nRF9160 Output Power and VSWR (load pull)

[Stephen Snyder]

Hello,

We are working toward certification of our product and conducting TRP/TIS measurements.  The TRP and TIS are not matching up with the conducted output power and the passive efficiency of the antenna.  Our antenna Band 12 resonance is kind of narrow, and I'm wondering if the reason for this is that output power of the nRF9160 is degraded due to high VSWR / non-50 ohm loads.

Can you provide guidance on how the output power may vary with load?  It would be wonderful to get load pull with circles of constant power.

Thanks so much,

Steve

[ketiljo]

The PA have an internal feedback so it will adjust the power being delivered to the antenna. So the output power will not be completely dependent on the SWR of the antenna.

If you you use any of the AT RFTEST commands to setup the nRF9160 in TX mode, it will return a value. Divide this by 16 to get the output power in dBm. https://infocenter.nordicsemi.com/topic/ref_at_commands/REF/at_commands/production_test/xrftest_tx.html

[Stephen Snyder]

Thanks Ketiljo, that's useful.

But do you have any characterization data to share?  If we had a smith chart with constant power circles, we could have a target area that we want to be in when passively tuning the antenna using a network analyzer.  Without this there is more trial and error and the tuning process will go slower.

[Stephen Snyder]

Hi Again Ketiljo,

I've experimented with the XRFTEST At command and it seems to be giving me values that don't make sense.

Basically, I get the same output power returned whether an antenna is connected or not.  I've reproduced this on the nRF9160-DK.  I use a murata test connector inserted into J1 with nothing attached to it as my no antenna test.  This appear to happen with 16-QAM and CW modulation.  See attached for more detail.

I would expect that despite any internal feedback, the PA can't deliver 23dBm to an open circuit.  Please help explain.  This command would be very helpful for antenna tuning OR it would be helpful to have load pull data, but at the moment we don't have either.

Thanks for your help.

Steve

XRFTEST Antenna Power.csv

[ketiljo]

I'm sorry, but load pull measurements are not available as public information. But in any case, a wideband 50ohm match is recommendation. ANT pin mismatch up to VSWR3:1 (due to typically is considered to be still acceptable performance wise.

Post-PA RF power coupler directivity in heavy mismatch/open conditions degrades and power measurement will be impacted by reflected power. Adding 50ohm termination at ANT pin and re-do measurement would show actual power towards load in typical conditions.