How good are Bowers & Wilkins drivers? Let's test and see.

[@derrickpierce4440]

Thank you for this video, I’ve always wondered how a surround-less driver would perform. It would be great to see how that driver performs at different drive levels.

[@APSuk2]

Amazing what those people at Klippel can do with software, most of this video was way over my head but was very interesting to see what goes into measuring the performance of a driver.

[@keepingupwiththejones2933]

Loving that swappable faceplate. Really cool. I appreciate you explaining how the baffle size works in conjunction with the nearfield scanner. You have a great way of helping us understand complex audio issues.

[@ErinsAudioCorner]

Glad you like it!

[@biggityboggityboo8775]

I would be surprised if that blip at 1.2kHz was from cone edge/surround, the LF000264, as measured by Zaph, has no such resonance, and that's the entire point of the flat surround anyway, to prevent this from occurring. My own experience with the 6.5" FST also mirrors the absence of the FR associated dip, impedance and distortion blip. If this little driver has such a surround resonance then B&W really didn't do their R&D properly here. That issue aside it's actually pretty decent in my opinion. Above 500Hz, at 96dB, it has better distortion than the 10F. Very low 4th, 5th, 6th, a third harmonic that falls off the radar with increasing frequency and no real HD associated issues because of the cone break-up. This is very much like the big FST, with very low inductance, only in the 4" version it cannot be used anywhere near as low. Basically you need to use it an octave higher than its bigger brother, which is no surprise given it's much lower surface area. It is essentially a 90dB/4ohm driver, which is usefully more sensitive than the 10F. Depending on how loud you want it to go with very low distortion cross at 300 or 500Hz, 4th order, and as low as your tweeter will go. B&W always mis-use their drivers and crossover way too high. Can you tell me if the phase plug is removable?

[@kushkush-dt9nr]

can you test a bowes & wilkins Zeppelin ?

[@DanMarx]

Cool video! Always been curious to see the raw response of some B&Ws. In practice though I don't think it's that bad once you throw a crossover on it. For grins I took your frequency response and impedance plots and created .frd and .zma files and noodled around in PCD for a few minutes to see what this thing could really do. This driver is actually quite simple to work with and models very well even with simple crossover components. A single 0.33 mH inductor (6 dB/1st order) combats that rising slope near perfectly making the response flat and providing a decent 4th order acoustic slope at about 6 kHz. That's a bit high for most mids, so to bring it down to 4 kHz only takes a 0.45 mH inductor and 12 uF cap (12 dB/2nd order) to provide a nice 4th order acoustic slope at 4 kHz with a near flat response all the way down to 200 Hz. So in all fairness, despite the perceived raw FR plots, the driver is perfectly workable in a system. But like you said, it ultimately depends on what you want to do, and this driver will not fit the bill for every situation.

[@attainableaudio4533]

Of course once on an enclosure that could change to a more complex xo... or not like you said depends on the application and what you want it to do. It does look cool though:)

[@ErinsAudioCorner]

Cool stuff. If in the future you see my tests and you want the output data just ask. I can email it to you.

[@maxpower7504]

Love the setup time lapses

[@ErinsAudioCorner]

Thanks! I gotta be honest... I think they're my favorite part of this video. LOL

[@TylerStout]

Very interesting video. I think the idea of speaker design decisions is interesting which is why I'm excited to see Laurie Fincham back on. Thought this was interesting considering the center channel has crossovers at 350hz and 4khz so from ~1700hz to 4000hz it's beaming.

[@biggityboggityboo8775]

Yep, B&W screw you over with off-axis in all of their speakers.

[@Mike81111]

Love these tests and now we have more to watch 👌 This driver is clearly meant to only handle the midrange (about 300-2000Hz).

[@Clobercow1]

I think the driver was meant to be a mid-range, but also meant to have a particular look. Overall, it's not a great driver with the rising response and breakup.

[@pliedtka]

I would shoot for 450Hz or more. But I was expecting cone break up higher than 4.5k from B&W cone. This driver reminds me a bit of Focal 5" Polykevlar cone mids from the 90s, but the Focal's fs was in the 80-90s Hz and cone break up was higher in frequency, Q in 0.4 region, 90dB sensitivity and 8ohm.

[@JR-ho5qm]

I’m curious as to how CerwinVega speakers are able to have high sensitivity as well as high SPL. And good bottom end I may have miss understood you bought I thought you stated that it’s hard to get both

[@kristiankeller4335]

Hoffman’s Iron Laws? They get both good low end extension and high efficiency by using big drivers in big enclosures.

[@jf4828]

Just curious how one justifies a ~ $100k Klippel test tool? What are you using that thing for besides reviews? I’m not even sure if many manufacturers pony up for one of those:)

[@rob_66]

@@roger_isaksson Or you could buy a decent system for $50k and cd's, vinyls, downloads or whatever format you prefer for $50k. Then sit down, pour your favorite beverege and just listen..... and forget all about measurements😉

[@jf4828]

I’m not hating, just generally curious as I’d love to have one. I’ve gotten pretty good with my measurement routine but polar and listening window test are time consuming. Low frequency resolution without having to do ground plane out in the elements is good. I’ve given up on splicing near field and far field for higher resolution because it’s too hard to get right….

