[jswallac]

There seems to be a sea change towards low powered (2-10 watts) amps.  While I understand why when the amps are based on tubes, I am curious why people prefer low powered SS (digital) amps.  High powered tubes are very expensive, not so with SS.  

It seems that more power is like an option.  You do not need to exercise your option, but you can if you want.  You can use high efficiency speakers, but you are not limited to these.  You can play complex music at higher volumes and not worry about the amp running out of steam.  Like a car with a lot of horsepower, it can always be driven slow, but has the reserves should you need it.

So why the preference of low power SS?  What are the advantages.  Why is less suppose to be more?

[JeffB]

The general idea is the less circuitry, then the less the signal is affected.

[miklorsmith]

As a two-time Clari-T owner:

Small amps mean possibility of batteries, no AC, no switching supplies, no expensive transformers, no expensive power cables, no power conditioning.

Small amps mean fewer, smaller internal parts, i.e. cheaper and higher quality.  This also means simpler, as in closer to the "straight wire" ideal.

With my Zu Definitions, the 6 watts will play obscenely loud without strain.  So, any headroom issues are moot to the sane observer.  I did notice the sound is cleaner 2 floors away with my 45-watt amp though.

[JoshK]

There are quite a number of reasons, many are contreversial.  Jeff mentioned one.  While I am not necessarily a believer or non-believer of the low watt world, I may be able to shed some of the ideas.  Proponents of low power often suggest:

- low power amps tend to be more linear
- high power amps are very difficult to do right, those that are done right are uber expensive.

There are other reasons but I am not thinking of them atm.

[jswallac]

Thanks for the responses so far.  I do agree that in the "boat anchor" world as some call it, analog SS is expensive to get high quality.  But that seems to be changing with these digital amps.  Take the Teac, while hardly high power, it is much higher than the SI or clari-t.  It can still be battery powered.  Doesn't it therefore give you more options in speakers, provide more headroom, etc.?  Any drawbacks?

[JoshK]

Ultimately I think it is one of those things where you have to hear a really good low powered tube amps on a hi-eff speaker to see if you like it or not.  All the rest is philosophical.

[miklorsmith]

Some have commented that the Clari-T is sweeter on top than the Teac.  Others have said the Teac is punchier on the bottom than the Clari-T.  Vinnie's got my Teac and I'll have my own opinions at some point.

But, these are a microfraction of the small-amp genre.  You could drive yourself insane trying to get your brain around all the options.  This doesn't even address tubes, which have an allure all their own.

[JoshK]

One thing I do know is that I would never settle for a system that could only handle "little girl with a guitar" music.  Nope, not for me. Its got to be able to rock, have balls when needed and play large scale orchestral well (no compression, breakup or other puttering out) or it ain't hi-fi to me.  

I am going to try to build a "low powered set amp" soon.  I spent some time researching what kind would satisfy me and it looks like I have very few choices.  I will try one and see how it goes.

[bhobba]

Hi All

I think a number of issues are linked in here.

First amplifiers, unlike loudspeakers, have a subtle effect on sound.  Blind tests show that by and large amplifiers with low distortion and not driven to clipping or driving hard loads etc sound pretty much the same:
http://www.stereophile.com/features/113/index.html
This has led some people like Tom Nousaine and Arny Krueger to proclaim amplifier differences beyond low distortion etc are irrelevant and those that disagree are of the weirdo tweako brigade.  That is one extreme.  However those that hold to such views ignore that experienced listeners, in blind listening tests, can pick the differences between amplifiers. Indeed they can pick the difference between the even more subtle effects of cables.  It is precisely those highly critical listeners that audiophile manufacturers cater to.

