• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Exciting new line of fullrange drivers from Feastrex

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Some of you may remember my posts some time back about the "Exact" (now "mu:xacta") brand of fullrange drivers made by Mr. Takashi Sano in Japan. Last year he entered into a joint venture with IBI Co., Ltd. to start a high-end brand called "Belle Lab" (now "Feastrex"). Mr. Sano and IBI have since parted ways and Mr. Sano continues to make ever more awesome drivers on his own (he is now working on a 30cm fullrange that makes his previous drivers seem both puny and inexpensive by comparison) and IBI has continued making drivers on their own. The two manufacturers are now going in quite different directions with regard to their respective product lines. The purpose of this post today is to introduce IBI's new Feastrex line of fullrange drivers, but I thought I should mention the (now severed) former connection between IBI and Mr. Sano in order to help prevent any confusion, as there are some superficial similarities between some of the drivers.

From January 5 to January 8, IBI Co., Ltd.'s "Feastrex" brand shall be exhibited at CES in Las Vegas (Alexis Park, Booth No. [Room No.] 1362). Speaking from personal experience as a user of their fullrange transducers, I can confirm that Feastrex has some truly outstanding products. They not only produce the world's most powerful fullrange drivers (both AlNiCo and field coil models), but they also have developed a highly innovative technology, the Naturflux (TM) magnetic circuit, that is surely a major milestone in the history of audio. After hearing it, I believe many will concur with the general assessment of hearers that it is the finest fullrange driver ever produced.

As I write, the Feastrex English website is in the process of being updated, so if you go there right away the English pages may still be "under construction"; Your understanding and patience are requested. (http://www.ibi.co.jp/feastrex/english/index.html )

I have prepared JPEG images of the new Feastrex product brochure. The printed version is a single sheet of paper printed on both sides and folded into thirds (for a total of six faces). Please take the time to check out the brochure images and the website. Those specs are for real, folks! They paid good money to have proper tests done by a testing facility and I have seen the data printouts. Those tests were done on pre-production prototypes, and they have tweaked the drivers somewhat to address a few minor areas that needed improvement (the test results showed them what they needed to work on), but the description of the drivers does not go beyond what they can demonstrate, or what a user could demonstrate with his own drivers. I have heard all the drivers except the most innovative model, the one using the Naturflux magnetic circuit. People who have heard it say it is definitely better than the other Feastrex drivers, which blows my mind because I thought the Feastrex drivers were already as good as a fullrange driver -- or any speaker -- could get. So I am really looking forward to hearing these new drivers at CES in a few days. Anyone who has a chance to hear these at CES should definitely take the opportunity to do so.

Here are links to the images:

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.


Please feel free to share this information with anyone who might be interested.

Sincerely yours,

Christopher Witmer
Tokyo

P. S. I will also be at CES 2006 with Feastrex, acting as an interpreter. In the interests of full disclosure: I shall receive remuneration from Feastrex for my interpreting, but beyond that, I do not have any financial interest in the company, and I do not derive any financial benefit from sales of the products. I do not have any arrangement with them to receive a percentage on sales, etc. (The payment for my services as an interpreter shall be exactly the same whether they sell zero units or 10,000.) I am taking time off from my real job in a securities company to do this. It is high time these exotic drivers got some actual exposure outside Japan, and I am glad to see that it is finally going to happen.

P.P.S. The pricing -- where did I put that price sheet? Sorry, I can't find it. But if I recall correctly, the Naturflux drivers have a local price in Japan of 148,000 yen each. That includes local 5% consumption tax which would not apply to exports.
 
cdwitmer said:
but they also have developed a highly innovative technology, the Naturflux (TM) magnetic circuit, that is surely a major milestone in the history of audio. After hearing it, I believe many will concur with the general assessment of hearers that it is the finest fullrange driver ever produced.

If it is indeed the finest FR ever produced, that is a far more important distinction than the magnetic circuit. As far as I can see though, nothing fundamentally in the magnetic structure differs from a standard neodymium cup type other than the returns being bow shaped as opposed to straight. I'd love to see the FEA on the magnetic circuit of this driver in contrast to that. If there are definitive advantages of the Naturflux design, FEA should show it.
 
