Wednesday, May 12, 2010

Okay, now it works.....

Alright, that's better. As I said in my last post, the output of the 756 MHz LO board for my 902 MHz transverter (W1GHZ design) was only putting out -28 dBm, so something was obviously wrong. Now, to find out what it was.

Worst case was that I had blown all the MMICs when I accidentally hooked the power up backwards for a split second. (Did I really do that? Uh, well, yeah, I did.) Time to find out. So I pulled C4 and put a test probe on there and sure enough I had 0.5 dBm at 252 MHz out of the first filter stage. So at least I knew that A1 right after the crystal oscillator was working. Phew! Dodged that bullet.

I wasn't too keen on actually slicing the board to test right after the second amplifier stage, so instead I double checked the values of the bias resistors. They were okay. Could it be that running the bias to A2 through a leaded ferrite bead changed the resistance of that path enough to matter? Unlikely, but I removed the ferrite. Same deal.

So I decided to re-flow the solder connections to A2 and A3 and, in doing so, noticed that I had the bias resistor to A3 going to its ground pin, not its output, one via over. Ah ha! Moved that over and....Bingo! 6.68 dBm of beautiful RF at 756 MHz.

Next up......the 1152 MHz LO board for the 1296 transverter.

No comments:

Post a Comment