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.

TPS929160-Q1: FAIL-SAFE mode issue

Part Number: TPS929160-Q1
Other Parts Discussed in Thread: TPS929240-Q1

Hello,

I'm working with the TPS929160-q1 (with the EV Board), and I need it, after setting it up, to be in FAIL-SAFE forever (on the final application there will be no controler).

I'm changing in the EEPROM values of PWMs, and when I set FORCEFS to 1, values charge from EEPROM correctly (but as it says in Datasheet, FORCEFS directly goes back to 0). So, the solution is to set the Watchdog to be always in FS (from Ch to Fh). This is what I'm doing, entering EEPROM mode, putting C1h or up to F1h to the register 84h, and the component keep being in FS as wanted.

Unfortunatly, what always happen is that LEDs take the correct value from EEPROM for half a second, and then all shut off forever. I looked for all values changing with FS mode, but can't figuring out why all LEDs set in FS correctly and then nothing. By putting the correct value in 84h every thing comes back as it should be.

(All FSx registers are to 00h, with Jumpers correspondents).

Would you know what I'm doing wrong ? Why all PWMs output are suddenly all shutting off ?

Thanks you very much for your help, this would help me a lot !

  • Hi, Paul ,

    May I know which version of GUI you are using now? Is LED D40 on or off? Can you read status to see if any EER happens? If run in OFAF mode in fail safe mode, then all OUTPUTS will be off.    

  • Hi,

    - I'm not using the GUI, I'm coding my own software.

    - Actually yes, the LED D40 in on when every other shut off.

    - Nevertheless, th OFAF is currently to 0 (register 86h : 10h).

    - For errors :

              - A0h : 0x05 (Has encounter thermal issue, I don't think it's the issue here).

              - A1h : 0x39, everything looks fine.

              - A2h : 0x7B,

              - A3h -> AEh : 0x00.

              - AFh : 0xB4, I don't think it's related.

    Do you an error that could be related ?

    Edit : If it can help, when I CLEARFS, it lights on all and the same, directly lights off them (WATCHDOG is on F1h), so really an issue with the watchdog

    Thank you very much for your time !

  • A0h 0x05 shows CRC error. It will turn off all outputs in failsafe mode. 

    Please reprogram EEPROM with correct CRC. You may use EEPROM program tool to generate files which can be download from product folder. 

  • After reading again the Datasheet, I'm actually never setting anything to EEPCRC. (Understood that the problem was EEPCRC != CALC_EEPCRC). I don't really understand what I'm supposed to write there at the end of my programmation, "When the EEPROM design for production is finalized, the corresponding CRC code based on the calculation must be burnt to EEPROM register EEPCRC together with other EEPROM registers in the end of production line." (from Datasheet)

    I have to calculate and then put in EEPCRC the CRC of what calculation ? It's the point I don't get

    I compared every single CRC I send in EEPROM with thoses calculated with the Configuration Tool and they are the same.

    Once again, thank you very much for your help !

  • Paul, 

    Can you help read out all registers data for me? You may also use below tool to calculate CRC personally. 

    https://www.ti.com/tool/download/SLVRBM1

    I guess the EEPROM had been programed. Anyway, you can easily reprogram it. You may follow Figure 7-20. Programming Sequence.

    When writing data to registers, 87h CRC is also required. 

    You may also refer to sample code in product folder, which includes EEPROM programing. 

    along with user guide below 

     www.ti.com/lit/ug/snvu854/snvu854.pdf

    Regards, 

    XG 

  • Hi XiaoGuang1,

    It did solved my problem for a time. Unfortunately, I'm now using the TPS929240A-Q1 (not TPS929240-Q1), and everything works just fine (after changing the address with 8), except for the FailSafe mode.

    I'm entering FailSafe mode as usual, changing the CRC as it previosly resolved the issue I had, but now it keeps making a flag for CRC (even if 87h and AFh are the same). Is there something else changing with this "A" version I should know ? The way I enter EEPROM mode on the previous component is the same as this one ?

    By the way, sorry for not thanking you for previous answers, it helped me a lot !

    Regards,

    PV

  • Hi, PV, 

    There is just DEVADDR[3] difference between A and non-A version, and accordingly CRC value. The way entering EEPROM mode is the same. 

