GE Resistor

Ge resistors are specifically produced for low-temperature thermometry carbon and Ru02 resistors are commercial products for electronics. Pure carbon is not a semiconductor. The negative slope R(T) is due to the production process which consists in pressing and sinterization of carbon particles with glue. The resulting resistance is probably determined by the contact resistance between the particles. The cost of the carbon resistor thermometer is very low. Manufacturers such as Speer, Allen-Bradley and Matsushita have produced in the past carbon resistors for many years. Most of firms have now ceased manufacture, although their products may still be found in the storerooms of research laboratories. 

See, for example, Unemployment of Scientists and Engineers, GE Resistor 3 (January-February 1971) 1-2. 

Fig. 9.5. The electrical resistance R(T) for some typical thermometers. A-B denotes Allen-Bradley carbon resistor. Speer is a carbon resistor. CG is carbon-in-glass. CX 1050 is a Cernox and RX 202A is a ruthenium oxide from LakeShore. Ge 100 and Ge 1000 are Cryocal germanium thermometers [45]. 

Due also to their (amorphous) composition, the heat capacity of a ruthenium oxide resistor is much higher than that of a Ge thermistor of equal mass [61]. This negative property prevents the use of Ru02 resistors as detector sensors (see Chapter 15). 

Fig. 9.8. I-V curve for a Ge-NTD 31 (see Section 15.2.1.1) resistor at four different temperatures TB of the mixing chamber. Note the strong nonlinearity at low temperature.

However the carbon specific heat was not as low as the crystalline materials employed later. Also important, the resistor material exhibits an excess low-frequency noise. Nowadays, two types of resistance sensors are used to realize LTD NTD Ge sensors and TES. 

Figure 15.8 shows the thermal scheme of one detector there are six lumped elements with three thermal nodes at Tu T2, r3, i.e. the temperatures of the electrons of Ge sensor, Te02 absorber and PTFE crystal supports respectively. C), C2 and C3 are the heat capacity of absorber, PTFE and NTD Ge sensor respectively. The resistors Rx and R2 take into account the contact resistances at the surfaces of PTFE supports and R3 represents the series contribution of contact and the electron-phonon decoupling resistances in the Ge thermistor (see Section 15.2.1.3). 

Nevertheless the heat capacity of a carbon resistor was not so low as that of crystalline materials used later. More important, carbon resistors had an excess noise which limited the bolometer performance. In 1961, Low [61] proposed a bolometer which used a heavily doped Ge thermometer with much improved characteristics. This type of bolometer was rapidly applied to infrared astronomy as well also to laboratory spectroscopy. A further step in the development of bolometers came with improvements in the absorber. In the early superconducting bolometer built by Andrews et al. (1942) [62], the absorber was a blackened metal foil glued to the 7A thermometer. Low s original bolometer [61] was coated with black paint and Coron et al. [63] used a metal foil as substrate for the black-painted absorber. A definite improvement is due to J. Clarke, G. I. Hoffer, P. L. Richards [64] who used a thin low heat capacity dielectric substrate for the metal foil and used a bismuth film absorber instead of the black paint. 

The optimum operating temperature for minimum noise from the FET was found to be slightly higher than the operating temperature of the Ge(Li) detector. A thermal heater made from a 100-ohm carbon resistor was therefore installed next to the FET and operated at a current that heated the transistor to a temperature just above that of liquid nitrogen. 

At the request of Brattain, Pierce39 coined the name "transistor," an abbreviation of the neologism "transfer resistor." Later, Teal40 and Sparks41 made the first Ge BJT at Bell Labs in 1950, and Teal made the first Si BJT at Texas Instruments in 1954 [22], ushering in the "silicon age."... 

Carachteristographers are d.c. instruments made up by a current (voltage) power supply and a voltage (current) meter. They give a two quadrant V-I relation for the resistive device. The information supplied by such instruments is more exhaustive than that supplied by a.c. bridges, but the measurement is time consuming, and an inversion of polarity is necessary to detect the possible presence of partially rectifying contacts. In Fig. 9.8, an example of a V-I relation in the first quadrant for a Ge-NTD 12 resistor (see Section 15.2.1.1) is shown. Note that the power used in the measurements can be as low as 10-14W. 

Of course V r has the same frequency dependence as Johnson noise v ge. This noise voltage is due to the equivalent resistor R of the detector and amplifier, and is given by... 

---