how close in sound can a tube and ss amp sound ?

Yup!

All you have to do is clip the amp and you will know the difference between tube and transistor. That's the tip of the iceberg and the rest is already well 'documented'.

You will find many transistor amps that are lauded as having 'tube-like' qualities. But so far no-one is acknowledged as having found a means of causing a transistor amp to actually sound like a tube amp.

The bottom line is quite simple: if you want the sound of transistors don't get a tube amp and if you want the sound of tubes don't get a transistor amp.

Much discussion could ensue but that will be the gist of it.

The issue is going to be the distortion structure. Tubes in general favor the lower orders and depending on the amplifier technology can make a lot or a little of the even orders, usually with not a lot above the 5th.

(This is a general statement, for example our amps are fully-differential so there is no even-ordered harmonic at all until you approach clipping.)

Transistors for the most part are absent the lower orders. Instead they favor the odd orders and higher ones at that, usually the 5th, 7th and 9th. Now they do this only in trace amounts, but many tube amps don't make them at all.

The odd orders are heard as brightness and also makes the presentation seem louder.

The lower orders are heard as warmth, bloom, that sort of thing.

So either way you look at it, distortion is what we are trying to get rid of and as far as the tubes/transistor thing goes, its two ways of doing that with a different price. Fortunately topology plays a role too, so its not cut and dried- IMO the variations we see are what make the high end part of the industry so interesting.

Mrtennis, Human hearing is all there is in audio. It is what is perfect, it is the audio equipment that is not.

IOW, the ear/brain system has certain rules it follows. The closer the audio equipment comes to following those rules, the more it will sound like music rather than electronics.

Now in the case of odd ordered harmonics, which as at the crux or our discussion, the ear happens to be very sensitive. In fact, this is one of the things that the ear is *most* sensitive to. So its to your advantage to not distort the odd ordered harmonics- the result will be a harder, brighter sound if you do. Now to emphasize the point: the ear is *so* sensitive in this area that modern distortion instruments have trouble measuring the area that is important- often the distortions are buried in the noise on the test bench.

It happens though that tubes tend to make less of these distortions overall, although a **lot** depends on the topology of the amplifier, and so if a designer knows what he is doing, he can do it with transistors too- its just a lot harder. To avoid this distortion, you have to keep the circuit as linear as possible, without loop feedback.

This is a difficult task, but you can see that if you really want to be about it, pentodes and transistors will have more in common than triodes and transistors will. Certain semiconductors are quite linear- approaching that of triodes, and to make the best use of them the circuit will have no feedback and will be class A. There are not many transistor amps like that. Those that are are either very expensive (+$100,000) or very low power (Nelson Pass First Watt).

Now its a lot easier to design the odd ordered distortion out of the amplifier if you work with triodes. I am not saying that they are superior, I am saying they are easier. But the fact of the matter is that there is a lot more experience in the field working with triodes in this way, so you might want to consider going the other way (towards triodes, away from transistors) if you are looking for the best performance with your Quads.

Audiofeil, in this case Mr. Tennis is right in some circumstances. If the amplifier is being clipped only on peaks, its behavior with that clipping will have a lot to do with its sound.

If you have a transistor amplifier and a tube amplifier of equal power and gain, such that either is clipped on peaks only, what you will find is that if the tube amplifier is built right, instead of outright distortion or excessive brightness, it will simply seem to compress right before audible distortion onset.

If for example the speaker is 90 db, and the amplifier 50 watts, this clipping behavior can be pretty important as there can be plenty of opportunity to clip the amp. It is situations like this that have led to the idea that tube amplifier power is often thought to be greater than that of transistors.

Audiofeil, my point is that many people indeed *do* compare clipping behaviors, only thing is, they don't know that that is what they are doing. Clipping can occur without the amplifier being outright distorted. Tubes have a much more graceful clipping behavior, which accounts for the idea that we have all seen before that 'tube watts' seem to be more powerful than 'transistor watts'.

Mr. Tennis, the range that you are talking about is the realm of the odd-ordered harmonics that I mentioned earlier. It is very challenging to build a transistor amplifier that does not produce a coloration at these frequencies, due to the way our hearing perception works. On paper you will find almost any transistor amp compared to almost any tube amplifier to be perfectly flat in this region.

IOW, we hear distortion as tonal coloration. That is why two amps can measure flat and sound so different. Now there is another reason, read this:

http://www.atma-sphere.com/papers/paradigm_paper2.html

Quads are a Power Paradigm technology, which means if you used a Voltage Paradigm technology (most transistors) on them, you will get a tonal anomaly, in this case brightness.

Audiofeil, my point is that during that comparison, no-one is thinking that they are comparing amps that are clipping. They don't know that that is what is happening. If you have a speaker that is 89 db and a 35 watt amp, depending on room size and listening habits, **inaudible** clipping could be going on a lot.

The effect of inaudible clipping could be a change in tonality before outright clipping is detected by the listener. Or it could be perceived as compression, perhaps a strained quality of the amplifier during complex passages. It depends on how the amp is built, how robust the power supplies are, that sort of thing.

Mr. Tennis, I did give one example earlier- the Nelson Pass First Watt. Another would be the Ridley Audio amplifier. Last I heard that one was over $100,000. I might have mentioned this before too- making a transistor amp that does not distort odd ordered harmonics is not easy.

Maybe you should write one of your white papers to make it easier for the those of us living in the forest to understand. How about:

"The inaudible effects of clipping when comparing 2 amplifiers that aren’t clipping but would if you pushed them to clipping levels and the resultant perceptions of the listeners who infrequently listen at clipping levels".

It would be more about the 'audible effects of inaudible clipping'. However many have beat me to the subject, some by decades. The earliest I have read about this comes from Norman Crowhurst in the 1950s.

So far the closest I have heard a transistor amp to sound like a tube amp is one that sounded so neutral that it would be best to say it sounded like music the way a good tube amp sounds like music, without much in the way of artifact.

That amp is the Ridley Audio amplifier.