and industry since this machine is the only one in the United States having a capacity greater than 5 million lbf.

Vacuum Generator.-Installation of a highly stable pressure generator for producing pressures in the vacuum range 10-3 to 10-8 torr (0.133 to 1.3×10-6N/m2) has been completed. This facility, utilizing continuous flow of gases through variable conductances, will facilitate the Bureau's work in investigation of vacuum-measuring instruments and in establishing an extended calibration capacity.

High Pressure Reference Points.-In the measurement of high pressures the need for improved accuracy has resulted in greater attention to the establishment of reference pressures, analogous to the fixed points of the temperature scale. New laboratory programs for improved measurement of two such reference pressures have been completed. These are the melting pressure of mercury at 0 °C, in the neighborhood of 108,000 psi (740 MN/m2) and the crystalline transition pressure of bismuth-the bismuth I-II transition at 25 °C near 380,000 psi (2.6 GN/m2). Both of these measurements were made with specially-built dead-weight-loaded free-piston gages. The imprecisions associated with the new measurements are the lowest ever achieved. In the measurement of the bismuth transition, the hydrostatic pressure reached is believed to be the highest yet achieved with a liquid-filled free-piston gage.

ELECTRICAL QUANTITIES—DC AND LOW FREQUENCY

The Volt Standard Moved to Gaithersburg. The NBS volt standard, consisting of 44 standard cells of the saturated cadmium sulfate type, was moved from its old location in Washington, D. C. to the new Bureau facilities in Gaithersburg, Maryland. The new standard cell laboratory at Gaithersburg includes many features to facilitate research on standard cells and the calibration of cells for the general public, industry, universities, and scientific laboratories. While the old facility in Washington was operated completely on direct current, the new one at Gaithersburg utilizes alternating current and is housed in a shielded room.

Specification for Dry Cells and Batteries.-The ninth edition of the American Standard Specification for Dry Cells and Batteries was prepared and issued. The new edition includes, for the first time, specifications on alkaline manganese dioxide, zinc-silver oxide, and sealed nickel-cadmium cells; previous editions were confined to specifications for Leclanche and mercury cells and batteries. New tests were incorporated to cover cells and batteries for high-rate discharges required in appliances such as tooth

A. G. McNish, Chief of the NBS Metrology Division, carries a platinumiridium meter bar, which was the U. S. National Standard of length until 1960, into the administration building at the Bureau's new Gaithersburg (Md.) laboratories. There it was placed in a new standards vault, along with a which is the Nation's standard of mass.

duplicate meter bar and two platinum-iridium one-kilogram weights, one of

brushes, electric shavers, etc., for cells used in electric watches, for cells used in highway safety flashers, and for cells used in electronic photoflashers.

D-C Voltage Divider Calibration.-A simple, easy, and inexpensive technique for calibrating d-c voltage dividers (volt-boxes) to 10ppm has been developed. This modification of a method proposed by the Julie Research Laboratories requires a set of 13 resistors only nominally equal in value, together with a 3-decade resistance box and a microvolt detector to determine their ratios and the ratios of the volt-box under test. All standards laboratories can now extend the range of potentiometric measurements from 1 to 1000 volts, by using the detailed information and error evaluation published by NBS.

Peak AC-DC Comparisons.-Independent verification has been provided of the NBS ac-dc standards on which all a-c measurements of voltage and current depend. A long and careful series of comparisons has demonstrated excellent agreement between a peak ac-dc comparator, developed by the NBS Electronic Instrumentation Section, and the basic rms ac-dc transfer standards of the Bureau. Tests with a wave analyzer preceded by an appropriate fundamental rejection filter showed that at 400 and 1000

Hz the possible waveform error in the comparisons was less than
These results make possible a new calibration service

10 ppm.
for peak as well as rms ac-dc comparators.

ELECTRICAL QUANTITIES-RADIO

Quantum Electronics.-Calorimeters for the measurement of pulsed laser energy at one wavelength have been built and tested. Two calorimeters of dissimilar geometries have given results which agree within a fraction of a percent. These can also be used for the measurement of power from a CW laser. It is expected that before long these instruments will serve as the basis for an NBS standard of energy, a standard of power, and eventually for standards and calibration services of an increasing range of energy, power, and wavelength.

Frequency Measurements of Laser Light.-A submillimeterwave laser is being built with the objective of measuring its frequency, and it is hoped that this will be the first step in the long process of making direct frequency measurements of coherent light.

High Frequency Electrical Standards

CW Power Measurement Technique.-An accurate, wide-range, power measurement technique was developed which makes possible the determination of the net power delivered to a load of arbitrary impedance. All power measurements relate to a known arbitrary level by dimensionless ratios which are measured using a precision waveguide-below-cutoff attenuator. At 30 MHz, with power measurements extending from 10-2 watts to 10-14 watts, the corresponding maximum uncertainties ranged from +0.5 percent to 1.5 percent.

RF Current Comparator.-A new technique for measuring the ratio between two sinusoidal currents permits both magnitude and phase to be measured with high precision, and with negligible perturbation of the network. The basic principle involves the insertion of a special network at the point where the current is to be measured. An auxiliary coherent source is adjusted so that the inserted network appears as a short circuit (zero impedance) to the current being measured.

