Colin Stamp wrote:
On 30/09/10 14:11, The Natural Philosopher wrote:
Adrian C wrote:
On 30/09/2010 01:40, The Natural Philosopher wrote:
Adrian C wrote:

It's not a MOSFET amp.

well I am not surprised it blew up. Pretty amateurish design.

FWIW I've found an old Maplin catalogue from 1985 with the original
project. (The days when the catalogue had projects and data sheets
printed inside that you could actually dream & build from!)

http://www.scribd.com/doc/38419958/50W-HiFi-Power-Amplifier-Maplin

First paragraph of that reads ...

"A superb quality 50W power amplifier. We threw away all our technical
specification handbooks and designed an amp that just sounded
musically perfect. When we'd finished we found that we'd got a pretty
impressive technical spec as well."

except temperature stability ;-)

It seems stable enough. I've driven it pretty hard at times, including
80W RMS per channel continuous sinewave into 8R dummy loads for half an
hour or so when I first built it.


The design is utterly conventional expect the temperature coefficient of
the output stages is highly positive, which is rather dangerous.

I can write bull**** like the above too.

I am a bit worried about it's hum rejection and switch on thump
characteristics as well.


Distortion and frequency response will be relatively good, for the day,
but when you do this for a living, that's just the start of the game.
There is no overload protection whatsoever apart from the fuse, no
thermal cutout, no SOAR protection, and I am not convinced that they
have got the hum as low as they should..nor arranged for the feedback to
start working before any serious power is delivered to the speakers, to
eliminate 'thump' on switch on.

There's just a faint click at switch-on, and the power supply rejection
is fine. With the input shorted, I have to stick my head right into a
speaker to hear anything, then it's just some hiss as I recall - no hum.
Mind you, It's been a long while since I stuck my head in a speaker with
the input shorted.


The last ever 150W design I did had everything. Down to the point where
even in the hot African sun, it would deliver into any load we threw at
it, for as long as the loudspeaker cones lasted, was fully short, half
short, open circuit protected, and was on the ragged edge of heat before
the thermal cutout went.

It had no switch on effects at all, because there was no current
delivered to the outputs beyond a mA or so in order to stabilise the
feedback before any serious power was applied to the speakers.

It was about my tenth or eleventh design for high power professional
use. After that I ended up designing software instead.

I cant remember the specs, but short of MOSFETS it was as good as any
bipolar design on the market.

Mine has even less protection since I've shorted out the output fuse in
the name of lowering the output impedance and I've up-rated the power
supplies. I like to sail close to the wind )


Removing the fuse will not affect the output impedance whatsoever.

That is a function of the feedback in the amp, and is likely to be well
below the resistance of the RF stabilising choke and the loudspeaker
cables themselves.

Cheers,

Colin.