When I started out in Hi Fi all those years ago one thing I learnt quickly is that constructing audiophile quality flea powered amps is a doddle - it can be done relatively cheaply.  Constructing high power ones is much harder and much more expensive eg the cost of power supplies rises dramatically.  Amplifier costs have come down so that good quality (not great but good quality) high power amplification is now available at low cost (eg the 200w Behringer A500 for $180.00 street price).  Such products measure beautifully but to critical audiophile ears are lacking.  And that is one of the issues - to go beyond what measurements show we must actually listen and introduce an element of subjectivity.  When that is done then what constitutes your preference will vary.  For some what they are seeking is ultra low noise and extreme detail retrieval.  Battery powered flea powered amps provide that without breaking a sweat.  For others it is ridiculously low distortion - for those audiophiles class A designs appeal - and low power designs are much easier to construct for that.
http://www.siliconchip.com.au/cms/A_102474/article.html
And yes trained ears can hear such low distortion.  Leo Simpson, a well known electronic magazine writer, published an ultra low distortion class A amplifier some years back and was surprised that it sounded a good deal cleaner than conventional class A/B designs.  But achieving such low distortion can only be done with lots of feedback which has its own issues.  This leads still others to zero feedback class A valve designs and its 'musical' distortion some prefer.  The list is endless.  The point is it is much easer and less costly to meet those requirements in low power amps.

Of course going that route has its own problems - engineering is always about compromises.  One can not drive low efficiency speakers to high volume levels with low power.  Those that install systems for a living say that people rarely if ever go beyond about 110 db peak at the speakers.  For relatively low efficiency speakers of 85db this equates to peak power of over 200 watts.  Increase the efficiency to 96db and power in now a more modest 20w.  110db is really loud - so it is fortunate it is just a peak.  CD's are recorded with about 18db headroom so this would equates to 91 db average produced at the speakers or at usual listening positions of 3 meters about 81 db.  That is for mono - for stereo it would be 84db.  Power mowers produce levels like that (a hair drier and person shouting produces about 80db).  Would you like a power mower next to your ears in your listening position at home?  Would you even like a hair dryer? How would your neighbors react?  Indeed prolonged exposure at 85 db or greater can cause damage to ears:
http://hearing-protection.4ursafety.com/exposure-limits.html

This leads some like myself to never actually exceed 82db average at about 1 meter from speakers ie a little above the level of someone shouting about one meter from the speaker. This still allows the reproduction of peaks at about 100 db which is very loud - for chorus and orchestra in a concert hall peak levels can reach 105db so you are not missing out on much realism.  But rock music is another matter - levels at rock concerts can easily exceed 120db.  To realistically reproduce that you need power - and heaps of it.  Just watch your ears and neighbors complaining.  But getting back to the more sensible 82 db at 1 meter consider 85 db sensitive speakers.  Allowing for 18db headroom this gives a max of 100 db max (instead of the 110 max you had before - which only sounds half as loud BTW) you now only need a more modest 20 watts.  Raise the sensitivity to 90 db and you only need about 8 watts.  Raise it to 96 db and you need less than a puny 2 watts.

Dave Ellis also has some rather interesting facts about amplifier power:
http://www.ellisaudio.com/wattsratings.htm

So just to bring it all together:
Cheap amplifier power is now available to satisfy all but critical audiophiles.  The A500 Behringer for example will easily drive even relatively inefficient  speakers to the maximum spl levels experience has shown is rarely exceeded.
However those amplifiers lack when listened to critically by audiophiles.
It is simpler and cheaper to build low power amplifiers that meet audiophile criteria.
Listening at sane levels to moderately efficient speakers requires less power than you think eg for 90db speakers you can easily get by with 8 watts.

And once one breaks away from heaps of power than some truly excellent amplifiers can be had rather cheap
http://www.audiocircle.com/index.php?topic=15411&sid=8c9bf6f5e3b76d6cac69cfe2926452a3  


Thanks
Bill

[JoshK]

Very well said, Bill!  Somehow though those number don't capture something about power amps.  I don't know how to explain it but a much more powerful amp, say my previous 600wpc amp sounds more effortless than another 150wpc amp did.  It could be a lot of things like dampening factor, current delivery, etc. But the watts equation would say that I don't need much watts into my 91db/(w or 2.83v?) RM40s.  Experience says otherwise.  It may be the impedance down low is harsh and needs the current, i.e. the dampening to control it.  I don't know enough to say for sure, as I haven't measured and experience it myself.

