To SET or not to SET

[neo33]

Interesting read...

Excerpt from an avid SET owner:

"Most people have assumed that a 2 to 8 watt per channel SET amp wouldnt have the balls to get out of its own way. It couldnt possibly have any real bass, but we hear the midrange is to die for.

Most people would also assume the only way to hear any dynamics from a flea powered amplifier is to use big nasty horn speakers so really what is the point?

All of these assumptions are probably a side effect of the general direction that high-end audio has taken since the 1960s. Solid state has made high power affordable for everyone so the loudspeaker industry responded by making speakers less efficient to both reduce their size and flatten their response.

Despite popular belief, you do not need horn speakers to use or enjoy a SET amplifier. You can also find many horn speaker designs that sound wonderful, better than wonderful in fact, so the question then becomes this:

What are the advantages to using SET amps with high efficiency speakers?

To answer this, lets start with the advantages of a SET amplifier over any other type. A Single Ended Triode is the simplest circuit design there is, using the least number of parts. Typically this is a driver stage coupled to a single output device. Triodes do not require negative feedback, something found in most all push-pull circuits, solid state or tube. Negative feedback is used to lower distortion specs and in the case of solid state devices it is often the only thing keeping the transistors from exploding all over the inside of your amplifier. Feedback a problem? If you dont mind the time smear it creates and the resulting 2 dimensional sound stage, then no I guess its probably not.

Aside from the amplifiers superiority by simplicity, there is a more profound reason for using SET amplifiers. The magic predominately lies in the first watt. By magic I mean inner detail and most of the dynamics. For example, a pair of 96dB speakers playing with one watt of power against the average noise floor in your listening room (55dB) is 40 dB of dynamic range. (96 55 = 41 dB) Adding a second watt increases the dynamic range by only 3 dB. For every additional 3 dB you need to double your power. This should clearly illustrate that there is over 10 times the dynamic range in the first watt as there is in the second.

This brings us directly to loudspeakers. A typical loudspeaker today is 86 dB efficient with 1 watt. It also usually has a complex crossover that attempts to keep the frequency response and impedance flat. The crossover alone will usually dissipate a significant portion of the first watt as heat before it even reaches the drivers. To reach the same loudness level as the 96 dB speaker will with 1 watt requires over 8 watts on the 86dB speaker. If we used 2 watts on the 96 dB speaker the other would require 16 watts to keep up. If we used 4 watts on the 96 dB speaker the other would require 32 watts to keep up.

The problem here is resolution. If you cant hit a listening level with the 1st watt, youre not likely to hear whats happening in that 1st watt. For a driver to achieve a high efficiency its moving parts must be low in mass. That makes it dramatically faster or more accurate than a speaker with heavier moving parts. If you like inner detail and want to hear all of the textures and layers of a good recording you need fast, efficient and coherent speakers.

A good SET amp combined with a single full range driver with no crossover or a simple 2-way using minimal crossover parts on the tweeter only, has a purity and depth that you simply dont find in more conventional systems. It is a benchmark for coherency, and noted for its ability to create hauntingly real holographic sound stage. Bass and dynamics with this combination sound more realistic in part from the tremendous speed and in part from the coherency.

Ive consulted many people about their audio systems, and the most common complaints include dry somewhat fatiguing sound with a fairly boring soundstage followed by the realization that it simply doesnt connect you to the music emotionally like it could. Experience has taught me that by far the easiest way to get a liquid sound that becomes holographic with stunning clarity and detail, something that excites the listener, is to set him up with an SET and simple pair of efficient speakers. It also usually ends up being the least expensive solution.

The biggest conformation of this is reports from audiophiles who used to have several hundred watts and many thousands of dollars invested in show winning audio gear, but now report that even a good 2 watt SET on efficient speakers has better dynamics and weight which they find simply amazing. If youve ever observed how audiophiles rotate through audio gear during their lifetime you might also find it interesting that the ones who finally land on SET amps and good speakers seldom find anything they like better.

The bigger is better mentality that is directly connected to more expensive the better is certainly the handicap that stunts most audiophiles from discovering truly high fidelity sound. Will the guys at the audio salon laugh? Yes, they will. Will your fellow audiophiles laugh when they hear you sold everything and got a 2-watt amplifier? Yes, they will. Is this important to you? Only you can decide, but I would suggest it has little to do with high fidelity.

In closing, remember this It is dangerous to place amplifiers, speakers, cables etc., into neat little categories in an effort to make some sense of it all. Not all tube amps sound good. Not all solid- state amps sound bad. With the Internet audiophiles have been set free to research things in a far more unbiased way then ever before. Before the Internet there were only trade magazines, and manufactures literature to educate us all and it goes without saying fairly biased sources. Of course the Internet is full of misinformation, some intentional, some out of ignorance, but at least you can find all sides and make your own determinations."

[Champagne taste beer budget]

Thanks, neo, interesting read indeed. One question to those with more experience than I: Is his example of a 55 db noise floor really typical? Since most of his following statements are based on that assumption, I would think it a very important number to figure into any of the equations.