    What do you mean "entering FailSafe mode as usual, changing the CRC as it previosly resolved the issue " Can you help read out all registers data for me? If there is logic analyzer data for me, it is also appreciated. 

    If the fault flag can be cleared with CLRFAULT? 

    Regards, 

    XG

     

  • Hi XG,

    Here is a screen of all flag from A0h to AEh :

    I don't see any problem, and with the exact same configuration FailSafe mode works with the Demo Board (with the TPS929240-Q1, with DEVADDR[0] High and so frames sent to 0x01), but doesn't with our board. Is there something to connect in Hardware because of the High DEVADDR[3] ?

    Regards,

    PV

  • PV, 

    It seems no error happens as A0h is 0x01 and A1h is 0x2Dh. 

    Did you set FORCEERR to 1? You may clear FLAG_ERR by writing 1 to CLRFAULT. 

    What do you mean "now it keeps making a flag for CRC (even if 87h and AFh are the same)"? FLAG_EEPCRC  is 0 as you showed above. 

    Please help provide more info if I misunderstood. 

  • XG,

    I will summarize :

    - I tried what you suggested, but nothing happened.

    - Sorry for the confusion, I wasn't watching the correct thing. The screen of flags I gave you is the correct one.

    - I tried changing the TPS929240A-q1 with a TPS929240-q1, but still the exact same issue.

    - Everything works just fine with the same flags with the demo-board which contains a TPS929160-q1. Is there something I should change (at the addressing or hardware level) by using the TPS929240-q1, by taking into consideration that everything works on 16 channels of the demo-board ? My schematic is the same than the one on the demo-board.

    Thank you,

    Regards, PV

  • PV, 

    Can you attach your schematic? Can you also test it on EVM board? 

    I don't know about your question yet. Can you address your issue and expectation in detail? 

    shows that the device had POR event and run into FS mode now already. 

    You may clear fault flag with CLRPOR. If fault can't be cleared, then you need to check power status. I hope this answers your question.

    Regards, 

    XG 

  • µSherpa_TPS-1.pdf

    Schematics is attached

    I will explain what I'm doing, expecting and what happens.

    I'm doing a software to harmonize LEDs, using the TPS929240-q1. The goal is being able to put the perfect configuration we want with PWM configuration. Then, to save it, I put this values in the EEPROM, and then make the component keep being in FAIL-SAFE mode by fixing the Watchdog to 0xF1, and putting the correct CRC in 0x87.

    This way, I'm putting the configuration I need/want, and then by saving it when I turn OFF/ON the alimentation, the last configuration I did put in EEPROM is always the one showing (my system needs to work without the connexion with the software).

    All of that perfectly works on the Evaluation Board. On my PCB (with the schematic attached), when I'm saving the configuration, values goes correctly in the EEPROM, and after setting the WATCHDOG, no LED lights ON when I turn OFF/ON the supply, while it does on the EVM board.

    Flags that i gave you are the one showing when LEDs are supposed to be lighting ON.

    The EVM board has a 16 channels TPS, and my PCB has a 24 channel one. The schematic I have (that is attached) is the same as recommended in the 24 channel TPS datasheet. FS0 is given supply.

    I hope I'm now being clear with the description of my system and my issue.

    Thank you for your time,

    Regards,

    PV

  • PV, 

    1. What is ERR pin status? Is it always low or high?

     2. If there is fault, if it is able to be cleared? Like CLRPOR to clear POR error. If fault can't be cleared, please kindly check if there is fault happen. 

  • Hello,

    The problem was that channels that werent used had there enable to 1, and not to 0 (open circuit fault). ERR pin wasn't connected, thank you.

    It now works, but after being put in FAIL-SAFE, configuration is correctly set but when the tension is below 15V, all channels turn off. Would you have an idea ?

    Thank you, PV

  • Hi, PV, 

    If there is open circuit fault, FLAG_OUT and FLAG_OPEN should be set. I don't know why the data you sent didn't show them. 

    Again for the new, please kindly read out register A0h ~AFh, there must be err detected. Something like FLAG_SUPUV or FLAG_OUT error...