5340 + 21540 timing

[skajam66]

Hi,

The attached screenshot shows a section of an air-trace for the 5340 running a commercial LE Audio stack. From the rising edge of the logic inputs, it can be seen that the 5340 RADIO_START task is triggered before FEM_TXEN is asserted. This would suggest that there is active RF (maybe unmodulated) being emitted by the 5340 before FEM_TXEN is asserted.

Questions:

1. Is this timing normal? (I would have expected RADIO_START task to occur after FEM_TXEN is asserted)
2. How long after the RADIO_START task is triggered does active RF appear on the 5340 ANT pin?

Kind regards,

ac

[Bendik Heiskel]

Hi,

 This timing is not normal at all, when the start task is triggered the radio is already ramped up and the carrier is already started:

https://docs.nordicsemi.com/bundle/ps_nrf5340/page/radio.html#ariaid-title7

The FEM_TXEN should have been asserted "long" before the start task.

The nRF21540 needs a minimum of 27.5us from power down to TX, so the order the should be PDN pin asserted followed by TX at least 17.5us later. This should happen early enough so that the FEM is in TX mode before the radio is finished ramping up. The rampup time of the radio is 40us, so the startup of the FEM can be started at the same time as the TXEN task is started.

https://docs.nordicsemi.com/bundle/ps_nrf21540/page/chapters/device_control/doc/state_transition_timing.html

Here is a example plot of the correct timing of the PDN and TXEN pin in relation to RF out of the radio:

Channel 1 is showing a power detection connected to the output of the ANT pin of the nRF5340, the transition from high to low is due to the radio transmitting.

Ignore the first pulse on the TX_EN pin, this was a artifact of the measurement setup.

 

The behavior of the PDN pin is also not correct,  In normal operation the PDN pin will be low by default, keeping the nRF21540 in power down by default, only exiting power down for the radio events, TX/RX.  Here the nRF21540 is being kept on by default, and momentarily shut off during the transmission.

 

Best regards,

Bendik

[skajam66]

Hi Bendik,

Many thanks for your response.

On your last comment about the 21540 PDN pin, the screenshot that I sent was zoomed in so that you could see the timing to the 5340 and 21540 signals. Below, I have zoomed back out and you can see that the PDN pin is being deasserted between transmissions.

Kind regards,

ac

[Bendik Heiskel]

Ok, this looks more normal with regards to the default state of the PDN pin, but it is being de-asserted briefly during the transmission in the first plot. This will cause the nRF21540 to exit TX the state and return to PDN:

So something is off the way the FEM is being controlled.

[skajam66]

> So something is off the way the FEM is being controlled.

Agreed. I can look into this. Just wanted a second opinion.

Many thanks,

ac