[3dzapper]

Good article Neo, even if the author is a lowly Lowthar lover.

Rick

[popbumper]

Neo:

An excellent read, thanks much for sharing!

Popbumper

[hwatkins]

----------------

On 6/28/2004 2:44:19 PM Champagne taste beer budget wrote:

Thanks, neo, interesting read indeed. One question to those with more experience than I: Is his example of a 55 db noise floor really typical? Since most of his following statements are based on that assumption, I would think it a very important number to figure into any of the equations.

----------------

I am not attempting to debunk (interesting article as a whole), but the 'average' noise floor has me somewhat confused (that's the 12th thing to confuse me today). Noise floor is a bit more to me than average and is very much associated with device noise and as a whole the extraneous (nothing else happening) rest of the room noise. This comes from other (all) powered equipment as well as the source device. That means, to me, that the source and amplification devices are primary in whether the noise floor is average or not. So - take the room out of play and the average is still hard to find - maybe average for a specific source or device will work.

Lastly, as in all our audio stuff, the floor changes based on frequency. In my old school days (remember folks we are talking 27 years ago - with the way Congress is diddling with things the Laws of Physics could well have changed by now) I seem to remember that as the frequency climbed, the noise floor decreased in a very significant way. If that seventies tainted memory is somewhat accurate - average is a fairly inappropriate way to categorize noise floor.

Now - I ain't one to know this stuff well enough to know if I and the article's auther are on the same page. This may be a different definition of noise floor than I am thinking...

[3dzapper]

Henry,

From my old dormatory days*, I seem to remember the noise floor as being where the party was that night!

Rick

*USAF not Cow College. That came later.

[popbumper]

Henry:

If it helps clarify at all (noise floor), I believe he is referring to the direct or indirect causes of extraneous noise in the room which contribute to the overall "ambient sound". You probably realize this already, so let me address your question of "is 55dB typical"?

My "listening room" (I use this term loosely) is upstairs, above the garage. This is no "special" room, in that it has no acoustic treatments, and is non-isolated from the house (in that it has an attached stairwell, as well as a pair of windows looking out over the driveway.

During the time I was setting up my equipment and doing calibrations, without effort I was immediately confronted with the sound of the overhead fan (which is precisely over the seats), and I was able to capture this "noise" on my sound meter - a whopping 70dB!!!

My best time to listen in this rom is after 10PM, when most of the neighborhood noise is quelled, and the wife/kid have gone to bed. Interestingly, here is what I have noticed:

a) The fan sound is the most predominant. If it is NOT really hot in the room (upstairs, Texas, Summer time), I turn it off.

The AC outlets (on the ceiling), contribute quite a bit of noise, though certainly not at the level of the fan.

c) The computer at the back of the room has some nice fan noise as well.

d) With heavy drapes, and at night, the outside noise is pretty low.

I was actually surprised at what these all contribute to the noise floor. My conclusion is that 55dB would not be atypical, but it really depends on the room location/environment and what is in it.

Popbumper

[hwatkins]

3D- You Air Force folks know a bit about parties. "Flying" can be done on the ground afterall...

Pop - Yeah, you covered it in the right way. I haven't gone to your lengths to check levels, but the noise I see in my main lisenting room (downstairs) mostly comes from the electronics and computers. I would put it at about 40db (the puters are some distance away), but not having measured I may have to accept the 55db. That said - if we are talking individual rooms it is misleading to use a specific average. I would be more comfortable with an average range.

My point could be more succinctly put as - When extraneous issues specific to an environment create the noise floor, it is inappropriate to use a unique average to reinforce your point. Perhaps as an anecdote I would be more comfortable with the article's conclusions.

BTW - I think certain SETs are cool (as I do PP - tube or SS). I can't get my ear to tell me that single FRs produce the sound I like. They are a bit like my old victrola - amazing what does happen, but not how I prefer it.

[jnorv]

Nelson Pass' take on SETs.

http://www.passlabs.com/pdf/seclassa.pdf

Jim N

[Deang]

As yet, very few other single-ended solid state amplifiers are available

on the market. This will change as the demand continues to increase

and as other designers learn how to build them.

In the meantime, transformer coupled single-ended triode amplifiers are

the alternative, using very large gapped-core transformers to avoid core

saturation from the high DC current. These designs reflect more

traditional thinking in single-ended amplification. They suffer the

characteristic of a loosely coupled transformer, more limited wattage,

and higher measured distortion than their solid state counterparts,

however they still set the standard for midrange lucidity, and are not to

be dismissed.

It would be interesting to see if any of use would actually prefer the sound of solid state in this configuration over some of the better tubed push-pulls.

[3dzapper]

"3D- You Air Force folks know a bit about parties. "Flying" can be done on the ground afterall..."

"And you can fly,

High as a kite if you want to.

Faster than light if you want to.

