Frequency offset in %XRFTEST CW testing

[Norbert Braun]

Hi,

I noted that the CW generated in the %XRFTEST TX test seems to be offset from the requested frequency. Specifically, issuing the command

AT%XRFTEST=1,1,20,8500,-50,0,4,0,0,0,0,0

generates a CW wave with a frequency of approximately 850.255 MHz (255 kHz off). Changing the requested frequency seems to leave the offset constant.

I am using a PCA10090 devboard (v0.8.5) with modem firmware mfw_nrf9160_1.2.0. Another devboard with modem firmware mfw_nrf9160_1.0.0 shows the problem as well.

Switching to generating a modulated signal, e.g. with

AT%XRFTEST=1,1,20,8500,-50,0,0,12,0,0,0,0

generates a spectrum that is nicely centered around 850.0 MHz, so this does not seem to be a hardware issue.

Is this a known issue, or perhaps expected behavior? I could not find anything in the documentation of the %XRFTEST command.

Best regards, Norbert

[Håkon Alseth]

Hi Norbert,

 

 

Is this a known issue, or perhaps expected behavior? I could not find anything in the documentation of the %XRFTEST command.

My apologies for the inconvenience, and thank you for reporting this back to us!

Given that the second command with modulated output is spot on frequency-wise; It sounds like you have stumbled upon a issue on our side. I will check internally and confirm the issue, as I unfortunately do not have access to a spectrum analyzer today, and get back to you soon (a couple of working days).

 

Kind regards,

Håkon

[Håkon Alseth]

Hi Norbert,

 

Sorry it took a bit of time before I got back to you. I can confirm the behavior, and it is actually done on purpose (in order for certain functionalities, like reporting power output from the AT command), but we have forgotten to properly document this offset. Future revisions of the AT document will have this behavior documented.

 

Again, thank you very much for taking the time to report such oddities back to us, and helping us improve!

 

Kind regards,

Håkon