Maximum RSSI value of -42 dBm

[FormerMember]

Greetings,

During a connection between two nrf51422s running on S120 (2.1.0) and S110 (8.0.0) with a connection interval of 7.5 ms I get only a maximum RSSI of -42 dBm, even when the two antennas as almost touching. When taken further apart the RSSI decreases to a minimum of about -100 dBm.

The RSSI measurement is started by

sd_ble_gap_rssi_start(m_conn_handle, BLE_GAP_RSSI_THRESHOLD_INVALID, 0);

And in a routine in between the two connection events I get the RSSI by the call

sd_ble_gap_rssi_get(m_conn_handle, &rssi_val);

What could be the reason for this behavior of RSSI not increasing beyond -42 dBm? With bare metal coding for advertisements and scanners I've noticed RSSI greater than -30 dBm too.

Thanks and cheers!

[Øyvind Karlsen]

Hi,

Is this on the development kit or using a custom board?

The performance in what is almost guaranteed to be near field will give odd results. If it is a custom board it can be because the matching is poor. Also when you are holding metal object like the development kit close to the antenna you will affect the RF performance.

A minimum of around -100 to -80 dBm is sensible, as the nRF51 has a maximum sensitivity of -90dBm in 1mbps mode. The swing of +/- 10dBm is due to RSSI measuring inaccuracies.

Best regards,

Øyvind

[FormerMember]

One is PCA10028 and the other is PCA10031. The same result is on our custom board too. Even with the erratic behavior of RSSI, -42 is quite low.

[FormerMember]

Even more confusing developments: At tx power of -20 dBm for both, the RSSI was -85 at 1 ft distance and -45 at contact. With tx power of 0 dBm for both, the RSSI was -65 at 1 ft distance and -45 at contact.

Next with the same hardware I made one device advertise (SD 110) and the other scan (without softdevice) and measure RSSI. I get the same result. At 1ft distance the RSSI changes based on the tx power of the advertiser. At contact, the RSSI measured by the scanner cannot go above -45. Can the hardware have bugs like this? At the moment I don't have any other dev boards with me to test.

[Øyvind Karlsen]

Theoretically you will see double range for each 6dB of RSSI steps, following your example that means -71dB for 2ft, -78dB for 4ft, -84dB for 8ft, - 90dB for 16ft. This makes me think your numbers are reasonable.

If you are struggelig with your hardware, please submit it for a review at our closed support system MyPage.

[Luis]

Any update on this matter? I am seeing samer behaviour using nRF51DK and module Core51822, which I believe is very popular (for being so cheap) and I expect it to have a bit tunned/optimized antenna (although I could be wrong). When both antennas are closer than 10cm, I always get RSSI -43 dBm, and that is odd, don't you think? It does not matter if antennas are 10 cm away or touching, I get -43 rssi reading, very stable value inside that range. So I think it is saturating somehow?

[FormerMember]

Not from my side. With the few devices that we've played around with, some of them DK, this behaviour is prevalent. The max RSSI is around -45 to -40 dBm. I also feel that they are saturating somehow.