A digitally synthesized source designed to provide two sinewave outputs with an accurately known adjustable phase shift in the second channel is described. A brief description is given of the bridge with which it will be used. The source requirements and design are discussed. A testbed wire-wrapped version demonstrates stability and noise of 0.5 p.p.m. (r.m.s.) over a 1-h period.<>

A power standard for calibrating high-accuracy wattmeters and watt-hour meters at 50/60 Hz is described. The technique is a modification of a previously described approach that utilizes a power bridge based on a current comparator. This standard was used in a recent international comparison of 50/60 Hz power.

Several groups have worked on the characterization of four terminal-pair (4TP) capacitance standards at high frequencies. This paper describes a variation of the technique to predict a capacitor's frequency characteristic. The method is sensitive to regression parameter selection and the paper gives a detailed analysis of the techniques used to calculate reasonable values for these parameters. The results of the analysis of the capacitor frequency characteristic prediction method's sensitivity to exponent parameter variation have shown that this sensitivity is a major uncertainty component in the uncertainty analysis of NIST's capacitance standard measurement system.

The Electricity Division at the National Institute of Standards and Technology (NIST, formerly NBS) has implemented a system to characterize capacitance and dissipation factor for four terminal-pair (4TP) air dielectric capacitors at frequencies from 1 kHz to 10 MHz. This paper describes an extensive uncertainty analysis of the measurement system. The analysis has been divided into three areas: 1 kHz capacitance measurements; network analyzer impedance measurements (covering frequencies from 40 MHz to 200 MHz); and a mathematical extrapolation algorithm that regresses the high-frequency characterization down to frequencies of 10 MHz and below. This algorithm is referred to as the capacitor frequency characteristic prediction (CFCP) method

A special purpose ac voltage divider system having voltage ratios of 600 V, 480 V, 360 V, and 240 V to 120 V has been developed to extend the voltage range of primary electric power calibrations from 120 volts to 600 volts at power frequencies of 50 and 60 Hz. The system consists of a special two-stage resistive divider compensated with an active circuit, thereby reducing the error contributions to below 1 /spl mu/V/V. The developmental goal to realize ac voltage scaling within 5 /spl mu/V/V uncertainty in a device verifiable with dc resistance ratio measurements has been attained.

Low frequency errors of thin-film multijunction thermal voltage converters are estimated using a simple model based on easily measured parameters. The model predictions are verified by measuring the converter's frequency characteristic using a digitally synthesized source.

A wideband wattmeter for measuring active power over a frequency range of dc to 500 kHz is described. The wattmeter is based on the three-voltmeter method in which three rms voltage measurements are used to calculate power. The wattmeter active power uncertainty is estimated to be within 0.03% from dc to 20 kHz and within 1.5% to 500 kHz.

Low-frequency tracking errors of thermal voltage converters are described and estimated using circuit models. A digitally synthesized source is used to confirm ac-dc differences in the 0.001 Hz to 40 Hz range.

A step down procedure is described for establishing voltage standards in the 1 mV to 100 mV range at frequencies between 10 Hz and 1 MHz. The step down employs low voltage thermal voltage converters and micropotentiometers. Techniques are given for measuring input impedance and calculating loading errors.

A wideband wattmeter for measuring active power over a frequency range of dc to 1 MHz is described. The wattmeter is based on the three voltmeter method in which three RMS voltage measurements are used to calculate power.

An automatic inductive voltage divider (IVD) characterization method that can measure linearity by comparing IVD's with different structures is suggested. Structural models are employed to decompose an error vector into components that represent each divider. Initial tests at 400 Hz show that it is possible to assign independent errors due to the binary and decade structures with a 2/spl sigma/ uncertainty of 0.05 parts per million (ppm) at the measured ratio values.< >

An impedance bridge that compares two-terminal standard inductors to characterized AC resistors in the frequency range of 10 Hz to 100 kHz is described. A dual-channel, digitally-synthesized source and sampling digital multimeter are used to generate and measure relevant bridge signals. A linear interpolation algorithm is used to autocalibrate the bridge to a 1 nF gas dielectric capacitor. An intercomparison of the new bridge with existing measurement standards conducted in the low audio frequency range shows agreement of 50 to 200 parts in 10/sup 6/ for inductors for 1 mH to 10 H.< >

A binary inductive voltage divider (BIVD) is compared with a decade inductive voltage divider (BIVD) in an automatic IVD bridge. New detection and injection circuitry was designed and used to evaluate the IVDs with either the input or output tied to ground potential. In the audio frequency range the DIVD and BIVD error patterns are characterized for both in-phase and quadrature components. Differences between results obtained using a new error decomposition scheme based on structural modeling, and measurements using conventional IVD standards are reported.<>

A digitally synthesized source (DSS) designed to calibrate low-frequency (0.1 Hz to 1 kHz) digital voltmeters and thermal converters is described. The DSS output voltage, frequency, and waveform are programmable over the General Purpose Interface Bus (GPIB). The rms value of the output voltage is calculated, with an uncertainty of less than 5 ppm, by measuring the dc voltage of each of the steps used to create the waveform.<>

An impedance bridge that compares two-terminal standard inductors to characterized resistors is described. A dual-channel digitally synthesized source and sampling digital multimeter are used to generate and measure relevant bridge signals. A linear interpolation algorithm is used to autocalibrate the bridge to a 1 nF air-dielectric capacitor. An intercomparison of the new bridge with existing measurement standards in the low audio frequency range for inductors 1 mH to 10 H is reported.<>