Proposal for a compact three way

[Evil Twin]

I spend many hours in the confines of my transport shuttle and in cramped quarters managing space based construction sites- it has occurred to me that with the introduction of new drivers by one particular manufacturer, SB Acoustics, a new standard may be possible for a compact sealed three way monitor, which could be extended to a larger floor standing ported or dual woofer system.

The fundamental concept is derived from the NeoD CC center/mains- using a Duelund crossover concept.


The Duelund crossover uses a single equation with three sections to describe a complete all pass three way concept tunable by the choice of a single gain coefficient- with a low value for aleph, the design devolves to an LR4 two way; with increasing values of aleph, the midrange element assumes a larger and larger portion of the spectrum, normalized to a center frequency.





In this case we see a transfer function optimized for the RS52 as the midrange element, and the resulting levels and crossover frequencies at approximately 600 Hz and 3.5 kHz; the response of the RS52 is ideal enough that response shaping networks as might be needed for a cone midrange element are unnecessary. Another factor favoring the RS52 is that no rear enclosure volume is needed.



The idealized network looks like this:




and the system conceptual response as follows:





Actual measured response of the RS52 on a typical baffle should not be difficult to work with:





Some modifications to this concept may be required, such as using a modified LR3 characteristic (as developed for my NatalieP design) to handle the time offsets of the woofer and midrange. In any case, compared with other three ways, the RS52 is used more as a filler or bridge driver. This places some significant performance requirements on the woofer and tweeter drivers.


For the woofer, a new model looks very promising, the SB Acoustics SB23NACS45-8:










For the tweeter, once more SB Acoustics appears to offer a high value proposition:







Samples of these drivers are now on the way to my acoustics lab in the Seinar facility- I look forward to investigating this proposal further...

[Carl V]

Morel makes a soft Dome Midrange of similar size. MDM55
Do you have any experience with this driver, comparing & contrasting
with the Dayton unit?

Last Month a Classmate in KC had me over to hear his new
INDUCTION DYNAMICS speakers. It appears to use a nice
inverted Titanium Focal Dome tweeter, an ATC 3" Dome Mid.
and a 6 1/2" Kevlar woofer, 10" (sub)woofers. he has similar
in wall set-ups for Center & surrounds. the XO points are
similar to what you have proposed.

The Sound was Very good. just a FYI.

[BobEllis]

Plugging the SB23NACS45-8 into WinISD pro A, it models virtually identically to the RS-225S-8 in a 2 cubic foot sealed box. An F3 of ~52 Hz with a Q of .745 or .772. I look forward to hearing your impression of it, particularly in comparision to the RS-225. Both seem to hit Xmax at sealed with a 2nd order crossover at 65 Hz, giving passband SPL of 112 dB, with the RS225 pair needing 200W to the SB pair's 150W.

I know it's tough to make value choice for others, but it would be interesting to hear if you think that it is worth $70 more than the RS-225 or if there is a significant listening difference. I have nice inert 2 cubic foot boxes I am ready to turn into bass bins for a W15CY/T25CF MT. I have a set of RS225s, but am willing to switch for better performance and repurpose the RS225s. Thanks, ET.

[dwk]

I was going to post asking whether these were intended to be a Modula design like the NeoDCC, but Bob has probably answered that question.

[Evil Twin]

You have seen to the heart of the matter, and this is why drivers are in transit as I write for full evaluations. Besides LF distortion, I am quite interested in midrange behavior and issues arising from out of band behavior, such as resonance amplification of distortion products. The effort in suppressing out of band output must be considered. While speculation on these matters is an amusing way to pass time, only hardware measurements can provide the information I need. The Qts and cone/motor design make it slightly more interesting for a sealed system than the RS225.

Given that you already have the RS225's on hand, and the specifics of your application, it's unlikely that the extra expenditure for the SB Acoustics SB23NACS45 delivers value- only a radical increase in price might deliver an interesting upgrade in performance, but with other drawbacks. This could be the case of for the Wavecor SW223 with it's increased SD and higher Xmax- but this comes at a very substantial price in purchase cost and reduced efficiency.

It will be very interesting to see measured performance- the SB23ANACS45 may even have potential for a two way approach, though that is not planned at this time. Still, it's an interesting possibility, particularly with a waveguide loaded tweeter.

