Confusion about the isolation of <RFC to RF1/RF2> and <RF1 to RF2> ?

Figure 11 shows the isolation between RFC to unselected RFx channel. For example, green trace (RF1) is representation of RFC to RF1 leakage when RF2 channel is selected.

Figure 12 shows the leakage between RF1 and RF2. Here, green trace represents the situation RF1 to RF2 leakage when RFC to RF1 path is selected.

All RF pins are seeing an 50ohm load during this measurement. 

Hello,

Thanks for your quick reply. however, I'm still confused on this. Why these two isolation is not the same, the switch isolation is linear and reciprocal I suppose.

for example, when measuring the RFC to RF1 leakage, the switch is set to RF2 selected. The RF2 port is connected to 50ohm load, and RFC is injected a signal, e.g., 0dBm at 1Ghz. Then, using spectrum analyzer to measure the output at RF1 port, you get the figure RFC to RF1 isolation. however, keep switch RF2 selected, just change the load and instrument connection (RFC -> load, RF1 -> signal source, RF2 -> spectrum analyzer). You measure RF1 to RF2 leakage, and get a different number. 

Note: based on the simplified diagram in datasheet, I'm expecting the isolation should be same regardless the signal direction.

That's my question, thanks a lot!

Regards,

Jeff

The isolation is expected to be different considering layout difference between RFC-to-RFx path vs RFx to RFx path differences.

Sorry, I couldn't image how this could make difference. The layout is there. Just changing the signal direction, it will get the different number. That's very interesting result. Hope you understand my confusion.  ;-9