Читать книгу Handbook of Microwave Component Measurements - Joel P. Dunsmore - Страница 58

Directional couplers separate the forward and reverse waves in a transmission system (see Section 1.3). A directional‐coupler is classically defined as a 4‐port device, often with a good load on the fourth port, as shown in Figure 1.33; but in practice a load element is almost always permanently attached. The directional‐coupler has four key characteristics: insertion loss, coupling factor, isolation, and directivity. In fact, directivity is related to the other three factors in a specific way.

(1.90)

Figure 1.33 Directional couplers.

Most couplers have a nearly lossless structure so that the directivity is nearly equal to the isolation/coupling, but for lossy structures, such as directional bridges, the earlier definition provides the proper description. In fact, consider the case of a directional‐coupler with 20 dB of coupling, 50 dB of isolation, and 0.05 dB of insertion loss, setting the directivity at nearly 30 dB. If a 10 dB pad is added to the input, as shown in Figure 1.34, the isolation is increased by 10 dB, the loss is increased by 10 dB, and the coupling stays the same. Thus, the simple but incorrect definition of directivity as isolation/coupling would yield an increase of 10 dB.

Figure 1.34 The effect of attenuation at the input of a coupler.

In fact, a better way of looking at directivity is the ability of the power at the coupled port to represent a change in reflection at the test port. Again considering Figure 1.34, if a signal of 0 dBm is injected into the input port and a full reflection (an open or short) is applied to the test port, the coupled port will show a power of about −30 dB (10 dB loss, plus a full reflection, plus 20 dB coupling; here the isolation term is ignored for the moment). If a load is applied to the test port, the signal at the input sees a 10 dB loss and 50 dB isolation for a value at the coupled port of −60 dBm. The difference between the open and the load is 30 dB; hence, the directivity is 30 dB, and adding the pad at the input has no effect.

In practice, the output match of a directional‐coupler is critical, and the test port mismatch can dominate the directivity. This signal flow is demonstrated in Figure 1.35. Mismatch at the output of the directional‐coupler affects directivity on a one‐for‐one basis. This mismatch is combined with the coupler input mismatch to create the overall source‐match. The source‐match affects the power measured at the coupled port when measuring large reflections at the test port. This “source mismatch” causes some reflected signal from the test port to re‐reflect from the input port, reflect a second time off the test port termination, and add or subtract to the main reflection, causing error in the coupled port power.

Figure 1.35 Coupler with mismatch after the test port flow graph.

However, the output mismatch is a direct error and causes reflection back into the coupler, thereby adding directly to the coupler directivity.