KeithR's "Dream Speaker" Search

[KeithR]

Tim, that's a really interesting question. I'm going to stir on that today and post a response later. I've heard some really interesting speakers:

Gamut RS3i and RS5i
Wilson Audio Sasha DAW
Rockport Altair II
YG Hailey 2 and Sonja 2
Avantgarde Duo XD and Duo Mezzo
Stenheim Alumine 5
Tune Audio Anime
DeVore Fidelity Orangutan Reference (ha!)

@tima - following up to your question after some thought.

I prefer YG speakers over the other less efficient types and Avantgardes on the horn front. I understand that may sound a little weird since they are such different designs, but both over-achieve compared with the competition.

I think Stenheim may get into that tier as well in the next few years - exciting, young company - but I still prefer the YG.

At the more reasonably-priced level, I still feel DeVore is fantastic. The new Zu Definition coming early next year may sway that belief however.

 

[KeithR]

They look like your Devores!

It really is a design extension of Devore. Sealed hardwood cabinet with chambers ala Devore inside, but using custom drivers (widebander up top, all wood midrange), minimal crossovers, and active bass w/ dsp. Also consistently demo'd with my favorite music genres. The W13 seems tube-friendly enough at 89dbs/6ohms with active bass, but not sure.

The off-putting thing about the Boenicke line is use of hifi "doodads" like Stein, Bybee, etc.

Yes, I realize it doesn't fit your design preferences, but Brankos with dual 15" refrigerator cabinets aren't happening

 

[jeff1225]

It really is a design extension of Devore. Sealed hardwood cabinet with chambers ala Devore inside, but using custom drivers (widebander up top, all wood midrange), minimal crossovers, and active bass w/ dsp. Also consistently demo'd with my favorite music genres. The W13 seems tube-friendly enough at 89dbs/6ohms with active bass, but not sure.

The off-putting thing about the Boenicke line is use of hifi "doodads" like Stein, Bybee, etc.

Yes, I realize it doesn't your design preferences, but Brankos with dual 15" refrigerator cabinets aren't happening

What's wrong with refrigerator cabinets with dual 15" speakers!

 

[christensenleif@msn.com]

What's wrong with refrigerator cabinets with dual 15" speakers!

maybe that they´re not hornloaded

 

[morricab]

Really? I think the speakers Ron mentioned as well as YG Sonja and AlsyVox panels "do it all" pretty well...

Some larger 4+ way horns as well... Sounds like the General's pnoe can too, but if so it's a unicorn.

The Alsyvox do most quite well ...maybe not large scale music as good as a huge horn or box but they will do it with very high resolution. Small stuff works very well too.

 

[morricab]

he's not going to be happy with my 41" tall Boenickes

funny thing is most say they scaled huge at Munich.

W13se+? Big sounding moderate sized speaker...not the last word in neutrality but definitely resolved and great tone.

 

[morricab]

I can understand that. All of the big cone speakers I cited have two 15”ers per side or the surface area equivalent thereof.

I have come to agree, and this is one of Keith’s philosophies, that big dynamic driver systems which also are highly sensitive (e.g., EA MM7, Goebel Divin Majestic) get you closer to the jump factor and fast rise times and dynamics of horns.

Well, if Keith got Boenickes then the big cones, high sensitivity philosophy just flew out the window . That said W13s have two large 13 inch car audio subs actively driven that can do big bass. Heard them several times at Sven’s place.

 

[morricab]

Wow - when I saw those I said Joachim Gerhard lives!

Very different sounding from Joachim’s work.

 

[morricab]

Another reason to not care about efficiency is prioritizing "pistonic" action of the driver, meaning the driver cone doesn't bend in-use. This is probably why YG speakers need big amps.

There are a lot of speaker designers who feel vintage woofers and paper cone drivers sound bad vs alternatives that may have much stiffer cones but lower efficiency. You can get some efficiency back with big and expensive motor design, but you're not getting 100 dB woofers with "pistonic" cones, just not happening! There is another way to have light coned woofers and reduce distortion though, you simply limit excursion by using more drivers, which also increases efficiency, but now you're looking at a larger speaker.

IMO speaker design without having an amplifier in mind makes no sense. For example, bass with high sensitivity mid and highs... you need to either design the bass section to be extremely efficient so the speaker can be powered by the same amp as the mids/highs or design the speaker for bi-amping. Yes, there are examples of speaker designers padding down the mids and highs over 20 dB (!!!) to meet the woofers but I think this is a bad idea!

