[Niteshade]

What's your favorite tube amp power rating? Mine happens to be around 30 to 50 watts per channel. I do have a question for consumers: Some people change to a lower wattage amplifier when moving into a smaller house or moving their system into a smaller room. The words, "don't need all that power any more" are said all the time. Here's what I don't understand: The power provided by the amp is 100% dependent on where the volume control is set. A 100 watt amp can just as easily produce a max of 1/2 a watt. I don't want to put size constraints into this topic, since as stated, most of the time it's a power concern.

[JoshK]

This topic isn't necessarily a lab related topic, more of a general audio topic. 

But I agree, that I never understood some of the logic behind moving to smaller amps.  The only rationale I can see is that there is a perception that it is easier to make a sweeter sounding lower wattage amp.  I.e. if you are using DHT or triode mode or class A or SET you are likely to be constrained with smaller watts.

The topic of how many watts is enough is one I find interesting.  I find that it varies wilding from person to person.  However, the small watt guys argue infatically (sic) that others don't need more watts.  I can only surmise that some only listen quietly in small rooms to little girl in a guitar music. 

In my case my speakers I am building are 96+ db/watt from 200hz up (actively crossed ~200hz, so the amp won't see <200hz content).  I am building a 10watt amp currently.  We shall see if it is enough.  I am skeptical.

When we visited bugleboy's house and listened to his Avangarde Duo's, which I think are 102db/watt from around 200hz up, actively crossed also, when played many flea watt amps.  There were none of the <8watters that I didn't hear clipping often.  Of course we listened pretty loud, louder than I wouldn't often listen, but still I like to be able to turn it up sometimes. 

Complexity of music is another kicker in my limited experience.  I like to listen to a lot of progressive metal, say Dream Theater or Tool for examples.  A lot of DT's music has some dynamic complexity that makes it like a crescendo in orchestra pieces, but more constant throught the songs, thus not as infrequent. So I am not so sure flea amps are in my future.

30 watters seems like enough for 96db.  With Jim's 93db/w Tannoys, 30 watts was just barely enough IMO.  For most non-party listening it is fine, but when I get to rockin out in my living room, I'd suspect I'd want a little more.  The extra 3db/w is probably that little bit more.

The Karna's I am building are close to 30 watts.  Its a PP 300B, so one would first think 16watts, but it's driver stage can drive the output into class A2 territory of ~30watts without clipping or higher distortion.

[DaveC113]

I have an amp that makes 6-7 wpc, more using UL mode... with 93 db speakers I have more than enough power. The system will hit 105 dB peaks (with the sub) before any audible distortion... at the listening position, in triode mode. A vast majority of the time, peaks in the mid 90 dB range are all I require, which the amp will provide with 2-3 W. UL mode makes enough power for listening at loud levels in the rare times I want that much volume. The only (very rare) exception is music with a lot of extreme low frequency or subsonic information. I only have a 10" sub and 4.5" single drivers that run full range, and the sub is x-overed at around 50 Hz.

[Scott F.]

I've got a fair amount of experience with low powered amps driving speakers of varying sensitivities in varying room sizes.

First let me say there are no absolutes when it comes to mating amps to speakers and then to rooms. You must take into consideration how loud you listen. If you like it loud, you must get the proper balance between sensitivity and amp power. On the other hand, if you listen at moderate volumes you open up your speaker and amp choices.

I listened to an 8 watt KT88 amp triode strapped driving 88db speakers in a 10x11 room for months and never felt it lacked enough power even for loud listening sessions.

In my other two larger rooms I listen to either a 2A3 or 300b both SET's driving either Altec A7s or Lowther PM2As (cut off at 125hz in an OB). Both of those systems will easily hit 105db peaks....and I'm sorry but this completely flies in the face of the little girl with guitar theory. In fact put my 25wpc JoLida in front of the Altecs and I can hit 110 db peaks if not higher.

I don't mean to be combative but the notion the SETs can't play loud without clipping and are only good for intimate music is just plain wrong. If you don't believe it, climb on a plane and come to my house. I'll be happy to demonstrate all I've said to you by dropping the needle on some NIN, Rob Zombie, Snoop Dog or any other band you want.

It's all in the implimentation.