[Evil Twin]

I was going to post asking whether these were intended to be a Modula design like the NeoDCC, but Bob has probably answered that question.

This is not a point which is fully answered- it would likely be feasible to do these in as small an enclosure as a standard 0.75 cut foot PE enclosure (Modula approach), but it would be far more esthetically satisfying to produce a custom enclosure and front panel more in line with the advanced builds- think of it as a stand mount Ardent, perhaps, 1-1.5 cu ft. - 30 liters or a bit more. Not needing a midrange enclosure, the overall cabinet volume would be much smaller than an Ardent. Just the top 1/2 to 2/3. Or imagine the Avalon mixing monitor scaled up by 40-50% with a dome midrange. And not finished in pickup truck bed liner....

[dwk]

Well, a full cabinet project would definitely be interesting, but for me would be a much longer term proposition. We recently moved, and while I now have a nice oversized 3-car garage which will ultimately allow a nice shop space (1 bay fully dedicated, with the ability to spill into the other 2 as needed), I still have to clear out the remaining crap from the move, insulate, probably upgrade power etc. so it'll be a while before I can consider anything that ambitious. I think I could manage to rout some baffles though :-)
I actually haven't been around HTGuide or audio in general in quite a while, but am now trying to figure out speaker plans. I actually came back partially to survey available designs for the PE cabs, and the NeoDCC was one I always wanted to try but didn't seriously consider until the tweeter was NLA. So, given the timing of this new endeavor I just had to ask.

I'll follow along with interest either way, since this looks like a very attractive project on a low (for you) budget.

[Evil Twin]

Yes, my perception is that this project does have a debt to the NeoD CC, though at this point I expect to implement a somewhat different yet still moderate price crossover. Testing the SB Acoustics tweeter will be interesting; there are indications it will play far out of it's price class performance wise- it will be interesting to see if it can play as low as it's response curves suggest with reasonable distortion figures. With 1.2mm of P-P xmas, it may be another class leader for small format tweeters. The off axis behavior is far above average, and should match well with the RS52.

This is a configuration that looks very promising on paper, and can be prototyped quickly in a Parts Express cabinet.

[Face]

That SB is an excellent tweeter for the price, I've used them before and was very happy with the results. Pairing them with a dome mid will give you a lot of flexibility as far as slopes and a crossover point.

[Evil Twin]

It is interesting but not surprising that that your experience confirms my intuition in this case. I agree that that should be substantial flexibility in the midrange to tweeter transition in this configuration, and plan to take advantage of exploring that. To manage the time offsets for the midrange to woofer, the same approach as the NatlieP should prove effective.

Given the tooling available in the Seinar research facility, we will test and optimize both a conventional enclosure as well as the more interesting approaches.

[5th element]

http://www.audiogurman.com/1.htm#sb26stcn

There has always been the Peerless/Tymphany HDS series of aluminium coned drivers, these don't seem to have attracted any attention. The same goes for their other HDS drivers too, which is odd because of the few drivers measured they've always performed really well, especially down low.

[Face]

I've spoken to a few people who've tried the HDS Al coned drivers and none of them were pleased with working with them.

[5th element]

I've spoken to a few people who've tried the HDS Al coned drivers and none of them were pleased with working with them.

Not to hijack the thread, but I did see some measurements for the 4" that showed very good linearity except for some really ugly distortion amplification products at 3kHz. This is what you'd expect for a metal coned driver with its primary resonance at 9k though, requiring a nice steep filter at around 1/4 of that frequnecy (2.25kHz). But no harder to work with than any other metal cone.

SB on the other hand seem to have taken a different route altogether with regards to the metal coned resonance issue, by placing ribs in the cone, a little like what YG acoustics have tried with their milled cones. It will be very interesting to see Jon's independent measurements of how effective these actually are.

[Face]

http://medleysmusings.com/sb23nacs45-8/

[5th element]

Yeah I've seen that, FR and HD are missing though, which is really what's of interest to me as it will show exactly how effective those ribs are at dealing with the problematic issues of stiff cones.

[BobEllis]

Posted in error

[JonMarsh]

Yeah I've seen that, FR and HD are missing though, which is really what's of interest to me as it will show exactly how effective those ribs are at dealing with the problematic issues of stiff cones.

No substitute for actual measurements!

