better tweeters. I'll be starting with the budget version, but if there is a C79 upgrade, or even an ER18+C79 upgrade, I could see following that path.

[bigg]

Driver substitution on the budget model

Hi Jon,
I was wondering if the Seas Excell T25C-001 (E0006) tweeter would work in the budget version instead of the Seas 27TBFC/G? I realize that it would add about $140.00 to the cost of the budget pair.

Thank you for the wire info. :thanku: Thank you for your time. :thanku:

[JonMarsh]

It wouldn't be hard to do a midrange upgrade, it would be accomplished by dropping in the C79 crossover with some adjustment to the bandpass. I could also easily see the attraction of a ER18RNX + C79 + 27TBFCG, and can provide that as a "stock option". I agree with your feeling that the C79 is more immediately justifiable as a primary expense, considering the frequency range it occupies. Still, I have to say the RS100-4 is quite remarkable for it's price class, especially in HD distortion in the lower part of the range. I expect it to turn in a very respectable performance. I'm surprised that almost no one else has used them, considering the price- maybe it's just an unsung hero waiting to be discovered. :W


The tweeter certainly is a part to agonize over; its not like I see a "perfect" solution here with any of the proposed choices, and at times I could almost talk myself into the C30N, but I'll just probably reserve those for the next project, as the form factor would allow a future upgrade on that one if I get really wacky. :W

Just listening to the swept waveforms, I'm pretty impressed with the Illuminator. I hope that YG Acoustics and some other big boys have picked it is a valid endorsement of it's acoustic potential; since I have them, already paid for, why not? Also, anyone could substitute the D2904/66000 and they would work very nearly identically, based on the test data, at some savings. But the 27TBFC is clearly the price/performance leader. Saving some money there and possibly putting it into crossover components could be a good tradeoff.

[bigg]

The RS100-4 does look like a quality unit so I'll stick with that for now, as well as the RS180S-8s. I was still interested in the Seas tweeter swap you didn't address. I have never been a fan of metal tweeters. And I was going to use the Seas Excell T25C-001s in another design that I've decided not to do. They both have magnetic fluid and most of the T/S parameters seem to be fairly close. I don't expect you to do a complete crossover redesign just for my swap. Any thoughts?

[cjd]

The 27TDFC has historically been a drop-in replacement for the 27TBFC/G...

[JonMarsh]

The RS100-4 does look like a quality unit so I'll stick with that for now, as well as the RS180S-8s. I was still interested in the Seas tweeter swap you didn't address. I have never been a fan of metal tweeters. And I was going to use the Seas Excell T25C-001s in another design that I've decided not to do. They both have magnetic fluid and most of the T/S parameters seem to be fairly close. I don't expect you to do a complete crossover redesign just for my swap. Any thoughts?

There are some tweeters that are pretty easy to swap back and forth because of similarities in the SPL response curves and impedance- an example, the Vifa XT25 with the RS28a with the SS98000.

The impedance curves and general response profile look like you could probably swap in the T25C-001, with maybe a bit of tweak to the tweeter level. Since you already have the T25C-001, that would make some sense- the only other thing to take care of would be the mounting hole rebate- 110 mm versus 104 mm.

[JonMarsh]

The 27TDFC has historically been a drop-in replacement for the 27TBFC/G...

That is certainly the budget choice if he didn't already have a pair of the Excel's.

I've become only slightly disenchanted with the 27TDFC; it's pretty benign compared with a lot of tweeters from it's day, but the high order HD is a bit higher than others, including the 27TBFCG (that's also an issue in most samples for the T25C-001), and the CSD is somewhat hashy at a low level in the 10-20 kHz area (where the one T25-001 I've measured was quite clean); makes for a little bit of the soft dome "fuzz". The D26CN55 is more neutral to my ears and by measurement, but doesn't go low enough for this application, anyway.

