I'm attempting to interface an AD6630 differential gain block to an AD6650 GSM narrow-band receiver at an IF of 170.6 MHz. Last year when the board was being designed, it was impossible to get 170.6 MHz differential GSM SAW filters, so I had to use Triquint 855912s, which are single-ended. Consequently, there are RF baluns all over the place in the receive chain. When testing the board, I see a nicely amplified 170.6 IF signal at the output of the AD6630, but the signal is obliterated somewhere between the input of the 2nd SAW filter at the AD6630 output and the AD6650 analog input (see attached schematic pages).
There's no signal to speak of at C333, the 47 nF cap on the 50 ohm side of the AD6650 RF input balun (T6). Is there an error in the wiring of T6? At the time, I know that I tried to follow Figure 38 on p. 27 of the AD6650 datasheet, but I now notice that the values of T101 and T102 in Figure 38 are not actually specified. Am I missing something obvious?