Audio Note amplifiers - why is it so difficult

It’s a great tube but it doesn’t sound anything like an old pentode EL34. Nothing could touch the lush mids those old tubes could produce.

 

I wish I could afford a quad of NOS EL34's to put that to the test! I am glad there is an affordable current production tube I can live with.

 

Someone correct me if I’m wrong but aren’t the efficiency specs Audio Note publishes pretty overstated? Not that the speakers aren’t great just that they aren’t magically efficient. I personally really like them but not with 5 or 10 watt amps. Prefer them with 30wpc + when it comes to most of the music I listen to.

 

Unfortunately people read Stereophile and the problem with Stereophile is that they don't measure speakers properly. You will often see a LOT of loudspeakers, not just An but DeVore and a host of others get a lower rating.

Speakers are sold in pairs - STEREO - and makers typically measure a PAIR of loudspeakers. Stereophile measures 1 loudspeaker in a quasi-anechoic chamber - in other words they measure the speaker outdoors or in the center of a room because they can't afford an anechoic chamber.

So let's say Stereophile measures a speaker at 90dB. That's fine - but whenever you add the second loudspeaker (this applies to ALL makers not) you add 3dB. Then another 3dB if the speakers are placed in corners. Indeed two speakers placed in corners adds a whopping 18dB of gain to the bass. So a speaker that is 40hz free standing will be 22hz in the corner. This is why many subwoofers are placed in corners.

So the AN E in Stereophile is 92.5. Add the second speaker = 95.5dB and the corners and you get another 3dB (minimum) for 98.5dB.

Martin Colloms who, unlike JA, actually has an engineering degree and founded Monitor Audio and writes books on loudspeaker design and was the testing and measuring guy at Stereophile - also measured the AN E and he got Audio Note's numbers.

When Stereophile has a discrepancy they should send an e-mail to the manufacturer and ask them how they arrived at their figures. Martin Colloms Hi-Fi Critic measurements kind of makes Stereophile look like they don't really know what they're doing.

 

These complaints about Stereophile’s measurement methodology are OLD. Whatever efficiency advantage there would be with a pair of speakers in a room would apply to ALL speakers they test, not just Audio Note. The whole point is Stereophile is using a consistent methodology for testing speakers so the measurements for one model can be fairly compared to those of another model. Now you can question the value and usefulness of the test results, but at least they have the benefit of consistency from one speaker to another.

The fact of the matter is the measured efficiency of Audio Note and DeVore speakers does not support the nanufacturer’s claims. However, this in no way contradicts the real-world fact that both companies’ speakers perform very well with low-power tube amps. There is simply more to real-world speaker/amp compatibility than measured sensitivity alone.

 

Agreed! The GL reissue KT-77 sounds terrific. Is it as good as a carefully matched set of NOS (choose your brand) tubes from the 1950s or 1960s? Beats me. Probably not but the reissue still sounds great and matched pairs are available from reliable sources at a fair price.

I can remember buying quads of NOS EL-34s (both Mullard and Sylvania) at the local Lafayette store back in the 1970s. No need for mail order! But it’s been way too many years since I last bought NOS EL-34 tubes to try to compare them to the GL reissue KT-77.

My present amps use 6L6GC outputs and I have collected several NOS sets and compared them to Tung Sol reissues. My favorite is a set of old GE tubes from the 1960s but my second favorite would be the Tung Sol reissue. I prefer the TS over classic RCA blackplates and SED Winged C Based on my limited experience, I believe reissue output tubes have come a long way in the last 10 or 15 years and are now a viable alternative to NOS. That’s a huge benefit because typically output tubes are the ones needing replacement most often. For all other tubes in my system including rectifiers I still stick to NOS.

 

The guys at Deja Vu said they could convert my amp from using EL34's to 6L6 tubes, I declined as I really like the sound of my amp with the KT77's but I was/am curious. From what I understand the differences would be a bit less power and a slightly softer sweeter sound?

 

Less power for sure but I would not expect a 6L6GC to sound softer and sweeter than a GL reissue KT-77. First of all, the GL KT-77 is already soft and sweet in the highs. The 6L6GC tubes I have tried sound more different from each other than just about any other type of tube, so it’s hard to generalize. But if pressed I would say the 6L6GC tends to be more linear across the spectrum; bass is tighter and not as warm sounding as EL-34 or KT-77, and the treble is fast, clean and more extended. That said the classic RCA 6L6GC blackplate is more like a KT-77 than the other 6L6GC tubes I’ve tried.

