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.

TPS65235: TPS65235 Detection error problem

Part Number: TPS65235

Hello, we currently have a case of using TPS65235, and the customer has reported a problem: When 22K is turned on, the cable detection mechanism of TPS65235 will cause false positives.
The simple schematic diagram of our board is as follows. Simply put, when the customer finds that no cable is plugged in, turn on 22k, and the register CABLE_GOOD will switch back and forth between 0/1, which means that the cable detection mechanism of TPS65235 is misjudged , But there is no such problem when 22K is turned off. Please help analyze the problem, is there a way to fix it from the software level, or it needs hardware rectification.

Steps to reproduce:

  1. Use Stingray 4.15.59 or later (excluding "special" versions with disabled LNB polling)
  2. Make sure Stingray has started in Factory mode
  3. Setup Stingray logs. Open UART console and type the following commands:
  1. log mute off
  2. log Error
  3. log named DVBSTuner Info
  • Provide LNB power is enabled and 22kHz tone is on. Go to Antenna editor. Select testAntenna0 and make the following settings:
  1. LNB Setup: Custom
  2. DiSEqC: None
  3. Tone 22 kHz: Enabled
  4. Voltage mode: Normal
  • Save settings. Go to testAntenna2 and do the same settings.
  • Go to "STB Settings" / "Antenna" and disable all checkboxes here
  • Make sure no any antenna cable is connected.
  • Go to "Manual search". Select "ABS 1" (will activate RF IN1). Enter any valid frequency. Don't connect any cable. Observe UART logs for ~1 min
  • Select "Apstar 6" (will activate RF IN2). Observe UART logs

          Actual Result:

          For RF IN1. False-positive Cable detection is triggered with random interval:

                   [18/01/2021 13:19:12.505] [Info] {tunerExecutor:serverTuner} [DVBSTuner] serverTuner connected: true

       [18/01/2021 13:19:12.510] [Info] {tunerExecutor:serverTuner} [DVBSTuner] serverTuner connected: false
       [18/01/2021 13:19:23.828] [Info] {tunerExecutor:serverTuner} [DVBSTuner] serverTuner connected: true
       [18/01/2021 13:19:23.833] [Info] {tunerExecutor:serverTuner} [DVBSTuner] serverTuner connected: false
       [18/01/2021 13:19:27.602] [Info] {tunerExecutor:serverTuner} [DVBSTuner] serverTuner connected: true
       [18/01/2021 13:19:27.607] [Info] {tunerExecutor:serverTuner} [DVBSTuner] serverTuner connected: false
  For RF IN2. False-positive Cable detection is triggered with random interval.             

         

       [18/01/2021 13:22:24.073] [Info] {tunerExecutor:clientTuner} [DVBSTuner] clientTuner connected: true
       [18/01/2021 13:22:24.076] [Info] {tunerExecutor:clientTuner} [DVBSTuner] clientTuner connected: false
       [18/01/2021 13:22:27.850] [Info] {tunerExecutor:clientTuner} [DVBSTuner] clientTuner connected: true
       [18/01/2021 13:22:27.855] [Info] {tunerExecutor:clientTuner} [DVBSTuner] clientTuner connected: false
 
  BUT after several minutes there is no Cable detecton and behaviour looks normal.

         This is specific only for RF IN2

        Expected Result:
         No any cable detection is triggered.

         Test case 2

         Go to "Antenna editor" and disable 22 kHz tone for testAntenna0 and testAntenna2.
          Repeat the steps above.

         Actual Result:
          No any cable detection is triggered.

  • Hi, Switch 

    1. Could you send me the schematic for review? 

    2. CABLE_GOOD bit status is determined by load current on cable. 

    If CABLE_GOOD value switches back and forth, I think the load current might be at the boundary of threshold. 

    Could you measure the load current when it switches back and forth? 

      

  • Hi Ma,

    Below is the schematic

  • Hi, switch 

    The schematic looks okay. 

    Confirm with you: 

    When there is no cable plugged in and turn on 22k tone signal, what is the loading of LNB0_PWR? could you measure it?  

  • Hi  Ma,

    When the capacitor C6022 (0.1uF) is removed, there will be no misjudgment. Please help to analyze it, thank you!

  • Hi, Switch 

    As I said before, CABLE_GOOD bit is related with output current.  

    I think when turn on the tone signal, the LNB output will source and sink Cout, when Cout is bigger, the output current would reach the CABLE_GOOD threshold, then CABLE_GOOD value will switch back and forth. 

    So I think it is normal. 

  • Hi ,

    Thanks,  Is there a way for the software to raise the threshold current to get around this problem?

  • Hi, switch 

    No, we cannot raise the threshold by software. 

    Questions: 

    1. If the CABLE_GOOD switched forth and back, what problem will it cause? 

    2. If we add a dummy loading > threshold, then CABLE_GOOD will keep 1, could you accept this for your application? 

  • Hi :

        I'm Ezra from GIEC, I hope to communicate with you directly about this issue, so as to be more efficient.

        1, If the CABLE_GOOD switched forth and back, what problem will it cause ?

           ----We will not be able to use insert detection properly,If the user opens 22K without any cables inserted, it will be repeatedly prompted to plug in, pull out, plug in,pull out...That's not right.

        2. If we add a dummy loading > threshold, then CABLE_GOOD will keep 1, could you accept this for your application? 

           ----If we do this, insert detection will not work,right ? we hope to keep this function.

       At present, if the threshold current is only 5mA, and the software cannot be configured, is it a chip design flaw ? Do any of your other customers report similar problems ? 

       Looking forward to your reply. Thanks .

  • Hi, 

    No, I never encounter this issue before.  

    Does every board have this issue? 

    I check the TI reference again, it is not recommend to add FB6001 and C6022. 

    So work around is remove C6022 and short FB6001.