I don't know whether they are using FEA in house or not, but I suspect they are not. One of their staff did tell me that they were not able to simulate the actual results they obtained with their magnetic circuit, and that the level of performance came as a surprise to them. Can you (or anyone) recommend a suitable finite element analysis package (I assume 3D modeling would be called for) that would be affordable for a very small outfit and could run on a desktop computer of reasonable power? And it would be nice to get an idea of the learning curve that goes into using such a tool . . .

Thanks!

Chris
 
Ok, here it is:

I specified an alnico magnet capped with shaped permendur (giving some benefit of the doubt, since I didn't see permendur specified in the writeup). The rest of the yoke is pure iron.

As you can see, there is indeed a flux bottleneck in the permendur below the gap, where flux density surpasses 2T, which is nearing the limit of permendur. If the pole cap is not permendur, this restriction would be quite a bit more serious. As it is, I wonder about the wisdom of encouraging saturation anywhere other than the gap faces, though such is all to common in motor design, usually because of penny pinching.

Bill
 

Attachments

  • feastrex 1.jpg
    feastrex 1.jpg
    63.2 KB · Views: 8,723
BTW,

Can you (or anyone) recommend a suitable finite element analysis package (I assume 3D modeling would be called for) that would be affordable for a very small outfit and could run on a desktop computer of reasonable power? And it would be nice to get an idea of the learning curve that goes into using such a tool . . .

I use FEMM, a free and easy-to-use magnetic FEA application that is slim, free, and easy to learn. I highly recommend it. It can handle planar (2d) and axisymmetric problems.

Cheers,

Bill
 
Thanks very much, Bill. I will pass this on to the manufacturer. This particular driver does not use Permendur. However, their actual measurements of gap flux density are higher than what the diagram suggests . . . it makes me wonder about how accurately their drawings represent the actual internal construction of the driver. (I note, for example, that their drawings do not indicate a shorting ring, even though I know they use them.)

I'm sure you know far more about drivers than I do, but I certainly endorse your view on the need for saturation at the gap. For optimum performance, saturation at the gap is essential. On the other hand, if achieving saturation outside the gap area does anything useful (especially if the gap itself isn't saturated), then I don't know what that might be . . .

Gratefully,

Christopher Witmer
 
On the other hand, if achieving saturation outside the gap area does anything useful (especially if the gap itself isn't saturated), then I don't know what that might be . . .

I was wondering about this, too.

Unless its gap faces are always saturated (practically impossible with healthy LF signals in the VC) there is a certain amount of field reversal and iron modulation in all dynamic bass and wide-band drivers. This flavors the sound to varying degrees, perhaps even in a pleasing manner in some cases.

Faraday rings can help with iron modulation only in the mid/treble and up as they couple with the VC field less and less efficiently as frequency drops. I can think of only three real ways manufacturers are currently combating LF iron modulation. In decending order of effectiveness: ATC's SLMM, 18Sound's AIC, and JBL's Differential Drive.

In the case of this Feastrex driver, at least the flux bottleneck is high up the pole, quite near the gap, which I imagine would allow less iron modulation than if it were at the base of the pole, as in so many other drivers.

Also, by reducing flux in the gap, it probably allows a higher Qts and therefore more bass.

Anyway, none of this is to say that this Feastrex doesn't sound great. I just raise the question about the flux bottleneck because it goes against conventional wisdom, and I'm quite sure it wasn't a cost-saving maneuver in this case! :)

Bill
 
Hi Bill,

Nice work there, but I wouldn't be sure the illustration you based the sim on is anything more than just to give a visual of the concept.

Firstly the specs state 1.8T in the gap vs the simulated 0.5-0.7T. Also for something where cost is not a huge constraint (I assume) they would have missed some simple angles in the motor design that would achieve more symmetrical fields too.

Also the sim. design is cut through the centre which could be a reason for the magnitude of the bottleneck. What if you draw the other half slice of the motor, hence full thickness pole piece as in the mfg. dia.
 
Firstly the specs state 1.8T in the gap vs the simulated 0.5-0.7T.

You believe everything a manufacturer tells you? I don't. I wonder how they determined they had 1.8T in the gap.