Coaxial Calorimetric RF Power Standard.-A new dual-load calorimetric power meter was developed as a reference standard in the range 2 to 100 W. The maximum uncertainty of the calorimeter is 0.35 percent at frequencies from d-c to 4 GHz.

Pulse Voltage Standard Extended.-The measurement range of the NBS low and medium level pulse voltage standard has been

extended upward to 1000 V, from the former 100 V limit. This standard measures the peak voltage amplitude of pulses with durations as short as 10 nanoseconds. Maximum uncertainty is 0.3 percent; a 1 percent calibration service has therefore been announced. Work is in progress to extend the range downward to 5 mV and upward to 50 kV.

Near-Zone Field Intensity Meter.-Prototypes of two new nearzone field intensity meters were developed and field tested. They were designed to cover only the two ends of the 150 kHz-30 MHz band, but proved to be adequate for the entire band. One resembles a tuned-rf voltmeter, while the other is a heterodyne-type voltmeter with a swept local oscillator. The meters are capable of measuring field strengths of 0.1 to 1000 volts/meter with very little perturbation of the field. This latter property was achieved by building all of the rf circuitry within the antenna or probe and using a specially developed highly resistive transmission line to transmit the d-c output to a remote indicator.

Broad-Band, High Directivity Coaxial Coupler.-A coaxial coupler has been developed for use as a reflectometer in impedance measurements and as a comparator in calibration of rf power meters. The directivity of the coupler is at least 50 dB over octave frequency bands, and auxiliary tuners are not required. The maximum VSWR of the coupler lines is 1.01 up to 4 GHz. More than 40 requests for design information and complete drawings on the coupler have been received from domestic and foreign laboratories and manufacturers.

Improvements in RF Voltage Standards and Measurement Techniques. The maximum uncertainty in the reference standard bridge used for rf voltage measurement has been reevaluated for frequencies up to 100 MHz and a new figure of 0.3 percent assigned. At 1000 MHz, the new figure is 2 percent. These values for maximum uncertainty were verified experimentally by comparison with power standards. It was also shown that use of the NBS-designed tee-junction allows calibration of voltmeters with a precision no worse than 1 percent in the 500 to 100 MHz frequency range.

Improvements in AC-DC Ratio Measurement.-Improvements in rf voltage bridge measurements over the past few months have given a reproducibility of the ac-dc ratio to ±0.15 percent at 100 MHz and 1 volt levels. Four different bolometer mounts were used in checking the agreement, in addition to a comparison with power measurements. The mounts included: (1) two negative temperature coefficient thermistors, (2) one negative. temperature coefficient thermistor and a "zero" temperature co

efficient resistor, (3) one positive temperature coefficient thermistor and a "zero" temperature coefficient resistor, and (4) a Bolovac. The resulting measurement uncertainty is estimated to be within ±0.3 percent, a considerable improvement over the previous value of +1 percent.

High Frequency Impedance Standards

Inductive Voltage Dividers.-Inductive voltage dividers having a 2 to 1 ratio of input voltage to output voltage were developed. Deviations from nominal are less than 0.1 ppm from 1 kHz to above 1 MHz, as measured with a limit of uncertainty of ±0.025 ppm on a bridge specially developed for this purpose. An experimental 1 MHz voltage attenuator having a 42 dB range in 6 dB steps was constructed from seven 2 to 1 voltage ratio transformers. Theoretical considerations and preliminary measurements indicate the uncertainties in attenuation to be less than 0.005 dB per 6 dB step. Also, two seven-decade inductive voltage dividers and several 20-section single-decade dividers were constructed for evaluation from 30 kHz to 100 kHz. Initial checking indicates the presence of systematic errors at 50 kHz of the order of 10 ppm, for which corrections can be made.

Low Impedance Measurement.-Low impedance devices may now be measured accurately with a precision admittance bridge at radio frequency. The technique involves measuring the low impedance in series with a 2-port (three-terminal) impedance which is evaluated as a pi network. By this technique, at 3 MHz, 0.1 mF capacitors have been measured with an uncertainty of about 1 percent, and 0.01 mF capacitors have been measured with an uncertainty of about 0.1 percent. Recently developed precision coaxial connectors and accurate high-frequency capacitance standards are vital to this technique.

RF Capacitance Measurements.-The use of precision coaxial connectors now permits the evaluation of capacitance from 1 to 1000 pF in the 1 to 5 MHz range, with uncertainty limits from about 0.2 to 0.01 percent + 0.002 pF. Preliminary intercomparison between the computed effective capacitance of parallel plate capacitors and coaxial line standards have confirmed these limits.

Electrodynamic Ammeter.-The coaxial radio frequency electrodynamic ammeter developed at NBS operates on the electric motor principle-that of a torque produced on a shorted copper ring by a changing magnetic field. It has a theoretical maximum uncertainty of ±0.5 percent of rf current for the TEM mode with a high standing-wave ratio.

Improved Calibration Services.-Precision has been improved for all services, with an improvement in overall accuracy in several