 

I think you are spot on with this comment.

I very much like the KT77 over the modern production EL34's, it is just a better sounding tube, with nicer bass.

But, I don't think it sounds at all like the NOS EL34's.

 

I said that. It applies to all manufacturers not just Audio Note. Second speaker adds 3dB - so if manufacturer says 90dB and Stereophile says 87dB then it is highly likely the manufacturer's stated spec is for a pair of loudspeakers. When there is a match - the manufacturer likely measures the same way JA of Stereophile does. Manufacturers do not have to conform to Stereophile's lazy one size fits all CHEAP measuring standards. Stereophile didn't invent them. They have developed the tests akin to testing cars on rollers - and that may tell you something - it doesn't tell you how a car handles at 100kph in corners. DeVore claimed 96 and got 91dB in Stereohile - AN Eclaimed 98dB and got 92.5dB in Stereophile. The interesting thing though was that Audio Note took more flack for not meeting spec (though it is still 1.5dB more sensitive than DeVore) while DeVore took no flack in the comments.

I actually don't have a problem with their test so long as it compares the same kinds of speakers. Free Standing speakers are all designed to try and get away from walls - they are trying to take all room boundaries OUT of the equation. All of these speakers want you the listener to sit in some sort of equilateral triangle and preferably in the near field. That is probably 97% of all loudspeakers. And you are right these speakers are being "Fairly" compared as such.

It does not fairly compare omni-directionals, corner loaded speakers, planar speakers due to their radiation patterns are also suspect in Stereophile's one size fits all test.

And my question is this - why are you taking Stereophile's measurement as the sole and ONLY word on the subject? When you have just said they have a testing protocol - one size fits all test. Speakers are not one size fits all.

I am sorry that you can't be bothered to look up what a corner does for speaker measurements - facts exist whether you choose to believe them or not. Audio Note posts their measurements for their intended use - 2 speakers - corner loaded - in room. They meet spec - Hi-Fi Critic and others have measured and confirmed they meet spec.

 

Yes, I am well aware of the effects of corner loading. I was reading books on acoustics and speaker design before you were born. Since you are so upset with Stereophile’s test methodology, you should contact the editors and try to persuade them to change. I will simply point out the purpose of a magazine performing any measurement tests at all is not to validate the manufacturer’s claims but rather to provide the readers with some objective measurement data that can be used to compare one speaker to another. And to be of any use whatsoever those tests must be performed in a uniform, consistent manner, even when the methodology puts a particular design at a disadvantage.

My personal opinion is that speaker measurements are of little benefit when purchasing speakers. Listening quality is the only issue that matters for me. But I recognize that other people do like to study test measurements and the data does produce occasional insights, so the tests are certainly worthwhile.

I have no affiliation with Stereophile but I believe they should be commended, not criticized, for their speaker testing efforts. They have accumulated test data on many speakers, perhaps a hundred or more, stretching over literally decades, and the tests have all been performed in a consistent manner allowing the data to be compared for different models. I don’t see other magazines, much less internet ezines, doing anything comparable.

 

I have no problem with that - if at the end of the review they make a note that the particular speaker - being corner loaded will receive the gains I noted with that photo. Granted many many years before Stereophile reviewed Audio Note I happened to see a lot of heated arguments between JA and Peter Q so I suppose bothering to send an e-mail to ask a question was off the table.

Indeed, I have made notes to JA on his inconsistency - which is the reason DeVore O/96 got upgraded from Class B to Class A. Since Art said the AN E should be class A and since both speakers in room measures just about exactly the same. The AN E is half the price and has a better finish (or at least as good) is more sensitive - you would think it too would get moved to class A - but that's a different issue.

I agree with you on the listening but people put a lot of stock into measurements. And when they don't support reality or the listening experience then it is somewhat pointless.

Again - if they correlated deeply with the end user experience they would hold value - HiFi Critic does measurements as does SoundStage in Canada - their measurements are deeper and equally unrelated to actual real world quality.

