• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 53 subscribers
  • Views 545 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

C1125 noise level too high while programming it in my own setup

Kilian Timmler
Prodigy 50 points

Hi,

We are using a CC1125EM together with a STM32 discovery board to receive an 2-FSK signal at 869.2125 MHz in transparent mode. While everything works fine so far I'm unable to reach the low noise level I can achieve with the same EM on the TRXEB. I'm testing this with the same signal generator at -80dBm. Under SmartRF Studio7 I get a -86dBm signal and a -128dBm noise floor while with my own register setup I'm getting only a 1dB difference between noise floor and signal at 80dBm while I'm also unable yet to set the RSSI offset correctly.

While my next step would be to retrieve my register settings so we can compare them for differences, I would like
to get some preliminary idea on where to look for any potential issues with my register settings.

My current focus is on the AGC settings but I have not found an issue with them yet.

I'm basically hoping for a pointer to an FAQ entry or a similar problem, which could resolve my issue befor I start extracting registers from our setup. Also I would like to know in which format I should post them.

Best Regards,


Kilian

  • 0
    TI__Guru**** 317180 points

    I assume you have tried to use the same register settings with the TRXEB and you own board to rule out hardware differences.

    Which datarate, deviation, modulation format are you planning to use? It could be that we have a suggestion to a starting point.

  • 0 in reply to TER
    Prodigy 50 points

    What we had in our initial setup that we build in a 2 week timeframe, is close to the settings I will attach here. That means that I'm dissatisfied with the agreement between our setup and the current TRXEB setup, but it worked as a simple prototype with the SNR issue and the register values were also TRXEB derived. So no I'm not using the same register settings right now. The GPIO pins are used differently but I will try to make the remainder identical in another test.

    The parameters are basically, 2.5kHz data rate, deviation is +/-2.5kHz, and the modulation used is 2-FSK. I'm generally using the 10kHz RX filter bandwidth, in our current setup I tried reducing it to 8kHz to get some SNR gain for the RSSI measurement (which doesn't help all that much).

    After all I wouldn't expect a 40dB difference in the noise floor. I'm wondering what could go that wrong. There might be an issue with power supply filtering that I haven't checked yet so if I have to put a llinear regulator in front of the EM VCC input please let me know.

    Regards,

    Kilian

    cc1125_regs_combined.xls
  • 0 in reply to Kilian Timmler
    TI__Guru**** 317180 points

    It's not clear to me if you have used the same hardware in both cases. This is how I understand your setup:

    1) EM from TI connected to TRXEB

    2) EM from TI connected to your MCU

    Have you checked that the center of your channel is at the same frequency as the SG? Do you have the same issue is you set the RX BW to 25 kHz (for debug) Not sure why this should be the case if you use the same EM but this is a very typical issue  with narrow RX BW compared to the signal bandwidth.

  • 0 in reply to TER
    Prodigy 50 points

    Your understanding is correct. The EM is identical, and also the cable to the signal generator.

    The receiver needs to perform a frequency correction for the LO offset, so far I'm trying to use the RSSI value for it. The graph inside/below shows the frequency offset in kHz and the RSSI value in dBm. This uses the 8kHz filter and the frequency stepwidth is 4kHz. The odd thing is that despite the programmed correction value in AGC_GAIN_ADJUST the result lies around -34dBm. The signal level is currently set to -80dBm @ 869.21250MHz

    rssi_80dbm.pdf
  • 0 in reply to Kilian Timmler
    Prodigy 50 points

    I tried the 25kHz suggestion it totally changes the whole situation.

    Somehow I'm not sure now how to properly set lower bandwidths, what did I miss? Anyway thanks for the suggestion, I feel much better now. The signal I'm using is just a CW signal btw.

  • 0 in reply to Kilian Timmler
    Prodigy 50 points

    I used a working 8kHz BW setup from RFStudio in my setup and was able to get a useable result with the same input as before. There is no  RSSI correction in place yet but the 40dB I was expecting I do get.

    So thanks for the support, I guess I should have been a bit more careful with the settings.