Tube power vs solid state power

[sprinter]

This is a great forum. I'm glad I ran into it. My question is going to seem naive for a guy with an electronics background, but it's driving me nuts lately. I'm an acoustic guitar crossover and don't know a lot about guitar stuff (it's really pretty bizarre frankly), but I find it very interesting.

So everybody says that a tube amp has more power than a SS amp with the same RMS power rating. Anybody want to explain that to me? I don't doubt that a 20 watt guitar amp sounds as loud as a 40 watt SS amp, I've seen it myself, but I don't understand it.

Thanks

Ron

[R.G.]

They really produce the same available power out, but the human ear hears the earliest bit of distortion from the tube amp as "louder". The SS amp goes from really clean to very noticeably distorted at a sharp threshold. The tube amp produces early distortion products that fool the ear into thinking that the sound is louder, not distorted.

[sprinter]

They really produce the same available power out, but the human ear hears the earliest bit of distortion from the tube amp as "louder". The SS amp goes from really clean to very noticeably distorted at a sharp threshold. The tube amp produces early distortion products that fool the ear into thinking that the sound is louder, not distorted.

Thanks. It got me thinking more about tube dynamics, A tube will continue to produce power beyond the clipping level (assuming adequate power supply), when a transistor just clips hard. Makes sense.

Are tube amps rated at a low THD like conventional amplifiers are? If that's the case, then at high drive levels the tubes would still be producing power well beyond their rating but with a lot of harmonic distortion, and that's a good thing. Guitar electronics is truly a bizzarro world...(but fascinating)...

[Steve Conner]

Yes, that's basically it. The good brands of tube amps were rated at the maximum unclipped output power, but there's about another 30% of distorted power on top of that, that is musically useful. Well, for rock music at least. It wouldn't be much good to an acoustic musician.

Designers of solid-state amps didn't know how to make their creations sound good overdriven until quite recently, but it is possible and then the "tube watts" thing no longer applies. Peavey started it with their Transtube technology, but now even hobbyists are doing it. I just saw a terrifying thing, a clone of a Plexi made entirely out of MOSFETs by some Russian guy. It is half the size of the real thing.

[Enzo]

As long as you understand that the claim is that a tube amp of the same power rating as a SS amp will SOUND louder, you'll be OK.

No one claims that 1 watt sounds louder from a tube amp than a SS amp. So it is not that 20 tube watts are louder than 20SS watts, it is that a "20 watt tube amp" is louder than a "20 watt SS amp." Basically you can turn them up farther without the sound turning to crap.

[sprinter]

MOSFET'S ? What's that doing, some kind of class D thing? Curiouser and curiouser...

[Steve Conner]

http://milas.spb.ru/~kmg/slpfet_en.html Here, figure it out for yourself.

As far as I can tell, it's normal Class-AB. The guy developed a MOSFET building block that emulates a tube, and he uses one for each tube stage in the 1959SLP. It even has an output transformer.

He's either a genius or completely insane, or both.

[sprinter]

Fet version of the Marshall Superlead Plexi 40W Here, figure it out for yourself.


He's either a genius or completely insane, or both.

Sometimes it's very fine line...

[Wilder Amplification]

Ah yes...the long debated (and much dreaded) "valve watts vs solid state watts" discussion.

RG is spot on. Watts are watts and they both put out the same actual watts. However, valve amps tend to accentuate musically pleasing frequencies that the human ear is more sensitive to which gives us the perception of "valve watts being more than SS watts".

Furthermore, the power rating of an amplifier is the "clean power" it will put out...i.e. the maximum power output with zero clipping. They put out much more power when they're clipping.

The difference is that when we drive a valve amp into clipping, it's sounds very musical and we love it, so we crank it more. However, with SS amps they sound very harsh when driven to clipping so we have a tendency to not want to turn them up so much. So the SS amp would only ever get ran at or just below its clean power rating whereas valve amps are routinely being ran well into clipping where they put out more than their clean power rating.

That's my theory on it anyway.

[JoeM]

Many years ago some attempt was made to quantify the apparent loudness of Hi-Fi amps that had the same rated (RMS?)power. From what I recall, the idea is that of two similar rated amps, one may put out short term peaks of greater amplitude, so that amp would actually play louder on music signals. This is based on the idea that a music signal isnt just a constant amplitude sine wave, but consists of short term peaks. I think they measured using pulses in the ms range. For example, a 15W amp that can put out 3db above rated power, would sound as loud as a 30 watt amp with no dB above rated power. I've always figured that our typical Tube guitar amp then must be capable of putting out short term peaks greater than the typical SS amps of the same power rating.

[sprinter]

I knew I asked this in the right forum. Good answers.

