3"or 4" driver with very good dispersion and high xmax?

Status
Not open for further replies.
Founder of XSA-Labs
Joined 2012
Paid Member
The Faital full rangers do sound quite nice. I have 3 3FE25 that I use in a trifield setup in my bedroom. The way they image is fantastic.

Finally someone with firsthand experience! I have been waiting for some feedback on these drivers for a while. I asked about them in this thread http://www.diyaudio.com/forums/full-range/246340-anyone-try-faital-pro-4fe32-3fe22.html

They have less xmax than the Vifa's but are more sensitive. They also appear to be made for backside mounting and look a little rough around the edges on the metal frame up front. Otherwise they look really good on paper.

I am thinking of trying the 4FE35 version which is only $25 and had larger xmax of all variants.
 
Alas, I have not heard a glass cone TG9... I can tell you that the paper cone TC9 sounds great.
The TG9fd sounds great as well. I haven't heard the TC9. Bought three of the TG9 for testing for a particular project, and tried the 7.3 only to get the modestly greater bass extension (greater Sd and Xmax). Both exhibit similar glitches between 1 and 2 kHz (the Peerless perhaps a bit better damped), and the dispersion curves from 2kHz to 8kHz all but overlap (as you would expect from the similar diameter cones). They both have a rising response above that (on axis) with pronounced resonant peaks . . . the 7.3 peaks are higher, but its dispersion is significantly wider. I'd say that the choice between them is very application dependent . . . they don't sound the same, but I'm not sure which I prefer. I still haven't made the final decision which to use, although at the moment the 7.3 is ahead on points. It also (I'm embarrassed to admit I care) just looks better with the woofer already chosen for this project . . .
 
I'd measure the Faitals myself before I'd put them before the vifa TC9's. Very few frequency response graphs published by the driver maker are very accurate. They always do smoothing, which is legitimate to a point (1/6th octave in my opinion).

I'd hate to have less Xmax than the TC/G 9's have (2.5mm peak), but the improvement in efficiency is a good thing for those who are driving with a single ended tube amp.

And yes, a 3 incher going from 200HZ to 7kHZ gives some of the best imaging I've ever heard. The Polk style holographic soundbar I built with the TG9's shows this particularly well.
 
Almost any metal FR driver in the 2 - 4 range will have large peaks around 10 kHz.
And better dispersion.
Just look at the Peerless alu drivers for example.
People saying that the difference between paper and metal is moot clearly haven't done their research.
Metal drivers have gone from being almost universally poorer performers, with the clear exception of Jordans, to being a contender. But there is still a clear difference in the overall signature of the sound.
 
Last edited:
Regarding the MA drivers. I cannot recall seeing (on this forum or any other forum) frequency response measurements performed by other than the manufacturer that are consistent with the published measurements provided by MA.

When I look at the frequency response measurements given in post #177, it is my clear perception that the published measurements are misleading. How can this driver be called full range? There are drivers out there, marketed as midrange, that shows more and smoother HF extension.
 
Thread title is "3"/4" driver with very good dispersion and high xmax?"
***
So if anybody has meaningful data (preferably off-axis response out to 90°) regarding the topic at hand ("3"/4" driver with very good dispersion and high xmax?") I would be grateful if he would post it.

The Aurasound NS3, mentioned above, isn't really a full-ranger (breakup at ~6kHz) but it's cheap and has more throw than anything else I know of in that size class. (Except the Dayton ND91, which is basically an Aura NS3 hot-rodded by Don Keele for his CBT36 design.) See Klippel measurements and some FR measurements out to 60deg here. I agree with Deward that dispersion is basically a function of piston size.

Compare with Klippel and FR out to 60deg of the ScanSpeak Discovery 10F, which is in some respects a fancier variant of that little Vifa mentioned earlier, here.

So you're claiming you know more about the operating parameters of the suspension, its tolerances etc. than its manufacturer does. Most impressive.

I don't and can't know what anyone else knows. I can and do know that someone is spewing a line here with this break-in nonsense, though. A line that given the measured performance of the product is very astute marketing!

