Lake Shore 350, Руководство пользователя

Страница: 36 / 192

22 c
HAPTER
2: Cooling System Design and Temperature Control
Model 350 Temperature Controller
No measurement cable of any length should be run from the Model 350 without a shield.
The shield provides a low impedance path to measurement common that prevents the
voltage change at the source from being seen at the measurement leads.
H-field noise inductively couples into measurement leads or other conductive loops.
Current is induced in the lead as if the noise source and lead are two halves of a trans-
former. The amount of coupling is related to the field change from the source and the
loop area. Tightly twisted leads inside the shielded cable reduce the total loop area
and minimize the effect of H-field noise. It is also effective to move the instrument
and leads away from noise sources because field strength reduces by the inverse cube
of distance.
Even with proper installation it is impossible to eliminate induced noise. High resis-
tance ranges are affected more because the leads create a more efficient antenna
when they are terminated at a high resistance.
2.5.2 Ground Loop
Noise Ground loops produce noise that acts similar to induced noise but the cause of the
noise is different. Current is generated in any conductive circuit or loop when that
loop contains changing current or magnetic field. In measurements this problem is
commonly referred to as a ground loop and generally results when improperly
grounded instruments are exposed to AC line current. Isolation reduces ground loop
noise by breaking the loop or circuit. The Model 350 has a well-designed power sup-
ply and uses optical isolation between the instrument chassis and measurement cir-
cuits. Careful installation is necessary to take advantage of the isolation.
2.5.3 Analog Circuit
Noise The current source and amplifiers inside the Model 350 make a contribution to mea-
surement noise. Circuits were carefully chosen to maintain good noise performance
over a wide resistance range without generating high leakage currents that could
self-heat sensors. Instrument noise, specified in TABLE 1-3, is dominated by amplifier
noise. When high resistance sensors (>10 k
)
) are measured the noise measured
becomes dominated by the thermal noise of the resistance. The thermal noise of a
sensor drops as the temperature is reduced, and at low temperatures the sensor ther-
mal noise becomes negligible, so high resistance sensors at low temperatures are not
impacted by the sensor thermal noise. The instrument noise (measurement resolu-
tion) listed in TABLE 1-3 was verified with a sensor cooled to 4.2 K, so that the result-
ing instrument noise best fits what will be seen in an actual measurement. Noise
measurements of high value resistances at room temperature will be significantly
higher than the noise measured at low temperatures.
2.5.4 Digital Circuit
Noise All digitally controlled instruments emit some high-frequency noise both through the
air and on their leads. Emission from a Model 350 is very low and will not affect sen-
sor measurements if the instrument is installed properly. The high frequency noise
can interfere with very sensitive measurements being made in the same Dewar. Inter-
ference can be minimized by separating the instruments and their leads outside the
Dewar and placing ferrite beads on the Model 350 measurement leads, near the
Dewar, to stop the noise from getting inside. Wrapping leads through the large beads
more than one time improves their attenuation of unwanted signals. Lake Shore sells
a clamp on ferrite bead as part number 2071.
A more common source of digital noise than the instrument itself is digital computers
used for data acquisition. Computers with their high clock speeds generate a wide
spectrum of noise and transmit it through line cords and interface cables. Isolation in
the Model 350 separates the measurement circuits from interface circuits but com-
puters must still be treated carefully during installation of a system.