Tony jumpered where he did, I jumpered the 5 volt output (directly
from the leg of the 7805) to the pin in the header where it was carried
off the PIO board. I believed the traces were too small and I had
damaged them when changing over from "pre-RFE" to "post-RFE". (3
board to 4 board stack). It is extremely easy to damage the pads.
Tony's approach to combining the analog and digital grounds of the DDS
is okay. The ECO on the web site uses plate throughs on the bottom of
the TRX board. I preferred this because it still allowed easy
disassembly of the stack. This was really needed while we were still in
the "let's hack some hardware" mode of the first two years. After we
completed the full understanding of the RX->TX transient (using the 0 dB
INA gain always when changing state), and then the changes in the RFE
front end, I view it as very unlikely for me to change this hardware
again and Tony's more permanent fix is a good alternative.
The DDS's really want these little "overclocker's" heat sinks on them.
Bob
Philip Covington wrote:
Looking at the picture a few of the crappy solder joints on my PIO
board were located at those pins on the header that Tony jumpered
around. I read about the heat problem with the DDS chip in an old
forum message and did some probing around when I first had the dropout
problem. It indicated that there was some problem on the PIO board
and not a problem with the DDS chip, but every time I put any
mechanical stress on the PIO board the problem would disappear. That
is why I just started resoldering all the bad suspect looking solder
joints on the PIO board. After running almost continuously for 3+
months I have not had any more problems since I resoldered those
joints... I would STRONGLY encourage you to inspect the solder joints
on the boards before suspecting the DDS chip, especially if your
SDR-1000 has the heatsink already on the DDS chip.
73 de Phil N8VB