[5th element]

Heh, as an ammendment to 'you can never have enough drivers' perhaps one should write - you can never have enough measurements (providing they are performed correctly). I don't really get the point in that extensive Klippel stuff. I mean I've seen some excellent looking Klippel stuff followed by crap FR and HD and some mediocre looking Klippel followed by excellent FR and HD. Not that I mind having all the boxes ticked, but as you say, you really need to measure the right thing to see just how a driver does. Fortunately HD and FR almost always show you the effects of any unwanted resonances or motor/suspension issues.

[Face]

Not totally clean, but actual measurements: http://techtalk.parts-express.com/sh...ros-here/page6

I'm sure you could ask him for specific measurements, he's a very nice and helpful guy.

[Evil Twin]

Not totally clean, but actual measurements: http://techtalk.parts-express.com/sh...ros-here/page6

I'm sure you could ask him for specific measurements, he's a very nice and helpful guy.

An interesting choice of drivers, one I can readily endorse... that should be an interesting system when he has finished his thoughtful development process.

At this point, I do have cabinets and baffles ready for basic measurements, including comparison with 8 and 4 ohm RS225, and like most of us here, trust what I see and measure by my standard procedures a little more than 3rd party efforts- however skilled.

[dwk]

Not to rock the boat more than I already have :-), but did you do a mental eval process on the mid or just gravitate to the RS52 based on past experience with the NeoD CC? I'm curious as to whether the Scan 10F would be worth evaluating. It would complicate cabinet construction due to the need for a chamber, but it appears that it has wider bandwidth, probably lower distortion, and more Vd (uh, and probably is more attractive). A bit more expensive, but you might well make that back up with a simpler xover. Based on Zaph's curves, dispersion seems to hold up well enough to probably work for the Duelund approach, at least to my untrained eye.

[Evil Twin]

These are reasonable comments from your side, but surely you did not confuse me with being a reasonable person? :W

The 10F is a very nice small cone midrange, but poses many of the same utilization problems as the W1337S Tangband- besides requiring a rear chamber and damping, it has a very small rear window envelope and needs special mounting techniques to really work well. The requirement for a rear chamber works against it for consideration for a Modula type design using a PE enclosure, as rear enclosure volume is critical and also being strained by the volume of a three way crossover.

The 10F, like the 1337S, has very nice dispersion for a cone midrange:




On the other hand, one can hardly complain about the RS52.




The distortion behavior in the intended usage band (800 Hz-3 kHz at near full level) for the RS52 is quite reasonable at 91 dB output:




(keeping in mind the actual levels with crossover, and the degree to which this is working more like a filler driver)


And even at 97 dB is very respectable:





Many DIY'ers on other forums say that they haven't been able to build a good sounding design with the RS52, but I strongly suspect this is a function of a crossover transfer function approach more suited to a soft dome midrange that just gets fizzy and rolls off on the top end, instead of considering crossover points and slopes in light of the resonance amplification which occurs with hard dome drivers, and manifests at lower frequencies. This is a shortcoming of many crossover networks for the Seas Excel drivers, too. For this reason I strongly recommend limiting the upper end cut off of the RS52 to ~ 3 KHz, where the time domain, distortion, and dispersion behavior are all excellent.

For cost and component count reasons, many 3 way designs resemble more the type of crossover filters used in the Finalists- with paper cone drivers with minimal resonance, this may work fairly well for tonal balance at low to moderate volumes. Note, for example, that the tweeter attenuation between 2 kHz and 1 kHz is only on the order of 4-5 dB.



I would describe this crossover overall as an LR1 three way. It is not my intention to create another Finalist type design or replicate something similar- it is fortunate that builders have so many prospective designs from which to choose.

Not being a reasonable person, I have to admit that the only person I'm really trying to please with this design is myself- as regards my expectations for power response, distortion, SPL capability, and size. This configuration might even benefit by graduating from the SB Acoustics 8" woofer to the Wavecor SB223





It remains to be seen how this driver performs up in the midrange- that question should be settled soon, as a test enclosure is nearly complete.

The cost and lower sensitivity would be something of a handicap for many, but for extracting the maximum performance from a given size enclosure (20-28 liters) this may be the best solution.


And of course, esthetics are a very personal matter....








In my case, this is most satisfactory...