[@klee268]

Klippel measurement system consists of many modules. NFS is one of the modules and shares some hardware and software with other Klippel system. I guess Erin has other modules for motor system analysis. I can say for certain that every major speaker manufacturer uses some modules of klippel for RnD and production. You can check the list of Klippel users at its web page. NFS is not a common and popular module because anechoic room can substitute it. But You can not build an anechoic room + automated measurememt system with 100k. Even a small one with 100Hz cut-off frequency will cost more than 200k. The room only. Of cause some measurement such as microphone measurements requires anechoic room though. I guess 100k NFS maybe too much for a "reviewer". I , Klippel distributer, still don't have the demo equipment of NFS. I thought Erin was running some consulting company at first. But if you are serious on SPL based analysis of speakers, NFS can be a "cheaper" alternative. Other simulated anechoic measurement methods have thier limitations. I think you already know them.

[@Cakebattered]

Now we will understand why many complain that B&Ws sometimes sound "Too Midrangy"

[@FSXgta]

this is the response without the crossover tho. I think B&W is more treble boosted than midrange

[@RennieAsh]

@@FSXgta it depends on the model too. I found the 683 S2 is quite smooth sounding in a way with more lower midrange and upper bass, treble and upper mid is good, even a bit laid back in upper mid. Then one of their new white cone speakers is more bright and "clear" like how people usually describe b&W.

[@mattm2684]

You said this driver is from a center channel. I b&w center speakers apart and even though they look exactly like the bookshelf speakers from the front. They are no way even close to the same tweeter or mid driver as the bookshelf drivers. I'd like see the same tests done on a driver from one of the b&w bookshelf or floorspeakers

[@ferdinandbardamu3945]

Somebody should send you the latest continuum mid-range driver from their 8 series line.

[@stevenswall]

Expectation: a driver with excellent build quality, great dispersion, plenty of power handling, and incredibly competent engineering... In every aspect except frequency response which they'll voice by ear and kick all the engineers out of the room for.

[@stevenswall]

Reality: it looks decent over 300hz and under 1000hz.

[@ErinsAudioCorner]

I'd say 2kHz is feasible. The issue, really, is the increasing treble which means the crossover would need to be more involved to get it to a point where you have a traditional acoustic crossover slope of LR2 or LR4. Which means more $$$. But if used with a DSP then it's more manageable.

[@biggityboggityboo8775]

@@ErinsAudioCorner Usually a rising response is very easy to crossover and requires no additional parts vs if it's flat, you just tune the values differently. A peak, on the other hand, is another story entirely.

[@ErinsAudioCorner]

@@biggityboggityboo8775 right. The combo of the not-too-well-behaved rise and the ringing at its peak are what would make it a bit tougher to remedy.

[@IliyaOsnovikov]

Erin, do you usually take the cumulative response measurements?

[@ErinsAudioCorner]

I sometimes do. It's honestly not something I pay much attention to, though, TBH. I know people love them but I get all I need to know from on/off-axis data.

[@yeahright3733]

Nipple near field scanner?

[@RennieAsh]

Nipple Klearfield scanner lol

[@KravchenkoAudioPerth]

Never enjoyed listening to B&W Kevlar midranges. You measurements tell me loud and clear why I have not been able to stand listening to them! Nice simple review of a mediocre driver.

[@RennieAsh]

Which parts don't you like? I though they are crossed over away from most of the issues?

[@KravchenkoAudioPerth]

@@RennieAsh Crossovers are tools that have their limits. The resonance at 4.5k is not a simple thing to null out in a crossover. If you flatten it out and you think it is gone you are mistaken. When you have a transient in the music around this point, and many types of music have significant energy in this region, you will again have a nonlinear driver that will be creating a resonance at this point. It will bleed through into what you are listening to. I ran into this with a few drivers I have worked on. These distortion effects are most noticeable when you don't hear them any longer. Once you have a very clean reproduction system you will notice these kinds of loudspeakers that hum along with your tunes. It is difficult for me to listen to nonlinear loudspeakers as I can readily hear these resonances.

[@RennieAsh]

Kravchenko Audio if you cross any driver at its breakup point it will do that. Do they let that peak through with volume or do they cross over much below that point? Almost every driver will have a breakup peak, aside from some poly cones or floppy paper etc

[@biggityboggityboo8775]

​@@KravchenkoAudioPerth Actually, with this driver, the 4.5kHz resonance is something you can simply notch out. It doesn't affect the distortion in any meaningful way, so if you notch it out properly, you aren't going to hear it. Besides this driver is not supposed to be used that high. As Erin mentions you start to get beaming, with a driver this size, just below 2khz. This is way below that resonance peak. B&W cross this over inappropriately but this is pretty much B&Ws entire ethos in a nutshell. Make good drivers drivers and then use them up way too high. I'm not surprised that you dislike B&Ws speakers as they are all quite poorly designed. B&Ws drive units are excellent though, they are just mis-used. The 800 series are the biggest offenders taking the 6.5" unit all the way up to 4-4.5kHz and with a very simple crossover!

[@JamesWilliams-gf8gm]

Great video. As expected this driver is a dud

[@ThePentosin]

Dud? How did you come to that conclusion?