nrf51 Radiated Emissions Fail

[user9513]

Hi, we are using nRF51822-QFAA in our project which is based on the PCA10000-nRF51822 Development Dongle 2.0 - layout of radio part is almost the same. I'm also using sdk v6.1.0 with s110_nrf51822_7.3.0_softdevice. Then I've prepared application which is based on radio_test_example with ability to configure radio to transmit on different static channels (top, middle and bottom) and send it to the lab. And finally the result is failed. As you can see in the graph attached (www.dropbox.com/.../RE1 FAIL 1-25GHz High CH.png, the harmonics of the transmitter exceed the limit. Marker number 12, the 2nd harmonic, is the highest measurement exceeding the average limit by +4dB. In addition to the harmonics, there appears to be a spurious emission at marker #11, 2565MHz which when the average was measured it exceeded the average limit by +7dB. So do you have any ideas what may cause that?

[Dave_couling]

There are a few potential causes:

1. Antenna design - RF is extremely sensitive to dimensions, ground clearance, substrate thickness, transmission line tuning ...etc

2. Balun: Did you use a discrete balun or chip balun (safer)

3. HFCLK: Verify that your HF crystal is oscillating at the correct frequency.

[FormerMember]

The layout of the RF part is very important for correctly suppressing harmonics. Could you upload a picture of the layout?

[user9513]

there is a print screen of RF part layout: www.dropbox.com/.../antenna_layout.png

[FormerMember]

The matching network is very sensitive. It is therefore important to follow the following:

• The distances in the matching network should be exactly the same as in our reference layout. That also includes the distance from ANT1/ANT2 to L1.
• There should be ground plane under the entire matching network. Ground as trace will not work. Instead of ground, it will act as a series inductor.