ADRV9002: Phase offset of RX1 and RX2

hsato said:

Will this phase offset always be 180 degrees?
Also, is it possible to set the phase offset of RX1 and RX2 to 0 degrees?

In theory no... There's no 180 offset AFAIK. Can you paste the other windows from the osc plugin to checl the settings you have?

There's also some debug patterns you can play with to make sure things work at the SSI interface level. Look at "Example of configuring SSI test modes" in the wiki...

- Nuno Sá

hsato said:

Will this phase offset always be 180 degrees?
Also, is it possible to set the phase offset of RX1 and RX2 to 0 degrees?

In theory no... There's no 180 offset AFAIK. Can you paste the other windows from the osc plugin to checl the settings you have?

There's also some debug patterns you can play with to make sure things work at the SSI interface level. Look at "Example of configuring SSI test modes" in the wiki...

- Nuno Sá

Thanks for the reply.

The settings for the IIO oscilloscope are as follows.
The transmit DDS is set to 0.0999 MHz. The received waveform measured a phase offset of 180 degrees.

And another setting was made for transmit DDS at 0.0110MHz. In this case, a phase offset of 180 degrees was also measured.

I thought it might not be a sampling mismatch, but rather that there is always a 180 degree phase offset between RX1 and RX2.

hsato said:

The transmit DDS is set to 0.0999 MHz. The received waveform measured a phase offset of 180 degrees.

Maybe I missed something but why are you setting the DDS? If I understood correctly, you're directly injecting a signal in the receivers right? Hence, I would just change both TX1 and TX2 to primed to not interfere at all with the reception.

- Nuno Sá

I have the mode as FDD.
I then set the local to the same frequency and loop back the transmit signal to the receive to check.
(The image at the beginning of this Q&A.)

I have also input the signal from a signal generator and the result was the same.
Could the problem be that I am checking in FDD mode?

Hmm, I just realized you have different DDS frequency between your TXs. Like that, I think phase offset is expected. I did a quick test and things look good:

And here are my settings:

Should also mention that I'm using the master development branch... If at the same frequencies, you still see the same behavior, please try the 2022_R2 branch.

- Nuno Sá

Thank you for your careful confirmation.
I also checked with the same connection and settings.
(The transmit waveform is branched so that it can be measured with an external oscilloscope.)

The phase offset measured by the IIO oscilloscope was 0 degrees.

What is interesting is that the waveform measured by the external oscilloscope has a phase offset of 180 degrees.

Does TX2 have a phase offset of 180 degrees relative to TX1?
Also, does RX2 have a phase offset of 180 degrees relative to RX1, so the final phase offset is 0 degrees?

hsato said:

What is interesting is that the waveform measured by the external oscilloscope has a phase offset of 180 degrees.

Hmm, not really sure why that number...

hsato said:

Does TX2 have a phase offset of 180 degrees relative to TX1?
Also, does RX2 have a phase offset of 180 degrees relative to RX1, so the final phase offset is 0 degrees?

Don't think so... It's expected for some offset between TXs and even RXs. It might depend on a lot of different things but I don't think there's any fixed phase offset between the ports.

- Nuno Sá

I measured the phase offset between ports at several local frequencies.
In most of them I measured a phase offset of 180 degrees, but I understand that this is not a fixed phase offset that changes with the settings.

Thank you very much for your kind response.

I have reviewed the EVAL-ADRV9002 schematic.
And I noticed that RX1A+(-) and RX2A+(-) have reversed polarity between balun and ADRV9002.
TX1A+(-) and TX2A+(-) have reversed polarity as well.

I should have checked the schematic first.
My apologies.