ADRV9002
Recommended for New Designs
The ADRV9002 is a highly integrated RF transceiver that has dual-channel transmitters, dual-channel receivers, integrated synthesizers, and digital signal...
Datasheet
ADRV9002 on Analog.com
Hi Team,
As part of the SSI interface verification, we have verified the Rx interface using the adi_adrv9001_RxSsiTestModeCfg_t feature. Both RAMP_16_BIT and FIXED_PATTERN are working fine without any issues.
After Rx validation, we started Tx validation using Tx loopback to Rx.
In this Tx loopback test, we observed that the Tx data is getting left shifted by 1 bit, and the 16th bit is seen on the LSB (it’s like a rotate shift left).
For example:
0x8000 is transmitted on both I and Q. When captured at the receiver side with loopback, the data seen is 0x0001.0x0010, and at the receiver, it’s seen as 0x0020. It’s always a 1-bit rotate shift left.What could be the issue causing this behavior?
Regards,
Siva.
Hi Tushar,
I would like to know what is the platform are you using for connecting to ADRV9001 is it TES or Linux? As mentioned in the in the post : Tx SSI tuning issue? , could you try to clear or reset SERDES on the FPGA. It can be the case where the clk being fed to the FPGA from ADRV9002 causes a hiccup in the strobe alignment. Another thing to you can do if your platform has the ability to experiment with the Clk delay within ''adi_fpga9001_SsiCalibrationCfg_t *ssiCalibration''
Regards,
Rahul
Hi RahulMushini ,
We are using custom Artix-200T FPGA for connecting to ADRV9002.
We are using RX clock derived from the RX interface for the TX path as well by enabling the attribute "USE_RX_CLK_FOR_TX = 1".
Along with this for sending a constant value 0x8000 on the transmitter we had probed the strobe_p and idata_p lines as well. the image is attached. Please let us know if this not how it should be going. (yellow - Strobe_p and blue- IData_p)
The capture was taken from the output of FPGA and just before it reaches the ADRV9002.
As for the resetting the SERDES, yes we can try resetting the SERDES once again in order to verify the hiccup.
Regards,
Tushar
