[FullRangeMan]

So, let's hear some recommendations for speakers that support low wattage playback.

Cames to mind Wright Lolita and Decware SEEL34 both 6W.
Also nice a pair of Decware EL84 as mono=6W.

[Duke]

Are there now any low-colored, high efficiency (say 95 dB/w/m), full range (F3 of at least 30 Hz) loudspeakers?

The internal box volume required to deliver an F3 of 30 Hz at an honest 95 dB/w/m is ballpark 8 cubic feet unless we take advantage of boundary reinforcement by placing the speakers up against a wall, in which case we can probably cut that in half.  But the sort of placement required to really take advantage of boundary reinforcement takes its toll on soundstage depth. 

So if we want good soundstage depth, 95 dB, and 30 Hz low end, but we don't want a huge box, handing off the bottom couple of octaves to subwoofage might make sense.   Maybe several subs... gotta face the elephant, you know. 

Imo what makes low wattage systems so special is that the amps have little or no global negative feedback, and the speakers have negligible thermal compression so their "voicing" doesn't changed with level.   For those who doubt the reality of thermal effects in the real world, Floyd Toole posted this on another forum: 

"in most home systems... most of the audible effects are from voice coil heating, which has a much shorter time constant [than magnet heating]. I just saw a test of a high-end audiophile speaker that in going from an average level of 70 dB (loud conversation, background music) to 90 dB (a moderate crescendo, or foreground rock listening) lost about 4 dB in output over about 3 octaves in the mid-high-frequency range. It became a different loudspeaker at different listening levels."

[JLM]

Thanks for the reality check Duke.  So would you happen to be cooking up for something along this line (F3 = 30 Hz, 95 dB/w/m, uncolored)? 

A couple of years ago you offered a 99 dB/w/m, 8 ohms nominal, 20 - 20,000 Hz, $6,000.00 system consisting of Planetarium Delta 15 mains (15 inch woofer, horn tweeter, rear firing "ambience" tweeter in a large stand-mount format), and 4 "swarm" subs with amp.  Very impressive (on several counts).  At that time you also mentioned kits (without the "ambience" tweeter but with options for larger cabinet to drop response without the "swarm" down to 30 Hz.  And recently you hinted at a newer version coming to 2015 RMAF.  (Sorry I missed you at 2015 Axpona, we did stop in but didn't see you there, had a buddy with me, it was quite busy, and lots of ground to cover in a day.)

The Planetarium Delta 15 system seems a bit daunting for the my mellow (aging) taste, 8 ft x 13 ft x 21 ft study, and worries about system/tube noises at higher efficiencies.  I'd have to reconfigure my study from a listening/office space to a dedicated listening room to accommodate the full Planetarium Delta 15 system and I'm not prepared to do that as long as I'm still doing part-time consulting.  While I've read Toole and appreciate your ideas including the "swarm" do you have anything in mind that would less impactful to ears, room, system, and wallet while reaching my goals?

[Quiet Earth]

Imo what makes low wattage systems so special is that the amps have little or no global negative feedback, and the speakers have negligible thermal compression so their "voicing" doesn't changed with level.

Yay! Somebody else answered the original question. (Good answer too.)


Freo-1 and JLM,

Do you guys really want to create a list of speakers for low watt amplifiers in this thread? That seems like it would be a different thread altogether. Why not start a new one about speakers or add to this one http://www.audiocircle.com/index.php?topic=105328.0
Or maybe you are saying that it is the speaker which makes low watt systems special. Ok, I can respect that opinion even though I do not agree with it.

JLM,

You keep using the word uncolored. How can we possibly agree on what is colored and what is not colored? What you consider uncolored is what someone else may consider difficult to listen to, or vis versa. And is there really such thing as an uncolored speaker? Who makes that determination? What about musical enjoyment? How do we get a consensus on that?
Also, I think you are looking for a speaker that does not exist. Just a guess.

[Duke]

Thanks for the reality check Duke.  So would you happen to be cooking up for something along this line (F3 = 30 Hz, 95 dB/w/m, uncolored)? 

What I plan to show at RMAF is in that ballpark, but with the Swarm for the bottom end, so more like 20 Hz.

A couple of years ago you offered a 99 dB/w/m, 8 ohms nominal, 20 - 20,000 Hz, $6,000.00 system consisting of Planetarium Delta 15 mains (15 inch woofer, horn tweeter, rear firing "ambience" tweeter in a large stand-mount format), and 4 "swarm" subs with amp....

