Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
In Reply to: RE: Zu Soul Supreme sensitivity posted by RGA on May 30, 2017 at 23:37:55
Corner placement might not have that large of an effect.
Per the standard, sensitivity should be measured under free-field conditions using pink noise, with the result weighted. The standard specifies that broadband sensitivity should be measured with a test signal having a two-octave bandwidth centered on 1 KHz (500-2000 Hz). Or, if a different bandwidth is used, it should be stated with the measurement. Toole recommends using a 300-3000 Hz test signal because it contains all the frequencies that contribute to the perception of loudness.
Boundary reinforcement boosts loudspeaker output only in the frequency range where the radiation pattern is omnidirectional. For a wide baffle box speaker like the AN-E, that's only in the bass <200Hz. Therefore, it won't have much impact on a sensitivity measurement made with either the ISO standard's proposed test signal or Toole's proposed test signal.
Obviously, if you used a white noise signal with no weighting, the boost below 200 Hz from corner loading would help raise the result, but then you have a useless number that can't be compared to anything else.
Audio Note is not the only manufacturer to advertise sensitivity or efficiency numbers that are optimistic. It's almost he norm among manufacturers of high efficiency speakers that use dynamic drivers in a box.
Take the Tekton Lore for example. It has an Audax TW025A28 tweeter whose sensitivity is specified by the manufacturer to be 94 dB, and a Eminence Legend B102 woofer whose sensitivity is 92.4 dB. Before you even consider crossover losses, it's obvious that the loudspeaker isn't going to come close to the 98 dB claimed by the manufacturer.
Compared to a free field measurement every boundary introduced will result in a 3dB rise in bass output.
So compared to free-field a speaker placed on the floor but away from walls gives +3dB;
placed on a wall another +3dB and corner placement adds a further 3dB resulting in a theoretically possible 9dB rise.
If you now measure SPL from 20-20kHz as a single figure it will seem as if the speaker is remarkably sensitive.
Less so if you measure SPL at 1kHz where corner loading has very little effect.
Perhaps read and see the measurement of Hi-Fi Critic for yourself if you are interested. Martin Colloms measured the speakers and got both the frequency and the sensitivity. Hi-Fi Choice Magazine and Audiophile magazine also measured the speakers and also achieved AN's specs.
It's not like 3dB either way is a big factor. High watt SS amp is dirt cheap - you can always look for that Rotel Power amp that is capable of 1kw at 1 ohm. Probably get it for $300 on the second hand market - then sensitivity is a moot point. And if you are buying a SET amp of the low watt variety chances are you are going to audition to make sure it has enough power to drive the speaker competently because as you note - you can't always trust what the sensitivity number is - since even if something is 95dB it doesn't mean that it will be easier to drive than than an 83dB LS-3/5a at 11 or 16 ohms.
But surely you have some common sense to realize that if a speaker is for 20+ years demonstrated with 6-10 watt amps and manages to on very many occasions win best sound at a show that the speaker is easy to drive. Even JA says the AN E is easy to drive.
So while this is a fun exercise in number crunching - real world application is more relevant to me.
And to that end - if you happen to live in California Steve Hoffman (mastering engineer for the Beatles, John Coltrane, Pink Floyd, Eva Cassidy, Ella Fitzgerald, Bob Dylan, Jackson Browne, Miles Davis, Peggy Lee, Wes Montgomery, Judy Garland etc) is having a get together where you can, on June 17th, go and listen to an AN E (not sure if it is either of Steve's pair or the dealer's) being driven on a 7 watt amp. Then bring your favorite measuring tools and hear for yourself. No cost - no selling - just listening. Special guests apparently and autographs for your albums and free food too. Not too shabby. If I lived there I'd go. Reel to Reel as well apparently.
Close computer screen/books and get off couch and experience what ease of drive means.
I never argued that AN speakers are not easy to drive. Nor Zu or Tekton or DeVore, etc.
Yet they're all fudging their numbers to claim high-90s when the reality is more like low-90s. There has to be a marketing effect from it otherwise they wouldn't do it. I'm guessing there's a market out there who want to believe they can get close to horn numbers with a box speaker.
Well that's fair enough. The AN E, DeVore and Tekton won't compete with the big horns on ease of drive - that's for sure.
Although Peter Qvortrup answered some of this in the High Efficiency forum:
Only 39 posts so get out the popcorn for those interested in how they measure and that other magazines achieve the results stated by the manufacturer.
"By the way, Martin Colloms asked me the exact same question when he reviewed the AN system 3 - 4 years ago, as he also had difficulty believing the efficiency rating, I showed a quick if somewhat simplified example of how we arrive at the stated efficiency ratings, which is done by combining 2 far field sound pressure measurements to get a room power response which is then used to calculate the efficiency, backwards so to speak.
Martin clearly agreed with the published 96dB/watt efficiency figure having tested the AN-E/SEC Silver, as he does mention in the review.
So here is a brief outline of the procedure, using the highly sensitive system we also use to match the drivers to the crossover and the reference, to get a near perfect match within pairs and to the reference,
1.) With the speakers in the correct corner loaded position, we take two sound pressure measurements at two different output power levels in two positions, using a broadband complex waveform (music), plus a series of sweep tones 15Hz to 25kHz, why use two different powe levels you may ask?
The reason for this is that I have found that many speakers appear to have an optimised behaviour at 2.83 volt input and their efficiency drops disproportionally if presented with less and in some cases more power, why this is, is a whole different question.
2.) We repeat this procedure at 3, 4 and 5 meters, using two speakers and a 2A3 or 45 and a 300B SET amplifier, each time noting the voltage output power from the amplifier and the corresponding sound pressure level.
We then compare to two reference speakers of "known" efficiency, measured the traditional way (as JA did with ours), then repeat the way we measured our speakers, compare the difference in efficiency at listening distance, average for distance and then work their room energy efficiency backwards so to speak.
We add/subtract the difference between these reference speakers and ours to get the figure we publish.
Yes, but it gives a far better and more accurate measure of what the real power transfer and equivalent sound pressure level is, and therefore closed to what the actual efficiency of the speaker is likely to be in situ in room, because if you take a 1.5 watt amplifier and you can get 100dB plus out of a speaker measured at 1 meter before audible clipping, then the speaker must be more than 92dB efficient, wouldn't you say..."
Sounds great, unfortunately the discussion is about sensitivity not efficiency and BTW , thats a lot of phooby Dust yaddy yada on measuring speakers , any speaker for sensitivity.
I guess it's all cool if you write in your own rules and science, anything pretty much goes and if you are only using 5-7 watts per channel on those AN speakers your pretty much dead to clipping IMO..
Crowded room on a 7 watt amp , thats clip city .....
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: