Close-In Spurs - Problem or not?

RE: Close-In Spurs - Problem or not?

michael — Mon, 06 Jul 2015 15:21:07 GMT

Reinhard, since then I haven't looked into this anymore. I could reduce the spurs by about 5dB by playing with bypass capacitors around the nRF51. The spurs did not seem to cause any problems with sensitivity or interferer testing and I was able to pass compliance testing without any problem.

RE: Close-In Spurs - Problem or not?

Reinhard — Mon, 06 Jul 2015 12:55:16 GMT

Did you resolve the issue with the close-in spurs? I'm experiencing similar spur problems - in my case located 166.5 kHz, approx -25dBc. I also experienced same behaviour on custom hardware as well as on PCA10001 evaluation board (battery supplied).

RE: Close-In Spurs - Problem or not?

Asbjørn — Wed, 28 Jan 2015 11:25:20 GMT

I'm a bit surprised that you measure this from the DK as well. How are you connecting the kit to the spectrum analyzer as these spikes shouldn't be there at least on the DK. As long as you are happy, we are happy:)

RE: Close-In Spurs - Problem or not?

michael — Fri, 23 Jan 2015 06:55:29 GMT

Thanks Asbjørn. It seems to me that the spurs are actually from the nRF51822 itself - but I may be wrong here. The reason why I think this is the case is that I measure the very same spurs also on the development kits operating from battery. Compared to my design the spurs on the reference boards are about 1-3dB lower. Nevertheless, they don't seem to cause problems with regulatory and ble radio compliance testing. I don't know how much they degrade the link budget, but it seems that for my particular appilcation I'm o.k.

RE: Close-In Spurs - Problem or not?

Asbjørn — Thu, 22 Jan 2015 11:34:48 GMT

I would try to locate the source of the ~67.2 kHz signal as this will enter the system as noise and could reduce your link budget. Do you have a motor or regulator operating at 67.2 kHz or do you have something else in your circuit that you know operates with this frequency?

Normally this kind of noise enter the radio through the power supply to the chip. So I would recommend you to use an oscilloscope and check the VDD pins on the chip. If you see the same noise there you could/should add some additional capacitance to decouple the noise. Add capacitors close the nRF51 chip.

Let me know if the noise seems to enter the chip through something else and I'm sure we'll figure out a different way to attenuate the 67.2 kHz noise.