The Planetarium Delta 15 system seems a bit daunting for the my mellow (aging) taste, 8 ft x 13 ft x 21 ft study, and worries about system/tube noises at higher efficiencies.  I'd have to reconfigure my study from a listening/office space to a dedicated listening room to accommodate the full Planetarium Delta 15 system and I'm not prepared to do that as long as I'm still doing part-time consulting.

The Planetarium Delta system didn't turn as many heads as I'd hoped, and I agree with your concerns about noise in uber-efficient systems.   That's one of the reasons why I've usually aimed for the lower 90's, which also makes the bass extension/box size tradeoff more manageable.  Given that I'm a fan of Atma-Sphere OTL amps, their little 30-watt model is plenty with that sort of efficiency. 

At that time you also mentioned kits (without the "ambience" tweeter but with options for larger cabinet to drop response without the "swarm" down to 30 Hz.  And recently you hinted at a newer version coming to 2015 RMAF.  (Sorry I missed you at 2015 Axpona, we did stop in but didn't see you there, had a buddy with me, it was quite busy, and lots of ground to cover in a day.)

Sorry I missed you too!  Will you be at RMAF by any chance?

The kits will not happen unless I totally change my business model.  A long conversation with a friend who has an MBA convinced me that realistically I can either be a kit manufacturer, or a finished speaker manufacturer, but not both. 

While I've read Toole and appreciate your ideas including the "swarm" do you have anything in mind that would less impactful to ears, room, system, and wallet while reaching my goals?

I'd rather have that conversation via messages rather than in this thread, so you'll be getting a message from me shortly.

[JLM]

Sorry for getting off topic (but it takes speakers to make it a system, yes?).

Duke does offer viable candidates, the Prisma in particular has F3= 30 Hz, 93 dB/w/m, 8 ohms).  Sonist Concerto 3 and Concerto 4 also qualify.  Of course many of the Tekton and Zu towers qualify.  But i'm still looking for more "reasonably priced" examples.


QE

Thanks for the link to the older thread, I'd forgotten about it.

Yes, vocabulary and sound don't always translate well.  Some of the negative colorations are stereotypical and need little qualification: bloated bass; etched highs, thin or forward midrange, tonal imbalance high/low, warmth.  Such colorations are deal breakers for me, so would helpful to know if candidates were "colored".

[Freo-1]

Yay! Somebody else answered the original question. (Good answer too.)


Freo-1 and JLM,

Do you guys really want to create a list of speakers for low watt amplifiers in this thread? That seems like it would be a different thread altogether. Why not start a new one about speakers or add to this one http://www.audiocircle.com/index.php?topic=105328.0
Or maybe you are saying that it is the speaker which makes low watt systems special. Ok, I can respect that opinion even though I do not agree with it.

JLM,

You keep using the word uncolored. How can we possibly agree on what is colored and what is not colored? What you consider uncolored is what someone else may consider difficult to listen to, or vis versa. And is there really such thing as an uncolored speaker? Who makes that determination? What about musical enjoyment? How do we get a consensus on that?
Also, I think you are looking for a speaker that does not exist. Just a guess.

Good point on using the other thread for a list of recommended speakers for low wattage applications.  We can certainly use the other thread to come with a list of recommended speakers that are low wattage friendly.  Could even become a sticky if desired.


Also, the points about speaker frequency response is interesting.  Although I'm keen to put together a low wattage setup, my present speakers are anything but low wattage friendly.  The ATC SCM-19's, at 85 db/w, are pretty linear, and can handle dynamics fairly well.  It's pretty hard to find low wattage friendly speakers that can provide the linear performance of speakers like the ATC's

I do think the approach Duke outlines (using subs) with low wattage friendly speakers makes a lot of sense.

[Duke]

One thing I forgot to mention is this:  The effect of a low-damping-factor amplifier's output impedance on the woofer's effective electrical Q (Qes) can, in some cases, be exploited as a sort of "free lunch" to get more low-end output than we otherwise would have.  If we can tailor the box tuning accordingly, we can get as much as 1/3 octave deeper bass extension.  However the same amp-speaker interaction mechanism can result in fat or bloated bass when a speaker designed for solid state is driven by a low-damping-factor tube amp (the devil is in the details).  Typically the amp gets blamed, but it's really a system mis-match issue; the same speaker or same amp would work just fine with a more appropriate partner.

