Signal transmission systems

This invention relates to signal transmission systems which use analogue to digital converter channels, for example channels capable of handling analogue signals with different ranges of amplitude and with different polarities.

An object of the invention is to provide means for at least partly nulling off-set errors due to component manufacturing tolerances and component value variations with temperature changes and with age so that commercial quality components can be used and yet accurate conversion can be achieved.

According to the present invention a signal transmission system includes an analogue to digital converter and at least one nulling circuit arranged to store an output from a part of the converter when that part has a null input and to add the stored output to the converter during a subsequent conversion of an analogue input.

Constant off-set error can be corrected by one nulling circuit which stores the digital output from the converter, and includes a digital to analogue converter arranged to derive from the stored digital output a nulling analogue signal, which is then added to the input to the analogue digital converter.

Where the converter channel has a switched-gain amplifier, so that it can handle analogue inputs with different amplitude ranges and there is an off-set error which depends upon the gain set, then a nulling circuit can store the output from the amplifier when the aplifier has a null input, and

The invention may be carried into practice in various ways, and one embodiment will now be described by way of example, with reference to the accompanying drawing, of which the single figure is a block.,circuit diagram of a signal transmission system.including an analogue to digital converter and an automatic nulling circuit.

The main analogue to digital converter channel includes a multiplexer 11 capable of receiving any of a number of analogue inputs at 12 in dependence on a control signal provided at 13 from an overall control logic unit 14, a switched gain amplifier 15 controlled at 16 from the control logic 14, to scale up the particular analogue input at 12, so as to make the most efficient use of the range of the converter, a mode amplifier 17 controlled at 18 from the logic 14 in accordance with the polarity of the analogue input at 12, a sample and hold stage 19 control at 21 from the logic 14 so as to hold the analogue signal it receives at its input constant during an analogue to digital conversion operation, and an analogue to digital converter 22 controlled at 23 from the logic 14 so as to produce a digital output at 'B' following the conversion period set by the logic 14.

That main converter channel shows imperfection in unwanted off-set voltages which together show an error in the digital output so that the output is not zero when the analogue input is zero. This error is known as "off-set error" '. It applies constantly components of the off-set error, namely the error up to 'A' which depends upon the selected gain of the amplifier 15, and the constant error between 'A' and 'B' are compensated for by respective measuring and storage systems.

The non-constant error uses a storage system consisting of a sample and hold stage 26 receiving at its input the output of the switched gain amplifier 15, a mode amplifier 27, and a switched gain amplifier 28, and each of those units is controlled at 29, 31 and 32, from the overall control logic 14.

The measuring and storage system for the constant error from 'A' to'B' includes a latch 34 for holding an output at 'B' from the analogue to digital converter 22, and a digital-to-analogue converter 35, and again each of those components is controlled at 36 and 37 from the logic 14.

Before the main converter channel is used to process an analogue input at 12, a nulling command cycle is initiated by the logic 14, and that has the effect of switching the input to the switched gain amplifier 15 at 41, and the input to the mode amplifier 17 at 42 to earth potential or null input, at which the off-set error will appear as the output.

Thus in the first measuring storage system, with the gain of the switched gain amplifier 15 set at a high value of, say, 200, then if the off-set voltage in the amplifier 15 is 'e 1' a voltage of 200e₁ at 'A' will be delivered to the input of the sample and hold stage 26. The choice of the gain of the amplifier controlled to hold a previously measured amplified off-set error signal.

However, if preferred, the nulling command cycle can be performed under control of the logic 14, before every new input is applied to the analogue multiplexer 11. Alternatively, a nulling command cycle can be initiated from time to time to re-set the correction voltages in dependence upon changing operating temperature, and ageing of component.

The off-set errors will be correctly nulled out of succeeding analogue to digital conversions irrespective of the selected gain and mode in the main channel.

It should be noted that since the fixed part of the correction signal is derived from the output of the main analogue to digital converter 22 that error can only be resolved to the resolution of the main converter. However that converter will be a high accuracy converter, because it is only with high accuracy converters that the off-set errors are significant.