Speeding through the universe,

Thinking is the best way to travel"

Rick

Source: The Moody Blues. I don't want anyone getting the idea that I am a poet!

[Colin]

This means horns: "For a driver to achieve a high efficiency its moving parts must be low in mass. That makes it dramatically faster or more accurate than a speaker with heavier moving parts. If you like inner detail and want to hear all of the textures and layers of a good recording you need fast, efficient and coherent speakers."

and PWK said it too

I have heard tube amplifiers that sound bad: North SE, old radios, etc.

of course, he doesn't say anything about the lack of mid- and low bass with singel-drivers, or adding a sub, but nice article anyway...

[skonopa]

----------------

On 6/28/2004 3:59:42 PM hwatkins wrote:

My point could be more succinctly put as - When extraneous issues specific to an environment create the noise floor, it is inappropriate to use a unique average to reinforce your point. Perhaps as an anecdote I would be more comfortable with the article's conclusions.

----------------

I am guessing the author of that article just picked a typical number out of the air to make a point. Since the article seemed to be aimed at a more generic audiance, not just people that may have a "special" room setup for listening, the number he picked really did not seem to far off.

My own listening room is the upstairs living room. It is opened to the dining room as well as the kitchen. Thus, I have the refrigerator noise, the fan noise, the windows opened (with the inherent "outside" noises - Rt 3 is less than a 1/4 mile away, a very busy highway), noise from a computer in the dining room, as well as the noise from the electronics themselves. I did not realize how loud that harddrive in my PVR unit was until I was sitting in there on a completely quiet day. I've taken a reading with my rat-shack SPL meter and I got anywhere from 50 to 60dbs when just sitting in the room doing absolutly nothing and with nothing really turned on that is not normally on all the time, such as the refrigerator.

On the otherside of the token, when I spent a week in Avon, N.C., which is on Hatteras island, just north of Cape Hatteras, it was unbelievably quiet! I was laying in bed in the bedroom of the house and it was so quiet that I could hear my own blood flowing through my head. It was almost like standing in that acoustic chamber there at Klipsch HQ. I'd be willing to bet the ambient noise was not even 30db. There was absoululy no traffic noise. The ocean was to far away to hear the surf. I've never experience something that quiet before outside of an "artifical" environment specifically setup to be that quiet.

As for the SET (to get back on topic), I really cannot comment as I've never heard SET before. Despite what some of the nay-sayers may say on here, I'd certainly would like to try them out. Would be interesting to try a pair on my RF-7s, compared to SS B&K amp packing 200 watts/channel.

[neo33]

"Problem with push-pull amplifier designs associated with crossover distortion have been discussed elsewhere at length, and one of the primary results is non-monotonicity. Class B and many AB designs have distortion products which dramatically inclease with decreasing signal. This is reduced greatly by Class A mode, but crossover distortion remains as a lower order discontinuity in the transfer curve.

For reproducing music as naturally as possible, push-pull symmetric operation is not the best approach. Air is not symmetric and does not have a push-pull characteristic. Sound in air is a pertubation around a positive pressure point. There is only positive pressure, more positive pressure and less positive pressure.

Descriptions of push-pull often illustrate this type of operation with a picture of two men sawing a tree by hand, one on each side of the saw. Certainly this is an efficient way to cut down trees, but can you imagine two men playing violin?

By contrast, push pull Class A circuits have two opposing gain devices producing the output signal, and though it is industrially effective and efficient, it is not the most delicate way to amplify the signal. Push-pull circuits give rise to odd ordered harmonics, where the phase alignment reflects compression at both positive and negative peaks and crossover nonlinearity near the zero point."

Hmmm...

[Deang]

Nice try Neo, but I'm not taking the bait. Enjoy your amps.

[neo33]

Baiting what, Deano? I am just pondering Nelson's PP thoughts.

[NOSValves]

Bunk from both directions in one thread geezzz

[Parrot]

----------------

On 6/28/2004 4:43:39 PM neo33 wrote:

but can you imagine two men playing violin?

----------------

I sure can, especially when I hear them with good amplification through Klipschorns:

Jascha Heifetz

Yehudi Menuhin

[neo33]

You're bunking from one end and Deano is bunking from the other end, right Craigo?

[hwatkins]

This reminds me of years ago as two warring factions that worked for me were diiscussing (arguing intensely actually) about Token ring and Ethernet - which would be the preferred solution. From an analogy standpoint - token ring was the winner - could you just imagine all those 'collisions' with ethernet. I almost let analogies sway me, but the logic finally started getting through, Token Ring would be bandwidth restricted and higher overhead for management. Both worked just fine.

I buy into the simplicity and the beauty and such with SET. I don't buy into the gospel the SET is the handsdown winner in the amplification wars. There are some strong arguments that support both PP and SET camps. I have heard both sound extraordinary.

I think that most of us here really think much along the same lines. As I raise my hand I am reminded to keep my analogies relevant (excluding when they are trying to be humorous).

Talk to the Hand...