Dear Sir:
We have another about AD8339, could you help ?
 
There is DC bias problem on the outputs of AD8339, even when there is no signal input to AD8339.
The following figures show three consecutive ADC reading(1st/2nd/3rd) on our RX board, you can see there is a strange transient response during about sample 1~100, and there is a DC bias after sample 100, the DC bias are different on each ADC reading.
 
We had found :
DC bias would be changed if RSET signal(4LO divider reset) is issued
DC bias would be changed after each re-configuration of 20-bit shift register
If RSET and 20-bit shift register is kept un-touched, DC bias would keep on the same level