• State Resolved
  • Locked Locked
  • Replies 18 replies
  • Answers 2 answers
  • Subscribers 54 subscribers
  • Views 364 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

CC1352P: Continuous RX in SmartRF studio, possible bug

Arthur Thompson27
Expert 1800 points
Part Number: CC1352P

Hi, I am trying to take some measurements related to the CC1352P receiver selectivity performance.  I am seeing some strange results when I look at the Continuous RX signal level in Smart RF Studio.  

Rather than attempt to format the post for the forum, I created some slides and am attaching them.

Smart RF Studio Continuous RX bug.pptx

Please take a look at the slides and try to help me understand the strange behavior and whether it is a bug.

Thanks,

Arthur

  • 0
    TI__Genius 11810 points

    Hi Arthur, 

    We are working on reproducing this bug and will get back to you shortly. 

    -Simon

  • 0
    TI__Genius 11810 points

    Hi Arthur, 

    We have not been able to reproduce your issue. Can you verify that both LPs have a stable supply voltage when you run the test? Can you send us the exact settings you are using in SmartRF Studio? 

    -Simon

  • 0 in reply to Simon J
    Expert 1800 points

    Hi Simon, can you share pictures of your findings from when you tried to reproduce the issue?  I'm wondering which part you could reproduce, eg. you see the noise floor of the 902.4 RX go up when a 902.8 TX is turned on? And you also see the noise floor of the 902.4 RX go up when a 927.6 TX is turned on?

    I attached a file for each launchpad that I got from "Save cfg" in Smart RF Studio.  Hopefully that gives you the exact settings.  

    Thanks,

    Arthur

    902p4RX.xml
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    <?xml version="1.0" encoding="ISO-8859-1"?>
    <!DOCTYPE configuration SYSTEM "C:/Program Files (x86)/Texas Instruments/SmartRF Tools/SmartRF Studio 7/config/xml/configdata.dtd"[]>
    <dcpanelconfiguration>
    <Devicename>CC1352P_PROP_PG21</Devicename>
    <Description>Saved configuration data</Description>
    <Patch>
    <Define>RF_MODE_AUTO</Define>
    <Cpe>rf_patch_cpe_prop</Cpe>
    <Mce></Mce>
    <Rfe></Rfe>
    </Patch>
    <commandsettings>
    <Command name="CMD_TX_TEST">
    <Field name="commandNo">0x00000808</Field>
    <Field name="status">0x00000002</Field>
    <Field name="pNextOp">0x00000000</Field>
    <Field name="startTime">0x00000000</Field>
    <Field name="startTrigger">0x00000000</Field>
    <Field name="condition">0x00000001</Field>
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    927p6TX.xml
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    <?xml version="1.0" encoding="ISO-8859-1"?>
    <!DOCTYPE configuration SYSTEM "C:/Program Files (x86)/Texas Instruments/SmartRF Tools/SmartRF Studio 7/config/xml/configdata.dtd"[]>
    <dcpanelconfiguration>
    <Devicename>CC1352P_PROP_PG21</Devicename>
    <Description>Saved configuration data</Description>
    <Patch>
    <Define>RF_MODE_AUTO</Define>
    <Cpe>rf_patch_cpe_prop</Cpe>
    <Mce></Mce>
    <Rfe></Rfe>
    </Patch>
    <commandsettings>
    <Command name="CMD_TX_TEST">
    <Field name="commandNo">0x00000808</Field>
    <Field name="status">0x00000002</Field>
    <Field name="pNextOp">0x00000000</Field>
    <Field name="startTime">0x00000000</Field>
    <Field name="startTrigger">0x00000000</Field>
    <Field name="condition">0x00000001</Field>
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Arthur Thompson27
    TI__Genius 11810 points

    Hi Arthur, 

    Here is my results:

    As you can see, no difference. Have you tried to test this inside an RF shield? Note that I am located in a county where the 900 MHz band is licenced. Are you sure that you don't have any other transmitters nearby? Have you tried to power the LPs from two different laptops?

    -Simon 

  • 0 in reply to Simon J
    TI__Genius 11810 points

    And here is an RX sweep showing RSSI at different channels:

  • 0 in reply to Simon J
    Expert 1800 points

    Hi Simon,

    You have not shown the RX sweep for a TX on at 927.6 MHz.  Could you please update the screenshot above to include that measurement as well?

    Can we compare Smart RF studio versions and Launchpad versions?

    SmartRF Studio 7 2.20.1

    Launchpad CC1352P1 HW Rev. B.

    Thanks,

    Arthur

  • 0 in reply to Arthur Thompson27
    TI__Genius 11810 points

    Arthur, 

    I was not able to get 927.6 MHz in the same screenshot, however, I was not able to see any signal anyway - all I was able to see was the noise floor. 

