Hi Alex,
your application sounds cool :)
So, amplitude stability over ten seconds isn't usually a design goal –
how strong are these amplitude modulations?
Also, I'd have a standard visualization that I'd usually share here:
offset tuning
$f_\text{RF}$ is the frequency of the actual (physical) LO on the
daughterboard, whilst $f_\text{offset}$, as you've guessed, is done by
multiplying the data stream (after interpolation to the 100 MHz master
clock rate) with a "digital"
$e^{j2\pi\frac{f_\text{offset}}{\SI{100}{\mega\hertz}}}$ in the FPGA,
before giving it to the DAC.
Now, that "digital mixer" is implemented using a CORDIC with an
inherently limited phase accumulator bit width – I'd see a reason for
that to experience some inaccuracy, but that would look more like a
millihertz frequency offset, I guess.
Also, note that I'm assuming you're either Using a UBX or WBX
daughterboard – as far as I know from the top of my head, these are the
only two that actually go down to 50 MHz. Maybe, on the other hand,
you're using a BasicTX or LFTX instead – in that case, there's no LO, so
$f_\text{RF}\equiv 0$. So: what daughterboard /are/ you using?
Best regards,
Marcus
On 02.08.2017 23:33, Alex Pesch via USRP-users wrote:
> Hello -
>
> I'm attempting to use my N210 in a somewhat unconventional application
> that requires a pure (real) sine wave output, with some frequency
> modulation around 50MHz. This signal will be used to resonate a
> physical device within a small band around 50MHz.
>
> However, I'm finding when the center frequency is set to 50MHz, and I
> monitor the resulting waveform on an oscilloscope, I get a 50MHz tone
> that is slowly amplitude modulated (within a ~0.1 Hz envelope). This
> behavior does not occur when I set the center frequency to 0Hz, and
> feed in an arbitrary intermediate frequency... Unfortunately, I cannot
> use this technique to achieve a 50MHz output frequency due to
> bandwidth limitations.
>
> I suspect the LO is getting set such that there is some time-variant
> phase offset being introduced (rounding error?), but am unfamiliar
> with the internals of the device.
>
> Any tips/suggestions would be appreciated - I would like the RF output
> to simply resemble that of a signal generator at 50MHz.
>
> Thanks in advance,
> -Alex
>
>
>
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
_______________________________________________
USRP-users mailing list
[email protected]
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
