Focal CSD Measurements

more_vert
Focal gets a lot of hype in online headphone discussion forums these days. For those unfamiliar, Focal utilize a novel 'inverted dome' diaphragm for their drivers. . . one of the known issues for this is that they hard clip at higher volumes, but this is not often an issue for reasonable listening levels. I am curious if there are other unspoken limitations, given the apparent novelty of this driver approach. Despite a humiliating fiasco for Utopia manufacturing precision (https://www.innerfidelity.com/content/focal-utopia-update), Tyll had an enduring fascination with the brand. I personally think this informs a lot of the current enthusiasm. I have recently been looking at spectral decay plots, something that unfortunately Tyll could not produce on his otherwise legendary test rig. It appears to me that the inverted domes could have a similar persistent issue with spectral band ringing in the upper mids: Focal clear CSD (https://www.soundstagenetwork.com/index.php?option=com_content&view=article&id=1823:focal-clear-headphones&catid=263&Itemid=203)
search
Their Elegia measurement shows some resonance around 3.5K as well, but it appears relatively better damped than for the clear (https://www.soundstagenetwork.com/index.php?option=com_content&view=article&id=1965:focal-elegia-headphones&catid=263&Itemid=203):
search
I had a fresh look at the otherwise useless idiots at SBAF measurements of the clear specifically
search
Despite the lower spectral res of this CSD, we see similar resonant peaks in the upper mids. . . In a hobby where folks claim to differentiate between amps/DACs at >120 dB SNR, it seems a bit telling that relatively pronounced -20 dB ringing has gone without objection on these 'end game' cans. . . Thoughts? Is this another fundamental (if relatively unobjectionable, given the resonance window) shortcoming of the 'inverted dome'?
(Edited)
7
61
remove_red_eye
535

search
close
lupi900
27
Jun 23, 2020
> In a hobby where folks claim to differentiate between amps/DACs at >120 dB SNR, it seems a bit telling that relatively pronounced -20 dB ringing has gone without objection on these 'end game' cans. . . People in this hobby LOVE assuming dynamic drivers are flawless, I've had Redditors/ASR posters argue that >2% on a HD650, 1990, etc is okay. But will gladly bash a ER4XR being 0.7 ~ 1.7% bad despite them oddly being unable to tell a loudspeaker with 8 ~ 20% bass distortion or colouring issues in lower mids. I think it gross this is okay for $4000 but the ER4SR/HE400i mesure better while being only <$350.
kjling
129
Dec 26, 2018
search

search

search

search

And even more ignorant people judge them for it. I will now be turning off your comments, congratulations on being the third person I done that with.
nick_t
227
Aug 30, 2019
Well its a fact unless you have any valid argument that back the claims. Evidence and well constructive statement.
kjling
129
Dec 26, 2018
IMHO, measurements, specs are all well & good... but ultimately, this hobby is so subjective that no amount of charts will inform us of what we like to hear on the day. E.g. i do like the benchmark HD6xx especially on a tube amp (which will always measure worse than SS amps) but I love my Elex and LOVE the Utopia. Yet I also enjoy the Audioquest NightOwl Carbon, a super laidback warm headphone and my LCD2C. A modded Hifiman HE-560 (ZMF auteur suede non perforated pads) is heavenly, and my current muse is the Hifiman Arya. I also have a ZMF Verite Ltd on the way. My likes are super diverse, i might even be accused of being easy to please. Yet I am not impressed at all by the HD58x, K7xx (only good for gaming due to extra wide soundstage), which others will sing high praises for. My 1 cent - YMMV, just try if possible before u buy, and go with what u like to hear. Don't let anyone tell you otherwise.
Neubrineo
27
Dec 22, 2018
WOW! Louder sound takes longer to drop below 55db than quieter sound, this is real mystery why its like that, we should hire Sherlock Holmes or Columbo to investigate this phenomena.
Resonance has a pronounced effect on decay rate regardless of the initial or final value, it's fundamental to its nature. So, no, when its part of a resonance like this quieter sounds might take longer, despite being from a lower initial value. This is ringing (resonance), possibly as a predecessor to well known hard clipping, neither of which you seem to understand based on your cute quip here, Columbo.
(Edited)
pslayer1
272
Dec 19, 2018
I haven't noticed any 'ring' in my Elex but I have gotten them to clip on some audio tracks. I thought I broke them, at first and was going to return/exchange them. Then, upon research, I found that this was 'common'. I love the Elex and listen to them every day. I just wish the clipping wasn't a thing.
phoenixsong
1209
Dec 26, 2018
Time to give your ears a break haha! Hopefully they'll be able to appreciate lower level listening after that :p
pslayer1
272
Dec 27, 2018
As the immortal Ricky Bobby once said....
search

jaydunndiddit
3214
Dec 18, 2018
Out of curiosity, do you have much experience listening with Focal headphones? As a pretty solid fan, I can tell you that ringing is not as prevalent as you may think and it's more akin to the tonality of other metal treated drivers. So if you're used to that sort of tonality, you know what to expect here. Some people are turned off by it but I've been listening to these types of drivers for years so for me it's the norm. The over excursion issues is hard mechanical clipping. Focal has been pretty outspoken about this, the why, etc. AFAIK, the Clear and Elegia are the only models that don't exhibit this. The Elex may be the worst offender followed by the Elear. And trust, I have to really give the Elex some power to push them to that point. I only really notice on songs with a heavily distorted low-end when the volume is turned way up. Personally, I don't think it's an issue in the real world. Especially with the Clear. They live up to their namesake for a good reason.
(Edited)
kjling
129
Dec 28, 2018
neither the WA7 nor the iTube2 uses the Tung Sol 5998.. take my word for it, and look out for a chance to try it on an amp that runs it one day :) Or sth like the Glenn OTL with 6x 6BX7/6BL7 or 2x 6336 power tubes.
Themselves
3
May 14, 2020
Bought a pair of the Clear Professionals to try this out. They clip, just as much as my two Elex do. :(
Tuckers
19
Dec 15, 2018
I would also think that the type of metal they use could be a culprit. The Elear, Clear and Elegia use the same Alminum / Magnesium metal drivers. The Utopia however uses Beryllium. Personally I've never been a fan of aluminum in drivers. I do own the Elear though, and I do hear that peak in a few types of music.
(Edited)
I doubt the transducer precision would compare, but there are a lot of quite nice audio analyzer apps for Android / iOS that can use your phone mic. I use 'Sound Analyzer App' for Android, I like that it also gives you a spectral breakdown. I would think it would tend to overestimate the SPL per channel, since it's hard to isolate the cross channel speaker from the phone mic. . . still quite useful to level match for A/B (and don't kid yourself, any listening comparison without level match should be treated as very suspect)
This is a good point, and I think something we should generally keep in mind when comparing THD specs at 90+ dB! The particularly dramatic Soundstagenetwork Focal measurements are at 94dB, maybe no surprise then that there is some 'rattle' showing at relatively low (and narrow) levels, not even hard clipping per say.
Megazine
511
Dec 15, 2018
Great work!