GB904018A - Improvements in or relating to circuit arrangements for multiplying functions in theform of electrical signals - Google Patents

Abstract

904,018. Colour television. PHILIPS ELECTRICAL INDUSTRIES Ltd. Aug. 18, 1958 [Aug. 21, 1957], No. 26506/58. Class 40 (3). [Also in Group XXXVI] An electronic multiplying device for multiplying a a function of a first variable by a second variable comprises two amplifying elements in series, a further amplifying device being connected in parallel with one of said two elements, an analogue voltage representing said first variable being fed to the control electrodes of the two identical elements and an analogue voltage representing said second variable being fed to a control electrode of said further element, the output at the junction between said two amplifying elements being an analogue of the required product. In the embodiment of Fig. 1, two identical triodes 1, 2 are connected in series and have a voltage V 2 applied to their grids while a pentode 5, having a signal V 1 applied to its control grid, is connected across triode 3, it being proved mathematically in the Specification that the output at A (V A )= - S p R A V 1 (where S p is the mutual conductance of pentode 5)= - constant f.(V 2 ).V 1 , R A being the anode load resistance of pentode 5 (R A is approximately equal to <SP>1</SP>/S which depends on V 2 where S is the mutual conductance of 1, 2). The valves may be replaced by transistors. The invention is applied to colour television to provide a correction signal for colour matching in addition to the γ correction, V 2 representing the brightness signal V y =αV R +#V g +#V B (for the correction of the signals V R , V g , V B ). The signal V y representing the brightness signal is applied to triodes 1, 2 while signal V R is applied to pentode 5 (separate circuits are needed for each of V R , V g , V B ) to give an output representing the required corrected signal V R <SP>1</SP>. If the system is used for reproducing films or slides, a γ unequal to 1 may be desirable, and colour errors may occur. The γ correction referred to is therefore done in dependence on the brightness signal in order to match the contrast range of the television system with that of a film to be scanned, the normal γ correction for non-linearity of tubes not being shown. In order to avoid colour errors each colour signal may be multiplied by a factor k(V y ) so that the corrected signals assume the form V<SP>1</SP> R = k(V y )V B , V1 G =k(V y )V G and V1 B =k(V y )V B so that the brightness signal becomes αV<SP>1</SP> R + #V G + #V<SP>1</SP> B = k(V y )#αV R + #V G + #V B ) = k(V y ).V y . This must correspond as far as possible with V<SP>1</SP> y = V γy . The multiplication factor is therefore V y (γ - <SP>1</SP>). Three circuits as shown in Fig. 1 are used and the V R , V g , V B signals applied to the control grids of the respective pentodes and V y to the triodes. It is stated and proved in the Specification that R A and therefore V y must lie between certain values and the optimum value of R A and therefore of V y is derived from a circuit as in Fig. 4 where the maximum permissible value of R A is fixed by means of a pulse signal V p of amplitude V p max. = V y max. and R A can be varied by moving tapping 15 of an arrangement as in Fig. 4 which interpolates between V y and V p . The circuit shown in Fig. 6 combines a circuit 21 as shown in Fig. 1 with a circuit 16 as shown in Fig. 4. The maximum signal V y max. is obtained from tapping 15 and applied to the grids of triodes 1, 2, the uncorrected signal V R being applied to grid 6 of pentode 5. The corrected signal V<SP>1</SP> R is derived at 38.