We are generating RF with a redpitaya (external clock) using the NCO bloc (40bit accumulator, 12bit or 16bit LUT) and we see quite strong spurs at 5 MHz and harmonics offset from the desired carrier frequency:
Here the programmed DDS-output is 20MHz and we see a second harmonic at ca -40dB from the carrier - probably ok. However the spurs at multiples of 5 MHz from the carrier are unwanted and we would need some narrowband bandpass-filter to remove them.
The 125 MHz differential clock we are feeding the redpitaya looks reasonably clean - probably not the source of the spurs at 5 MHz offset?
If anyone has seen this before and there is a fix I would be interested - thanks. If this is just a feature of the redpitaya and NCO design then we will try with ca 1-2 MHz wide bandpass filters on the output of the DDS.
We are generating RF with a redpitaya (external clock) using the NCO bloc (40bit accumulator, 12bit or 16bit LUT) and we see quite strong spurs at 5 MHz and harmonics offset from the desired carrier frequency:
Here the programmed DDS-output is 20MHz and we see a second harmonic at ca -40dB from the carrier - probably ok. However the spurs at multiples of 5 MHz from the carrier are unwanted and we would need some narrowband bandpass-filter to remove them.
The 125 MHz differential clock we are feeding the redpitaya looks reasonably clean - probably not the source of the spurs at 5 MHz offset?
If anyone has seen this before and there is a fix I would be interested - thanks. If this is just a feature of the redpitaya and NCO design then we will try with ca 1-2 MHz wide bandpass filters on the output of the DDS.