4. Results
The software was changed to set the I2C address to 0x62 as per the data sheet for our Si5351. After connecting the logic analyzer, and learning that there was no acknowledgement, and trying to make additional corrections, it became clear that there was some other issue in the board.
Dr. Frohne helped determine that the issue was with one of the FETs that we had soldered several connections to. The best ways to fix the error are:
- Design a new PCB, have it made, solder everything again.
- Use one of the extra broken PCB’s to start again, knowing which fixes are needed.
Either way requires a lot of time desoldering and then soldering again. Ultimately decided to make sure Jared’s board worked, which wound up requiring a large amount of collaboration. The board in it's current, broken state are shown below.
We used the logic analyzer to confirm that on this board, the Si5351 communicates with Arduino. We used an oscilloscope to confirm that we could set the phases using Arduino code. On the signal generator we hooked up to the board, we found the sensitivity to be less than 0.1uV. Also, with Quisk set up, we found the image rejection to be approximately 3 dB. Though the most important result, is that we were able to tune Quisk to pick up a signal with the antenna connected.
A video of a signal we picked up can be found here.
And a picture of the fully constructed and working board can be found below:
Finally, here is a picture of Quisk running, and picking up a signal:
