Pros and cons of valve amps

[anthonyTD]

And I respect that.

I'm afraid our agreement parts company there Anthony I will say q=1/2cv^2 though A big enough PSU on a SS amp nullifies it though...

[Arkless Electronics]

Higher voltage rails... Yes, that was the one! I can't comment on the technical points you've made, save to say that, like you, I'm convinced that there's more going on, in that respect, than soft clipping, which is confirmed by my extensive listening experience in that area, and using a multitude of valve and SS amps over the years

It would be interesting to hear Paul, Nick, Alan and/or Andrew's opinions on this matter, or indeed any of the other registered 'techies' on the forum

Marco.

It would be great to have some more input yes. So long as we stay technical and away from subjectivism. Technical fact can be tested, measured and proven. As soon as it gets purely subjective all bets are off.... anyone can claim anything, no matter how ridiculous and without any basis in fact and "anything goes".

1A X 1000V = 1000W just as 1V x 1000A = 1000W so on the face of it high voltage cannot be it. However, the way in which valve amps use the voltage could have a link. High output impedance and poor damping etc means that valve amps are much less of a true Voltage source than SS amps. Hence they can't do the 50W into 8R, 100W into 4R, 200W into 2R that a really meaty SS amp can do. BUT, the SS amps output will not rise at all into a higher impedance neither. It can to some extent though with some valve amps and hence if a speaker is rated 8R but rises up to 20R at some frequencies (not unknown at all) and especially if this is in the mid range where our ears are most sensitive, the valve amp could produce over its rated power at some frequencies in a speaker dependent way. When operating under these conditions the distortion is likely to be really high as well, probably 10%+, and the distortion could help make it sound louder.

The waters are muddied on this by the fact that the valve amps most likely to soft clip in a big way, and sound subjectively louder due to this, are the same ones as are most likely to exhibit the effect I mentioned and the ones generally with the worst overall measured performance!

On the other hand a Radford with its high NFB and excellent measured performance can be expected to be most SS like in all this.

Another thing needs mentioning here. Most valve amps are class A and have huge power supplies relative to their output power. If we are comparing a 25W valve amp with a 25W SS one then the 25W SS amp needs to be something like a Mark Levinson 25WPC class A SS amp to give an even playing field.

Anthony, is the above where you're coming from on the high Voltage bit?

[montesquieu]

The Meishu is a lovely sounding beast. So you rate it as warmer sounding than the Oto SE?

s.

With the right speakers the Meishu can sound nice. With my Tannoy 15in Monitor Golds it sounded absolutely dreadful. (No damping factor worth speaking of, high output impedance - all that you don't want with a speaker that needs to be controlled like a big Tannoy).

[Marco]

It would be great to have some more input yes. So long as we stay technical and away from subjectivism. Technical fact can be tested, measured and proven. As soon as it gets purely subjective all bets are off.... anyone can claim anything, no matter how ridiculous and without any basis in fact and "anything goes".

So we've all just to shut up while you guys sit and talk in a 'foreign language'?

It's ok, I know what you mean. It would definitely be interesting if we had more input from the 'techies' here, but if I read something that I can relate to through listening experience, and happen to agree or disagree with it, then I'll comment accordingly!

Marco.

[Marco]

With the right speakers the Meishu can sound nice.

Who the fuck wants "nice"? Get yer slippers on then and light your pipe. Certainly NOT me... I *demand* REAL!!

Marco.

[anthonyTD]

Damping factor is another reason that is often put forward as the trump card, in favour of sold state amps over valves.
And yes, I agree, if we were feeding a perfect speaker load, the higher damping factor of SS amps would indeed be a major advantage, as far as bass and controling unwanted cone movement.
But think about this for a moment; most real world speakers have something in the way of the drivers that the amplifier will see first, in the case of the Woofer, or bass unit, you will have an inductor, and in the high frequency unit, or Tweeter, you will have at least a capacitor, and almost always a resistor.
Therefore; it can be seen that the huge diffrence in damping factor figures between valve amps and Solid state, becomes much less of a concern when considering the final load, ie; the Loud speaker!
As far as other reasons why valve amps tend to sound' watt for watt more powerful, and imediate than solid state amps of similar power output, consider this; A valve amp does not require negative feedback to enable it to function. The primary purpose of Negative feedback is to quieten the amp and provide some damping to the speaker. And infact' In comparison to SS amps, the amount of negative feedback that can be applied to a conventional valve amp circuit is quite small, any more than is necessary will push the circuit into oscillations and instability.

A pushpull Tetrode configured valve amp, [similar to Marco's amp] with no negative feedback, will attempt to provide a constant current at the secondary of the output transformer. As the Impedance of the speaker increases so does the Voltage from the output transformer. therefore; The output Voltage from the valve amp will rise in an attempt to maintain constant current to the speaker. If an imaginary speaker had no internal Reactance and behaved as a perfect 8ohm resistor, across a desired frequency spectrum, there would be IMHO little auditory difference between a valve and solid-state amp. All cone speakers are highly Reactive and their Impedance varies over the frequency spectrum, Therefore, because Valve and solid-state amps operate in opposite ways to how power is delivered into a changing Impedance, the resultant audible difference is very noticeable. Therefore With most cone speakers, many solid-state amps appear to sound flat and lifeless in comparison to a valve amp of similar power output etc.


