Matching Power and Speakers -- Much Ado about Not-that-Much? (Tube amps and speakers)


Interesting conversation at the Part Time Audiophile's The Occasional Podcast.
There, tube amp maker Justin Weber of Amps And Sound makes these comments:

At about 24:03 in, they have an interesting conversation about power in tube amps and speakers' needs.

Justin makes an interesting comment about power and speaker efficiency. He's careful to caveat that room size might be an important differentiating factor, but for many (most?) audiophiles (with small-medium size rooms), there is just not that much to worry about regarding amplifier power and speaker matching. He is not touching on "synergy" in some larger sense -- he's just addressing the "power needed to drive speakers" question.

If I'm understanding the upshot of what he's saying, it's that lower-power tube amps made with quality transformers (and well made generally) have a very good shot at doing an excellent job for a much wider range of speakers than is typically assumed in conversations audiophiles have.

The whole interview is interesting, but here's the interesting bit -- I did my best to transcribe it, but go listen for yourself!

"Almost always, you need vastly less power than you realize....I've seen 1 watt power whole rooms and big rooms with moderately efficient speakers....What's a practical standard? 30 watts should do 99% of everything for everyone, and 15 watts should do 95% of everything for 99.9% of everyone....

Most audiophiles have small listening rooms.... [A more powerful tube amp may sound better, but] I think that’s a matter of it having a better output transformer and [that] output's transformer's core actually having a flatter frequency response and going lower.

So [in those cases where a more powerful amplifier is used] I don't know that it's a question of producing more power -- that you need more power -- as much as the transformer [in] the more powerful amplifier is [instead just] a better transformer."

Not sure what folks here think about these claims. Perhaps they seem so obvious as to almost not need repeating. But there are so many conversations about speaker sensitivity and watts that do not mention the quality of the transformers or which seem to overstate the importance of how powerful an amp is.

If Justin is right, then many, many pieces of advice related to "how many watts do you need" are basically wrong.

 

SOURCE: 
5 Things To Consider When Buying A Tube Amplifier
March 13, 2023 Brian Hunter Occasional Podcast 1
URL: https://parttimeaudiophile.com/2023/03/13/5-things-to-consider-when-buying-a-tube-amplifier/

128x128hilde45

@atmasphere 

Uh, just so we’re clear: current cannot exist without voltage; when voltage and current are present there is power.

Yeah yeah, of course.  I was referring to the published specifications.  Was trying to use short hand.

There’s also cases of speakers rated for 4 ohms which are 2 ohm speakers and they don’t need power so much as current. You tend to get higher current with bigger amps (not universally true).

@erik_squires

Uh, just so we’re clear: current cannot exist without voltage; when voltage and current are present there is power.

1 Watt equals 1 Volt divided by 1 Amp. That’s the Power formula and its inescapable.

The art of music reproduction in the home lies somewhere in between. I have to wonder if you or Ralph love music because if you really do, you would know that the reproduction of music is not subject to pure measurement.

@fsonicsmith1

The disconnect between sound and measurement used to be quite real. But in the last 30 years the technology has become available which makes your conclusion above false.

What has lagged behind the technology though is our understanding of how the ear/brain system perceives sound. As a result, even though we can make the measurements, most people including those that make the measurements don’t really know what is important about them or how to interpret them so you can predict how the circuit will sound. I spend some time over on ASR and I see this a lot over there.

I also see that both the measurement and subjectivist camps are equally guilty of confirmation bias.

As you point out, the sound is everything- its the proof of the pudding. Personally I fall into a middle ground category which is unpopular with both the measurement crowd and the subjectivist-listen-only-to-your-ears crowd. Daniel von Recklinghausen of HH Scott fame first stated my position:

If it measures good and sounds bad, it is bad; if it measures bad and sounds good, you’ve measured the wrong thing.

@8th-note 

I am sorry but you are beyond harsh, you are just extreme. Extremism is to be avoided 99% of the time. 

Frankly, Mr. Karsten borders on the same end of the spectrum but at this point I believe he is just too hell-bent on the scientific approach. It amazes me that he ever built tube amps. 

The art of medicine needs strict adherence to the scientific approach because there is no room for subjectivism. Art itself, visual, musical, or in letter form is not subject to the scientific approach. The art of music reproduction in the home lies somewhere in between. I have to wonder if you or Ralph love music because if you really do, you would know that the reproduction of music is not subject to pure measurement. If it make you feel any better, I think the same of John Atkinson and Kal Rubinson. 