[bhobba]

Very well said, Bill!  Somehow though those number don't capture something about power amps.  I don't know how to explain it but a much more powerful amp, say my previous 600wpc amp sounds more effortless than another 150wpc amp did.  It could be a lot of things like dampening factor, current delivery, etc. But the watts equation would say that I don't need much watts into my 91db/(w or 2.83v?) RM40s.  Experience says otherwise.  It may be the impedance down low is harsh and needs the current, i.e. the dampening to control it. I don't know enough to say for sure, as I haven't measured and experience it myself.

Hi Josh.

I experienced the same thing myself with my Axis LS88's.  It has 94db efficiency so a 40W amp should drive it no problemo.  I had a 100W rotel connected and my friend had a 40w low distortion audiophile pioneer - which aside from the power was reputedly slightly better than the rotel.  At matched volume levels his amp sounded weedy. Why?  We did an experiment - we biamped them.  This produced a bigger improvement again.  Conclusion - they were sucking more current at bass crossover (in this case about 300hz) than can readily be supplied by a low power amp.  This was confirmed when I tired an ME that is noted for delivering lots of current.  Better again - but not as good as the Rotel biamped.  It came as no surprise when I contacted the manufacturer (great guy BTW); he confirmed at crossover it went down to 2 amps - a hard load for any amp.  Bass can have a lot of power so if the speakers crossover in the bass region and it dips low then your amp is required to deliver a lot of current which can have an audible effect.  Solution - try passive bi amping it with a high power monster for the bass and a lower powered high performance one for the rest.  You could even make the one driving the bass an el cheapo like the Behrenger A500 - I have heard its bass performance is rather good even by audiophile standards.  Even better if money is not a problem why not try a DEQX and stick a A500 channel on each actual bass speaker and better quality amps directly connected to the VMPS ribbons.  That would probably really be something because by actively bi/tri amping amplifiers have effectively four times the power:
http://sound.westhost.com/bi-amp.htm
'If we now we separate these signals again - prior to the power amp - and using an electronic crossover, we have an entirely different situation. (Note: It is assumed for the sake of this article that the crossover frequency is near the halfway point between the two discrete frequencies of Figure 2.) Each signal is now supplied to its own 100W amplifier (there will be but a hint of the other frequency still visible on an oscilloscope, since the filters are not "perfect") and thence to the loudspeakers. The amplifiers are not clipping, both signals are reproduced at their original power, and the effective result is that we are emulating a 400 Watt amp with two 100 Watt units.'
With 200 watts driving each bass driver from a Behrenger that would equate to an effective amplifier power of 1600W which would beat your bass drivers into submission.  Combine that with better quality amps for the midrange and tweeter and DEQX room equalization - I salivate at the thought.  I would use those nifty 8 w jobs from the Boulder cable to drive the ribbons which due to the biamping would be equivalent to 32 W which at 91db sensitivity (although I suspect with the crossovers removed their sensitivity would rise) should be fine (you would be able to reach 105 db peak - the same as in a concert hall).  Hey - its only money.  BTW I am looking at going the DEQX route myself.  

Give the passive biamping a try - I would be very interested in your results.

Thanks
Bill

[warnerwh]

I totally agree with you Josh.  High powered amps sure sound more relaxed and if there's any shortcomings my ears don't notice it.  I suspect alot of amps under 100wpc with 90db or less sensitive speakers are clipping regularly by people who listen to any more than moderate volume and I don't think they know it.

[lcrim]

Right now I'm watching and listening to Cream's Reunion DVD on a system w/ 1.8 watts per.  The speakers are 95dB efficient.  The dynamics are to die for.  
I like the way tubed class A and AB amps sound through properly efficient speakers, at least for now.   It took me awhile to work this out.  The low power SET, high efficiency, analog source thing does it for me.
I agree with the thinking that the cost of getting great sound with high power and lower efficiency speakers may be prohibitively expensive.  But I'm not sure I would spend a lot more even if I had it to spend.  Well, maybe.