Foam surrounds rot, paper cones become brittle and shred, "plastic" cones lose plasticizers and become stiffer, compression tweeter diaphragms crack from fatigue . . . all sorts of mechanical changes occur in loudspeaker drivers over time and with use. Do you really believe none of that ever happens? Why would you deny that a hundred hours of flexing might change the properties of a "new" spider?

My experience is that drive-units aren't wear items in the way that, say, tires, brake pads, or shoe soles are. Let's throw the myth of break-in (as opposed to breaking) to a real expert who's actually done the relevant test.

Dr. Floyd Toole said:
In parts of the audio industry, there is a belief that *** loudspeakers need to “break in.” Out of the box, it is assumed that they will not be performing at their best. Proponents vehemently deny that this process has anything to do with adaptation, writing extensively about changes in performance that they claim are easily audible in several aspects of device performance. Yet, the author is not aware of any controlled test in which any consequential audible differences were found, even in loudspeakers, where there would seem to be some opportunities for material changes. A few years ago, to satisfy a determined marketing person, the research group performed a test using samples of a loudspeaker that was claimed to benefit from “breaking in.” Measurements before and after the recommended break-in showed no differences in frequency response, except a very tiny change around 30–40 Hz in the one area where break-in effects could be expected: woofer compliance. Careful listening tests revealed no audible differences. None of this was surprising to the engineering staff. It is not clear whether the marketing person was satisfied by the finding. To all of us, this has to be very reassuring because it means that the performance of loudspeakers is stable, except for the known small change in woofer compliance caused by exercising the suspension and the deterioration breaking down—of foam surrounds and some diaphragm materials with time, moisture, and atmospheric pollutants.

It is fascinating to note that “breaking-in” seems always to result in an improvement in performance. Why? Do all mechanical and electrical devices and materials acquire a musical aptitude that is missing in their virgin state? Why is it never reversed, getting worse with use? The reality is that engineers seek out materials, components, and construction methods that do not change with time. Suppose that the sound did improve over time as something broke in. What then Would it eventually decline, just as wine goes “over the hill”? One can imagine an advertisement for a vintage loudspeaker: “An audiophile dream. Model XX, manufactured 2004, broken in with Mozart, Schubert, and acoustic jazz. Has never played anything more aggressive than the Beatles. Originally $1700/pair. Now at their performance peak—a steal at $3200!

Sound Reproduction at 353. Emph. added.

Same with Magneplanars, and the same "break in" recommendation from the manufacturer. All "idiots", I suppose . . . :rolleyes:

I concede that a drum head suspended in magnets may be different from a cone/dome. (Electrostats do literally break quicker than regular speakers, too.) However, considering how Maggies sound, I'm not sure your characterization is unfair...
 
My experience is that drive-units aren't wear items in the way that, say, . . . shoe soles are.
Ah yes, but . . . how about the shoe? Never bought a new pair of boots that were less than completely comfortable when new (blister inducing, even), but after a few weeks on the trail "broke in" and became like a second skin? It was (and is) one of the great advantages of leather as a shoe material (as opposed to the no break in, no never not ever, synthetics of today).

I suspect that surrounds don't "break in", at least not past the first few thousand flexures (which should come pretty quick). Metal cones clearly don't. The place(s) where "break in" might happen are the spider (which in my experience really does change properties with use) and perhaps the flexible bond between the cone and the former (which in the MA drivers is supposed to give at least some high frequency decoupling). Toole's "general rule" may well apply to Harmon products, which are designed to work "out of the box". MA says their speakers need some exercise to reach their long term performance plateau. I think it's reasonable to assume Toole knew how Harmon products behaved, and that MA knows how MA products behave.
 