[bigg]

tweeters

Yeah, I see that the H1212 is a couple of db more efficient than the Excel. I guess I'll just wait and see what your final choice is going to be. :unsure:

[JonMarsh]

re-padding the tweeter is not a big deal- the RS based system will be about 1 - 1-1/2 dB lower sensitivity anyway, due to driver parameters.

[bigg]

tweeters

Since the system is lower in efficiency, is the Excel T25C-001 actually a better match?

[JonMarsh]

Mmmmm, padding a tweeter is never a problem, in my experience; higher sensitivity means you have more latitude to work with. Also, the tweeter has to cope with some BSC in it's range, too, due to the cabinet design, so sometimes some extra sensitivity is desirable so that in the lower part of the range you have enough nominal output level even with baffle step fall off. With this faceted type design and a varying front panel launch width to raise the baffle diffraction frequency, the behavior isn't as simple as for a rectangular front baffle. This has to be taken into account along with the planed filter target function versus frequency.

[bigg]

tweeters

OK. :?? I'll just wait on this one. :Z I might try the 27TBFC/G on another project. See how I like it. 8x) Rock on! ;b>

[Bear]

It wouldn't be hard to do a midrange upgrade, it would be accomplished by dropping in the C79 crossover with some adjustment to the bandpass. I could also easily see the attraction of a ER18RNX + C79 + 27TBFCG, and can provide that as a "stock option". I agree with your feeling that the C79 is more immediately justifiable as a primary expense, considering the frequency range it occupies. Still, I have to say the RS100-4 is quite remarkable for it's price class, especially in HD distortion in the lower part of the range. I expect it to turn in a very respectable performance. I'm surprised that almost no one else has used them, considering the price- maybe it's just an unsung hero waiting to be discovered. :W

Jon - I was thinking about the statement, above, and was curious whether the specific value of aleph and the center frequency might change this somewhat. For example, as you lower aleph, as I understand it, the mid becomes more and more of a "fill" speaker. It still occupies a hyper-critical passband, but it ultimately becomes something of a bridge between the woofer and tweeter. As a result, would you eventually get to the point where something like a C79 would be total overkill versus something like an RS100?

[JonMarsh]

Jon - I was thinking about the statement, above, and was curious whether the specific value of aleph and the center frequency might change this somewhat. For example, as you lower aleph, as I understand it, the mid becomes more and more of a "fill" speaker. It still occupies a hyper-critical passband, but it ultimately becomes something of a bridge between the woofer and tweeter. As a result, would you eventually get to the point where something like a C79 would be total overkill versus something like an RS100?

Well, I can see why at first blush this reasoning seems, well, "reasonable". But the situation is a little more complicated than that, because of how this also affects the requirements on the woofers and tweeter. (a second point is that the RS100 has a pretty reasonable Xmax, a bit more than the C79 - my only "complaints" about it are higher distortion in the range above 1 kHz- it's response otherwise is very manageable).

Let's go back to the graphic published earlier in the thread, originally part of the NeoD CC thread, with a number of target function curves with different values of aleph. It's easiest to look at it in a practical context.




This is only four values of aleph illustrated, and you can use pretty much anything you want, and of course calculate values in between. But I think this is sufficiently fine grained to be illustrative.

For the value of aleph = square root of 2 = 1.414..., the function devolves to a two way LR4, with the black curves.

For an aleph = 2, the situation is akin to what you describe, more of a filler driver for the midrange role- the peak level is only -6 dB, and the -20 dB levels are ~400 Hz 8500 Hz, in this case for a NeoD style design with CF=1600 Hz. But what happens to the woofer and tweeter range? The woofer level at the CF=1600 Hz = -12 dB, and so is the tweeter level- not bad for the tweeter, but the woofer and tweeter must both make up about 3 dB of the unity output level, so they must be running well and tracking phase OK. If we move that down to the Ardent center frequency, that means the tweeter level is only down 12 dB at 900 Hz, which is challenging for many tweeters- especially if you'd like speaker comfortable with 100 dB levels - that's 88 dB at 900 Hz. That's potentially a bit ugly...