[dlneubec]

Hey ET,

Just another thumbs up on the SB26 neo tweeter. I'm currently working on two projects (one 2-way, one 3-way) that use it and will be doing crossover design for another guy with this tweeter. Hard to beat it from a price/performance/flexibility standpoint, IMO, though probably not a candidate for a waveguide solution.

Also, I'm using the dual magnet SB12MNRXF25-4 midrange in two project and give it a very high bang for the buck rating.http://meniscusaudio.com/sb-sb12mnrx254-p-1401.html Of course, this is not a driver that works with your goals on this project. I'd like to see SB do a dome midrange, maybe a little larger in size than the RS52 with ability to cross lower.

It is interesting but not surprising that that your experience confirms my intuition in this case. I agree that that should be substantial flexibility in the midrange to tweeter transition in this configuration, and plan to take advantage of exploring that. To manage the time offsets for the midrange to woofer, the same approach as the NatlieP should prove effective.

Given the tooling available in the Seinar research facility, we will test and optimize both a conventional enclosure as well as the more interesting approaches.

[Evil Twin]

Your feedback and comments are appreciated Dan, it's clear that SB Acoustics is making a strong effort along a number of fronts, and the SB12MNRXF25 is clearly part of that.
What will be interesting to see shortly is how the distortion behavior of the SB26 Neo tweeter compares with others of the same or more conventional form factors. In a three way, requirements are more relaxed, but the measured response looks promising for two way systems also.

The SB Acoustics woofers were received by the Import/Export traffic manager yesterday, and I find a physical examination to be enlightening. There are clearly different sets of tradeoffs compared with the Dayton RS225, in fundamental design and execution. The most obvious difference is in the fundamental motor configuration, with the RS225 using a center post which aids cooling and eliminates the build up of pressure under the center dust cap (as there is none), but this loses slightly in SD and requires more care in the cone to control breakup modes, as the strengthening of a center cap is not present. The RS225 uses a 38 mm VC diameter, fairly standard in this size class.

On the other hand, the SB23NACS45-8 uses a conventional dust cap, and has a very large diameter vent in the rear of the motor structure, in part made possible by the larger diameter 45.5 mm voice coil assembly used, a design feature that will also usually increase power handling and may stabilize cone behavior.



Another interesting construction feature is the use of symmetric lead outs on opposite sides of the VC and frame; some designers claim this reduces the possibility of rocking motions of the cone with long excursion. A simple push test does confirm the kipple report that the suspension compliance may be a limiting factor for useful excursion and distortion, more so than the nominal VC Xmax and motor linearity. This is surprisingly common, though perhaps the designer believes that the outer 50% of travel will rarely be used and the design is better served by a suspension which will keep the conde and VC more centered. This is all speculation, until some actual audio measurements can be completed.

A physical examination also confirms the specification sheet dimensions, and highlights that this cannot be used interchangeably in the same cabinet and mounting holes as an RS225, as it is substantially larger... an inconvenience for some, but SB Acoustics seems to have put the additional size to good use.

[JonMarsh]

Those could even work in a "lower cost" version of the Wavecor Ardent, I suppose, though with some substantial crossover modifications.

[5th element]

Jon, when measuring the SB23NAC, I don't suppose you could measure it and an RS225 one after the other, so as to provide a direct comparison. This would be most useful, I think, so we could easily see which is the better of the two and in what areas, lower bass vs mid bass vs midrange capability etc. I have high hopes for the SB, as I think we all do. The trouble is, the RS225 is already a very good performer and it would most likely take a direct comparison to actually show which driver is better and for what.

[Evil Twin]

Without question that is the ONLY way to conduct the comparisons with a rigorous control of parameters- even then, it will only be fully valid for the specific drivers.

I have extra baffle sets on the way, to machine for each driver, and expect to test the RS225, (8 and 4 ohm versions), the SB Acoustic SB23NAC, and the Wavecor SB223BD01 in the same configuration and conditions. This should prove to be interesting... testing will be quasi-IB near field and 1M from the test enclosure in an elevated configuration. Generally, the near field test is the most useful, though the far field can provide a basis for design preparations for enclosures with the same baffle width.

[5th element]

Sounds like a plan. One has to remember that the Wavecor is very very black too!

[Evil Twin]

Sounds like a plan. One has to remember that the Wavecor is very very black too!

