In article , Andy Evans
wrote:
Good morning, Jim, and indeed a fine sunny one.

Here also. :-) The weather people on TV/radio keep telling me that today
is cloudy and colder. I am beginning to suspect that none of them live
anywhere near us or bother to look out of the window! :-)

[snip]

However for the same reason I might personally tend to prefer a SS
design precisely because it might give less scope for component
variations having an audible effect.

This is interesting, because I'm sure many people like valve amps
because they CAN make audible changes by swapping componants. This gives
them an interactive role in their audio passion, which they appreciate
since it gets them more involved. I'm sure I don't need to point out
that 'interactive' is one of the buzz popular appeal subjects of recent
years.

In itself, that seems fair enough. However it does concern me for two
reasons.

One is the reason I have already implied in previous postings. That it may
mean people experiment for long periods and never really know *why* one
arrangement 'sounds better' than others, or if a slight change might be
markedly 'better'. (I am putting 'better' in commas here to indicate that I
am keeping the definition vague beyond meaning 'preferred by the individual
involved'._

The other is that if the reasons for the preference were understood, it
might mean 'better' sound could be obtained more reliably, and with a much
lower cost and effort.

The obvious example for me is the effects of an amp having a frequency
response that is not flat, and an output impedance that interacts with the
load to further alter the response. If these factors produce what it
desired, and the user is aware of this, then similar 'tonal changes' could
be easily provided on a better basis with no need for the expense and
effort implied by having to experiment to develop your own valve amp, nor
the risks of instability or unsafety.

Alas, although valve amps and low feedback are 'in vogue', things like
'tone controls' are largely regarded as the spawn of the devil. This seems
a pity to me as - decently implimented and carefully applied - they can
often improve the sound (in my opinion/experience).

with your detailed construction notes which prove that substituting
componants has no effect on the sound.

How could a set of notes 'prove' this?

Exactly, I was being ironic. I thought my ironies were more transparent!

OK. :-) I must bear in mind the USA dictionary definition:

Irony = contains some iron. :-)

I'd be interested to know how you verify or test your own perceptions
in this area

Being 'lost in space' is part of the creative process

Yes. Can also be a PITA though... :-)

[snip]


For a production model clearly more verification and safety testing
would be required, but I'm strictly talking DIY here. The experienced
amateur builder would, however, not be happy with 'sounds good'. He/she
would want to know as much as possible WHY a componant change made a
difference, and further than that how it could be repeated in different
circumstances. One way to do this would be to build an infinite number
of prototypes and make careful notes. This is exactly what most
inventors did - e.g. the Wright brothers with their many prototype
'planes.

Yes, many inventors have proceeded that way. However although at times this
is unavoidable, if it continues it implies an 'unscientific' method and can
become very wasteful of time and effort. May also mean that better
solutions are missed due to simple lack of understanding.

However when they built a powered heaver that air flying machine they
didn't have much in the way of succesful previous examples to analyse. Nor
did they have the analytic methods we do nowdays. This situation is
somewhat different to what is available to anyone who wants to develop a
*commercial* amp these days. People have been making audio amps for yonks.
All kinds of designs. And the books and journals like JAES are filled with
reams of analysis and results. Seems a waste to ignore them. Although I
agree that at times there is no substitute for simply trying things out for
yourself with no preconceptions. Need a balance here...


At a later stage, hopefully, would come theory and equations
(as with aviation models) but the DIY builder may largely work by trial
and error. Not to be dismissed out of hand - it has its merits.

Agreed. Keep doing things that way myself. Seems a perfectly valid part of
the 'mix' when appropriate.

If I need a set of scope measurements and distortion tests for a circuit
I go down the road to my designer friend who designs commercial products
and has a proper bench setup. I have a scope but using it is still on
the 'to do' list of projects. So my verification is partly subjective
and partly intelligent trial and error and careful notes of results
achieved, voltages, componants, etc, tests which I do with a multimeter
and analyse in Excel, using equations derived from books like Morgan
Jones, Menno van der Veen etc and data on Duncanamps (valve data, power
supplies) and other audio sites. I also have a valve tester to test the
valves I use to see how they perform, so I at least know that.

Above seems fair enough in general terms to me. :-)

In my experience it is one thing to have an impression that a change
has had an effect. It is something else to test this over a period of
time in a way that can reliably rule out uncontrolled variables

This is true of so many things in life - I'm sure we could think of
numerous examples. I wouldn't even try to claim I'd eliminated all
uncontrolled variables, though one does try to use as much methodology
as one can.

My key worry here is the one of 'uncontrolled variables'. I know that
'unconscious bias' is also a worry, but since most of my own mind seems
'unconscious' I'm not on firm ground arguing about that. ;-

My problem is that I am aware of all kinds of factors which cause
variations in the sensitivity, etc, of my ears, variations in speaker
behaviour, variations in the acoustic properties of the air, slight head
movements, etc, etc. As a result I am very wary of simply 'using my ears'
unless it is in a carefully controlled and repeated way, and ideally with
some blind cross-checking with others. This is particularly the case when a
design might be commercial or for use by others. The problem here is not
'trusting my ears'. but of being able to be confident as to the reasons
*why* something did (or did not!) make me feel there was an audible change.

I have on many occasions not noticed changes which later became evident. Or
heard changes that later seemed to evaporate. Hence I am wary of factors
which affect this and which were not controlled and obvious at the time.

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