[bhobba]

I totally agree with you Josh.  High powered amps sure sound more relaxed and if there's any shortcomings my ears don't notice it.  I suspect alot of amps under 100wpc with 90db or less sensitive speakers are clipping regularly by people who listen to any more than moderate volume and I don't think they know it.

Hi Warnerwh

I supose that depends on what you count as moderate levels.  It is dangerous to listen at 85db and above for extended periods so to be safe it should always be less than that.  CD's are recorded with on the average 18db headroom.  It rises a bit above that for classical music but 20db is a pretty safe bet.  Thus it will not clip if your maximum normal listening power is 100 times less than maximum power.  If you are 3m from your speakers reasonable listening would not exceed 80db (this is the same as a person shouting right in your ear) which corresponds to each speaker delivering 87db or being fed with a half a watt.  50 watts is enough for that.  It is only if you start listening at dangerous levels like 85db average you run into problems with 90db speakers and 50w.  Lower than 90db - yea.  Also your amp may not be able to deliver that power into real life speaker loads - which is what I suspect is happening to Josh.  But then again if you stick at sane levels ie 80db and less at the listening position you should be ok.  I listen mostly to vocalists at a level of not more than 80db at 1 m from the speakers.  Normal singing level is 75 db so to produce a person singing at you from about where the speakers are then 95 db max is fine - for 90 db speakers that corresponds to not quite 4 watts. Going above that would simply not be realistic.  Consider classical music - it reaches 105 db max in concert halls - which corresponds to 85 db average with 20db headroom - which is dangerous for long periods so I would wind it down a bit - to about 80db - so 50w is still enough.  It is if you want to listen to rock music at 120 db you have problems - but your ears and your neighbors will also have a say.  

Thanks
Bill

[warnerwh]

Most of the time I listen to an average level of about 80 db. If I turn it up at all with my 90 db efficient speakers 100wpc is going to clip. I can't see not having plenty of power unless you can't afford it. IF your speakers are  more than 90db sensitive, say 93 then 100wpc should be minimum. I heard a pair of La Scalas recently powered by 35 wpc Dynaco tube amp. Dynamics were excellent and I could tell the amp never clipped at high volume. The sensitivity the guy said is 106db at 1 watt!

[GHM]

There's definitley a difference in amplifiers sound. I seem to gravitate towards amplifiers of 50 wpc or less. They always seem more transparent to my ears and definitely quieter as a whole..there are exceptions though.
How about transient response? I've heard high wattage amplifers that do not posess the speed and response of some lower powered amplifiers.
 
My speakers are 89dB ..50 wpc is more than enough to drive them to levels above what I consider sane. With only one inductor ,one resistor and one capacitor in between the driver and the amplifer..the speakers are an easy load.
I'm sure distortion plays it's role as well. An amplifier with less distortion will give you a sense of more volume as the music is much clearer and tangible.The background in the music  is also much more alive.

One has to consider what their amplifier must push current through as well.
If I used speakers with 10 pounds of crossovers in them that claim 89 dB efficient. I doubt 50 wpc would do them justice. I noticed this while doing some listening test with Paradigm 100s at a friend's home. We used several different amplifiers.The more brute the amplifier the better the speakers sounded.
Even with all the power, a speaker like this never had the transparency and presence of some speakers I've heard with minimum to no crossovers using low wattage amplifiers. There's just no getting around minimum parts  and less circuitry. Even the best parts are not as good as not having them at all in the signal path if designed to be so. I apply this to all audio equipment. If it's over engineered I'll pass on it.

A speaker that uses no crossovers has it's drivers directly coupled to the amplifier. Basically it's like having active speakers. There's give and take with anything in this hobby. You must decide what's more important for you and what you can live with. A person's musical taste will also be a deciding factor. I'm not a huge fan of rock ..so needing a speaker/amplifier to play at 120dB is unnecessary for me. Not to mention I would like to stay in the hobby more than a few years before my hearing goes out.

[bhobba]

Hi GHM

There's definitley a difference in amplifiers sound....