Yes, this clearly conveys a significant point advantage to the Wavecor over the SB Acoustics before any data is recorded...

[JonMarsh]

Another 8" driver you ought to seriously consider testing is the 8" SS Revelator- about 9-10mm of usable Xmax, good symmetry in BL product and compliance, and an actual Q that is quite in the right range for a sealed box system.


http://medleysmusings.com/scanspeak-...aluminum-cone/


Hard to argue with those results... and I think I know where a pair of them are in storage...








Of course, that takes things pretty much out of the range of a fair fight with the RS225, but then, so does the Wavecor.

[5th element]

Well it's always worth testing the expensive stuff along with the more budget orientated offerings, just to get a feeling for how things are actually performing. This then lets you see what exactly it is you're paying for when you go boutique, or how poor the improvements actually are when going boutique.

I've observed Medley's musings over on the PE forums with regards to his opinion on the various drivers that he has just tested. He has a need for a good 8" in some design, but seems to think that the SB wont cut the mustard.

It surprises me the number of times people forget how small the increase in net dB is when talking about xmax figures and that going from the SB to the Scan (if going by the BL curve), will only give you an additional ~3dB. This, to me, is a tiny amount and if the SB really isn't going to cut the mustard then you need to think differently than just another 8".

He was also asked if he was going to run distortion sweeps of said 8" drivers, citing that he didn't think this was important and he also didn't think that it was important to test how graceful the drivers are once driven beyond xmax. He seemed to think that going beyond xmax ran too much of a risk of damaging the driver, where obviously there should not be any risk unless one exceeds the actual xmech figure (not always provided).

Now again, this is only my opinion, but even if one has large requirements of said system, we rarely call upon them to actually go to those limits. Yes I want my system to be able to reach 105dB at 20Hz and also reach 110+dB peaks when i want to crank it, but I do not see this as every day listening by any means. Under the high SPL circumstances I do not mind if I exceed the Xmax figure providing nothing blows up and it isn't disturbingly obvious (such as a coil whacking a backplate) because it probably happened during a HT explosion, or a very loud musical passage, where the SPL was so intense my ears hurt. Here it is important for me to know quite how my drivers handle going beyond their standard linear capacity, especially if a driver is spec'd to have more useable xmax than I actually measured. If it's suspension or coil offset that we're talking about then it's highly liklely that you're only going to see a gross increase in 2nd harmonic before the higher orders shoot up when driven harder, this I could live with.

Under every day circumstances I would much rather have the driver that exhibited lower overall distortion, when run nowhere near it's limits, than the driver that would handle the odd SPL peak with more grace. But to find out which driver that is you need to actually measure it's distortion performance when it's producing 90dB or so.

Of course I don't know what his requirements are, but aparently he's going to pack up and sell the 8" SB without even running a distortion sweep, apparently he's not going to run distortion sweeps on any of the current 8" drivers he's testing. This to me seems a little short sighted, especially as one of the reasons he's running the blog and having purchased the testing equipment in the first place, is to provide useful data to us DIYers.

[JonMarsh]

These are all very good and valid points.

The Kippel data for the SB driver is not particularly promising, partly because the resting point is not centered with regards to the motor structure- either a design issue, or more likely, a quality control problem in the manufacturing implementation. Note that the Discovery SS 8" woofer he tested also had alignment issues, but the Revelator didn't. This can be a key issue for small signal distortion at normal listening levels, even- 90 dB and less. It will show up in higher than anticipated levels of HD2- we do see that in a lot of drivers, it's relatively rare to see HD2 and HD3 well matched, or HD2 lower, but with good execution, this is how it should be.

Wavecor put a lot of effort into making their new balanced drive motor symmetrical- but it will also require correct suspension design and centering to get good performance numbers. The early prototype RS225's were pretty good as regards execution and decent as regards design. This comparison will be interesting to see.

[5th element]

Presumably a push-pull implementation should help to correct for that.

[JonMarsh]

It could to some degree, I suppose- like the Linkwitz mounting for dipole woofers- but it's better to just do things right to begin with- it's just aligning the stuff correctly in the beginning. That's one of my pet peeves about many SS tweeter models- having HD2 up to 20 dB higher than HD3! I prefer to see something like the better Seas tweeters, say the Millennium Excel, where the difference may not be more than 6 dB for most of the range, or good samples of the RS28a, which track very closely from 2400Hz up.

