Thingy:91X vs Thingy:91 BLE performance

Hi,

We are evaluating the Thingy:91x and found that the BLE range is significantly decreased compared to the Thingy:91. The nRF5340 has a lower maximum TX power but when testing at the same TX power (0dBm), and using the throughput sample, we see that at around 20m the throughput is about 17% lower.

Is there any way to get the Thingy:91x to have close to the same BLE range as the Thingy:91?

Best regards,

Emil

Parents
  • Hello Emil, 

    Are you able to share how your tested the performance? Are you able to share results with us? I have forwarded your inquiry to our hardware team working with the Thingy:91X.


    Kind regards,
    Øyvind

  • Hi Øyvind,

    I have attached the code that I used. It is the throughput sample from SDK v2.6.4 modified to work on the Thingy91 and Thingy91:X.

    I tested using a simple setup in the image. All of the devices have been configured to use 0dBm TX power.

    Only one of the peripherals were powered at a time.

    With the Thingy:91 the throughput was 109 kbps and the Thingy:91X was 90 kbps. This test was prompted by the fact that in our application the effective range was halved.

    The Thingy:91 is HW 1.6.1 and the Thingy:91Xs are 1.0.0 dated 2024.48.

    Best regards,
    Emil

    0525.throughput.zip

Reply
  • Hi Øyvind,

    I have attached the code that I used. It is the throughput sample from SDK v2.6.4 modified to work on the Thingy91 and Thingy91:X.

    I tested using a simple setup in the image. All of the devices have been configured to use 0dBm TX power.

    Only one of the peripherals were powered at a time.

    With the Thingy:91 the throughput was 109 kbps and the Thingy:91X was 90 kbps. This test was prompted by the fact that in our application the effective range was halved.

    The Thingy:91 is HW 1.6.1 and the Thingy:91Xs are 1.0.0 dated 2024.48.

    Best regards,
    Emil

    0525.throughput.zip

Children
  • Hello Emil, 

    Here is the initial response I got from our HW experts regarding the observations you had. Please note that the Thingy:91 X is primarily designed around nRF9151 and LTE performance. 

    1. Antenna: On Thingy:91 X The antenna is optimized for dual-band operation, which may require performance trade-offs in certain scenarios

    2. RF switch and diplexer: Sharing the antenna with Wi-Fi introduces additional components in the signal path. An RF switch is used to alternate between 2.4 GHz Wi-Fi and BLE, followed by a diplexer that integrates 5 GHz Wi-Fi support. 

    3. PCB layout: On Thingy:91 X BLE performance was not a primary focus and was therefore deprioritized during the initial antenna evaluations, which did not include a dedicated BLE antenna in trials with various vendors. As a result, the nRF5340 was positioned on the back side of the PCB, requiring the RF signal path to traverse multiple vias and follow a more complex routing to reach the antenna.

    Combining BLE and Wi-Fi on the Thingy:91 X gave another type of antenna for 2,4GHz compared to Thingy:91, which gives about 17% performance reduction compared to the Thingy:91 and is considered a reasonable outcome. This corresponds to approximately 0.8 dB in signal loss, with the RF switch alone accounting for around 0.5–0.6 dB of that.

    Hope this gives an answer to your question. 

    Kind regards,
    Øyvind

Related