5 years, 7 months ago.

SX1276MB1MAS 433MHz range and noise problem


We have a problem with SX1276MB1MAS and ST32F411RE Nucleo boards. The SX1276PingPong tests at 433 MHz band are very poor in range, compared to 868 MHz.

With a spectrum analyzer we can detect a well formed FSK spectrum signal with power levels of -70 dBm near the receiver board. But the demodulator receives the frames with errors. The theoretical sensitivity is about -105 dBm. The band is a bit more noisy in 433MHz, but this not explains this diference in range compared to 868 MHz.

The antenna is connected on the LF side. We are using the 433MHz antenna that comes with SX1276MB1MAS. We bought another SX1276MB1MAS board to discard a hardware problem, but the behaviour was the same.

We have corrected some issues in SX1276PingPong mbed application. One of those is that in callback function "OnTxDone" from DIO0 isr, it needs a delay of 2 mseconds before sending the transceiver to sleep. If this wait is not implemented, the transmitter could be sleeping when is yet transmitting the CRC bytes and the receiver could detect a CRC error. This bug seems that is only affecting to 433MHz.

However, even with the improvements, behavior at 433 MHz remains worse than expected.

We have tested with other Nucleo boards (NUCLEO-F334R8) and the problem persist.

We suspect that the problem could be in the transmission branch, because if we put attenuators in the receptor antenna, the RSSI measured in demodulator can reach -100dBm. One might conclude that the signal is already noisy when leaving the transmitter itself. But we aren't sure of that.

Is there any error in SX1276 transceiver at 433MHz that is not documented in the errata-sheet?

Is there any design bug in SX1276MB1MAS or antenna tunning?

Does 433MHz band need any especial register setting?

Does anybody have had these same problems with 433MHz? I have found a post in openlora forum that could be related with this issue: http://openlora.com/forum/viewtopic.php?f=12&t=651&hilit=433

Thank you and regards,


Question relating to:

2 Answers

5 years, 3 months ago.

we also have a problem with 455MHz any response yet?

I haven't received any answer.

posted by Gilen Tell 14 Sep 2016
5 years ago.

I never used Lora transceivers but based on my radio development experience, very often some jamming sources are the cause of this kind of issues, 1/ This can be external jamming at very low level that can't even be seen with a spectrum analyzer, So the best way to check this, is to connect directly the two transceivers through attenuators and coax cable without using antennas. If it fix the issue it means that noise or jamming is received by the receiver..

2/ The jamming can be internal source coming from the development kit you are using, especially at frequencies matching with clocks harmonics. Did you see special frequencies at which this problem occurs? Is there any specific relations between all these frequencies that may help to understand the issue?