[Evil Twin]

This discussion has become even more interesting, but if we are going to pit woofers priced at over $300 against the RS225, then shouldn't the tweeter be fair game also for unreasonable comparisons? Why not pull out the pair of D29004/7100-02 which I've been holding in reserve- this should be either a Revelating Illumination, or an embarrassing upset... which that proves to be we will have to see. Measurements by Treols Gravesen show fairly good distortion behavior, but not as good as the 6640.

[5th element]

There had been some talk on the PE forums about offset actually improving linearity to a certain degree, where one thing compensates for another.

Zaph does have measurements of the 7100, where its HD2 is almost as low as the Millenium, but with far far lower everything else. I mean I suppose it depends what you go after, and whether or not Scan are engineering something into the mix so that the rest of the HD products go down at the expense of HD2. That tends to be a trend in most of the newer scan tweeters, where they have high HD2, but extremely low everything else. Speaking of HD2 though, the SEAS DXT has lower levels of HD2 than most. Whether or not this is audible I don't know, but it does sound very nice.

Some claim that distortion isn't audible providing the driver isn't poorly designed, whatever that really means. Nevertheless, it has always been my experience that if you go from a system with HD2/3 around -40 to -45 dB, to a system that improves upon those figures by 10+dB that it does sound cleaner/clearer as a result. And no, we don't listen to sine waves, but I can always hear higher levels of HD3, especially around 1-3kHz, when I'm running distortion sweeps and if I can hear it there, then I most certainly don't want it to be there when listening to music, regardless of how much masking there is involved.

Edit - speaking of 8 inchers though, it's a shame the Usher 8955A was discontinued.

[Jonasz]

This discussion has become even more interesting, but if we are going to pit woofers priced at over $300 against the RS225, then shouldn't the tweeter be fair game also for unreasonable comparisons? Why not pull out the pair of D29004/7100-02 which I've been holding in reserve- this should be either a Revelating Illumination, or an embarrassing upset... which that proves to be we will have to see. Measurements by Treols Gravesen show fairly good distortion behavior, but not as good as the 6640.

So now when the subject is tweeters do you ET, or anyone else for that matter, have any idea why the SS Illuminator 602010 and siblings don't get more love? They have a desireable small form factor, very respectable low distortion and still doesn't cost an arm and a leg, $122 at Madisound. According to Zaph's measurements they even seem to be able to cross low when they have lower distortion in the 1-2kHz range than proven performers like Millenium or the RS28. What am I missing? :??

I vote for 602010, RS52 & 8857 (or 8851 since I have them and could replicate your build :yesnod: ), after all the tweeters protective grill matches the RS52 most satisfactory... :yesnod: and why not spend a little extra when you put up so much time in building anyway...? :yesnod:

[dwk]

Well, this is escalating out of control rather quickly. No measurements yet, and already the budget is getting blown All we need now is an Accuton mid.......

I know the RS225 is a well-known quantity, but wouldn't the RSS210 be a closer analog to the Wavecor as a budget candidate for a sealed system? I know I shouldn't put too much stock in the cosmetic similarity, but they also seem to be rather similar in purpose. In 1 cu ft sealed the 210HF model rather similarly to the provided charts of the Wavecor in 21L, and appear fully capable of extending up into the 500+ range for crossing to a mid. IIRC you had a pair, but they got shelved early on due to not comparing too favorably in construction/venting to the Wavecor, but having run them in the PE 1.0 MTM cabs under the Xds monitor for a while I suspect they'd work well as a budget substitute - certainly providing more extension and output than the 225s when run sealed. They may not quite prove capable of extending high enough for a Duelund transfer function, though. The RS225 is just as black, though, so in that regard they're a push.

I went back and have read through a lot of the older Ardent and driver eval material. I now see that my earlier question on the Scan 10F had been covered previously. Profuse apologies for dredging it back up. It's too bad the Anarchy went out of production at an inopportune time - based on some of those older discussions it seems like the idea of an Anarchy-based Ardent variant might have been explored. That sounds like it might have been an interesting undertaking.

