摘要:

The most outstanding and useful application of the reversible thermal and electrical effects, which occur at the junctions of certain semiconductors, is the application of the Peltier effect to small-scale cooling. The theory of Peltier cooling is outlined, together with methods of producing couples, or multi-couple units, from the basicp- andn-type thermoelements available commercially. Simple measurements can be made during construction to ensure that units perform in a predictable manner. To illustrate the particular problems involved in thermoelectric measurements, an account is given of rapid-check methods and of the more sophisticated techniques used to ascertain the quality of thermoelements and couples. The design of a manually-variable power supply capable of producing smooth direct current for a range of composite cooling units is considered, with a view to obtaining a compact reliable unit by using transductor control. A constant-temperature enclosure is described which is intended for use as a cold-junction reference for conventional thermocouples in place of the more usual ice/water reference used for temperature measurements. This is an example of a simple on-off control of cooling power. Finally the design and performance of a dew-point hygrometer are given in order to illustrate the potentialities of thermoelectric cooling applied under proportional servo-control conditions.

DOI:10.1049/piee.1963.0182

出版商:IEE    

年代:1963

数据来源: IET

摘要:

A system is proposed which uses a linear network in the forward path of a second-order on-off control system to achieve time optimum response for step inputs. The system, here referred to as the ‘error-network system’, is compared with several other proposals for various input signals. Among the other proposals is a novel function-generator system which achieves optimum switching for both step and ramp input signals. The error-network system compares well in performance with more complex systems.

DOI:10.1049/piee.1963.0183

出版商:IEE    

年代:1963

数据来源: IET

摘要:

The paper states briefly how Maxwell's geometric mean distance is used to reduce the self-inductance of a thin conductor to that of a pair of equivalent filaments. It then develops a method for splitting the standard Neumann double integral, used for finding the inductance of such filaments, into two sets of infinite series. The idea is first used for the known case of the circular loop, taking only the first terms of the sets of series to demonstrate their rapid convergence, and is shown to give the standard result. It is then applied to the hitherto unsolved case of the elliptical loop to derive two formulae, each valid for ellipses of different proportions. The accuracy of the second, which is semi-empirical, is examined, and finally the results are summarized. As the formulae appear in terms of the elliptical integral of the second kind, a simple formula for calculating the numerical value of this function over the range of interest is given in the Appendix. The formula expresses this in terms of the parameter of shape appearing in the inductance formulae, and may thus be used in preference to performing an additional calculation for the modulus that would be required to find the figure in a table of elliptical integrals.

DOI:10.1049/piee.1963.0184

出版商:IEE    

年代:1963

数据来源: IET