{"title":"A monolithic companding D/A converter","authors":"J. Schoeff","doi":"10.1109/ISSCC.1977.1155668","DOIUrl":null,"url":null,"abstract":"COMPANDED PULSE CODE MODULATED (PCM) transmission of voice signals has become standardized through widespread use of the Bell system p-law and the CCITT** A-law transfer characteristics. Until now, all codecs (eoder/decoders) for these communication systems have been fabricated in either discrete or hybrid form and have been relatively expensive. This paper will present a newly developed monolithic digital-to-analog converter specifically designed for compression and expansion of signals according to the existing PCM standard. This converter, however, is not limited to PCM communication, but may be used in other areas such as data acquisition, servo controls, data recording, telemetry, voice synthesis, log attenuation, secure communications, sonar, and many other applications which require a 12-bit plus sign dynamic range and the convenience of an 8-bit digital code. When used in a telecommunications application, the companding DAC is a complete PCM decoder, with metal options for p-law and A-law. A one-half step decision level for encoding is provided within the circuit and controlled with the encode/decode logic input. This current offsets the entire transfer characteristic one half step, regardless of the value of the output current. The outputs are multiplexed for time sharing of one DAC for both encode and decode operation. The DAC settling time is 500 ns, and i t will decode more than 32 PCM channels in 125 ps, which is the sampling period at 8 kHz. In a shared encoder it will convert eight channels, assuming a 1 0 ps sample and hold acquisition time. The outputs are high impedance, high compliance current sources and will interface with most balanced loads. The reference inputs will accept a fixed reference or a positive or negative multiplying input. The transfer characteristic of the companding DAC is shown in Figure 1. The output consists of eight positive chords and eight negative chords, each containing sixteen steps. The slopes of these chords are binarily related with the chord at the origin having steps equivalent in size to those in a 12-bit converter. The step size is a nearly constant 3.2% of reading throughout most of the dynamic range, which corresponds to approximately 0.3 dB per step. Each successive chord endpoint is 6 dB below the next higher endpoint for every chord in the A-law specification, and follows this for most chords in the p-law. The dynamic range, or ratio, of the full scale to the smallest step size is 72 dB for the p-law version and 66 dB for the A-law unit. The electrical specifications for the circuit are summarized in Table I.","PeriodicalId":416313,"journal":{"name":"1977 IEEE International Solid-State Circuits Conference. Digest of Technical Papers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"1977 IEEE International Solid-State Circuits Conference. Digest of Technical Papers","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISSCC.1977.1155668","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 13
Abstract
COMPANDED PULSE CODE MODULATED (PCM) transmission of voice signals has become standardized through widespread use of the Bell system p-law and the CCITT** A-law transfer characteristics. Until now, all codecs (eoder/decoders) for these communication systems have been fabricated in either discrete or hybrid form and have been relatively expensive. This paper will present a newly developed monolithic digital-to-analog converter specifically designed for compression and expansion of signals according to the existing PCM standard. This converter, however, is not limited to PCM communication, but may be used in other areas such as data acquisition, servo controls, data recording, telemetry, voice synthesis, log attenuation, secure communications, sonar, and many other applications which require a 12-bit plus sign dynamic range and the convenience of an 8-bit digital code. When used in a telecommunications application, the companding DAC is a complete PCM decoder, with metal options for p-law and A-law. A one-half step decision level for encoding is provided within the circuit and controlled with the encode/decode logic input. This current offsets the entire transfer characteristic one half step, regardless of the value of the output current. The outputs are multiplexed for time sharing of one DAC for both encode and decode operation. The DAC settling time is 500 ns, and i t will decode more than 32 PCM channels in 125 ps, which is the sampling period at 8 kHz. In a shared encoder it will convert eight channels, assuming a 1 0 ps sample and hold acquisition time. The outputs are high impedance, high compliance current sources and will interface with most balanced loads. The reference inputs will accept a fixed reference or a positive or negative multiplying input. The transfer characteristic of the companding DAC is shown in Figure 1. The output consists of eight positive chords and eight negative chords, each containing sixteen steps. The slopes of these chords are binarily related with the chord at the origin having steps equivalent in size to those in a 12-bit converter. The step size is a nearly constant 3.2% of reading throughout most of the dynamic range, which corresponds to approximately 0.3 dB per step. Each successive chord endpoint is 6 dB below the next higher endpoint for every chord in the A-law specification, and follows this for most chords in the p-law. The dynamic range, or ratio, of the full scale to the smallest step size is 72 dB for the p-law version and 66 dB for the A-law unit. The electrical specifications for the circuit are summarized in Table I.