Post Go back to editing

Adrv9009 RX data error

Category: Software
Product Number: ADRV9009
Software Version: vivado 2021.2

Hi ,

I have acuired data from ILA by using the default adrv9009 app. The RX connected a terminator so the signal should be near zero.

But the waveform acquired has some pulses inside as the pictures below. It looks like adc sampling error and MSB occurred '1' bit wrong value.

only happened at one sample

The report seems all calibration is ok.


Hello
rx_clkgen: MMCM-PLL locked (122880000 Hz)
rx_os_clkgen: MMCM-PLL locked (61440000 Hz)
tx_clkgen: MMCM-PLL locked (61440000 Hz)
rx_adxcvr: OK (4915200 kHz)
tx_adxcvr: OK (2457600 kHz)
rx_os_adxcvr: OK (2457600 kHz)
talise: Device Revision 192, Firmware 6.2.1, API 3.6.2.4294967297
talise: Calibrations completed successfully
rx_jesd status:
Link is enabled
Measured Link Clock: 122.882 MHz
Reported Link Clock: 122.880 MHz
Lane rate: 4915.200 MHz
Lane rate / 40: 122.880 MHz
LMFC rate: 3.840 MHz
Link status: DATA
SYSREF captured: Yes
SYSREF alignment error: No
tx_jesd status:
Link is enabled
Measured Link Clock: 61.441 MHz
Reported Link Clock: 61.440 MHz
Lane rate: 2457.600 MHz
Lane rate / 40: 61.440 MHz
LMFC rate: 3.840 MHz
SYNC~: deasserted
Link status: DATA
SYSREF captured: Yes
SYSREF alignment error: No
rx_os_jesd status:
Link is enabled
Measured Link Clock: 61.440 MHz
Reported Link Clock: 61.440 MHz
Lane rate: 2457.600 MHz
Lane rate / 40: 61.440 MHz
LMFC rate: 3.840 MHz
Link status: DATA
SYSREF captured: Yes
SYSREF alignment error: No
tx_dac: Successfully initialized (122882080 Hz)
rx_adc: Successfully initialized (122882080 Hz)
rx_os_adc: Successfully initialized (122882080 Hz)

I checked the calibration mask as below

Is there any method to improve it?

Parents
  • Hi, can you please let us know what board are you using? is it ADRV9009 eval board and custom board?

    can you please capture the FFT spectrum and share it

  • Hi RR4,

    I am using our custom board.

    This is RX2 connect to TX2 and set RFon . So I think the low frequency is LO leakage.

    If I connect RX2  to a terminator. The waveform should be near zeroes but it looks like some pulses in the acquired signal. I have changed the AGC mode to manual.

    I connect the adc signals from rx_fir_decimator to ILA.

  • Can you please terminate both the RX ports of ADRV9009 and capture the FFT spectrum instead of time based capture, it helps to understand the spur signals in Rx capture.

  • Hi,

    I have done this at RX2 to a terminator and the first pic is time domain and the second is frequency domain. I did not put I and Q both because the phenomenon of IQ is the same. RX1 has no connector on PCB.

    The erroneous data was obtained by decoding packets from the ADC side. It appears to be some kind of internal data transfer error within the ADC. Based on the calibration mask, it seems that the ADRV9009 does not perform any PRBS-like validation. This is the part that makes me suspect there might be an issue.

Reply
  • Hi,

    I have done this at RX2 to a terminator and the first pic is time domain and the second is frequency domain. I did not put I and Q both because the phenomenon of IQ is the same. RX1 has no connector on PCB.

    The erroneous data was obtained by decoding packets from the ADC side. It appears to be some kind of internal data transfer error within the ADC. Based on the calibration mask, it seems that the ADRV9009 does not perform any PRBS-like validation. This is the part that makes me suspect there might be an issue.

Children