In article , David
wrote:
Personally, I'd prefer to listen to music, not to any added distortion
at audible levels - irrespective of its order for THD. :-)
Just a thought - I have always found that power supplies in "high" (more
than 50 watt) powered amps. are able to drive the amp to just over the
quoted output (for a given distortion level) then run out of steam. All
the "low" power amps (8 and 15 watts / channel) I have owned have had
comparitively massive power reserves from the PSU which means that they
will drive way above their rated output (when abused) but with rapidly
increasing ammounts of distortion (and heat).
My experience is that it is not quite as simple as the above.
The main point of having a high power amp is that you should then never
reach clipping. This being the case, it should become irrelevant to the
user how much dynamic headroom the amp has.
I've not noticed any specific correlation in this with rated power. However
I can see that when the rail voltages approach the limits for the devices,
then the designer is tempted to use stabilised rails. When you do this, the
dynamic headroom may be reduced - although as indicated above, this simply
may not matter in use. If you are going to voltage clip a 200Wpc amp with
no dynamic headroom, then you will probably clip most 15W amps even if they
have some headroom. Once you clip, the results cease to be hifi. Must admit
I'm not personally persuaded that we should factor in any idea that an amp
may be better if it "sounds better when clipping". :-)
For a 15W rated amp to match a stabilised-rail 200W rated amp it would have
to have quite a high dynamic headroom value. Off-hand I can't think of any
that would do this. Can't recall any values, but my vague recollection is
that with floppy PSUs the values for headroom I've seen were only a few
dBs. Nothing like 6dB if we are comparing short-term rms values.
FWIW I have personally tended *not* to use stabilised or 'stiff' PSUs in
high power amps. Thus in my case I'd expect a '200Wpc' amp to deliver
somewhat more than this for short periods of time, and have a distinct
dynamic headroom. For the reasons you give, I do prefer an amp to have a
much higher short-term power/voltage/current ability than its continuous
rating - even when rated 200Wpc. :-) For similar reasons, I am less than
enthused by any active limiting/control protection circuits. If they
operate it implies you need a bigger amp, or they are needlessly in the
way. 8-]
Low power amp meeting a transient peak may, or may not, suffer the same
degree of TID but will almost certainly show far fewer power ripple
effect following the peak and will maintain composure.
Again, my experience is more complex than the above. :-)
If the amp clips, the waveform is distorted. Having more power available
may help avoid this.
If the amp has the required slew/current ability for the load, and has a
standard input network, TID should not arise. The designer should have
dealt with this.
If the amp acts correctly and tends to reject rail fluctuations (below
clipping) then it should not show much ripple under load. Again, the
designer should have dealt with this.
None of the above in my experience correlate much with the power rating if
the amp is reasonably well designed. (Except, of course, that greater power
ability may help avoid clipping.) I'd agree that making a 'good' high power
amp can be more of a challenge, though, as it requires better components
and more care.
One of the reasons I have always liked the sound of old Sugden amps is,
I suspect, due to the massive over-spec'ing of power supplies in P50s,
A48s etc. (although they had poor channel tracking at low levels which
made for frustration when trying to listen (very) quietly) TID, THD -
which is more obvious in the real world?
TID should not arise if the amp is correctly designed and used. Despite
Otala and others I must admit I found TID to be something of a red herring.
Can't recall much about the early Sugden class A amps. However they do have
a problem with limited currents, and with their early o/p devices were not
IIRC entirely unconditionally stable. Could work very nicely within their
limits, though.
I find ALL solid state amps to have a degree of "grain" in the sound -
Possibly tube amps just bury the grain in the output t/former? (havn't
really listened to t/formerless valve amps)
Again, my experience seems different to yours. :-)
Slainte,
Jim
--
Electronics
http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc
http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio
http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc.
http://www.st-and.demon.co.uk/JBSoc/JBSoc.html