[JLM]

One thing I forgot to mention is this:  The effect of a low-damping-factor amplifier's output impedance on the woofer's effective electrical Q (Qes) can, in some cases, be exploited as a sort of "free lunch" to get more low-end output than we otherwise would have.  If we can tailor the box tuning accordingly, we can get as much as 1/3 octave deeper bass extension.  However the same amp-speaker interaction mechanism can result in fat or bloated bass when a speaker designed for solid state is driven by a low-damping-factor tube amp (the devil is in the details).  Typically the amp gets blamed, but it's really a system mis-match issue; the same speaker or same amp would work just fine with a more appropriate partner.

Good point.  I always blame the newest component.   

[G Georgopoulos]

One thing I forgot to mention is this:  The effect of a low-damping-factor amplifier's output impedance on the woofer's effective electrical Q (Qes) can, in some cases, be exploited as a sort of "free lunch" to get more low-end output than we otherwise would have.  If we can tailor the box tuning accordingly, we can get as much as 1/3 octave deeper bass extension.  However the same amp-speaker interaction mechanism can result in fat or bloated bass when a speaker designed for solid state is driven by a low-damping-factor tube amp (the devil is in the details).  Typically the amp gets blamed, but it's really a system mis-match issue; the same speaker or same amp would work just fine with a more appropriate partner.

Hey Duke,what are you talking about,system match comes from good components say a good amplifier and a good speaker,there is nothing electrically mismatch there, Qes is for speaker design specs,the amp has to make sure the audio signal is right,first time i have heard an amp design has to take Qes into consideration in the design of an amp,Duke a speaker is a load to an amp!!!remember that!!! if a speaker sounds good has nothing to do other than the speaker itself first and foremost.

I dont agree with you on this 

[Duke]

Hey Duke,what are you talking about,system match comes from good components say a good amplifier and a good speaker,there is nothing electrically mismatch there, Qes is for speaker design specs,the amp has to make sure the audio signal is right,first time i have heard an amp design has to take Qes into consideration in the design of an amp,Duke a speaker is a load to an amp!!!remember that!!! if a speaker sounds good has nothing to do other than the speaker itself first and foremost.

I dont agree with you on this 

You are on the right track to observe that the speaker is simply the load for the amp, but the missing piece of the puzzle is, not all amps behave the same.

Most solid state amps approximate a "voltage source" - that is, they tend to deliver constant voltage into the speaker's impedance curve. 

Most tube amps approximate a "power source' - that is, they tend to deliver constant wattage into the speaker's impedance curve.  (In the real world tube amps don't do this into extreme loads, just as solid state amps don't deliver constant voltage into extreme loads.)

It is possible to design a solid state amp to behave like a tube amp, and vice versa, but we're going to ignore those exceptions for now.

So suppose we have an 8-ohm speaker whose impedance curve has a 16-ohm peak at the 50 Hz bass system resonance, a broad 6-ohm dip between 150 and 400 Hz, and a 24-ohm peak at the 3 kHz crossover frequency (a 4-to-1 spread in the impedance curve is about normal).   Let's assume the amp (either one) is putting out 1 watt (2.83 volts) into our speaker's nominal 8-ohm load.

Into this speaker's impedance curve, our solid state amp puts put 1/2 watt at 50 Hz, 1.33 watts from 150 to 400 Hz, and 1/3 watt into 3 kHz crossover region.  But it's okay - the speaker was designed with that kind of amplifier behavior in mind.

Now let's drive this speaker with a tube amp that puts out 1 watt across the board into whatever impedance the amp presents.  Compared with the solid state amp, when driven by a tube amp this speaker is 3 dB louder at 50 Hz; about 1 dB softer from 150 Hz to 400 Hz; and 5 dB louder at 3 kHz.   We are very likely to conclude that tube amps are terrible because now our speaker is boomy in the bass, weak in the lower midrange, and screams like a banshee in the lower treble region, right where our ears are most sensitive! 

On the other hand, if this speaker had been optimized for tube amps, we'd conclude that the solid state amp was terrible. 

The way to design a speaker to work well with both amp types is, keep the impedance curve as smooth as possible and, if it's a ported box, allow for user-adjustable port tuning. 

While my example made some simplifying assumptions, the basic concept that it illustrates is quite valid.   I've measured the frequency responses of speakers in the design stage and they follow the pattern described above according to what type of amp is used.  The real-world differences are usually not as severe as in my example but are still enough to very significantly change how the speaker sounds.

Now you don't design an amp for the speaker's Qes, but if we know in advance that a speaker will be used with an amp that has a low damping factor, we can calculate how much that changes the effective electrical Q of the woofer (which showed up in the example above in the 3 dB bump at the 50 Hz system resonance), and design the speaker enclosure accordingly. 

