SU423237A1 - METHOD OF CODE ANALOG TRANSFORMATION - Google Patents

Description

one

The invention relates to analog-digital computing and electrical measuring equipment.

The method can be applied to code-to-analog conversion and for dividing the voltage in digital voltmeters, analog and analog-digital computing devices.

There is a method of code-analog conversion by author. St. No. 185591.

The technical implementation of this method is significantly complicated by increasing the accuracy of the digital-to-analog conversion by decreasing the resolution of the output voltage setting. The converters constructed by the described method have a practical discreteness of the output eg E „/ 11

wives within - that in

jV chi /

In many cases (for example, to build accurate multi-decade compensators or digital voltmeters) is not enough. L - obsh, its number of resistors in the decoder, her grid. The proposed method differs from the known one in that, in order to increase the resolution of the conversion, the pulse signals carrying the dynamic unit code are further extended in time by an amount proportional to the sum of the weights of the lower bits of the source code.

FIG. Figure 1 shows schematically the principle of the code-analog conversion by the proposed method; in fig. 2 is an example of a circuit implementation of the method.

Pa figs. 1 shows: I is a possible decoding device; a - decoding device keys; U2 (t) is the voltage at the output of the grid; II - the schedule of switching on the keys (the higher bits of the code are represented by the inclusion of the keys during the whole time of the tact junior - by the inclusion of one key for a part of the time of the tact (E; on-volt voltage); III - voltage graph at the output of the decoding grid, if .5 and i2 () i:) -

 2I 5 15 U

 - E; span of a variable component on

equals; IV

grid exit schedule

L

The keys, if the code in high-order bits is zero. The voltage at the output of the decoding grid in this case is shown in graph V.

The impulse of switching on the key at the input of the decoding grid can, in general, expand at the end of the pulse, at its origin, or on either side.

Thus, with respect to the higher bits of the source code (the number of the higher bits is determined by the number of co-motivations in the decoding grid: for example, with eight resistances, the number of the higher bits of the binary code is 3) the conversion process is similar to the auth process. St. No. 185591, but the lower bits are converted by pulse width modulation of pulses on the output of the decoding grid.

The achieved effect is very significant, as it allows, without complicating the circuit of the analog part, to perform an exact conversion of the multi-bit code.

In this case, the following positive properties of the method according to ed. St. No. 185591:

- the errors of the resistance of the decoding grid are eliminated;

- averaged residual parameters of keys that carry a dynamic unit code on the input of a decoding grid, which helps to reduce the convergence of conversions, and also allows you to simply eliminate the influence of residual parameters of keys.

In addition, the possibility appears, without increasing the number of resistances of the decoding grid and keys, to increase the resolution of the transformation (i.e., the number of bits of the source code) by introducing a pulse width modulation of the coding-carrying pulses. The maximum allowable resolution can be characterized by the ratio

I::.,

T

where til is the minimum duration of the codon carrying impulse determined by transients of exact keys at the input of the decoding grid; T is the period of circulation of the code at the input of the decoding grid. Thus, this method provides high precision code-analog conversion without complicating the analog conversion circuit.

The device for code-analog conversion using this method contains (Fig. 2) a pulse generator 1, a decoding grid 2, an averaging device 3, keys 4 at the input of a decoding grid with memory elements, a number of pulses 5 generating a trigger pulse for 4 keys; the coincidence circuit 6, generating the returning impulse keys 4; shift registers 7, 8, or counters with decoders that distribute triggering and returning pulses alternately to keys.

Subject invention

Method of code-analog conversion by author. St. Up to 185591, characterized in that, in order to increase the resolution of the conversion, the pulse signals carrying the dynamic unit code are further extended in time by an amount proportional to the sum of the weights of the lower bits of the source code.

WITH

 0 (2 ofj efS

PTT