diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Sound quality of amps at 0.3 watts vs @ typical 5 watts (http://www.diyaudio.com/forums/solid-state/54190-sound-quality-amps-0-3-watts-vs-typical-5-watts.html)

rick57 25th March 2005 01:10 PM

Sound quality of amps at 0.3 watts vs @ typical 5 watts
 
Nelson Pass addresses +/- 95db speakers with First Watt (with a SS class A approach). :)

Im close to entering a new paradigm, building high efficiency speakers (100 dB) mounted in horns, to give about 105 dB, ie + about 10 dB on high efficiency speakers. Ive read remarks like this:
FACT: The first few watts are always the juiciest. In every amplifier those first few watts, or milliwatts, are the best sounding watts, which is why it is so desirable to keep an amplifier operating in that range. This explains why, for decades, our Japanese bros have preferred high efficiency speakers...especially for their transistor gear. (Harvey Gizmo)

In this situation, 15 watts is plenty. :cool:

* How does the relative sound quality of amps at 0.3 watts differ from the relative sound quality at 5 watts?

* How can we assess the sound quality of amps for 105 dB speakers, auditioning with in most situations, average 90 dB speakers?

Cheers

KarlDL 26th March 2005 02:02 AM

Rick,

Here goes one person's perspective. Others will differ.

I believe that what most of us seek is, first, aural transparency, and, second, graceful degradation, when the first is beyond realization. Transparency is maximized when the inherent nature of the audio signal is fully preserved. If such is impossible, then we want our amplifier to degrade in a sonically-acceptable manner.

This principle presents substantial peak voltage and current demands on amplifiers feeding <90dB-efficient speakers. It's safe to presume that, if an amplifier is capable of passing peak signal levels 20dB greater than "average" (or RMS) measured audio levels (on highest musical crescendos), it's rather "transparent" and degradation is a minimal issue. When the efficiency drops below 90dB and you want to "listen loud", the required peak power capability for absolute transparency quickly exceeds 500 Watts!

Audio enthusiasts address this problem along two paths. Some seek higher speaker efficiency, while others seek graceful amplifier degradation. The latter is inherent in the best vacuum tube implementations and the more esoteric MOSFET solid-state amps. Few seek boundless amplifier power, as excursions down that path have generally led to blind specsmanship and ignorance of crossover and other distortions at levels far below the clip points of the circuit.

I'm a (broadcast) systems engineer by day and that's the approach I take to my audio activities. I will define the audio level range I prefer to listen at (in SPL terms), translate that to amplifier output power (Watts or dBW) via a speaker's efficiency, then add 15-20dB of headroom to determine the appropriate amplifier output power rating. If the latter exceeds the bounds that I consider reasonable, then I will seek more efficient speakers.

Inherent in this approach is a presumption that the amplifier is transparent within its rated output power range. All of us can cite classic examples of products not sounding very good at lower power levels. But that's just not true for better product.

The other approach is to trust only ears and find the most sonically pleasing combination. This is laudable, but highly subjective. Perhaps one chews up more experimental time than would occur if an initial cull based on systemic principles were used.

It is the experience of listening - with speakers of moderate efficiency - that drives the conclusion that "the first few Watts matter most". That makes sense from a systems-engineering perspective, as well, as the amplifier remains at such levels, on peaks, for most non-head-banging listening.

If the amp performs well in the single digits on the bench and has either (a) adequate headroom or (b) graceful degradation, then it will provide a pleasureable listening experience with speakers of moderate efficiency. Otherwise, it's grungeland.

A near-certainty is that vacuum tube and single-ended MOSFET amps will always improve their sound as abosplute output level is reduced. More conventional solid-state amps may well exibit "valley" behavior; sounding deficient at low and high levels and best in between. This is why one should match the SS amp to the speakers' efficiency and desired listening levels carefully.

I will always seek awesome headroom, with rated power output of the amp slaved to speaker efficiency. Others will seek graceful degradation, probably involving lower power-capability and MOSFETs or "valves". Neither is the sole truth. Listen to many combinations and pick your preference.

rick57 26th March 2005 07:25 AM

Interesting comments Karl

I agree 100% that listening to the options is the best way.

However I am going to China in May, and have an opportunity to get a SS/ class A or tube amp there and then at very low cost. Or I listen later and maybe make a Pass design.
The speakers wont be built & XO sorted by May, so Im going to best estimate my most likely best answer, and if not a Pass design, take advantage of the low prices - I may not be back there for 5 or 10 years.
If Im a little wrong, Ill keep it; if Im a lot wrong, Ill sell it, and probably not lose money, in fact from what I hear about prices there, I could make a small amount, to offset the hassle.
> speakers of moderate efficiency - drives the conclusion that "the first few Watts matter most"

I have chosen higher speaker efficiency. Theres also the fact that the amp will only be driving > 250 Hz, which may easily halve(?) power requirements. 5-10 watts is enough, so the first 500 milliwatts matter most

> A near-certainty is that vacuum tube and single-ended MOSFET amps will always improve their sound as absolute output level is reduced.
> More conventional solid-state amps may well exhibit "valley" behaviour; sounding deficient at low and high levels and best in between.
What have those of you with 105 dB speakers found best?
> All of us can cite classic examples of products not sounding very good at lower power levels. But that's just not true for better product.

So what Im asking here, is which SS/ Pass design averaging the use of maybe 0.2-0.8 watts, and max 5-10 watts, might/ should sound good/ best?

Ie What is the better/ best design for those *low levels?


All times are GMT. The time now is 06:10 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2