In article , Chris Isbell
wrote:
I am currently working on changing the electrolytic capacitors in my
Quad 33/303 amplifier. After nearly thirty years of daily use, they are
probably at or beyond the end of their useful life.

So far, I have replaced the four large electrolytics in the power
amplifier. (Two are for power supply smoothing, connected in parallel,
and two are for decoupling the speaker output.) These were last replaced
in 1985 after the original components leaked and damaged the amplifier
board underneath. Renewing these parts has produced a marked improvement
in the sound quality. (The replacements have larger capacities than the
originals - from 2,000uF to 4,700uF for smoothing and 10,000uF for
decoupling.)

You may already be aware of what I'm about to say, however two points occur
to me which you might like to bear in mind. I can't recall the details of
the 303 off-hand, though.

The larger smoothing caps may mean a higher peak current through the power
diodes. Might be useful to check their rating, and perhaps uprate them to
avoid the risk of a failure at some point due to high current inrush at
switch-on.

Changing the output cap value may alter the details of the electronic
resonance between this capacitance and the speaker inductance. In general,
this may be 'good news' but this will depend upon the circumstances.

The next job is to replace the remaining 'small' capacitors in the
preamplifier and the earlier stages of the power amplifier. Having spent
a evening looking through the RS and Farnell catalogues I am looking to
use solid aluminium capacitors wherever possible. (Farnell sell Sanyo SG
series 'audio grade' devices.) Where suitable solid aluminium devices
are not available I am considering 'ultra low' ESR electrolytics with
low leakage, primarily intended for switch mode power supplies. (Typical
part numbers are 664-492 and 769-794 respectively.)

Comments on these choices would be appreciated.

Better grade of caps is probably fine. One point may be to check if the
caps have a can that is connected to one lead, or if it is 'floating'. This
may affect field coupling to components around the cap. I've encountered
examples (not in a 303) where this affected stability, etc.

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