[dwk]

So now when the subject is tweeters do you ET, or anyone else for that matter, have any idea why the SS Illuminator 602010 and siblings don't get more love? They have a desireable small form factor, very respectable low distortion and still doesn't cost an arm and a leg, $122 at Madisound. According to Zaph's measurements they even seem to be able to cross low when they have lower distortion in the 1-2kHz range than proven performers like Millenium or the RS28. What am I missing? :??

Good question. Not that my system plans have ever been too firm, but I've always eyed this tweeter as being rather attractive - seems like an ideal pairing with the 10F which has been the combo I've been thinking of trying to use. It doesn't seem to get much attention though. Of course, the 10f seems to be under-utilized based on measurements too, so maybe the problem is that there are only so many projects and a lot of driver candidates to fill them.

[5th element]

I don't know why they don't get more loving. Perhaps it's because if you want, or need to go neo, there are lots of other and cheaper options that also perform well. There's the DQ25, the SB dimple dome, the already discussed SB non dimple dome, the now discontinued, but if you can get it, D26NC55. Not to mention there's the small format XT25/19s DX20 style tweeters, if you don't need to cross low.

To throw another very black driver amongst the already crowded room, there's the U22REX/P-SL from SEAS. I had totally overlooked this driver. I too am on the look out for something decent in an 8" formfactor that could be paralleled or quadded up, either sealed or vented, in a smallish enclosure, for decent sensitivity. These are sensitive and work very well in small ported (for an 8") enclosures. I've been simming with them for the past few mins and rather like what they do. Sure, they don't go tremendously deep, but you do get a solid 90+dB on the dial, which is never something to be sniffed at. Their only caveat is the reduced xmax - only 4mm one way, but they do go out to 10mm one way before damage occurs, so there's plenty of room for the odd excursion out of xmax.

The only reason why I'm drawn to this driver is because of it's shorting rings and the measured performance of it's smaller brothers. The nominal 5" version doesn't have shorting rings, but SEAS have obviously done something to the motor over the standard prestige drivers as it has noticeably lower distortion, especially down low.



The 6.5" version otoh does have shorting rings and with the greater SD offers even nicer low end performance.



If the 8", with its shorting rings and even greater SD, is engineered to the same standards, it'd probably be quite a performer. I suppose these are similar to the ER18, which has extremely good performance when used in the bottom of a three way, especially given its price.

[Evil Twin]

So now when the subject is tweeters do you ET, or anyone else for that matter, have any idea why the SS Illuminator 602010 and siblings don't get more love? They have a desireable small form factor, very respectable low distortion and still doesn't cost an arm and a leg, $122 at Madisound. According to Zaph's measurements they even seem to be able to cross low when they have lower distortion in the 1-2kHz range than proven performers like Millenium or the RS28. What am I missing? :??

I vote for 602010, RS52 & 8857 (or 8851 since I have them and could replicate your build :yesnod: ), after all the tweeters protective grill matches the RS52 most satisfactory... :yesnod: and why not spend a little extra when you put up so much time in building anyway...? :yesnod:

It happens that a pair of these are also available for testing- consider it added to the list... this should be interesting... very interesting...

[fbov]

So now when the subject is tweeters do you ET, or anyone else for that matter, have any idea why the SS Illuminator 602010 and siblings don't get more love? ...

They do, indirectly, in the Modula MTM CC. I used the D3004/602010 as a replacement for the original D2904/6000-01 with excellent results. I agreee it would be a nice visual match to the RS52, but I prefer to leave ET to his evil musings, as I like what comes out.


HAve fun,
Frank

[5th element]

I thought I'd post this as a follow up to the small discussion on Push-Push vs Push-Pull loading. I decided to run a quick test using my XLS subs, first in Push-Push configuration and then in Push-Pull. These are in small sealed enclosures and I ran the test at a voltage level that would cause them to reach roughly 1/2 their xmax value at 20Hz.

To say that the results surprised me would be an understatement. The lower of the traces is the drivers loaded push-pull where they show a considerable reduction in 2nd harmonic distortion. I didn't bother to show any of the other harmonics, as the change was far less drastic, but the overall trend was a slight lowering of each HD product.

[JonMarsh]

:T pretty significant below 100 Hz! I'm wondering how the Wavecors would test in that mode...

[JonP]

Indeed... I've always read that drivers in push-pull cancel out distortion from unevenness in suspension and motor, but I don't think I've ever SEEN someone do it and measure the results.

Nice!!