Do you know the only thing worse than old men with fixed sets of belief and value systems? The answer is old ENGINEERS with fixed belief-systems. 

I do think audiophiles are taught to chase big amps far too much and you can get amazing results with some sweet small amps and a half-decent speaker.

There’s also cases of speakers rated for 4 ohms which are 2 ohm speakers and they don’t need power so much as current. You tend to get higher current with bigger amps (not universally true).

I’m thinking of models from older Legacy and the KEF Reference1 which I’ve recently seen measurements for and going "wow."  The current requirements of those speakers is just far above what a lightweight amp which otherwise would sound fabulous.

Take a CJ Classic 60.  Pair it with a Fritz speaker and it's going to sound fabulous.

@8th- note I suggest you listen to a bit more of the interview before judging him so harshly.

Mr. Weber is ignoring a significant swath of high end speakers that have low sensitivity and a difficult impedance curve. Wilsons and Thiels are two prime examples. In the more extreme cases the speakers also have phase issues which makes them even more difficult to drive. Anyone who reads Stereophile and pays attention to the sensitivity, impedance, and phase measurements is aware that many speakers require a robust amplifier to operate at their best.

In my case I have a pair of Thiel CS6 speakers driven by a Krell KSA 300S (2400 watts into 1 ohm) and it takes everything that amp can produce to drive my speakers to concert levels. Because of the punishing impedance curve (around 2 ohms for part of the audio band) a tube amp will act as an unintended tone control on these speakers.

It's interesting that high end speaker companies are still designing speakers that have low sensitivity and a low, variable, impedance curve. This apparently is a tradeoff that some speaker designers are willing to make to get the sound they are after. These speakers need a solid state amplifier that can double it's output as the impedance drops by half.

Mr. Weber designs tube amps that cannot run speakers like mine. Sure, they will sort of work but they will sound anemic and the frequency response of the speakers will not resemble what the designer intended. His statements are uninformed and obviously intended for people with little experience in matching amplifiers and speakers. As this subject comes up over and over again I am disappointed that people like Weber, who should know better, espouse this drivel.

If I'm understanding the upshot of what he's saying, it's that lower-power tube amps made with quality transformers (and well made generally) have a very good shot at doing an excellent job for a much wider range of speakers than is typically assumed in conversations audiophiles have.

The issue is deciBels. The ear hears on a logarithmic curve; deciBels are a logarithmic unit of volume. 3dB is barely perceptible to the ear in terms of volume increase but requires double the amplifier power. So the difference between 30 Watts and 300 Watts is only a doubling in the perceived sound pressure.

This is why if you need more than 100 Watts to make your speakers work, you'll probably need 600-800 Watts or more to get the sound pressure you're looking for without clipping the amp. If you need more power than that the speaker is really impractical or your room is really large.

OTOH we have SETs which are a unique phenomena. Because they use no feedback (and so are 10% distortion before clipping), if you want to get the most out of them, your speaker has to be so efficient that the amp is never asked to do more than about 20-25% of full power. At power levels above that the amp will make more higher ordered harmonics on transients, causing the amp to sound 'dynamic', since the ear uses those harmonic to sense sound pressure.

That makes SETs impractical on most loudspeakers and is why horns made a comeback in the 1990s. SET users get by with so little power because the amp can sound 'loud' when pushed, so the user tends to think its plenty of power. When using a cleaner amplifier the volume tends to go up... but if such a user goes from a 7 Watt SET to a 30 Watt PP amplifier, they may find themselves running the 30 Watt amp out of gas because they are trying to get the same 'loudness' as they got before, if they don't understand that the perception of 'loudness' was caused by distortion.

I have his Nautilus. It drives both my Devore O/93's and Spendor D7.2's very well. I do have a small listening room. And after about six months of using it to drive the loudspeakers it has been relegated to headphone use.

The Nautilus is a beast of an amp build-wise. It weighs about as much as my 10x higher wattage ARC Ref 80S. I think what Justin is saying is that speaker driving power is much more complicated than watts. 

Does the Nautilus sound the same or similar to my more powerful tube amps (also have an ARC Ref 150 SE)? No. And the differences go beyond mere sonic signature. I will concede the low-end grunt of even moderately low bass goes up as I go from the Nautilus to the Ref 80S to the Ref 150 SE. The Nautilus on the other hand has a sonic purity thing going on that is not there with the higher wattage amps. 

