snip my own stuff and a couple of inane comments by others,


Professor, What you described was a horror story for a layman. With all said
and done, how do the bloody contraptions sound?


The sound is "interesting", goodish, and I will see what my customer
thinks when we
sit down to a couple of CDs next Saturday using his 120W Musical
Fidelity A3 amplifier.

But they are not loudspeakers, just quietspeakers.
An amp designed for 50 watts into 5 ohms doesn't have enough
voltage headroom produce enough level.
The limiting factor is bass levels.
if the bass panels had twice the area, much less membrane excursion
woulod be needed,
and this would allow the FRED effect, full range electrostatic driver.

I think perhaps if I abandoned the need for deep bass production below
100Hz
on these units, I could safely apply twice the signal, and get a lot
more volume.
But then my customer needs the extra bass speakers.

If anyone buys the kits as is, I'd recommend they take the stators to a
guy who can cut out shapes in sheet plastic, and have them
create perforated 0.8mm thick plastic PVC sheet to the same pattern of
the steel stator holes.
This could then be glued well on the inside of the stators to "hold off"
the membrane
from trying to stick to the stators.

But I wouldn't stop there.

I'd use a larger stator to membrane distance, same as Quad ESL57,
and use similar materials.
Trouble is by the time I describe all the things one SHOULD do to make
the kits world leading kits,
one wouldn't bother to buy the kits at all, unless ER Audio incorporate
what I am saying.
I suspect they won't, perhaps because of the stocks they want to sell.
They won't alter everything just because some self taught
audio lunatic in Canberra wants to tell them how they ought to design
their product.
What I am suggesting is rarely ever welcomed as something to fertilize
minds
to make something better grow. Starting about NOW.

The Quad ESL57 treble arrangement has 1/2 the distances of the bass
panels, and much lower EHT and 1/2 the signal
drive levels, so the mid/treble panel is smaller, lower C, and really,
methinks its a better design for treble.
However, the clearance distances and EHT could be higher, like in Quad
ESL63,
but then the signal drive must be higher.

There are 5 things that are basically balanceable against each other,
area of the panels,
bandpass of the panels,
distances between stators and membrane,
signal voltage drive,
EHT levels.

Coating of the membrane is critical, but easy to do once youv'e done a
couple.

The design of the step up tranny is critical, and the higher the voltage
ratio,
the higher the Z ratio, and Quad's ESL57 potted and waxed SUT is a
lesson to us all about
how to do things right.

In ESL63, Quad made things easier for themselves by using TWO trannies,
one for
each phase applied to stators. Its vitally important
to keep insulation thickness between interleaved windings as thick as
possible
to curtail shunt capacitance which appears at the input as Z ratio x
Cshunt at the sec.

Braun made ESL57 variety in a panel size that was 1.5 times as high as
the old '57.

This is 1/2 way to having a stacked pair of 57.
ppl who have stacked ESL57 seldom complain about bass being missing,
so why not just have a similar design to '57 that is about 1.5M high,
same width.
The work involved isn't much more than with a single panel equal in area
to '57.
Capacitance rises though when the area increases, and thus its harder
for some amps to drive them.

But careful SUT design avoids the worst of the C.
The only panel needing to be directly connected to the SUT sec
is the small treble panels.

The real enthusiast would never buy any kit, he'd draw up his design
and make his calculations, after learning enough, and just do it.

Apart from the SUT, the expenses for materials for two large sized
panels
can easily be less than $300.

Getting anyone to wind the SUT with perhaps 10,000 turns of 0.2mm dia
wire
for the secondaries is a bother, so the enthusiast will make a lathe and
wind his own.
This is what I have done about OPTs.
Using a large core of about 5Kg with large window area permits less
turns,
so less self capacitance of windings, and less leakage L
and thicker P-S insulation.

The SUT for ESL is very much like an OPT, except that its step up, not
down,
and that the V ratio is much higher, so the capacitance must be very
low.

The material costs in a 6Kg tranny are about $15 per Kg max, or $90.
The rest is labour, so for the enthusiast who refuses to go to pubs,
watch TV,
dither about on the PC all day and night, he will find the time
as was done in the past to do it all himself.

We have become a lazy plug and play society.

When I was 29, I built all my own furniture, a double garage
which is a fine workshop now, all in my spare time after work
in a regular day job. I also found time to build a decent pair of
speakers,
I was BURSTING with energy i could only be proud of, and
all the idiots at the pub and stupid TV shows at night meant nothing to
me.
I doubled the size of the house I bought in the next few years after
that.

My energy and clear minded approach probably saved me
an enormous sum of money paid to others.

Patrick Turner.





















west