    I used SmartRF Studio 7 2.22.0 and a rev. B LaunchPad. 

    1. Have you tried to use two different laptops? One transmitting and one receiving? 

    2. Can you verify that you have a 3.3 V on the 3V3 connector on both LPs?

    3. Have you tried to run this test conducted? You can do so by moving C482 to C483 footprint connect a coax cable on J7.

    -Simon 

  • 0 in reply to Simon J
    Expert 1800 points

    1. Using 2 different PCs for RX and TX, but still keeping the LPs close to each other, I saw the same results as before.  The RX RRSI went down lower while the TX was set to 927.6MHz, and the RX RSSI went back up once the TX was disabled.  I updated to V 2.22.0 for this test, just to match your setup.

    2. I don't see its name labeled on the silkscreen but if you're referring to the 3-pin header near the SMA connector that's labeled GND-GND-3V3, then yes it is 3.3V on both boards.

    3. Yes I could try this at some point.

    (4.) I also asked a colleague to try this same test, with 2 different boards.  Will let you know if they see the same behavior as you, or me.

  • 0 in reply to Arthur Thompson27
    TI__Genius 11810 points

    1) How close do you have them? What happens if you move them away from each other, a meter or so?

    2) Yes, that was the connector. With 3.3 V your supply should be ok. 

    3) 4) Please let us know your results.

    -Simon 

  • 0 in reply to Simon J
    Expert 1800 points

    Hi Simon,

    There is only about a 2 inch gap between both boards.  You can see a picture of it in my Powerpoint above. 

    My colleague ran the same test and saw the same results as me.  Turning on the 927.6 TX is actually lowering the noise floor reported by the 902.4 RX.  See picture.

    Thanks,

    Arthur

  • 0 in reply to Arthur Thompson27
    TI__Genius 11810 points

    Arthur, 

    What happens if you:

    1) Change the receiver frequency? 

    2) Run the same test on another location? 

    3) Try with another LaunchPad?

    -Simon 

  • 0 in reply to Simon J
    Expert 1800 points

    Hi Simon,

    1. I have also tried a test where I RX on 927.6 and TX on 902.4.  The observation is the same - the 927.6 RX noise floor goes down when the 902.4 TX turns on

    2.  I have seen this behavior at my home office and in the lab at work

    3. I have swapped the RX and TX roles between my two boards, the observation is still the same.

    Are you consistently unable to see the lowered noise floor in any number of different trials on your end?

    Thanks,

    Arthur

  • 0 in reply to Arthur Thompson27
    TI__Genius 11810 points

    Hi Arthur, 

    Unfortunately, we have not be able to reproduce your observations at all, which makes it very hard for us to provide efficient support. 

    1) Have you tried different PHYs? Are you observing similar behaviour then? What if you try the 868 MHz band?

    2) What happens if you reset the RX device after you have turned off TX on the other device?

    -Simon

  • 0 in reply to Simon J
    TI__Genius 11810 points

    Hi Arthur, 

    The combination of the high background noise (-80 dBm) and the strong interferer (20 dBm) is likely why we see this behaviour at your end (The noise floor at my end is in the -100 dBm range). 

    If you move the LPs away from each other this should improve. Also, by using a lower data rate where you can use a more narrow RX filter will also improve. 

    -Simon 

  • 0 in reply to Simon J
    Expert 1800 points

    Hi Simon,

    I'm very curious about what phenomenon I'm seeing here.  Is it a case where the strong interferer is reducing the AGC of the RX radio MORE than the amount of interferer signal that gets into the RSSI plot?

    I.e. am I seeing my high noise floor go lower as the RX AGC turns down?

    If the radio can sense this interferer, and reduces its RX gain accordingly, why isn't the interferer energy being displayed in the plot in Smart RF Studio?

    Thanks,

    Arthur

  • 0 in reply to Arthur Thompson27
    TI__Genius 11810 points

    Hi Arthur, 

    The strong interferer is saturating the front-end. When the interferer is 20 MHz away the AGC go down to the lowest gain setting but the energy is not visible in the plot since it's far outside of the channel (filtered out further down in the signal-chain). When the interferer is closer to wanted, you see that the rssi is increasing due to spectral widening (some energy gets into the channel). 

    -Simon 

  • 0 in reply to Simon J
    Expert 1800 points

    One more follow up question, is there any option for me to configure that filtering (further down in the signal-chain) so that it doesn't prevent me from being able to see this interference?  It doesn't seem useful to hide this interference from view if it is still going to be affecting our performance.

  • +1 in reply to Arthur Thompson27
    TI__Genius 11810 points

    Hi Arthur, 

    The filtering I referred to is the channel you have setup the receiver to listen at. You need to change channel if you want to "see the interferer". 

    BTW - This phenomena actually has a name: Desensitization

    -Simon