To complicate the issue a bit more, in a linear, flat response amplifier (e.g. stereo amp) driven only linearly, is there any advantage sound-wise with a tube amp? I know this an age-old debate. My current take on this issue is that in the "olden days", transistor amplifier technology probably did sound inferior to a good tube stereo amp, but that a modern well designed SS audio amp should be just as good in sound quality at the speakers. But I'd like to hear more experienced, objective opinions on this.

Ron

[Steve Conner]

Yes, this is an age-old debate, like life after death, what exactly the government does with all that tax money, etc.

Tube hi-fi amps have a lower damping factor than solid-state ones. This means that the frequency response when used with a speaker is no longer flat: it looks like the speaker's impedance curve. Usually the result is a bass boost and a presence boost, which can sound subjectively better, but it's not "hi-fi" in the sense of accuracy. With some speakers it can sound plain bad. This is why the single-ended triode weenies like their Lowthers and other single-driver speaker systems: the impedance curve is more benign, so they sound acceptable with low damping.

Tube amps have more distortion, which is supposed to sweeten the sound and make it more musical, or something. But if you have a 30W tube stereo, a comfortable volume might correspond to about 1-5W of power. You're not really going to get much distortion at that low level, unless you're one of these single-ended triode weenies who loves what it does to Diana Krall's voice, or whatever audiophile approved pap he is listening to. If you like to crank drum'n'bass or metal really loud, though, you would prefer a solid-state amp.

I have both, a 120Wpc MOSFET amp that I built years ago, to the Alexander current-feedback topology which was fairly modern at the time, and a 30Wpc Crown tube amp, 1960 vintage, that I restored. Either will go loud enough to piss off my neighbours without any gross distortion. To be honest, I like the sound of the MOSFET one a little better. The Crown gets a little mushy at really high volume, but the MOSFET amp will go as loud as you can stand, and still stay completely clear, with great bass until you reach the point where the woofers themselves can't take any more. But it's an ugly beast with a noisy cooling fan, so it got banished to the kitchen.

Solid-state amps don't necessarily sound bad when clipped. It's easy to end up designing one that does (my MOSFET amp sounds horrible) but it's not mandatory. I just finished building a 40W hybrid guitar amp, with an EL84 driving a pair of output transistors through a transformer. You can really clip the output stage quite hard without even noticing.

It's mostly held back by the little 8" speaker. Plugged into a 4x12 it would be deafening.

To address JoeM's point: Yes, this is true in a sense. Tube amps have bigger power supplies than solid-state ones of equal rated output, because tubes are less efficient. The power supply needs to supply the rated output, plus all the wasted power. This gives them a little extra headroom that is normally never used. You can't see it when testing with a dummy load, but they can drive more power into a high impedance load, like a speaker at resonance, or up in the presence band where the voice coil inductance is starting to limit things.

Again, that would make the tube amp seem louder, fuller and brighter. You can achieve the same by just giving your solid-state amp a bigger power supply with higher voltage than normal, but that costs money which isn't reflected in a larger wattage figure for marketing.

[sprinter]

Many years ago some attempt was made to quantify the apparent loudness of Hi-Fi amps that had the same rated (RMS?)power. From what I recall, the idea is that of two similar rated amps, one may put out short term peaks of greater amplitude, so that amp would actually play louder on music signals. This is based on the idea that a music signal isnt just a constant amplitude sine wave, but consists of short term peaks. I think they measured using pulses in the ms range. For example, a 15W amp that can put out 3db above rated power, would sound as loud as a 30 watt amp with no dB above rated power. I've always figured that our typical Tube guitar amp then must be capable of putting out short term peaks greater than the typical SS amps of the same power rating.

Hey, is that where that misleading marketing term 'music power', as in "100 Watts of pure Music Power!!", came from? I haven't seen that used for a long time and never did know where they got that from.

[JoeM]

I dont know. But even Fender used the term "music power". For example, in a Fender catalog I have (1972), the Deluxe Reverb has "20 watts RMS, 44 watts peak music power".

[sprinter]

Steve Connor: Nice answer. Thanks. I never thiought about the speaker's inductive load having a large impact on the power stage. I guess if you wnat to be very careful, you would have to select a good amp/speaker match. That could be a challenge.

I started this thread because I'm getting a Fender Mustang III modeling guitar amp, which sounds pretty good with it's own 12" celestion, but it has stereo effects and a stereo line out in front of the power stage which I want to amplify into a pair of good speakers. So I'm researching the whole business because I don't know anything, but I only want to do it once. The problem is that I love the big sound of 15" speakers even at low volume, so finding a good speaker set at a low enough price for me may be a problem although I have no problem buying used. Also low power is fine; As I always say, I don't need big power, just big sound.

I'm not an audiophile but I could sure turn into one. I just got a pair of Sennheiser headphones after my old Sony's wore out and I couldn't believe what I had been missing! I thought the Sony's were pretty good, but these stunned me. Now I'm reasessing my whole 5.1 setup in addition to the guitar thing.

I'm open to any thoughts on either. You guys are a really helpful bunch here.