[time-nuts] Re: Eloran long test from now to August or September.

[Jim Lux]

Exactly.. That does work, although it requires opening up your RTL-SDR (and on V3, in HF receiver mode, the RF input is directly coupled to the Q input, without going through the front end downconverter.). It didn’t work well.
And ultimately, I was trying to figure out something that would work at Ku band, so I could “calibrate out” the delay through the LNB, the coax, and the RTL-SDR.  
The basic technique has been around since the 1960s. Â Bracewell used it at Stanford, and a similar scheme was used for the Jansky VLA in New Mexico.


On Tue, 26 Sep 2023 19:24:21 +0000, jeanmichel.friedt at femto-st.fr wrote:

> I did it with 4 RTL-SDRs, and I’ll scrounge up the measurements. My first attempt was just to
> couple the 1pps from a GPS module into the inputs - that’s not easily detectable (I was hoping that
> the fast edges would be sufficiently broadband to be detectable in the sampled data stream -
> nope..)

See attached: I capacitively inject the 1-PPS on the RTL2832 input (either I or Q), not
the R820(T2) radiofrequency frontend. This allows for timestamping on the rising edge
of GPS 1-PPS the datastream fetched by the ADCs of the RTL-SDR.

Best, JM