RX sensitivity test (S2 and S8) on DUT 2 DUT mode

[Gordon Liu]

Dear Sir,

I'm use 2 nrf52840 chips to test sensitivity by nRF-DTM-1.0 tool.

I find out it's work on 1M and 2M but S8 and S2 can work.

The PER can't get 0% when I only use 25 dB attenuator between 2 DUT ( conducted in the shielding room).

Could you share other  tool to test sensitivity of S8 and S2???

PS: I don't use any equipment(Anritsu, Keysight...) on production.

[Kenneth]

Hi,

Can you try using the built in DTM in the nRF Connect for Desktop?
https://www.nordicsemi.com/Software-and-tools/Development-Tools/nRF-Connect-for-desktop 

I am not sure what PER value you are measuring, but since the packets are longer when using S2 and S8, it may be simply that the first and last packets are not received due to the receive window is too tight.

Best regards,
Kenneth

[Gordon Liu]

Hi,

I have use "DTM tool  of nRf connect 3.6.1".

I find out S8 and S2 still have packet loss 350ea ~100ea when I send 1000 packet.

PS: I use the Lite-point IQXel to capture the signal pattern of a S2 packet.  

[Kenneth]

Can you also try with two DK as shown here?

When using 30byte payload with S2 there will be a packet sent every 1.25ms (when using S8 every 3.125ms), so that means I should receive about 8000 packets in 10seconds when using S2. That seems to be correct here.

[Gordon Liu]

Hi

Thanks for your responses.

I find out the Rx test tool that need more timing accuracy of receive packet.

So, I change the capture pattern (RF signal) that be 100 packets be a pattern file.

I can test S2 and S8 now, but S2 sensitivity less S8 sensitivity  5 dB.  It's not make sense.

Do you have any comment??

And Please share the send time per  a packet  for 1M and 2M (30byte).

 

[Kenneth]

Hi,

From the datasheet I can find sensitivity values:
https://infocenter.nordicsemi.com/topic/ps_nrf52840/radio.html#unique_544454632

Looks like there is a typical 4dB difference between 125kbps(S8) and 500kbps(S2). So that is very close to what you measure of 5dB.

Looks like your measurements are a bit worse compared to datasheet:

Difference 1Mbps datasheet vs. your measurement: 1-2dB
Difference 2Mbps datasheet vs. your measurement: 0dB
Difference 125kbps(S8) datasheet vs. your measurement: 3-5dB
Difference 500kbps(S2) datasheet vs. your measurement: 4-6dB

I think maybe you have some low level noise in the system (e.g. USB) that may impact the lower sensitivity levels here. If the loss was due to cable and connectors I would expect them to influence all measurements. You could for comparison use 2 DK's and a step attenuator between the two DK and see how much attenuation you can have until the packet loss increase to about 30%.

For 1M and 2M they are both 625us when using 30byte payload.

Best regards,
Kenneth