In my speaker I chose bi-amping with a woofer that has a very stiff paper cone and a massive motor with a lot of excursion. It ends up being 94 dB, and it has very, very low distortion. For example my Pioneer S-1EX have the TAD 7" carbon fiber woofers, 2 per side just like the TAD Evolution tower speakers, but they sound quite warm and distorted vs my 15" woofer. I also chose a higher excursion model as a compromise because I only want to use one woofer. It requires a 5 cuft box to achieve it's low frequency extension and that's as big a most folks are going to want their speaker to be. Add a 2nd woofer and a horn? Now it's massive and nobody is going to buy it (relatively).

Sooo... imo if you want a very high sensitivity woofer section with low distortion and low frequency extension... if you want all that it's simply going to be large and expensive. No way around it. But today's woofers are AMAZING. At least some of them. IMO you can find a good compromise in a modern woofer driven with a massive class-D amp. It gets a bad rap because of all the poor designs, but it's not the design, it's the implementation. It definitely CAN work, and it can work extremely well.

Our Dynamikks Athos 10 has the option of being half active...then the coax can be run wide open at 97db/watt. The 10 inch pro woofer has a HUGE voice coil, hardly moves even with heavy bass tracks and growls down low like I have rarely heard even in big 15 inches. Fully passive it is around 92db...

 

[morricab]

Here is the convention that I follow (not saying this is necessarily the one and only valid convention):

The term efficiency refers to how loud the speaker is at one meter for a given wattage input, typically one watt. So if a speaker is 92 dB efficient, it will do 92 dB with a 1 watt input. (This is NOT the most precise use of the term "efficiency", but I do think it is the most useful.)

The term sensitivity refers to how loud the speaker is for a given voltage input, typically 2.83 volts. So using our 92 dB efficient speaker in the previous example, IF it is a 4 ohm speaker, then its sensitivity is 95 dB because 2.83 volts into 4 ohms is 2 watts. If it is an 8 ohm speaker, then its sensitivity is 92 dB because 2.83 volts into 8 ohms is 1 watt (which is why we normally use 2.83 volts). If it is a 16 ohm speaker, then its sensitivity is 89 dB because 2.83 volts into 16 ohms is 1/2 watt.

The following gets a bit more geeky.

A more precise way to express the term "efficiency" would be as a percentage... energy (watts) input vs energy (watts) output. A 1% efficiency corresponds to 92 dB, and a 10% efficiency corresponds to 102 dB, and a 100% efficiency corresponds to 112 dB. But, even this is not as precise as it might be. What is missing is, the radiation pattern.

Imagine a garden hose with an adjustable nozzle. The same amount of water comes out whether the pattern is wide or narrow, but the more narrow the pattern, the greater the on-axis pressure. Same thing with loudspeakers.

A horn is one way of trading pattern width for increased on-axis sound pressure level. Making some simplifying assumptions, if we have a 92 dB efficient cone with a 90 degree radiation pattern, and we mount a horn in front of it which narrows the pattern to about 30 degrees, we will have reduced the pattern's area down to about 1/9th of what it was, and correspondingly the on-axis sound pressure level will have increased to about 101 dB. (This tradeoff relationship between pattern width and on-axis SPL explains how we can attain on-axis efficiencies up around 112 dB from drivers whose actual electrical efficiencies are well below 100%).

IMO, your first part is kind of confusing to the layman and they should stick to sensitivity. Your second part is more technically correct where a 92dB sensitivity = 1% efficiency, which the rate of conversion from electrical energy to acoustic energy....the rest is wasted as heat. It is the low efficiency that has real issues for dynamics compression as heat has the knock on effect of impacting sensitivity. I think this is probably one of the main culprits affecting perception of dynamics.

 

[microstrip]

I have come to the conclusion that there is nothing to be gained innately from low sensitivity speakers — unless there is something about the sound resulting from the low sensitivity design which more than compensates for the low sensitivity (which I think does not promote “jump factor” or fast rise time).

Yes, but in order to increase sensitivity you can loose in some other aspects, such as directivity or sound coloration.

IMHO the optimal sensitivity is 93.5 dB/W/1m ...

 

[microstrip]

Here is the convention that I follow (not saying this is necessarily the one and only valid convention):

The term efficiency refers to how loud the speaker is at one meter for a given wattage input, typically one watt. So if a speaker is 92 dB efficient, it will do 92 dB with a 1 watt input. (This is NOT the most precise use of the term "efficiency", but I do think it is the most useful.)