Suppose our zero-global-feedback SET amp has an output impedance of 4 ohms, giving it a damping factor of 2 (referenced to an 8-ohm load).  And suppose our "8 ohm" woofer has a Qes of .30, and a DC resistance of 6 ohms.   The new effective Qes of the woofer, when driven by that SET amp, is ((4 + 6)/6) x .30 = .50, which is quite a bit higher, and would call for a different tuning strategy for best results.  However when the dust has settled, we might well end up with usefully deeper bass extension. 

The reason people go to the trouble to find speakers that work well with these strange low-powered tube amps is, their distortion profiles are much more benign to the human ear as long as they aren't pushed into hard clipping, even though on paper their distortion measurements look a whole lot worse.   Briefly the metric used to measure amplifier distortion, THD, is a dismal failure at predicting subjective preference, but that's another subject for another day. 

[viggen]

less is more?  not just in terms of watts but the electronics in low wattage systems are generally made with fewer parts hence less ways the electronics get in the way of the music.

[G Georgopoulos]

You are on the right track to observe that the speaker is simply the load for the amp, but the missing piece of the puzzle is, not all amps behave the same.

Most solid state amps approximate a "voltage source" - that is, they tend to deliver constant voltage into the speaker's impedance curve. 

Most tube amps approximate a "power source' - that is, they tend to deliver constant wattage into the speaker's impedance curve.

It is possible to design a solid state amp to behave like a tube amp, and vice versa, but we're going to ignore those exceptions for now.

So suppose we have an 8-ohm speaker whose impedance curve has a 16-ohm peak at the 50 Hz bass system resonance, a broad 6-ohm dip between 150 and 400 Hz, and a 24-ohm peak at the 3 kHz crossover frequency (a 4-to-1 spread in the impedance curve is about normal).   Let's assume the amp (either one) is putting out 1 watt (2.83 volts) into our speaker's nominal 8-ohm load.

Into this speaker's impedance curve, our solid state amp puts put 1/2 watt at 50 Hz, 1.33 watts from 150 to 400 Hz, and 1/3 watt into 3 kHz crossover region.  But it's okay - the speaker was designed with that kind of amplifier behavior in mind.

Now let's drive this speaker with a tube amp that puts out 1 watt across the board into whatever impedance the amp presents.  Compared with the solid state amp, when driven by a tube amp this speaker is 3 dB louder at 50 Hz; about 1 dB softer from 150 Hz to 400 Hz; and 5 dB louder at 3 kHz.   We are very likely to conclude that tube amps are terrible because now our speaker is boomy in the bass, weak in the lower midrange, and screams like a banshee in the lower treble region, right where our ears are most sensitive! 

On the other hand, if this speaker had been optimized for tube amps, we'd conclude that the solid state amp was terrible. 

The way to design a speaker to work well with both amp types is, keep the impedance curve as smooth as possible and, if it's a ported box, allow for user-adjustable port tuning. 

While my example made some simplifying assumptions, the basic concept that it illustrates is quite valid.   I've measured the frequency responses of speakers in the design stage and they follow the pattern described above according to what type of amp is used.  The real-world differences are usually not as severe as in my example but are still enough to very significantly change how the speaker sounds.

Now you don't design an amp for the speaker's Qes, but if we know in advance that a speaker will be used with an amp that has a low damping factor, we can calculate how much that changes the effective electrical Q of the woofer (which showed up in the example above in the 3 dB bump at the 50 Hz system resonance), and design the speaker enclosure accordingly. 

Suppose our zero-global-feedback SET amp has an output impedance of 4 ohms, giving it a damping factor of 2 (referenced to an 8-ohm load).  And suppose our "8 ohm" woofer has a Qes of .30, and a DC resistance of 6 ohms.   The new effective Qes of the woofer, when driven by that SET amp, is ((4 + 6)/6) x .30 = .50, which is quite a bit higher, and would call for a different tuning strategy for best results.  However when the dust has settled, we might well end up with usefully deeper bass extension. 

The reason people go to the trouble to find speakers that work well with these strange low-powered tube amps is, their distortion profiles are much more benign to the human ear as long as they aren't pushed into hard clipping, even though on paper their distortion measurements look a whole lot worse.   Briefly the metric used to measure amplifier distortion, THD, is a dismal failure at predicting subjective preference, but that's another subject for another day.

When you say ss put voltage and tube puts power,I dont agree with you,because both supply power to the load,Duke dont take it personally,you're very good at speakers,but this i still dont agree with you on this..

