[USRP-users] X310 Transmit Burst Issues
carel.combrink at gmail.com
Wed Mar 30 02:58:04 EDT 2016
> Did you ever try that patch I sent? If so, did it help?
I finally got around to spend time to test the patch that you have supplied
for the transients. When we started looking at the frequency of the
transmitted data, it appeared that there are some settling time on the
frequency of the initial transmitted data between bursts, before the
frequency gets locked. We suspected that it might be due the components
getting switched off and on again and then having to lock again. The patch
seemed to have fixed this start-of-burst frequency issues. I have not yet
looked on the oscilloscope how the data looks but on the Real-Time Spectrum
analyser it looks a lot better. To quote my RF Engineer: "It looks very
Some more information from the Real Time Spectrum Analyser: On burst start
the frequency can be off anywhere between 2 MHz and 15 MHz (what we
observed). If it is off by the latter, it takes about 75 us for the
frequency to stabilise and about 175 us for the phase to stabilise.
With version 3.9.3 that was recently released, I am assuming that I will
need to re-apply the patch if I want to keep the functionality. If yes, is
there a reason why this is not incorporated into the main branch since
other users also finds it useful and it solves a few issues related to the
device operating in a burst mode. This can be in the form of perhaps a
software setting or mode or make it the default.
Regarding my initial issue (the reason for the thread), the buffer issue,
was any time spent on further investigating the buffer issues that I have
seen and described.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the USRP-users