I like to see how the speakers are doing in the real world - I tend to look at what sorts of speakers get traded in for what sorts of speakers. Same with amps - people tend to move from SS to tubes to SETs. Progressively worse measurements - speakers like a KEF or Revel will get traded for an Audio Note or DeVore or Harbeth - not generally the other way around. So you see these people buying the gear that gets the best measurements to then sell them for gear with weak measurements. It would be cool if every dealer on the planet would keep track for the next 5 years - took speaker X in and sold them speaker Y. I suspect speaker X (will be Dynaudio, B&W, KEF, Revel, Paradigm, PSB, Monitor Audio and other that have a similar shape and speaker Y will be all the myriad fatty speakers.

 

Yep, that was my evolution too.

 

Thank you for the very detailed reply! I think I am in a pretty sweet spot running my amp with the KT77's so will stay pat. Though if a pushpull 2A3 amp that I can afford comes my way someday...........

 

I don't pay any attention to any graphs on any of the audio gear that I own.

I don't even pay that much attention on how a piece of gear might sound on its own. My concern is how well it will work in my system, together with the other gear that I am using.

While I do agree with this statement. I also feel that measurements were first established to verify any claims that the manufacture makes, with reference to the performance of the gear that they are offering for sale.

Consistency is always important to be make any kind of intelligent comparisons.

At the end of the day, whether something is tested one way or by another method, it may or may not be of any particular importance, since testing of any piece of audio gear does not determine how good it actually will sound.

Moreover, each different piece of gear has a different performance "graph". Now take all of these different graphs and place them together in series and predict how the outcome will sound? Not happening.

 

Stereophile measures using 2.83V/1m, which is not at all the same from using 1 watt/m, which a lot of manufacturers use for their spec. Using 2.83V usuallly translates to 1w/m at 8 ohm, meaning a speaker may see more or less than 1 watt based upon its impedance characteristic.

Go to Zu's website and look at its measurements for Druid V. Druid V is spec'd at a 16 ohm nominal speaker and Sean Casey will tell you it behaves more like a 19 ohm speaker. Zu specs Druid V at 101db, but at 1w/m. Their measurements graph shows this but they also publish the 2.83V measurements which are around 9 db lower because the speaker is seeing less than half a watt. This is what Stereophile measures at, and hey guess what, they also measured Soul Supreme (also 101db/16 ohm) at 91.5 db.

https://static1.squarespace.com/sta...b0f7c27aeb8f8c/1375636573869/Druid-5A_C+S.pdf

The O/96 is a 10 ohm nominal load (7.1 minimum), so it is seeing less than 1 watt, and therefore it is reasonable that Stereophile measured it at less than its spec of 96db. The AN-E is a nominal 6 ohm load (3.9 minimum), and therefore has less of an excuse for measuring lower than spec'd. This is why AN "took more flack".

John DeVore explains it better than I can:

The simple answer is that JA measures speakers using a solid-state amp. The classic solid-state amp will output half the wattage into twice the impedance. The AN/E is 4-6 ohms, while the O/96 is 7.5-12 ohms. JA's measurement amp is putting out nearly twice the amount of wattage into the AN/E as the O/96.
Read more at Manufacturers' Comment ​

 

I still have a pre-Galahad M6 Phono and a pair of Empress Silver amps purchased during my ANNUAL period.’ They are absolutely lovely together and a grand march as the Empress are rather high gain and have been known to ‘trap’ the owner of a modern M3,5,6 and E/xx HE at the bottom of the vol pot’s range. Above 7 o’clock may blow you out of your chair. This’d a problem with the older Ms.

 

This is correct - The AN E is less efficient than the O/96. 4ohms effectively draws twice the power from an 8 ohm rated amplifier

So if the SS amplifier is humming along using 3 watts on the O/96 and the 0/96 hits a frequency where it drops to 7.5 ohms the amplifier will still draw 3 watts.

On the AN E - the same amplifier and same 3 watts is humming along at 8 ohms then hits a bass passage where the speaker drops to 4ohms - the amplifier will be asked to draw "double" the power to meet the impedance - thus it will require twice the power of the o/96 for 6 watts.

Or to put it simpler - an 8 ohm nominal speaker is usually rated 1 watt per 1 meter. When the speaker drops to 4 ohms - it is 2 watts per 1 meter. The problem for the AN E is that it is a 6 ohm speaker - not 4 not 8 - it dips to 4 but it sits mostly above 6. So what to do what to do - if you buy a tube amplifier - and let's face it - Audio Note speaker owners are 99% buying a tube amp - the tube amp has 8 and 4 ohm taps. On the 8 ohm tap that speaker will draw 1,5 watts at 6ohms. In the 4 ohm tap the amplifier will be using 1 watt at 4 ohms and .75 of a watt at 6 ohms. So it will be slightly harder to drive than spec on the 8 ohm tap and slightly easier to drive on the 4ohm tap.