Valve amp; power is directly proportional to speaker Impedance, therefore' power increases as the speaker Impedance rises.


Solid-state ; power is inversely proportional to speaker Impedance, therefore' power decreases as the speaker Impedance rises.

[montesquieu]

Who the fuck wants "nice"? Get yer slippers on then and light your pipe. Certainly NOT me... I *demand* REAL!!

Marco.

The point I was making Marco is that the Meishu can sound 'nice' ie unobjectionable in an undemanding setting, but sounded absolutely shit with the Tannoys. A design like that is kneecapped by its limitations for the majority of real-world situations - something utterly predicted, as it happens by how it measures.

My experience has been that if something measures well, it generally sounds prety good at least within its design parameters. If something measures badly it will usually show up at the ears in short order.

Neither your Copper Amp nor my Radford was, as Jez put it, designed by 'ear' but instead are the product of solid design, measurement, and high quality bits where there are required.

Where I take issue is some of the kit around the place (especially the stuff coming out of China recycling proably ill-understood old circuits) that is designed on voodo principles rather than solid engineering. Got to be single-ended, got to have no NFB, got to be triode, and so on and so on. Yes, it's possible to make a good amp this way, with with expensive enough transformers and power supplies. But most of the amps following this fashion are, frankly, a bit shit and unless used with 100db horns or other kit of that ilk, this becomes apparent very quickly.

Valve vs SS, I've settled on valve but I'm open to SS as an approach, I'm particuarly interested in how Class D is evolving ... I started another thread asking 'is it there yet', I'm not sure it is but perhaps it's not that far away.

[Arkless Electronics]

So we've all just to shut up while you guys sit and talk in a 'foreign language'?

It's ok, I know what you mean. It would definitely be interesting if we had more input from the 'techies' here, but if I read something that I can relate to through listening experience, and happen to agree or disagree with it, then I'll comment accordingly!

Marco.

Marco, My subjectivism, and as I keep on saying I am a subjectivist when it's justified, is in just this sort of thing. We KNOW something is going on we CAN measure it BUT we can't entirely equate what we measure with what we hear. This is what I would call "normal" subjectivism and it is essential if we are to correlate what we hear with measurements and thereby move the whole game forwards. It's the subjectivism of "voicing" an amp, the subjectivism of hearing a cart you know to be superb but after re-installation it's sounding off and you know the VTA, tracking force whatever is off.

Where we're from different planets is in things like fuses where we can easily scientifically prove that it does diddly squat, even if we measure to 6 decimal places of accuracy, and where all theory and common sense also says it could not possibly have any effect

[Arkless Electronics]

The point I was making Marco is that the Meishu can sound 'nice' ie unobjectionable in an undemanding setting, but sounded absolutely shit with the Tannoys. A design like that is kneecapped by its limitations for the majority of real-world situations - something utterly predicted, as it happens by how it measures.

My experience has been that if something measures well, it generally sounds prety good at least within its design parameters. If something measures badly it will usually show up at the ears in short order.

Neither your Copper Amp nor my Radford was, as Jez put it, designed by 'ear' but instead are the product of solid design, measurement, and high quality bits where there are required.

Where I take issue is some of the kit around the place (especially the stuff coming out of China recycling proably ill-understood old circuits) that is designed on voodo principles rather than solid engineering. Got to be single-ended, got to have no NFB, got to be triode, and so on and so on. Yes, it's possible to make a good amp this way, with with expensive enough transformers and power supplies. But most of the amps following this fashion are, frankly, a bit shit and unless used with 100db horns or other kit of that ilk, this becomes apparent very quickly.

Valve vs SS, I've settled on valve but I'm open to SS as an approach, I'm particuarly interested in how Class D is evolving ... I started another thread asking 'is it there yet', I'm not sure it is but perhaps it's not that far away.

Good post Completely agree with that.

[Marco]

Marco, My subjectivism, and as I keep on saying I am a subjectivist when it's justified, is in just this sort of thing. We KNOW something is going on we CAN measure it BUT we can't entirely equate what we measure with what we hear. This is what I would call "normal" subjectivism and it is essential if we are to correlate what we hear with measurements and thereby move the whole game forwards. It's the subjectivism of "voicing" an amp, the subjectivism of hearing a cart you know to be superb but after re-installation it's sounding off and you know the VTA, tracking force whatever is off.

Completely agree, and out with of designing equipment, I use precisely the same form of subjectivism, when assembling sonically (and musically) rewarding hi-fi systems. Virtually none of my current system has been evaluated or chosen on measurements. I simply used my experience, gut instincts and ears, and that approach has been successful and served me very well for the last 30-odd years

Where we're from different planets is in things like fuses where we can easily scientifically prove that it does diddly squat, even if we measure to 6 decimal places of accuracy, and where all theory and common sense also says it could not possibly have any effect...

Well... Only if it's 100% certain that you're measuring for the RIGHT thing, using the RIGHT apparatus, which is actually capable of providing the results needed!

In my view, that's not always the case with what's often ignorantly dismissed as "foo"... We need to be cleverer, and look into things much deeper than only that which is immediately obvious [confirmed by currently accepted wisdom], or simply what appeases our 'belief system'/sensibilities (scientific or otherwise)!

Anyway, we're veering wildly off-topic, so let's leave that subject for another day

Marco.