New H-field rx is coming along well.  Design changes are
completed - it has a lower system noise, wider bandwidth,
significantly lower common mode response, and a much better
line driver.  Like the new E-field rx, it also has a built-in
calibration port.

Currently testing the design changes on breadboard and it
performs just as per the Spice model, except for a higher than
expected system noise.   Should be just below 0.1fT in 1Hz
but I'm measuring 0.46 fT.   Maybe this specimen of LT1028
is a bit noisier than others?  It's my last one but I've
got more arriving hopefully today.   I'll pick the best two.
Even with this mysterious extra system noise, it's still 20dB
below the natural noise floor after sferic blanking.

Apart from that it is working perfectly, just as per design
and looks very good at 137 kHz too.

Hopefully will be able to start the cutting and drilling and
filing and soldering this week.  That's my least favourite part
of the job.

What I won't discover until it's finished and installed, is
whether it actually cures the excess noise at 8kHz and 5kHz
that we've seen lately.  I haven't been able to identify the
cause of the relatively (compared to E-field) poor S/N on the
existing rx.  Hopefully it is a low level of intermod in the
line driver.  If so the new rx will not suffer that problem
and we'll see another step improvement in the 8270 spectrogram.

--
Paul Nicholson
--