The D3004/662001 at 92 dB




The 27TBFCG at ~90 dB.





These are two of the strongest tweeters with regards to LF distortion.

The 29TAFW is more typical of what's out there in the real world:



Definitely happier staying above 2500 Hz.


Going back to the "master plot", the 2.828 aleph (blue) brings the mid level up to - 3 dB peak, still needs some fill in by tweeter and woofer at the center frequency, but here center frequency woofer and tweeter level are down to -18dB, a nice level to assure that whatever they do above or below that in the frequency range is not of too much consequence. The nominal crossover points are about 750 and 3800 Hz, at - 6 dB, which is about at the edge of what the RS52 will do on the low end and what's possible to get a good tweeter and midrange summation by using a small form factor tweeter. It's a little bit of a juggling act, but it works for the most part, based on how the system sounds and measures.

Now, considering the current working target function for the Ardent,



the nominal crossover points are 300 Hz and ~2700 Hz, which stays well within a nice working range for the mid, with the -20 dB points at 100 Hz at 8 kHz; the dispersion of the mid will narrow above that upper crossover point, but the tweeter takes over more of that range. nominal level for the mid is 80 dB @ 100 Hz for a nominal maximum playback of 100 dB.

At 96 dB nominal level (with some roll off below 1 kHz) the C79 holds up OK, though I could wish it did better in the low end, perhaps (there's a reason they make the C90-T6).



At 90 db nominal, 80 dB output at 100 Hz, things don't look bad.




At 93 dB output, the RS100 holds up pretty fair, but again we could wish it did better below 300 Hz.



Even at 87 db nominal, about 83 dB at 100 Hz, it is climbing a bit...



But remember, this is what we need when the midwoofers are doing 100 dB; this a single ER18RNX at 97 dB:



And its providing the bulk of the output level (20 dB higher), so the midrange distortion products are reduce in the overall mix by 20 dB. But I can see the argument for moving this up in frequency a bit, lowering the aleph a skosh, and edging the CF back up to 900, especially for the RS100. But to avoid LF intermodulation of too much of the vocal range, I think I might like to keep it where it is, for more "normal" listening levels.

Is that the optimum trade-off? I don't know- but there are other advantages to the lower crossover frequency, including the driver spacing not being as much of an issue with regards to managing the phase transition and the minimum distance everything sums OK.

That's why I will be hooking up my Aragon X3 to my Praxis test PC, which has a four channel Maudio card, and emulating and comparing variations on the crossover frequency. I don't want to build and test two versions of the passive crossover, as I did for the NeoD CC.

We must remember this is not the ultimate solution.... just a nice set for the family room that may wind up in the bedroom someday.

Now, THIS is a midrange driver.... at 99 dB output.



The Chancellor and ET think they have this fully under wraps, but there are rumors beginning to emerge of a secret project underway, called "DeathStar", using technology formerly developed in Isiris Nebula. I'm a bit skeptical about what I'm hearing, so I wouldn't take it very seriously at this point. Probably just vaporware.

[Bear]

Jon - Always educational, and always very much appreciated! I had forgotten which target curve you were looking at (Fc and aleph), and yes, the tailing up in HD for the RS100 would definitely point towards a different driver if one were available.

Bill

[JonMarsh]

Yeah, but it's all relative- and at reasonable levels, as a three way, and considering how far down the midrange level is at that point- the RS100 will sound pretty good. Especially for just $25! Of course, that's where the whole thing seems a little wacky, as the crossover and other stuff will cost much more than that mid driver. But as the question was raised in the driver testing thread, what can you buy that's actually better for $50 or $100? It's not clear to me- even the Visaton TI-100 at $189 is not a clear winner between it's 6-8 kHz break up modes, and sharply riding HD above 1200 Hz or so. The TB W4-1337 fares more poorly in HD at the lower frequencies, too. At twice the price of the RS100-4.

