Instrumentation & Measurement Magazine 26-3 - 49

Fig. 3. Feedthrough losses for different metals (Cu, Al, Fe, Ti) as function of
the feedthrough cross-section Ac
(I = 10 A, lc
= 100 mm, ΔT = 220 K).
cross-section Ac for an assumed current of I = 10 A, a fedthrough
length of lc
= 100 mm and a maximum temperature
difference of ΔT = 220 K.
A notable point about these losses is that they become minimal
by optimized choice of the Ac/lc
ratio, as seen in (2). Taking
(1) and (2) into account, the minimum achievable power dissipation
of a material (3) depends on the product of the specific
electrical resistance of the material and its thermal conductivity.
They are thus independent of the feedthrough length.
Minimum Cooling Temperature Depending on
Test Specimen Losses
When measuring test components or power converters, the
power dissipation will result in an additional heat input in
the testing chamber, causing the minimum cooling temperature
to rise. To measure this loss-dependent increase, starting
from the lowest cooling temperature of Tca
trated in Fig. 4, a power dissipation PV
= -191 °C, as illusis
applied via a power
resistor in four steps and the resulting minimum cooling
temperatures T1 and T2 as well as the case temperature Tc
of
the power resistor are recorded. For this measurement, the
power resistor is placed in the center of the testing room and
thus flowed around by the cryogenic nitrogen gas. As seen in
Fig. 5, the measured cooling temperature shows only a minimal
increase of a few degrees up to a power dissipation of
25 W.
For a power dissipation of 100 W, such as a 10 kW power
converter with an efficiency of 99%, the minimum achievable
cooling temperature in the testing room is Tca
= -175 °C. To determine
the heat transfer coefficient from the case of the power
resistor to cooled ambient αca
according to (4), the case temperature
Tc of the power resistor is required in addition to the
cooling temperature Tca of the testing room. As shown in Fig.
4, for a power dissipation of PV
of the power resistor is Tc
= 100 W the case temperature
dissipating surface of the load resistor A ≈ 130 cm2
transfer coefficient from case to ambient results in about αca
65 W/m2
forced convection.
May 2023
= -55 °C. Taking into account the heat
, the heat
≈
K. This value is a typical heat transfer coefficient for
Fig. 5. Measured cooling temperature Tca
IEEE Instrumentation & Measurement Magazine
(crosses) with trend line in the
testing room vs. additional power dissipation of test specimens.
49
Fig. 4. Measured minimum cooling temperature Tca
temperature Tc
1.59 mm).
 
ca
PV
AT T ca
c

Operating Options of the Cooling Chamber
for the Investigation of Test Specimen at Low
Temperatures
Depending on the sample to be characterized, the testing
chamber can be operated in two different options. For
the low-temperature characterization of power converters
with non-negligible power dissipation, an operating mode
with successively adjustable ambient temperature of the
test specimen from room temperature to the minimum cooling
temperature is recommended, as shown in Fig. 6. In this
mode, the cooling temperature is kept constant over a specified
soak time to allow the test specimen to reach thermal
steady-state at each test temperature before a measurement is
started. Neglecting the selected soak times of 20 minutes, the
minimum cooling temperature of -190 °C is reached after approximately
70 minutes if a maximum temperature gradient
of 4 K/min is set. The cooling temperatures at the measuring
points T1 and T2 in the wind tunnel are almost the same.
(T1, T2) and case
of the DUT due to adding power dissipation (capillary diameter:
(4)
Instrumentation & Measurement Magazine 26-3

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 26-3