Further a lot of tube amplifiers have weak output transformers so the AN E at louder levels with bass will need an amplifier with superior output transformers than the O/96. Though I would say that owners of these expensive speakers are probably not running $500 tube amps from China.

But this didn't address the corners. Audio Note never said the speakers were the easiest load in the world (although it is still a very easy load just not a supremely easy load) Let's also not get too bogged down in this as even JA said about the AN E " Even so, I don't think the speaker will make significant demands for current on the partnering amplifier, which, in combination with the high sensitivity, makes it very suitable for use with flea-powered single-ended amplifiers."

The AN E is a little more sensitive than the o/96 - the 0/96 is more efficient than the AN E.

Placing the AN E in the corner will not raise their efficiency but it will raise their sensitivity.

It's still really basic math - Stereophile 92.5db - Add corner 6db (3db from side and 3dB from back wall) 92.5dB+6dB = 98.5dB.

Or take AN's figure of 98dB - subtract the side and back wall gain -6dB = 92 dB.

 

I see AN now has a bigger number of steps in their stepped attenuators. My M3 with the older 24 steps is fine with my Empress amps but my speakers are not HE version and I am quite loud at 9 o'clock - what surprised me a bit was that the Phono Stage seems virtually as powerful as the line inputs. On most of my older amp the phono sections was a lot quieter - had to turn the volume up much more to equal CD - that doesn't seem to be the case with the M3 . I find there is enough steps at 24 for me but I understand why people would want to upgrade to the new volume pots with 36 steps or so. I might do the volume pot upgrade but I am having no issue with the Empress Silver or the PureAudio One (as power amp). I am a bit lazy - If I am going to pay to have volume pots changed - I will probably just wait and get an M6 or M6 Two Box Signature.

But with Hong Kong the way it is - I am being a bit more cautious with what I spend.

 

I'm not sure what this word salad means, but it translates to me as "we're still undefeated if you don't count injuries, the refs, bad playcalling, and the scoreboard"...

 

I didn't think this would be so hard to understand but as a teacher - it is often the case that any given topic needs to be taught in a few different ways so that the kids will get the concept.

John DeVore is correct on a very finite number of amplifiers. What John says - The AN E needs twice the power of the o/96 applies in a technical sense but not in a real world sense.

And it doesn't typically apply to good SS amplifiers. I will try and explain this as simply as I can.

Let's say you own the Accuphase A-36 Stereo power amplifier - and you have your preamp (doesn't matter which one) and you put the preamp to exactly the 9 o'clock position. The AN E will be louder than the O/96 by 1.5dB

That is because the Accuphase amplifier is rated 30 watts into 8 ohms and 60 watts into 4 ohms - it makes zero difference if the AN E is less efficient than the O/96 because the amplifier is essentially making up the difference. Any time you see an amplifier where the amp is rated 100 watts @ 8 ohms and 200 watts at 4ohms - the efficiency advantage disappears because 95%+ of all amplifiers sold (that audiophiles will buy) on the audio market are 4 ohm capable and have a higher watt figure into 4 ohms than at 8ohms.

Where the efficiency advantage comes in is with "some" tube amplifiers that only have 8 ohm taps. Some of these amplifiers are not 4 ohm stable.

One of the complaints about SET amplifiers is they are not particularly good with speakers that have large impedance swings or low impedance - low as in below 3ohms. 4 ohms min is still pretty easy to drive.

Art Dudley uses Shindo power amps I believe - i looked up the Shindo Cortese a 300B power amplifier 8 watts presumably 8 ohm tap - there is no 4 ohm tap from what I can see on the back of the amp.

This particular amp will fit with what John DeVore is saying. The amplifier likely isn't able to put out 8 watts at 4 ohms - it may only put out 2-4 watts at 4 ohms. So in this case the AN E becomes TWICE as difficult to drive as the O/96. 1 watt from the Cortese into the DeVore and you get 91dB - but for the AN E to produce it's 92.5dB rating would require 2 watts.

