Mixers: Measuring Performance

Measuring mixer performance

Mini-Circuits policy for product measurement and preparation of specifications is straightforward: present the facts so the design engineer can evaluate product data for his or her circuit and system requirements and arrive at a proper design decision. For mixers, which involve control of three signals at different frequencies, particular care is necessary.

A 50Ω broadband system is used for all factory measurements on mixers. This uniformity offers the customer a convenient and consistent means to obtain correlation with data generated by Mini-Circuits. Figure 3 presents conceptual block diagrams of test setups for measuring the performance of Level 7 mixers. In actual practice, both for engineering development and for production testing, highly integrated and automated equipment is used to perform the measurements and document the results.

Conversion loss measurements

For conversion loss measurements, Figure 3 (a), fixed attenuator pads are connected to all three ports so the mixer sees 50Ω at the frequency of interest and all significant harmonics.

For the IF power measurement, a low pass filter is included before the power meter, to reject all responses other than the desired IF. For example, any LO leakage has no effect on the conversion loss measurement since it does not reach the power meter. If a spectrum analyzer is used instead of the power meter, the filter can be omitted.

Isolation measurements

When measuring LO-RF and LO-IF isolation a pad is placed between the generator and mixer in Figure 3 (b) to ensure a 50Ω-impedance. Also, a 50Ω termination is connected to the IF port when measuring the power at the RF port, and vice versa.

When measuring isolation from RF-to-IF, normal LO drive and RF power in the linear operating range are applied. A spectrum analyzer replaces the RF voltmeter. This technique ensures measurement integrity: the power at the fundamental frequency of measurement is distinguishable from other components in the spectrum. For broadband applications, the effects of harmonics can be judged. For narrow-band applications, only the fundamental need be considered.

VSWR measurements

VSWR measurements are made under the same dynamic conditions that the mixer would encounter in practice, see Figure 3 (c). First, let’s consider VSWR measurement at the RF port. An LO signal is applied to its port and the unused IF port is terminated in 50Ω. The RF generator supplies an input level corresponding to linear mixer operation. With the mixer disconnected from the directional coupler, a reference level is obtained (all the RF power is reflected back). The amount of reflected signal depends on the directional coupler used; a 20 dB coupler would establish a reference level 20 dB below the RF input. Next, the mixer is connected to the output of the directional coupler. The spectrum analyzer acts as a narrow-band filter and allows observation at the RF input frequency. The RF power reflected back from the mixer is displayed and can be measured by the calibrated scale on the spectrum analyzer.

The VSWR at the IF port is measured in a similar fashion. In this case, the RF port is terminated in 50Ω.

VSWR at the LO port is measured with specified LO power applied (+7 dBm, in the example) via the directional coupler, and both the RF and IF ports are terminated in 50Ω; not shown in Figure 3 (c).

Broadband mixers, it should be noted, exhibit a different VSWR characteristic at different frequencies. Factors causing this include circuit resonances and changes in diode impedances as the LO power level changes. Also of importance is the fact that the input impedances of the various ports are load dependent, even though they are isolated from each other. At high frequencies this effect is more noticeable, because isolation tends to drop as frequency increases.

Two-tone, third-order intermodulation measurements

Two-tone, third-order intermodulation distortion takes place when two incoming signals arrive at the mixer RF port and interact with the conversion-loss non-linearity. See Figure 3 (d). High isolation and low harmonic content in the two RF sources are essential.

For details regarding intermodulation measurement refer to the article, “Improve Two-tone, Third Order Testing”.

Browse all RF frequency mixers