What can cause a drop of the bias voltage at the RX_EXT_LO_IN pin?

We have a design with two AD9361 and want to synchronize LOs by using an external LO. For this, I apply a sinus from a signal generator to the RX_EXT_LO_IN pins and set the appropriate registers. However, this doesn't work and while debugging we noticed that the bias voltage at our RX_EXT_LO_IN pins is only 0.03V instead of 0.5V as in the FMCOMM2 reference design. TX_EXT_LO_IN is grounded, but grounding the pin on the FMCOMM2 didn't cause a voltage drop. This voltage is generated chip internal, hence my question: What can cause a drop of the bias voltage at the RX_EXT_LO_IN pin?

Thanks!

- Thomas

  • 0
    •  Analog Employees 
    on Sep 20, 2016 7:58 PM

    Did you enable external LO mode?

  • Yes, I wrote 0x30 to the Rx LO Gen Power Mode register 0x261 via direct_reg_access.

  • 0
    •  Analog Employees 
    on Sep 21, 2016 7:01 PM

    What you are seeing could be due to damage to the chip or internal configuration. Do you see this issue on more than one board?

    Please double check your SPI writes against the SPI Write sequences below. Ideally you should be using ADI provided drivers for device configuration.

    The register setting for RX Ext. LO:

    SPIWrite 005,77 // Both TX and RX use LO signals coming from Ext. LO

    SPIWrite 050,0F // Enable RX LOGEN but power down RX VCO

    SPIWrite 057,0x // Enable both TX and RX Ext. LO (NOTE preserve D3:D0 if desired)

    SPIWrite 261,30 // Power down RX LOGEN divider

    The register setting for TX Ext. LO:

    SPIWrite 005,77 // Both TX and RX use LO signals coming from Ext. LO

    SPIWrite 051,0F // Enable TX LOGEN but power down TX VCO, plus TX VCO LDO, therefore

    external supply is needed for TX Ext. LO; If using TX VCO LDO for TX Ext. LO, set Reg050 = 0Eh)

    SPIWrite 057,0x // Enable both TX and RX Ext. LO (NOTE preserve D3:D0 if desired)

    SPIWrite 2A1,F0 // Power Down TX LOGEN divider

    When using TX/RX Ext. LO, the corresponding internal RF synthesizer should be powered down (0x015[6:5] = 2’b11).

  • If I do the SPIWrites for Rx Ext. LO the chip goes into a state where it just gives out the same sample over and over. The state continues even when writing back the old values. I write the registers via direct_reg_access in the following way:

    echo 0x005 0x77 > /sys/kernel/debug/iio/iio:device0/direct_reg_access
    echo 0x050 0x0f > /sys/kernel/debug/iio/iio:device0/direct_reg_access
    echo 0x057 0x0c > /sys/kernel/debug/iio/iio:device0/direct_reg_access
    echo 0x261 0x30 > /sys/kernel/debug/iio/iio:device0/direct_reg_access
  • 0
    •  Analog Employees 
    on Dec 7, 2016 12:22 PM

    The values are correct but you are writing in wrong device.

    iio:device0 is - ad7291

    Change it to correct device as below and try again

    iio:device1 - AD9361

    echo 0x005 0x77 > /sys/kernel/debug/iio/iio:device1/direct_reg_access
    echo 0x050 0x0f > /sys/kernel/debug/iio/iio:device1/direct_reg_access
    echo 0x057 0x0c > /sys/kernel/debug/iio/iio:device1/direct_reg_access
    echo 0x261 0x30 > /sys/kernel/debug/iio/iio:device1/direct_reg_access