Dear Team, One of our customer have the next problem with DAC9125, can you please help us?
We are using an Analog Devices AD9125 DAC.
It is operating at 100 Msps baseband sampling frequency. We are using 8x interpolation and the coarse mixer to upconvert the baseband by fs. We are using only one output.
We have a strange problem, which seems to be an analogue output interfacing or grounding problem, maybe the digital control is interfering with the analogue output.
The phenomenon is the following:
In the close range of a certain output signal level (around -40 dBFS) we get a huge ripple in the output spectrum. I have attached a spectrum analyzer screenshot.
Above this level and below this level the ripple is smaller, nearly unvisible. The periodicity of the ripple is equal to the inverse of the DAC delay.
We think that the digital data and the delayed analogue signal are interfering somehow.
We have experimented a bit and we have found that if we add a small DC offset to the baseband signal this ripple is greatly reduced or it nearly vanishes. Because of the coarse mixing this DC offset increases the level of the 100 MHz sampling clock by 30 dB and gets in the range of -20 dBFS.
Maybe the higher out-of-band output signal level has some advantage?
Do you have an idea what can be this phenomenon and how can we get rid of it?
Please find attached the schematic of the DAC circuit.
We need a little more information to help sort out this problem. It is not clear what configuration you are trying to use the AD9125 in. What is the input data rate 100MSPS or 12.5MSPS?
What filter modes have you selected for HB1, HB2 and HB3. HB1 is in mode 2 or 3? What about HB2 and HB3 modes. What input frequency and input level are feeding into the AD9125 data inputs when you obserb the ripple on the output.