A blog post

Music: Tubes and transistors, once more

Posted on the 19 April, 2011 at 8:27 pm Written by in Music

Question: will a high quality 2 x 80 watt mono block tube amplifier, based on 211 tubes, perform better with big difficult speakers like Dynaudio Consequence (a 5-way system), than the Krell FPB600 transistor amp?

Short answer: no.

I tested, borrowing a friend’s tube mono blocks, each weighing 50 kg or so, due to big oversize transformers. These were hardcore 80 watts, the amps should be able to drive difficult loads, or be much like a higher number of official transistor watts, say 150 to 200.

I noticed some interesting things on the way.

First, the tube amps would not get a decent volume in my ca 45 m2 room until I put the preamp volume at four o’clock – with the Krell, it reaches this level already at one o’clock. I first thought, well, this goes to show how power-hungry the Consequences are, but I also tested with a small Royd Sorcerer speaker, and the story was much the same there (or, to three o’clock). The amps need a stronger than usual input signal, which by itself should not influence the sound quality much. Possibly, my IO preamp sounds best cranked up, as long as it goes beyond 11 or 12 (with the Krell), it sounds fine.

I measured the decibel level in the listening position, music came alive around 75-85 decibel, much the same as with the Krell, although this required a higher than usual preamp volume.

Second, the tube amps are the first I have tried that seriously makes an attempt to make the Consequence speakers work. My small Ming Da 38 watts stereo tube amp does not work very well, making the speakers sound strange and thin.

However, even if “the hand” (amp) was now large enough to try out the “glove” (speaker), it was not a very good fit. The 80 watt tube amps did not “hold” the speakers in the way achieved by the Krell, not surprising, since the latter has perhaps x 2 or 3 in objective power level (hard to say exactly, many variables involved).

This was most noticeable on dynamic large scale music, but also on attacks, transients etc. The speakers sounded somewhat hollow, and the music somewhat strained.

In fact, after some hours of testing, my wife sat down to listen, and after an hour, concluded, “no, this does not work out, the sound is bad”. She used words like “hard, straining”; I would also say, “flat”.

Third, the experiment cast the Krell in a good light. Ken Kessler, reviewing the FPB 600 amp in Stereophile in 1997, was right when he said that the Krell was almost indistinguishable from good tube equipment in some key areas. The 2 x 80 tube monoblocs sounded remarkably “Krell-like” in suprising ways, for me, – quite analytical, much more frequency-linear and less “tubey” than my small Ming Da tube amp (less of a bell curve). The were less sweet than the Krell!

I have read reviewers writing of the convergence of transistor and tube, but have not experienced it for myself. The tubes sounded a bit harder and more analytical than the Krell – but this may be due to the fact that they were not quite powerful enough to drive the speakers properly, and perhaps also that the 211 tubes were not optimal. A main symptom was the hard, strained sound at high music volumes, and another that voices were not pronounced by the tube amps, as they should have been, but were instead more hollow, as if lost in the mix. A likely cause is power limitation, too few watts to drive the speakers properly, fill the “glove”.

One of my conclusions is, that the next time I suspect the Krell amp for adding transistor-related types of pollution to the sound, I should pause, and look for other factors instead.  The transistor hardness was not much notable in the test.

Why do I take the trouble of this kind of test? Heavy hifi is, well, heavy. Changing cables, carrying equipment, adjusting and listening takes time. This is by intention my hobby only. I have no work connection or commercial interests in the audio field.

One reason is that I often hear various kinds of noise and sound pollutions. I listen to music and play some, as a musician, and since my positioning in music is as participant, not just spectator, I can’t help looking for what to blame, or what could be set (more) right. This is “the audio bug”, but it has also helped me and my family to get much more out of music than we would otherwise have achieved.

If I suspect the Krell, should I be more careful, perhaps the source is somewhere else? A possible broader conclusion has to do with – in audiophile terms – component matching and system synergy. If you drive speakers right, they will sound less hard, more dimensional, improving the imaging, and so on, everything will fall into place.  You should test a component in its optimal system setting.