However unlike the Cortese - Audio Note's 300B is the Meishu also rated 8 watts but it has an 8 ohm tap AND a 4 ohm tap. SO you connect the o/96 up to the Meishu - 8 ohm tap and you get 1 w =91dB. You connect the AN E up to the 4 ohm tap - and feed it 1 watt and you get 92.5dB. Thus - again - nullifying the efficiency advantage.

It's kind of fun to talk about the reality is this - both speakers will reproduce DIO/Delerium/Slipknott, Black Sabbath, Motley Crue, or AC/DC to very loud ear damaging levels with ample bass from a quality 8 watt amp.

Both of these speakers are in the top maybe 1% of speakers sold for ease of drive on the audio market. So it's sort of a discussion looking for a problem.

 

This is not what John said at all. He said the AN-E sees twice the power of the O96 from a 2.83V signal from a solid state amp.

This is not true either. Stereophile measured the O96 at 91db at less than a watt. At a full watt it probably measures closer to its 96db spec. Which was the point of JA's measurement comments, DeVore's remarks, and Art's subjective listening comments that the DeVores were more sensitive than the AN-E.

You got that right. If you want to use the fuzzy math of 4 ohm taps, corner placement and a second speaker to get back up to 98db, knock yourself out. It doesn't matter. But neither does your claim that Stereophile somehow moved the goalposts on Audio Note.

 

Well then John is wrong - The AN E is not a 4 ohm speaker - it is a 6 ohm speaker - at SOME very few frequencies the speaker hits 4 ohms. The fact that the O/96 is more efficient has nothing to do with the results of the AN E. "The Audio Note AN-E Lexus Signature was one of the more sensitive dynamic speakers I have measured, at an estimated 92.5dB(B)/2.83V/m."

With the O/96 JA says "conforming to the specified 10 ohms. At the 2.83V level used to calculate the speaker's sensitivity, the O/96 will therefore draw less than 1W from the amplifier, which will have a negative impact on the measurement." Well without an actual measurements - how much less than a watt are we talking 0.97% of a watt or 0.25% of watt. It's pretty silly to hail Stereophile for using consistent measuring techniques when they can't even tell you the exact Sensitivity and have to "guess" or make "assumptions" about what is happening.

The O/96 is an easier to speaker to drive for lower power amplifiers.

QUOTE="Whoopycat, post: 22735045, member: 55314"]
This is not true either. Stereophile measured the O96 at 91db at less than a watt. At a full watt it probably measures closer to its 96db spec. Which was the point of JA's measurement comments, DeVore's remarks, and Art's subjective listening comments that the DeVores were more sensitive than the AN-E. [/QUOTE]

Subjective listening - well the DeVore is a free standing speaker - it is physically many feet closer to the listening chair than the AN E which is in a corner. Gee I wonder which might sound louder on an amplifier incapable of driving a 4 ohm dip in response?

I'm sorry if some posters can't understand what a 4 ohm tap on an amplifier is or how a room affects a loudspeaker.

Audio Note measures their AN-E loudspeakers with speakers placed in corners - their figure is correct. As was proved by Martin Colloms in Hi-fi Critic.

Stereophile's figure is also correct for the same speaker placed free standing in a room. Stereophile actually confirms the number as well whether they know it or not.

If you place the speakers free standing in a room you will get 92.5dB sensitivity - if you place the speaker in the corners you will get 98.5dB sensitivity.

There isn't anything fuzzy except in the minds of people who can't understand room corner gain. Hell I even showed you a nice little picture because maybe the words are too big for you.

 

There are so many technical misstatements in Richard’s postings that one hardly knows where to start. I decided a detailed response was not a worthwhile use of my time, so I will just leave it at that.

 

I looked up this topic on Audioasylum from 2006 shortly after Stereophile published the measurements. A competing speaker designer who now specializes in subwoofers (you know - bass) took Peter Qvortrup (AN) to task for the high sensitivity spec and bass spec (both of which were confirmed by Martin Colloms). Also in the discussion is competing amplifier designer and manufacturer and regular AA poster Donald North who steps in to correct the math. What is impressive to me is that people are still arguing this 13 years later. Brilliant marketing really - folks missing the forest to argue about the trees. That was convoluted. Now for an encore, tell us how you arrive at your -6 dB specification. - Duke - High Efficiency Speaker Asylum

Personally, none of it really matters - no one is exactly buying Single Ended tube amplifiers and wide baffle speakers with lossy cabinets (DeVore or AN or even Harbeth etc) for the "ultimate" in measured performance.