[Duke]

When you say ss put voltage and tube puts power,I dont agree with you,because both supply power to the load,Duke dont take it personally,you're very good at speakers,but this i still dont agree with you on this..

No, that's not what I said.  Here it is again:

"Most solid state amps approximate a "voltage source" - that is, they tend to deliver constant voltage into the speaker's impedance curve. 

"Most tube amps approximate a "power source' - that is, they tend to deliver constant wattage into the speaker's impedance curve."

Let me put it another way:

The power (wattage) that a solid state amp puts out tends to vary inversely with the speaker's impedance curve.  The power (wattage) that at tube amp puts out does not.   Therefore, the net frequency response from the speakers, resulting from the interaction between the amplifier and the speaker's impedance curve, will be different for these different types of amps.   My post above explains that difference in detail. 

George, since you disagree with what I've said, are you taking the position that tube amps and solid state amps behave the same into whatever impedance curve the speaker presents? 

Do you think the source impedance of the amplifier matters at all, and if so, what difference does it make?   

*  *  *  *

In my opinion, part of designing any audio component is taking into account how it will interact with the components that it interfaces with.  Speakers interface with three types of components:  The amplifier, the room, and the listener's ear/brain system.  All of these interfaces should be taken into account.   

[G Georgopoulos]

George, since you disagree with what I've said, are you taking the position that tube amps and solid state amps behave identically into whatever impedance curve the speaker presents?

Yes, if the zout of both is =<10th the speaker load 

[Duke]

Yes, if the zout of both is =<10th the speaker load 

I think I now see where our opinions diverge. 

The Zout of many specialty tube amps is > 1/10th the speaker load.  The examples in my long post above are in the ballpark in that regard, in particular the 4 ohm Zout described in the second to last paragraph. 

[G Georgopoulos]

I think I now see where our opinions diverge. 

The Zout of many specialty tube amps is > 1/10th the speaker load.  The examples in my long post above are in the ballpark in that regard, in particular the 4 ohm Zout described in the second to last paragraph.

aha,those specialty tube amps that behave as current sources/high impendance sources,yes they can make a good speaker sound bad,you can't match synergy of a good speaker,because amp impendance and speaker impendance would vary widely,with negative feedback you're closer to good sound minus distortion and low zout and better sound,it's not the distortion but tonal imbalance that would result from such non feedback amplifiers interacting with the speaker,a good speaker may sound very bad indeed.

[Quiet Earth]

It is interesting to me that the people with the strongest convictions about speakers do not own a low wattage amp.

[undertow]

What is the "Low Watt" threshold in place here for this question? Are we talking 2 watts, or 50 watts?

In any case the real magic does seem to come in the sweet spot of 98 db plus speakers using lower wattage tube or class A amps from my experience.

The real advantage I found in this configuration is something of a by product or side effect not brought up here yet... GAIN

You can lower the noise, and gain in a system substantially using efficient speakers with a low gain amp input more so than just adding the low power tube watts. Most tube amps create the most noise from the input tubes, much more noise comes from this than a high gain preamp has an issue with higher gain tubes. The other nice thing about tube amps and preamps to a degree is the choice to change tubes with slightly different gain types. Solid State your more or less locked in to one gain structure.

For example a 20 db gain preamp with tubes is virtually silent when done right same as a 10 db gain preamp, but if you hook up a TUBE amp with 30 db input gain on a 100 db speaker you will normally get some extra noise. And the fact you won't have much volume variation as its either loud, or quiet, hard to control getting smaller volume steps.

This again only works with high efficiency speakers. I think there is less difference between a 2 watt or a 20 watt output tube, its more about the total gain many times in the system that makes the difference in the case of low watt high efficiency systems, and it all depends on the overall design of the amp.

By the way I have found standard efficiency speakers of lets say 88 to 94 db perform well with 40 to 50 db of gain combined between the preamp, and amp. So if you have a 26 db amp, and a 20 db preamp your right about in the wheelhouse.

But if your running 100 db horns you want about 30 to 35 db total gain in your combination, so something like a 20 db amp, and 15 db preamp works excellent. Passive preamps are a whole other story as well.

This is of course assuming your using digital or analog sources something in the standard range 1.5 v to 2.0 v range.

[FullRangeMan]

What is the "Low Watt" threshold in place here for this question? Are we talking 2 watts, or 50 watts?

Gizmo used to say under 10W there is higher sound quality in tube amps due lower parts count, connections, lower V tension, transformers are smaller etc not to mention lower price.
I see this always looks correct so far.