Can AD converter AD9122 generate single test tone with the internal NCO?

Hi tony_tao ,

Thank you for your interest in AD9122. The NCO tuning word is configured in FTW [31:0] in Registers 0x30 to 0x33 (FTW_0 to FTW_3). After loading the FTW registers with the desired values, Bit 0 of Register 0x36 must transition from 0 to 1 for the new FTW to take effect.

Regarding your second question, I tried this on bench using this Wiki User Guide.  I didn't the connect the FPGA (ADS7-V2 controller board), I only configured the AD9122 through SPI using ACE (software developed by ADI). I used Fdac = 800MHz Interpolation x4 and generated a 20MHz output frequency through NCO. This corresponds to a FTW of 0x0CCCCCCD or 214748365 (in decimal). As seen below, the NCO output is still seen. For this case, the DAC was connected to the Balun outputs in the AD9122 Evaluation board, however the same should be seen if the DAC was connected to the ADL5375 modulator output.

However, once the FPGA has been loaded (using DPGLite), this sets the data lines (D0 to D16) to a known state which turns the NCO off. By then, it will be required to send data to the FPGA to output again through the NCO. Figure 57 in the datasheet provides the block diagram for the quadrature modulator. The I/Q data is mixed with the NCO output to generate the quadrature carrier signal, and translates the input signal to a new center frequency. 

May I ask your target application for this? 

Best regards,
Marco

Hi Macro

              Thank you for your experiment of AD9122 demo and measurent of output.

              Is there any  definite theory description of nco test tone can be generated from ad9122?

              We just  want to elimate of traditional ways of liquidometer transmitter with  only  AD9122  together with modulator to genertate singletone .

Hi tony_tao ,

Apologies for the delayed response. I tried the experiment again on another AD9122 evaluation board. Although I'm able to generate an NCO output again using the same procedure as in the first board, the amplitude of the tone was arbitrary whenever I either reset the AD9516 chip (Clock distribution) or perform a power cycle (i.e. turned off the power supply then turned on again). There were even instances where there was no output. 

The amplitude of the single tone was being random depending on the power-on values in the FIFO. Since there is no FPGA board connected, there should be no DCI detected by the AD9122 chip (although in some instances there is which is weird). There are FIFO warning fields in Register 0x18 [7:6] and the FIFO level is indicated in Register 0x19. The value in these fields were not constant during my testing thus the random amplitude levels in the NCO output. 

Due to this, the answer to your question is no. You can't output a single test tone with the internal NCO only. However, there should be a workaround to this: If you could tie the input data bus to a fixed DC value and find a way to toggle DCI several times so the value gets loaded into the FIFO, this would ensure repeatability that there is always an NCO output and the amplitude will always be the same. 

Best regards,
Marco

Hi JMMina,

                   Thank you for your effort to help me for the single tone questions,  I have received a satisfactory reply.