The term sensitivity refers to how loud the speaker is for a given voltage input, typically 2.83 volts. So using our 92 dB efficient speaker in the previous example, IF it is a 4 ohm speaker, then its sensitivity is 95 dB because 2.83 volts into 4 ohms is 2 watts. If it is an 8 ohm speaker, then its sensitivity is 92 dB because 2.83 volts into 8 ohms is 1 watt (which is why we normally use 2.83 volts). If it is a 16 ohm speaker, then its sensitivity is 89 dB because 2.83 volts into 16 ohms is 1/2 watt.

The following gets a bit more geeky.

A more precise way to express the term "efficiency" would be as a percentage... energy (watts) input vs energy (watts) output. A 1% efficiency corresponds to 92 dB, and a 10% efficiency corresponds to 102 dB, and a 100% efficiency corresponds to 112 dB. But, even this is not as precise as it might be. What is missing is, the radiation pattern.

Imagine a garden hose with an adjustable nozzle. The same amount of water comes out whether the pattern is wide or narrow, but the more narrow the pattern, the greater the on-axis pressure. Same thing with loudspeakers.

A horn is one way of trading pattern width for increased on-axis sound pressure level. Making some simplifying assumptions, if we have a 92 dB efficient cone with a 90 degree radiation pattern, and we mount a horn in front of it which narrows the pattern to about 30 degrees, we will have reduced the pattern's area down to about 1/9th of what it was, and correspondingly the on-axis sound pressure level will have increased to about 101 dB. (This tradeoff relationship between pattern width and on-axis SPL explains how we can attain on-axis efficiencies up around 112 dB from drivers whose actual electrical efficiencies are well below 100%).

As you say we are addressing conventions. Sensitivity is the usually quoted figure because it is easy to measure and it the traditional measurement. Voltage is easier to manipulate and measure than current!

Although efficiency tells us more about a speaker intrinsic physical properties, for most consumers the sensitivity is the more interesting parameter, although it should be combined with the impedance and phase curve or the equivalent peak dissipation resistance (EPDR) vs frequency information.

Amplifier power is commonly rated at 8 ohms, considering that power is limited just by voltage. It is acceptable for most solid state designs, but can be misleading with tube designs. Also people should consider that we do not have exact measured data for most speakers and word of mouth sensitivities and impedance's are often incorrect.

All in all, we often risk having big surprises in speaker/amplifier matching ...

 

[KeithR]

Very different sounding from Joachim’s work.

how so? the Virgo was one of my first speaker experiences - almost bought the 3 but couldn't swing it at the time.

 

[morricab]

how so? the Virgo was one of my first speaker experiences - almost bought the 3 but couldn't swing it at the time.

All of the Audio Physic speakers and later Sonics speakers from Joachim sound much more buttoned down and less dynamic (still wondering how Sven gets a good dynamic presentation from low sensitivity speakers) and not nearly as expansive sounding or tonally interesting...I am not a big fan of Joachim's designs and failed to understand how he made a name for himself.

 

[morricab]

how so? the Virgo was one of my first speaker experiences - almost bought the 3 but couldn't swing it at the time.

So did you get Boenickes or not? I would have suggested them for you to have a listen but never thought that they might be available in California.

 

[PeterA]

All of the Audio Physic speakers and later Sonics speakers from Joachim sound much more buttoned down and less dynamic (still wondering how Sven gets a good dynamic presentation from low sensitivity speakers) and not nearly as expansive sounding or tonally interesting...I am not a big fan of Joachim's designs and failed to understand how he made a name for himself.

Morricab, are you suggesting that low sensitivity speakers are incapable of delivering a good dynamic presentation? Or that it is just more difficult to do so?

 

[morricab]

Morricab, are you suggesting that low sensitivity speakers are incapable of delivering a good dynamic presentation? Or that it is just more difficult to do so?

That it is VERY difficult to do so...it's physics...and some psychoacoustics as well...

 

[asiufy]

I agree that it's difficult. But doable. It just requires more power than some people, particularly in this forum, are willing to have in their amplifiers.

 

[PeterA]

That it is VERY difficult to do so...it's physics...and some psychoacoustics as well...

Could you then define what you mean by "low sensitivity speakers"? I have heard systems that I would describe as very dynamic which have 90 dB speakers and pretty big amplifiers. Perhaps my idea of dynamics is different from yours. How does one measure whether or not a speaker is dynamic?

 

[morricab]

I agree that it's difficult. But doable. It just requires more power than some people, particularly in this forum, are willing to have in their amplifiers.

Power corrupts ...and doesn’t overcome thermal compression