Accurate measurement of intermodulation distortion for the MGA-72543 requires some precautions to be observed in the test setup. This discussion will focus on third order intermodulation products and calculation of the input and output intercept points.

There are three sources of error that should be considered when making distortion measurement on devices with relatively low intermodulation products. These are:

Test equipment interaction

The preferred method for combining the two signal generator outputs is to use a resistive power combiner. A resistive combiner has the advantages of being well-matched at all three ports and presents a constant, power-independent impedance to both of the signal sources as well as the device under test. In addition, a resistive combiner is frequency insensitive (broadband) and being virtually completely linear, creates no intermodulation products of its own.

One of the most important considerations in making IM measurements is to provide adequate isolation between the two signal generators. If the two signal generators are not well isolated from one another, their ALC circuits can interact in such a way as to produce excessive IM products. The best choice for isolating the signal generators is to place a ferromagnetic isolator between each signal generator and the power combiner.

A less desirable alternative is to substitute fixed attenuators, e.g., 10 to 20 dB, for the isolators. Although attenuators have the advantage of making the test setup broadband, the isolation is not as high and increasing the output power level from the signal generator to compensate for the attenuation may introduce additional distortion into the signals.

Placing a fixed attenuator (e. g., 6-10 dB) between the power combiner and the input port of the device under test is an extra measure that will ensure the test device is presented with a constant impedance source. Similarly, an attenuator between the test device's output will ensure a good match for both the spectrum analyzer and DUT.

Dynamic range of the test equipment

One of the challenges of measuring high intercept point devices is that of dynamic range. A wide dynamic range is necessary to simultaneously measure very low level IM products while in the presence of large fundamental signals. The dynamic range of the spectrum analyzer is limited on the low end by its sensitivity, or noise floor. The upper end of the analyzer's dynamic range is limited by the maximum input signal level for which the distortion products generated within the analyzer itself are below a level that would obscure the measurement.

Excessively high input signals to the spectrum analyzer may create internal distortion products that could be mistaken for the actual distortion to be measured on the device under test. To maximize the dynamic range of the spectrum analyzer, the input signal level to the analyzer should be kept as low as possible consistent with still being able to observe the third order IM products.

The usual quick check to determine if the analyzer in generating internal distortion due to input signals that are too high, is to increase the analyzer's input RF attenuator by 10 dB and verify that there is no change in the signal level of the IM products. If the IM product levels do change when the input attenuator is switched, the analyzer is generating internal distortion products and the input signal level is too high to make a valid measurement.

Quality and accuracy of the test equipment

Other factors affecting intermodulation measurement accuracy are related to the quality of the test equipment. Synthesized signal generators are the recommended choice for the two fundamental signal sources. For even greater measurement sensitivity, it may be desirable to select signal generators with low noise options.

Much of the accuracy of the measurement is related to the quality of the spectrum analyzer. Of key importance, analyzers should be chosen with adequate dynamic range specifications. Low noise options may also be useful in improving the sensitivity for observing low-level IM products.

Calculating the intercept point

After ensuring that an accurate measurement of the intermodulation products has been made, the third order intercept point is calculated as follows:

OIP3 = Po + DIM/2

where OIP3 is the third order intercept point (in dBm) referred to the output of the device under test, Po is the output power level of either of the two fundamental signals (in dBm), and IM is the difference in signal level (in dB) between the fundamental and the third order intermodulation products. The signal levels as observed on the spectrum analyzer are depicted in Figure 3.

Figure 3. Fundamental and IM Signals.

The input third order intercept point is then:

IIP3 = OIP3 - Gain

where IIP3 is the third order intercept point (in dBm) referred to the input of the device under test and Gain is the test device gain in dB.

Summary