Joel P. Dunsmore

Novel method for vector mixer characterization and mixer test system vector error correction

This paper presents a novel method for characterizing RF mixers, yielding magnitude and phase response for input match, output match, and conversion loss, and works for mixers which have reciprocal conversion loss and for which the image response can be filtered out. The characterized mixer is used to accomplish a full vector correction of a mixer test system, which can measure other mixers that are not reciprocal. A key contribution is phase and absolute group delay measurements of the mixer-under-test.

New methods & non-linear measurements for active differential devices

This paper presents new results for nonlinear measurements of active differential devices based on a novel source system. The system provides both true-differential and common-mode stimulus for CW and modulated RF signals, and presents results of non-linear operation of differential devices driven with such signals, which are compared with the same devices driven at the same levels using single ended drive and calculating the differential response. A remarkable result is that the latter method performs very well for devices having common-to-differential isolation.

Novel two-tone intermodulation phase measurement for evaluating amplifier memory effects

The amplitude and phase response of the two-tone intermodulation product, as the tone spacing is swept in frequency, may be used to predict memoryrelated distortion in power amplifiers. Previously, the methods used to measure the phase response were slow and limited to narrow bandwidths. This new measurement method is extremely fast, proposes a method for normalization or calibration of the phase response, and is extremely broadband, making it suitable for measurements of wide band amplifiers such as those used in satellite communication systems.

Characterizations of asymmetric fixtures with a two-gate approach

A technique for extracting asymmetric fixtures from a through standard and de-embedding the fixture effects from the composite measurement of DUT and fixture combination has been proposed. The technique consists of: 1) building a through standard which is the cascaded combination of the two fixtures. The two fixtures can be asymmetric in return loss but are symmetric in insertion loss and have the same electrical lengths, 2) extracting the fixture S-parameters from the measurement of the through standard, 3) de-embedding the fixture effects from the composite measurement and characterizing only the DUT. The 2-port single-ended case was investigated first and then we extended the technique to 4-port mixed mode. Deductions for extracting the S-parameters of the two fixtures are described in details and the extraction result was validated with simulations and measurements.

Gating effects in time domain transforms

Gating is well known in modern VNAs for improving or isolating responses from multiple reflections in the frequency and time domain. However, the particular details of how positions of selections of the gate start and stop values, as well as gate shapes, aliasing and masking effects are not well documented. This paper presents an explanation of gate band-edge effects due to renormalization effects, and masking effects, and provides recommendations for improving the effectiveness of gating.

Sampling-oscilloscope measurement of a microwave mixer with single-digit phase accuracy

We describe a straightforward method of separately characterizing up- and down-conversion in microwave mixers using a sampling oscilloscope. The method mismatch-corrects the results, determines both magnitude and phase, and uses a novel time-base correction scheme to improve the accuracy of the measurements. We estimate our measurement accuracy to be on the order of a tenth of a decibel in magnitude and a few degrees in phase. We use the method to characterize the magnitude and phase reciprocity of a microwave mixer.

New measurement results and models for non-linear differential amplifier characterization

Previously, the author described a system that provides true differential and common-mode stimulus for CW and modulated RF signals, and presented results of non-linear operation of devices driven with such signals; they were compared with results for the same devices driven at the equivalent drive levels using a single-ended drive and calculating the differential response. Those devices showed no difference, in differential gain terms, whether measured with true-mode or single ended measurements. New results on limiting amplifiers demonstrate a substantial change in the differential gain compression when measured using a true differential mode drive. A theoretical model is developed which considers circuit topology differences and predicts which devices will have non-linear characteristics that change depending upon the test methods. Finally, some difficulties with error correction as applied to differential, non-linear circuits are discussed.

A new noise parameter measurement method results in more than 100x speed improvement and enhanced measurement accuracy

A new method for noise parameter measurements is introduced, with better than 100times speed improvement over traditional methods. The setup is simple and easy to configure, and the entire calibration and measurement process is very fast, making dense frequency spacing practical. The new method produces smoother data with lower scatter, and the dense frequency spacing eliminates shifts due to aliasing and makes it easier to identify the scatter and outliers.

New methods for testing frequency converters with embedded local oscillators

Previous methods for characterizing the magnitude, phase and group delay responses of a mixer required access to the LO. In satellite systems the LO may be internal to the frequency converter and not accessible, nor perfectly stable. Here, a method is developed that can measure the response of a mixer with an embedded LO. This method works by phase tracking the IF of the converter output and using that information to determine the frequency and phase drift of the embedded LO, and compensate for any drift. The accuracy of the response is limited by the phase-noise quality of the internal LO, and methods are described for compensation of this noise. This method can provide the absolute group delay response, and on a typical Ka band satellite converter, delay accuracy as small as 125 ps have been achieved.

Complete Pure-Mode Balanced Measurement System

Previously, the first author has shown that the nonlinear behavior (e.g. gain compression or TOI) of some balanced amplifiers can be accurately measured using single-ended measurements: amplifiers that have a balanced stage before a non-linear stage; other amplifiers require true-mode drive to give correct differential and common-mode results. This paper describes, for the first time, a complete solution to the problem of true-mode measurements. The system and method included in this makes use of a dual-source VNA to produce either pure-differential or pure-common-mode drives in both the forward and reverse direction. A method for correcting drive signal errors due to mismatch between the VNA and the device under test is introduced, as well as a method to provide full balanced error correction to all terms. Its remarkable that small modifications to a standard single-ended 4-port calibration provides all the required error terms.