So, we'll see how the "budget" version works out. In the scheme of things the cost of the RS100-4 is almost in the noise floor compared with everything else needed- such as crossover components and clamps! The driver is practically free- it's all the necessities surround it that will add up. (on the C79 version, I'm ordering Cardas SE-9 cable for the connection to the driver- for the RS1800, I'll use some Kimber 4TC I have on hand. That sells for $8.50 a foot, so the cable to connect it will cost more than the midrange (OK, I'm a little schizoid by HTG standards I suppose...)

[augerpro]

So that last plot was the Jantzen? I like the combination of drivers. I think the Seas 29TAFW is a sleeper that will really perform over it's price class.

[Bear]

So that last plot was the Jantzen?

I'm pretty sure the very last plot was the Accuton C173n-T6-90 mid-range.

Yeah, but it's all relative- and at reasonable levels, as a three way, and considering how far down the midrange level is at that point- the RS100 will sound pretty good. Especially for just $25! Of course, that's where the whole thing seems a little wacky, as the crossover and other stuff will cost much more than that mid driver. But as the question was raised in the driver testing thread, what can you buy that's actually better for $50 or $100? It's not clear to me- even the Visaton TI-100 at $189 is not a clear winner between it's 6-8 kHz break up modes, and sharply riding HD above 1200 Hz or so. The TB W4-1337 fares more poorly in HD at the lower frequencies, too. At twice the price of the RS100-4.

So, we'll see how the "budget" version works out. In the scheme of things the cost of the RS100-4 is almost in the noise floor compared with everything else needed- such as crossover components and clamps! The driver is practically free- it's all the necessities surround it that will add up. (on the C79 version, I'm ordering Cardas SE-9 cable for the connection to the driver- for the RS1800, I'll use some Kimber 4TC I have on hand. That sells for $8.50 a foot, so the cable to connect it will cost more than the midrange (OK, I'm a little schizoid by HTG standards I suppose...)

Since you did bring up the idea of options, I was trying to think through what is currently known about the BOM, and a pair of RS100s is definitely dwarfed by the cost of just one C79. Someone looking at the two extremes would probably be willing to "pay up" for the SEAS woofers, since the cost there is relatively modest, but for those on a budget, the Illuminator tweeters and the Accuton mid each provide a step increase in the budget.

Of course, if the XO is $10k in parts, then the driver budget is all peanuts, no matter which version you choose! :E

[JonMarsh]

Yeah, Brandon, Bill called it right- that's the C173N-T6-90.



The thing everyone should understand is that while I've let Evil Twin loose on this project, and you'll find him dogmatic in some of his "suggestions" for materials and technique, the "Y" in "DIY" is making individual decisions about where to draw the line, what level of materials, processes, etc. one is willing to put into this. If built as it will be shown with removable base plates, it will always be possible to go back and revisit. It would be possible to start with a "base" RS180 +RS100-4 + 27TBFCG with just Dayton crossover parts and zip cord for internal wiring, and wind up with rebuilds with a fairly or ultimately "deluxe" version. If you're using an HT receiver with the colorations that their electronics volume control chips induce, there's no sense considering using Duelund VSF caps, maybe not even Jantzen Superior-Silver, though I'd still recommend Jantzen Superior-Z or Clarity SA as the base caps. But anyone is free to change that. Capacitors will be mixed and matched to effect and cost/performance benefit in the crossover, and I have a few options on LF coils that will be tested before recommendations will be made- as well as a experimental porting system that may reduce port resonance (golden ellipsiod cross sections, anyone?)