Last edited:
The real problem with the Alpair driver (besides the highly misleading vendor supplied magnitude response graph) is its off-axis behavior. A power response hole followed by a huge peak.
attachment.php
The 30 degree curve is actually somewhat better than one would expect . . . apparently the "button" tweeter is working, at least out that far. The 60 degree curve is exactly what one would expect from a cone that size . . . look at the TG9 at the same angle.
http://www.tymphany.com/files/TG9FD-10-08 Spec sheet Rve 2_0 090612.pdf
I don't know what you were expecting to happen when cone diameter becomes greater than the radiated wavelength, but now you know what actually happens. Off axis nulls. The 7.3 does a better job of controlling them through a reasonable listening angle than other drivers of that size, but out at 60 degrees and beyond you can't beat physics.

This is all something with which one becomes immediately and intimately familiar if you do dipoles, because off-axis behavior is the meat and potatoes of dipole design.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
The Vifa has about the same cutout dia as the 85wk so you may be able to do drop in replacement for sealed or nominal 10:1 TL alignments to check it out quickly. There really is not any break-in required that I can tell. I have measured drivers used daily for a year with a brand new one and the response is very consistent. Have fun with them - they are a great sounding unit for a nice low price.
 
Founder of XSA-Labs
Joined 2012
Paid Member
I don't know what you were expecting to happen when cone diameter becomes greater than the radiated wavelength, but now you know what actually happens. Off axis nulls. The 7.3 does a better job of controlling them through a reasonable listening angle than other drivers of that size, but out at 60 degrees and beyond you can't beat physics.

You can't beat physics but the TC9FD apparently is using physics to its full extent because that huge dip at 7kHz is not there. Here are results of the TC9FD measured for 0 deg thru 90 deg. Presented on same vertical -10dB to 110dB scale as the A7. This was measured at 1 m away by moving the mic and leaving speaker stationary relative to room.

(calibrated mic and DAC using Dayton UMM-6)

These are for the dual driver Nautaloss II:

402191d1393333563-3-4-driver-very-good-dispersion-high-xmax-nautaloss-ii-2-drivers-polar-1m.png


These are for the single driver Nautaloss I:

402193d1393334341-3-4-driver-very-good-dispersion-high-xmax-nautaloss-i-1-driver-polar-1m.png
 

Attachments

  • nautaloss-ii-2-drivers-polar-1m.png
    nautaloss-ii-2-drivers-polar-1m.png
    116.1 KB · Views: 5,129
  • nautaloss-i-1-driver-polar-1m.png
    nautaloss-i-1-driver-polar-1m.png
    115.1 KB · Views: 3,399
Last edited:
With all the respect due to your obviously very good measurements X (I have no query at all out to ~30 degrees since they concur well with others I've seen for this unit) -out to 90? I would have severe reservations to how accurate they genuinely are. I stress no offense etc. of any kind is meant by that. But you ask any manufacturer or pro-testing body whether they think high off-axis measures are likely to be particularly accurate without full industry standard test gear in an anechoic chamber. With very few exceptions, you'll get the same (polite) answer -viz. 'no'. YMMV as always of course & doesn't alter the fact that the above is clearly very good indeed in terms of linear FR behaviour.

With that said, the TC9 should certainly have superior HF dispersion to the Alpair 7, since, notwithstanding the latter being a good performer in this regard for a unit of its size, the Vifa is about 28% smaller, which is an advantage in this regard.
 
I have some reservations about > 30 degrees off, although I'd expect the TC9 to do well.

Probably -I wouldn't want to push it any harder, or even so hard -HD + potential FMD would not be pretty. Good driver though it is, it's still a 3 1/2in unit with about 5.5in^2 cone area so it can't shift much air. Not that it matters, I rather like them since while there are some things you might wish they did more of, they don't do anything annoying. Which is a bigger deal than it sounds / reads, if you know what I mean.
 
But they are dirt cheap. You could buy 2 or more for the same price as the "low end" MA drivers (CHX-70 and the small Alpair).
Angled arrays or line arrays (Roger Russell) would give you any kind of dispersion you'd want and better low end (two drivers should be able to hit 50 without straining too much and four would touch the mid 40s).
 
Last edited:
Status
Not open for further replies.