Home Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

[ Follow Ups ] Thread: [ Display  All  Email ] [ Propeller Head Plaza ]
[ Alert Moderator ]

Re: Why SETs?

Hi Tube guy,

I follow you and you are mostly right. However, you have to realize that there are quite a few mediocre to bad SET amps out there; just as there are many bad SS and PP tube amps. The problem is one of experience and hearing a really good one to prove the point.

So far the best amp I have heard is the KR audio Kronzilla monoblocks. These were parallel SET with no feedback but they are hybrids, ie. tubes only on the output. They have a jfet input and mosfet driver to the tube output. Of course the KR tubes are special in that they are new designed tubes to specifically move current at audio frequencies. So while they look like transmitter tubes due to their size in practice they are not. Now I am playing with the KR VA350i. It is a 30 watt integrated with the KR T100 tube. It is about the size of an 845 but can deliver a bit more power.

On top of sounding great they measure good. A SET that has been thoroughly optimized can have a good intrinsic linearity.

The other issue is keeping the amp with in a power range with which it is happy. For most SETs, even high powered ones, this is in the 10-15 watt range. Above this and you see high order harmonics creep in and the sound can harden. However, and this is what most people don't realize, you will almost never use this much power with a decent sensitvity speaker except on music peaks, where the loudness will mask the increase in the distortion (as long as its a mild increase).

The issue is really how linear is an amp at low power. Forget Audiohobbies comments on measurements. What he doesn't know about how to interpret measurements could fill a stadium. Take Soundstage measurements. WHile they can be pretty valuable, one has to realize that the harmonic spectrum measurement they make at 1KHz is done at about 10 watts of power. Now for a SET amp this is substantial and it is no wonder that sometimes they show a lot of harmonics. What also has to be noted then is how steadily the THD of a SET amp drops with decerasing power. Now at 1 watt the THD is often every bit as low as an SS amp but without all the feedback induced problems and the crossover distortion problems etc. etc. If they measured the spectrum at this power it is likely a very different picture in favor of the SET. The point is that they don't normalize the test to account for the basic limitations on power that a SET has.

I went to a McIntosh demo where they wanted to "prove" why you need their monster new 2000 watt three chassis (per "monoblock") amplifiers. It was an all Mc chain, including their big 95db/watt linesource speakers. It was in a big ballroom and we were sitting at least 6 meters from the speakers. They started off with some Mariah Carey!! Why? Because it was the most compressed music the guy could find. It had virtually NO dynamic range as the big blue needles on the Mc amps showed. At a comfortably loud level the needles read about 2 watts. That's right 2 watts. Here were 2000 watt amps (8000 peak) idling along at 2 watts and it was plenty loud enough. Now I haven't seen the measurements but I am sure the amps measure fine...especially near their max power. But what about at 1 watt? Probably the same but this means the RELATIVE distortion is higher at low power...the opposite of a SET. Many big SS amps have a decreasing %THD with increasing power or it is flat right up to clipping.

Then the demo guy put on a very dynamic recording of some classical with a loud gunshot or explosion (not 1812 cannons though). Through most of the music it was running at well under 1 watt!! Then the explosion and the needle almost hit 2000 watts! Here the guy said that you need the power to handle such dynamic headroom. However, the duration of this peak is for sure very short and so all this idling power is for what? 10 milliseconds of use? What was more interesting to me was that most of the time the music stayed at a very low power. THese speakers could have been driven to very satisfying levels with a 20 watt SET no problem. THat same peak would have made the amp badly clip; however, with no negative feedback the clipping would be instantaneous and then back to business as usual. You might not even hear it. Nearly every amp in the world would have clipped on that shot but so what?? Who cares if an explosion is reproduced cleanly or not? It might even make it more spectacular if it does. IMO, the McIntosh demo showed the exact opposite of what they were trying to prove.

It showed: 1) that most people's music (ie. rock and pop) is heavily compressed and requires very little power to play loudly. 2) With wide dynamic range material the average power required is very low and as long as big peaks are short then running out of power may not be very sonically deterimental. 3) Having all of this headroom is largely unnecessary and money is better spent on impoving the sonics of the design. 4) Since nearly all amps (except for these absurd McIntoshes) will clip at high peak level uncompressed music then how well an amp recovers from clipping is of the utmost importance. Typically a zero feedback amp will recover much faster than a high feedback amp.

This post is made possible by the generous support of people like you and our sponsors:
  VH Audio  

Follow Ups Full Thread
Follow Ups
You can not post to an archived thread.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]