neck tube temperature sensor

[roger smith x294]

Roland Reiss asked....

>Which type of sensor do you use and how exactly do you glue it
> to the tube?

It;s nothing fancy.  I use a simple 1N914 forward biased at 10 uA.
We calibrated a bunch of them from a single batch at the boiling point
of Nitrogen and at the melting point of water and found that they varied
by only about a degree or two from the mean.  The variation is primarily
"gain" rather than "offset" giving greater errors at room temperature.
We verified a few to be quite linear, and don't bother checking that any
more.  These are cheap and simple.  The cost is in the calibration but
we did a whole lot all at once to be more efficient.  If you try this
be sure to put the sensor in a plastic bag to keep it dry when "immersing"
it in water since the conductivity of tap water is not negligible.

The epoxy was thermally conductive, electrically insulative and vacuum
safe.  A tiny drop is applied to the neck tube on the vacuum
side just above the N2 flask and just below the point it exits the dewar
so that it is about 3cm from the warm end.  A twisted pair runs down to the 
detector mount where we have made a simple keyed socket from stackanle SIP
sockets.  Sorry - I am in Tucson at present and don't have the epoxy
at hand to check what it is.  It's blue!!

The sensor is glued *directly* to the neck for simplicity and good contact.
It has never separated due to the thermal shock from nitrogen hitting the
other side of the wall, which I think is remarkable.

See you in Garching in a few weeks.

Roger