I can not find frequency resolution specification in QA403 documents.
What’s the minimize frequency step of QA403 Generators?
I need to generate a sine wave with frequency 1000.03275Hz. Is it possible with QA403?
Even if you did set it to that number the internal clock is not accurate enough to meet that spec. You would need a frequency synthesizer with a precision reference or gpsdo reference. I’m not sure any go that low with that resolution. 500ppb.
What requires that precision?
Hi @Thomas, the sine waves are generated numerically, so the relative accuracy will be there. That is, a 1000.000001 Hz tone will be higher in frequency than 1000.000000 Hz tone.
But as @1audio notes, you need to also be concerned about the clock accuracy. The oscillator used for all audio timing is ECS-7050MV-491.52-BN-TR, and this has a frequency stability of +/- 50 ppm from -40 to +85C. At the bottom of the spec, the manufacturer notes this tolerance includes the initial tolerance, temp, supply voltage, the shock from reflow soldering and also aging.
And that 50 ppm error will translate to the 1 kHz tone, which means a max error of 0.05 Hz. Now, you won’t see this in measurements where the ADC is measuring the DAC after running through a DUT, because both will have the same error. It will appear perfect. But if you ask a PC DAC to generate a 1000.000 Hz tone, you will have the xtal errors compound. The PC might have a +/-100 ppm xtal and the analyzer has a 50 ppm xtal, and so the combined error is +/- 150 ppm.
Finally, note the xtal manufacturer also has a +/-25 ppm and +/-20 ppm option for this crystal. This means that the 50 ppm xtal will probably have a bimodel distribution. That is, all the 25 and 20 ppm performers are removed from the population and what is are to the left and right of the nominal. So, a pessimistic view would suggest the QA403 clock is probably either 15 ppm high or 15 ppm low, roughly.
3 Likes