Also the sim. design is cut through the centre which could be a reason for the magnitude of the bottleneck. What if you draw the other half slice of the motor, hence full thickness pole piece as in the mfg. dia.

My guess is, and Bill can confirm, that the model is axisymmetric in which case the math accounts for the part not shown on the other sied of the centerline axis. Looks like a very nice program, I used to use ANSYS for magnetic field calculations when I was working for an MRI S/C magnet manufacturer.
 
You believe everything a simulation tells you?

If I know the basis for the simulation, and have confidence in the analyst doing the simulation, then I have a good feeling for what accuracy and inaccuracy is present in the results. No simulation is perfect, the best simulations are only good enough to give an accurate result for a specific set of questions.

Do I believe a glossy marketing data sheet with one very detailed result but no T/S parameters (which are kind of fundamental properties)? To be polite, I will only say I am skeptical and would not spend the money on the drivers yet without more trustworthy information. I would spend the money on AER or Lowther drivers first with much more confidence.
 
MJK said:


Do I believe a glossy marketing data sheet with one very detailed result but no T/S parameters (which are kind of fundamental properties)? To be polite, I will only say I am skeptical and would not spend the money on the drivers yet without more trustworthy information. I would spend the money on AER or Lowther drivers first with much more confidence.


I have a pair of Japanes Exact drivers which are realted somewhat to the Feastrex drivers. It's an excellent driver much more transparent and faster than the Coral Beta 8 which I also own. I would say in addition to AER, Lowthers..... you can put the Exact and probably the Feastex in the same category. Ultimately which single driver you finally choose will depends on the type of cabinets and taste.. :)
 
I would say in addition to AER, Lowthers..... you can put the Exact and probably the Feastex in the same category. Ultimately which single driver you finally choose will depends on the type of cabinets and taste..

For me, the final choice comes down to who has at least one reputable US distributor and who has the best technical information available to determine if the driver will align with my interests. If they have a US distributor and the distributor is difficult to deal with then that also rules out the driver, I had this happen when I wanted to buy AER and Supravox drivers a few years ago. Personally, I would not order an expensive driver all the way from Japan period. There are just too many risks both technically and logistically. But that is just my opinion.
 
MJK said:


For me, the final choice comes down to who has at least one reputable US distributor and who has the best technical information available to determine if the driver will align with my interests. If they have a US distributor and the distributor is difficult to deal with then that also rules out the driver, I had this happen when I wanted to buy AER and Supravox drivers a few years ago. Personally, I would not order an expensive driver all the way from Japan period. There are just too many risks both technically and logistically. But that is just my opinion.

Fair enough, but do note they will be exhibiting at the CES over the weekend hopefully some reputable US dealer will pick up the line... If that happens, I'll really look forward to reading more reviews about the drivers.. :)
 
Hi Vikash,

Nice work there, but I wouldn't be sure the illustration you based the sim on is anything more than just to give a visual of the concept.

Firstly the specs state 1.8T in the gap vs the simulated 0.5-0.7T. Also for something where cost is not a huge constraint (I assume) they would have missed some simple angles in the motor design that would achieve more symmetrical fields too.

Also the sim. design is cut through the centre which could be a reason for the magnitude of the bottleneck. What if you draw the other half slice of the motor, hence full thickness pole piece as in the mfg. dia.

I agree that we shouldn't make too much of an illustration that may not reflect the full reality of the actual motor. Whatever the actual proportions, though, if the pole is drastically undercut (as they seem to illustrate) and there isn't significant tapering toward the gap, then there must necessarily be a flux bottleneck below the gap. Also, as Christopher mentioned, the pole cap is not in fact permendur. I can't see how they could achieve anywhere near 1.8T in the gap under these conditions.

Again, though, let me say that I still admire the drivers as audio objects d'art, whether or not they rigidly conform to spec, and I'd love to see more boutique manufacturers crop up. Actually, I'll likely join a new ultra-high-end effort later this year... (and we'll always supply lots of accurate specs and measurements). :)


My guess is, and Bill can confirm, that the model is axisymmetric in which case the math accounts for the part not shown on the other sied of the centerline axis.

Correct, Martin. :cool:
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.