In article ,
wrote:

Do you concur that the difference is in the IC ?

I think that's the likeliest thing. The two IC designs probably have
different "phase shift, to output voltage" sensitivities. The tank
circuit which was used with the original quadrature detector (a
two-coil type) is probably not quite what the new detector IC was
designed to use... and so a slight change in the tank Q might be all
that you need.

You might be able to research this if you care to. Look up the specs
for the tank coil with the AN377 used in the receiver. Then, look up
the specs for the tank coils suggested for use with the NTE chip, or
with the Sanyo or similar single-coil detectors. With lots of luck
and effort you might be able to compare the coil specs or sample
circuits and see where the differences lie.

Or, just try a swamping resistor and see if it makes the
difference... if so you'll know where the problem lies.

The print is available the Hifiengine if you'd like to have a look, or I
could Dropbox it. There is another coil coming off the tank.

Yeah, that was one of the significant changes in quadrature detector
designs over the years - getting rid of the need for the second
(usually fixed-value) inductor in the tank circuit.

The tank
has two adjustments, but only two connections. Kinda makes me think
adjusting that would affect the gain (output per deviation) but it
doesn't seem to be working that way.

No, I don't think it would... or, at least, not a big one. The two
adjustable cores (or trimmers) in the tank would adjust the resonant
frequencies of each part of the tank... but I think that any change in
the Q would be relatively minor. The Q for each of the two halves of
the tank transformer is going to be set by the values of its reactive
components, compared to the (external) resistor which is shunted
across them. To make a big change in the Q and phase shift (which is
what I think you need to do) without changing the resonant frequency
(which needs to remain at the IF frequency) you'll have to play with
the resistor.

What really gets me is the quality of the front end in these old tuners.
The new stuff seems to be junk.

A lot of people agree with you. There's a great deal of enthusiasm
out there for the "golden age of analog" FM tuners... expecially those
with real air-variable capacitors. A lot of 'em have better
sensitivity and selectivity than you'll find in modern products (in
which the FM section is almost an afterthought). Modern receivers
often tend to emphasize DSP and effects and 587-channel operation, and
cut corners on the basics :-(