DC/DC power consumption peaks after TX

[Sven]

Hello!

I have PCA10006 with revision QFAA nRF51422

I am testing the buck converter's ability to make the system consume less power when transmitting. Example is ant/broadcast_tx from SDK 11 with S210 ver. 5 soft device.

Seems like the RMS is really lower as the maximum is not that high as when DC/DC disabled. I am using the following API as:

APP_ERROR_CHECK(sd_power_dcdc_mode_set(dcdc_mode));

One thing that seems weird is the short peaks that happen at the end of the power consumption "wave". The yellow line shows the power consumption measured from the power measurement pins of the DK. The blue line is the actual power from the antenna (an external sniffer with power detection abilities). As you can see the actual power output from the antenna matches the power consumption. But can someone explain those short peaks after that?

Thanks in advance!

[run_ar]

What is the build code on this chip? QAAA??

[Sven]

The full revision - QFAAE0

[run_ar]

This is a rev 2 IC. So the dc/dc is not really supporter when running the softdevice.

[Sven]

Ok thanks, good to know, are you able to elaborate more? Is there somewhere (like PAN) said something about this?

Still, from a perspective of engineering, what those peaks could be caused of?

[run_ar]

I assume they are caused by the dcdc beeing turned off / coupled to the battery, so you will get a power distribution between the capacitors on the power source and the one on the dcdc. What power source are you testing with? If you test wuth a battery as source I would expect the spike to be reduced due to the internal resistance in the battery.

You should not use the dcdc with the second revision chip as it could affect radio performance. Regardless I would recommend using the latest revision ic for development as this is also the chip revision you will be able to buy for productions.