[lcrim]

Josh;
Your view that low wattage amps can't produce high outputs w/o clipping or audible distortion is just not correct.  I agree w/ Scott that matching your amp with speaker efficiency and the room size as well as your listening habits has more to do w/ a sucessful system than raw output.  I have a small Decware amp (EL84 strapped triode in my small (13'X15'X8') bedroom and its 1.8 watts will drive the Parker 95dB speakers to far louder than my ears can stand.  Another EL84 amp also strapped triode in my living room, the Glow Audio Amp One with all of 5 watts per side can drive the Omega Compact Hemps to measured peaks over 100 dB.  I've never encountered clipping or compression in complex orchestral music w/ either system. 

[Niteshade]

Sources such as DVD movies have a wide dynamic volume range. Going from a whisper to an earth-shaking BOOM! isn't uncommon. I have a pair of Klipsch RF-83's that are around 98db efficient and find that they play well with amps as low as 1-4 watts. However, that doesn't cut it for DVD action movies. About 10 watts a side and up does well and 30 to 50 watts/channel seems perfect for any program material I can throw at the system. The RF-83's will easily take high wattage without distorting too. A good tube amp, to me, anyway, sounds significantly better than a THX rated receiver even if only two channels are being used. Yeah- that's subjective, but the part about having enough power to run under highly dynamic volume scenarios isn't. At least with a big amp there's options. I build small amps and they're appropriate for many scenarios. It's all about getting the right tool for the job. I do not subscribe to the notion that small amps can have a more accurate 'first watt' response. Large amps can work just as well. It's all in the design.

[JoshK]

I just wanted to clarify.  Firstly, it is not my view that low watt amps with efficient speakers still isn't enough.  I don't have a view, as I stated I don't have that much experience.  However, the experiences that I have had were that the flea amps noticeably clipped in the listening sessions we were in, and this on 102db/w speakers. 

To date, I can think of 3 sessions where lower watt amps were used with efficient speakers.

1) SET 45 amps into OB Fertin speakers (~96db iirc) supplemented with 2x 15" woofers per side (woofers powered with SS amp).  Very noticeable distortion when playing Tool at modest levels.  Sounded amazing on Jazz and Hawaiian music.

2) as said before, Avantgarde Duo's.  Played a SE 45 amp, SE 2A3 amp, PP 2a3 amp, SE 300B amp and Nagra PP 845s.  The Nagra was the best of the bunch and the only one I heard not clip or sound highly distorted when playing more complex music.  The 45 SET was noticeably clipping quite often.  The 2A3 SET clipped less often but sounded quite distorted still when playing loud.  The PP 2a3 sounded "off" so it didn't get much air time.  I was out of the room for most of the 300B SET listening, but walked in near the end, so I can't really comment too much.

3) Rave at Jim's place with his Tannoy's.  The main amps being played were 30wpc Redwine and a 30wpc Jadis KT88 PP triode amp.  Their was noticeable distortion quite often when pushing the speakers loud, specifically when playing Tool (I like to torture systems).  I didn't notice any blatant clipping though. 

Again, I am holding my judgement until I hear more setups.  But so far I haven't heard such a system that didn't have some limits, even if the limits aren't often met.

I am building a 10watt SET amp, mind you so I am not close-minded on this issue. 

I am building a few different amps with different power ranges.  I can choose the amp for the music I want, or the mood I am in.   So I wouldn't rule out the low watt amps, as some of the low watt amps had glorious sound with the right volume and/or music.

[Niteshade]

I wanted to add a note for those who like low powered amps: Push-Pull amplifiers can be made without negative feedback, which causes high frequency rolloff and various other colorations. Sometimes, I admit, there is a reason to use it. Most high grade single ended amps do not have any feedback circuits and 90% of the Push-Pull amps made do (in order to make the amps work with the largest gamut of speaker systems). There's been a trend to install feedback on/off switches and that's a great idea.

I __really__ want to stress that push-pull systems do not insinuate lower quality standards. There does not have to be a power vs. quality tradeoff.

[DaveC113]

I __really__ want to stress that push-pull systems do not insinuate lower quality standards. There does not have to be a power vs. quality tradeoff.

Thats may be true, but a PP amp will cost (a lot) more than a SET using the same quality parts. The increased circuit complexity usually indicates a reduction in SQ vs a SET, but I'd agree there are too many variables to make a blanket claim.   

[Niteshade]

The cost per watt for a push pull system is considerably less than for a single ended amp.  My goal in designing any kind of tube equipment is to eliminate as many signal barriers as possible. As a matter of fact, I've made push pull amps that have less circuitry than some single ended ones.  Naturally, you're 100% correct that the fewer circuits there are within the signal's path, the better it will perform provided that nothing crucial to proper operation is excluded. Eliminating the negative feedback circuit itself reduces the parts count dramatically.