What I have always done with these kits is to do a progressive build. That is, start with the switching circuitry and voltage regulators, and then test that these areas are functioning. Then I do the LO and then finally the RX and TX strips . . . testing each stage in turn. This is a good way to minimise the debug time if there is a fault. In this case there were three minor faults. The first was an inability to tune the LO onto 1152MHz . . . or if it did, it wouldn't re-start. The clue in this case was the +8V rail was reading almost +9V and was caused by a discontinuity between the tab on the 7808 and the side-wall of the box (the centre pin is removed and the tab is soldered to the box). Once the joint was re-flowed, the oscillator behaved correctly. The second and third faults were both related to the TX strip. I could have connected the output up to my HP436A power meter (via a suitable attenuator) but I opted to just go with the detector output. However, there was no voltage showing on the detector output. A voltage check back through the TX strip showed 0V on the GALI-4. This turned out to be a component placement issue. I had placed the device slightly
too far from the side-wall and the underside of pins 1 and 3 were shorting to the land that the tab was soldered to and effectively shorting out the supply to the MMIC. Because of the proliferation of through-plated holes it was a bit problematic lifting and re-sitting the MMIC . . . once accomplished however, the voltmeter still showed no detector volts . . . hmm? This I have to conclude was probably a solder whisker. A check across the schottky diode indicated an unwanted return to ground . . . somewhere. Judicious use of fine solder-wick did the trick.