To some, this will seem more like an X-Files episode than engineering- unless you're willing to look at the nature of some subtle details, and use your ears as well as the relatively crude instruments we have. So, for those interested in this project, you're welcome to follow as far down the path as your interest or convictions will take you- or not, obviously. :W

For the rest, paraphrasing Obi Wan, "Move along now, these aren't the speaker projects you're looking for", accompanied by a not so subtle mind push. :B

[CraigJ]

For the rest, paraphrasing Obi Wan, "Move along now, these aren't the speaker projects you're looking for", accompanied by a not so subtle mind push. :B

...and not your average "caps" either. :rofl:



Enjoy your two weeks of speaker camp. :^x
Craig

[JonMarsh]

Hmmm, you know, i don't think we'll be using those big guys.... maybe just a few of the baby ones....

[Jed]

Hmmm, you know, i don't think we'll be using those big guys.... maybe just a few of the baby ones....

Looks like a bar of chocolate.

Also as noted in another thread, you might want to buy another pair of the C173-90 because they improved the top end HD profile. Now there is almost a complete lack of harmonic distortion above 2.5k.

[Bear]

Looks like a bar of chocolate.

Mmmmm. Scharfenberger!

Also as noted in another thread, you might want to buy another pair of the C173-90 because they improved the top end HD profile. Now there is almost a complete lack of harmonic distortion above 2.5k.

Given the cone size, wouldn't this start to beam a bit above 2k?

[JonMarsh]

Mmmmm. Scharfenberger!


Given the cone size, wouldn't this start to beam a bit above 2k?

Yes, certainly... that has a significant influence on the choice of tweeter and crossover points.




That's why the Isis uses this with the diamond version of the C30N, not the diamond version of the C13N.

[JonMarsh]

Looks like a bar of chocolate.

Also as noted in another thread, you might want to buy another pair of the C173-90 because they improved the top end HD profile. Now there is almost a complete lack of harmonic distortion above 2.5k.

Well, if I could get the Federal Government to give me a low interest "stimulus" loan for a new pair, that might be attractive... right now, I think I'll have to live with the "crummy" old ones, because the Imperial Treasuries need the money elsewhere... :roll:

Fortunately the rising distortion above 3 kHz is just 2nd, and the nominal amplitude response will probably be at a -10 dB point at 4 kHz. Compromises, compromises.

Of course, anyone that would like to donate a new set to further our research, that would be most appreciated. :W

[Hdale85]

Hmm are those the 600 dollar each ones or 300 haha

[Bear]

Hmm are those the 600 dollar each ones or 300 haha

$600 now, they've come up in price a bit over the last year or so.

[ThomasW]

$600 now, they've come up in price a bit over the last year or so.

+1

Accuton's price increases over the last 18 months is staggering.... 8O

[bigg]

Ardent?

One of the tests referenced the Ardent project and showed two woofers (12 inch I believe) two mid bass drivers and a tweeter. Is this right?

I thought the Ardent was going to have two woofers a mid bass/midrange and a tweeter. (TMWW) :unsure:

Also. $600.00 drivers? :E I hope there is going to be a middle line for us poor factory workers to upgrade to. Maybe I will use those T25CF-001 Excell tweeters with the base line Ardents. :later:

[JonMarsh]

Nah, this pricey Accuton is for something else- Evil Twin's DeathStar.

Check the first page of this thread and Ardent Speaker Camp thread- it's a configuration for 2-7" woofers, one midrange, one tweeter. Two build plans- budget and premium. Budget drivers are RS180-8, RS100-4, and Seas 27TBFCG. "Premium" drivers are Seas ER18RNX, Accuton C79/6 midrange, and dealer's choice on tweeter- my build is with the D3004/662001, but you could substitute a 66000 air circ or other similar impedance scan speak model, maybe you could even get away with the 71000 series. Or if you need to save money on the tweeter, to afford the midrange, go with the 27TBFCG and matching crossover, though you'll have to tweak LPAD values slightly.

[bigg]

Ardent?

Whew! Glad to have that cleared up. I got confused for a moment. :stupid2: So the C79/6 (I believe it has a new number/name), is really that good? I mean is it really that much better than the RS100? Thank you for your time.