I agree - but blind listening tests with low distortion amps spl matched and not driven to clipping or driving difficult loads show it is not night and day.  In fact they suggest there is no audible difference.
http://www.mastersonaudio.com/audio/20020901.htm
One needs to be experienced with the amps and have them in a system you know well to pick it.  I always suggest you buy the speaker - get a cheap amp like the Panny XR55 to power it initially - then at your convienience trial amps at home to get the one you want and make sure you can tell the difference between them blind.  After that you can investigate the even more subtle effects of cables.

How about transient response? I've heard high wattage amplifiers that do not possess the speed and response of some lower powered amplifiers.

I agree.  However I see no engineering reason for that to inherently be the case providing you are willing to throw humongous dollars at the problem.  But would any sane person pay the price of the product? - well audiophiles do some strange things at times.

Basically it's like having active speakers.

I believe that the dollars some are spending on amps/preamps would be better spent buying a DEQX as the preamp, removing the crossover and using low power high quality amps directly connected to the drivers - except for the bass - one needs a lot of power for bass if you are actively equalizing.  I recently had a discussion with someone who purchased SP continuums and wanted advise on amplifiers.  I suggested he look at lower power amps like the Charlize which is a steal at $90.00.  Maybe have a cheaper high power amp and a lower power one for critical listening.  Instead he went for Clayton amplifiers.  Fine amps IMHO - but the cost - $3500 new - and you still need a pre.  You can get a DEQX pre for $3500, rip out the crossover and drive the speakers directly with say a Nuforce for the bass and Boulder's nifty battery powered amps for the 'tweeter' (it should not be called a tweeter in SP speakers because they cross over at 650hz).  The sound IMHO would be streets ahead of any passive approach (largely due to room correction - but also contributed by the fact you no longer have crossovers etc to worry about).  I however did not suggest that to him as it is rather radical.

Thanks
Bill

[miklorsmith]

I haven't done any blind comparisons, but I've had the following amplifiers through this house just in the last 2 months and each sounded considerably different:

Onix integrated, Clari-T in two different guises, NuForce Ref. 9, AudioZone AMP-1.  Outside observers have been with me while I switched amps though none has heard all of them sequentially.  FWIW, neither have I.  When the system is adequately revealing, the difference is not an audiophile one at all.

And, the active XO/amp solution seems to be a good one for experienced users, but I wouldn't turn one loose on some average shmoe, as they wouldn't even have a clue what the sound should be nor how to get there.

[mcgsxr]

Other than sub amps, I have never had a powerful (by these standards) amp.  Perhaps I am missing out on something.  Perhaps not.

I have used a Sugden integrated (45 wpc at 8 ohms), two heavily modded Teac tripaths (each 30wpc into 8 ohms), several JVC receivers (around 70wpc into 8 ohms), have a Panny receiver coming (around 50wpc into 8 ohms as I recall) and for my main 2 channel system, I currently use a small JVC (again 30wpc).

For my 2 channel system, I am using DIY open baffle Visaton single driver speakers - about 94-97db efficient, depending on what I read.

For the HT (where the receivers come and go) I am using 85-86db efficient kits from DIY Cable (their KIT41).

Totally different sounds - both systems are good for what I use them for, and really tough to do an A/B in either system.

For me it is economics - the lower powered gear that I like the sound of (vs what I assume I might like the sound of in a more powerful solution) is just simply out of the $$ for me, at the moment.

To each their own, I suppose.

[JLM]

IMO the amp must provide a commanding grip over the speakers in order to reproduce the bass control and dynamics of real music.  I've listened to several SETs, other tube amps, digital amps, and of course solid state up to 200 wpc with speakers ranging from 86 dB/w/m to 104 dB/w/m.  It doesn't take 110 dB continous spls or anything close to a golden ear to hear what a commanding grip is like.

Public service annoucement:  Extended exposure to 85 dB or more will lead to permanent hearing loss and possible permanent pain.  So I critical listen (mostly jazz) at around 75 dB averages and less for classical.  The spl meter taught me that most live classical music is below 60 dB.

Size your amp to provide 10 dB headroom over your rock music playback levels, 15 dB over your jazz playback levels, and 25 dB over your classical playback levels (for your given speaker efficiency).