Interesting and informed replies, as usual. Thank you. It would be cool if Justin Weber could weigh in, too, but I don't know if he's on this forum.

I have 3 systems with 3 amps and 3 sets of speakers, I move it all around quite often. The 3 amps, Naim SS 50 watts, Vincent Class A hybrid pre / Power 150 watts, Cayin tube amp EL 34 tubes on triode 38 watts. All speakers are 89 dB and 4 ohms, which amp drives all of them best?  Guess, but it's not the Solid State or the Hybrid. 

It depends.
 

Assuming a general match with speaker sensitivity.

It depends on what characteristics you prize most. If you’re into slam… then more power more better. If your into musicality then you are more concerned with other characteristics and power costs money.

An additional generality. Solid state… more power is more important than tube amps. The really fast reaction times of SS can deplete current stores very quickly and they do not sound good when that happens. Tubes can breeze through stuff that an equivalent solid state amp might not sound so good.

If you own planar speakers… the amount of current / power rating is going to mater a lot. In general they require a lot of power.

I have owned massive amps for most of the last fifty years. But then I owned planar speakers… so it was critical. Also, you had to buy audiophile amps to get high current designs. I now own tube amps… operating in triad mode so only producing 60 watts per channel. I couldn’t be happier.

I guess, in reality, it is like other parameters… it depends.

Someone should by the new Schiit amp for a few hundred dollars and give it a try compared to what they have - I’d love to hear opinions on that. The GJALLARHORN is 10 watts of fun. 

pta is a fringe publication, I have read a lot of uninformed opinions there. But there may be something to power being overrated. 

It also depends on what type of music you listen to. Folk/Vocal Jazz/Pop don’t require the kind of dynamics you need to reproduce a concert piano or a symphonic orchestra. Also, if you’re very fond of rock/blues/pop, which feature a lot of bass where most modern speakers have a low impedance mode with high phase angle, no 15W amp out there can do anything. The instantenous current draw that a modern 4ohm speaker with EDPR of 1.5/2 ohms can be met by very few 100/150w tube amps, let alone a 15W amp. Keep in mind that a speaker that is 89 db sensitive does require usually much more than 1W to reproduce sound at 89 dB under realistic conditions. 
 

Tube amps are excellent at providing voltage not current, no matter what some random guy claims. That’s why a 83db sensitivity LS3/5a can be driven easily by a 30W tube amp, because of its high impedance (11 or 15 ohms depending on versions). Also, tube amps are very linear and noise free amplification devices, but at low levels, whereas most SS amps are more noisy with signals and behave more linearly with higher output.

But keep in mind that no amp works at its best when pushed near to or at its limits, so just because you maybe CAN drive a less-efficient pair of speakers with a tube amp doesn’t mean you’ll be getting the best out of the amp — or the speakers.  As always, proper matching is key. 

There is a caveat here.  Originally, the magic of tube amps was produced by SETs that made up to about 6 wpc.  Thus, the need for moderately sensitive speakers, say 94 dB and up.  

Then people who had already invested lots in less sensitive speakers wanted the tube magic, so much more powerful push pull amps were made.  Some of them are very very good (I have one) but many of us agree that the paragon of tube amp performance happens with just a couple of watts produced by a direct heated triode SET.  

So yes, you can power an 89 dB speaker with a Primaluna, but I'm aiming higher.

jerry

This has been visited any times. My opinion,for what it's worth is that to adequately power your speakers headroom is essential. Instantaneous power needed to avoid clipping is a huge factor as to providing dynamic range. Don't want my amps clipping, or to even come close. Hasn't everyone here at one time or another had their system uncomfortably loud past a certain point?

If you can afford headroom that is a huge advantage in many factors related to musical satisfaction. Yes, if you listen at lower levels this doesn't become a problem. 

 

He's making two, separate, claims:

1) About 15W amp power should suffice for most home systems

2) (IF) High powered tube amps sound better than the lower powered ones, it's probably because their output transformers are better -- i.e. transformer quality and not power is what makes the difference (if there is one).

I understand where he's coming from, but would add another parameter: current. It's not only the strength (watt) that makes a difference, it's also the "quantity" of energy provided that matters.

 

Otherwise, 15watts should be ample for speakers 91dB / 8ohms or better. At peak power you get over 100dB -- not bad!