Other Parts Discussed in Thread: LMX2820, USB2ANY
Hello,
We are trying to program the LMX2820EVAL using a Raspberry PI and have run into an interesting issue. We are still trying to debug the problem ourselves but would like any input you all can provide. When we program the LMX2820 through the raspberry pi, using the exact same register settings as TICs PRO software uses, we get the correct frequency, the lock detect goes high, but the output power is very low, ~-50 dBm. The left side of the picture above depicts the setup. We are only making 4 connections from the Pi to the LMX2820: SCK, SDI, GND, and LE. We are also using the Reference PRO board provided by TI to give a 100 MHz reference in. Both the LMX2820 and Reference Pro are powered by a 3.3V power supply.
However, we can bring the output power back to its nominal value simply by replacing the raspberry pi connection with the USB2ANY programming module, and doing nothing else. The USB2ANY doesn't even have the USB connection, however, simply making the 10 pin connection brings the power up to the expected 3 dBm we measured during our initial testing using the USB2ANY module entirely. Any thoughts as to what connections the USB2ANY is making internally to cause the LMX2820 to perform as expected? What state causes the LMX2820 to output the correct frequency but remain at -50 dBm regardless of the requested frequency?
As another clue/note: We use the same program (slightly different register values obviously) and programming setup as depicted on the left of the above picture for the LMX2595EVAL and it works as expected.
Happy to provide more context if needed,
Thanks in advance,
Sami
