LM36011: Inquire about behavioural anomalies in the LM36011 and request a circuit review

Part Number: LM36011

Tool/software:

Hi, TI expert

There is an inquiry from a customer regarding an abnormality in operation during review of LM36011.

- Application: Oral scanner

- The LED at the output terminal is an OSRAM Compact Light Source LED (KW CSLNM1.TG) applied to DLP.
- R139 in the circuit is applied as 0 ohm.

[Abnormal operation details]
1) Flash Mode does not work: According to the customer's F/W manager, the light does not turn on when operating in Flash mode.
The following is the FW init code.
This is the Flash mode setting.

-------------------------------------------------------------------------------------------

int8_t lm36011_initialize(lm36011_Regs_t* p_lm36011_Regs)
{
  int8_t ret = 0;

  // get current settings information
  ret |= lm36011_get_Reg(LM36011_ENABLE, &p_lm36011_Regs->Enable.byte);
  ret |= lm36011_get_Reg(LM36011_CONFIGURATION, &p_lm36011_Regs->Configuration.byte);
  ret |= lm36011_get_Reg(LM36011_LED_FLASH_BRIGHTNESS, &p_lm36011_Regs->Led_Flash_Brighness.byte);
  ret |= lm36011_get_Reg(LM36011_LED_TORCH_BRIGHTNESS, &p_lm36011_Regs->Led_Torch_Brighness.byte);
  ret |= lm36011_get_Reg(LM36011_FLAG, &p_lm36011_Regs->Flag.byte);
  ret |= lm36011_get_Reg(LM36011_DEVICE_ID, &p_lm36011_Regs->Device_Id.byte);

  // lm36011 Enable Register Setting
  p_lm36011_Regs->Enable.bit.Mode = 3;               // Flash Mode (M0=1, M1=1)
  p_lm36011_Regs->Enable.bit.Strobe_Enable = 1;      // Enable: Strobe 
  p_lm36011_Regs->Enable.bit.Strobe_Type = 0;        // Strobe Type: Level Triggered
  p_lm36011_Regs->Enable.bit.IVFM_Enable = 0;        // Disable: Input Voltage Flash Monitor (IVFM) 

  // lm36011 Configuration Register Setting
  p_lm36011_Regs->Configuration.bit.Torch_Ramp = 0;  // Disable: Torch Ramp 1ms
  p_lm36011_Regs->Configuration.bit.Flash_Timeout_Duration = 15; // Flash Time-out Duration: 1600 msec
  p_lm36011_Regs->Configuration.bit.IVFM_Levels = 0; // IVFM Levels: 2.9V

  // lm36011 Led_Flash_Brighness & Led_Torch_Brighness Register Setting
  p_lm36011_Regs->Led_Flash_Brighness.bit.Thermal_Current_ScaleBack = 1;  // Enable: If enabled, the LED current shifts to torch current level if TJ reaches 125 °C
  p_lm36011_Regs->Led_Flash_Brighness.bit.LED_Flash_Brightness_Level = 0; // LED Flash Brightness Level - minimum : 11 mA (default)
  p_lm36011_Regs->Led_Torch_Brighness.bit.LED_Torch_Brightness_Level = 0; // LED Torch Brightness Level - minimum : 2.4 mA (default)

  // set initialize settings information
  ret |= lm36011_set_Reg(LM36011_ENABLE, p_lm36011_Regs->Enable.byte);
  ret |= lm36011_set_Reg(LM36011_CONFIGURATION, p_lm36011_Regs->Configuration.byte);
  ret |= lm36011_set_Reg(LM36011_LED_FLASH_BRIGHTNESS, p_lm36011_Regs->Led_Flash_Brighness.byte);
  ret |= lm36011_set_Reg(LM36011_LED_TORCH_BRIGHTNESS, p_lm36011_Regs->Led_Torch_Brighness.byte);

  return ret;
}

-------------------------------------------------------------------------------------------------------

Q1) Does it not work with an external Strobe signal when operating in Flash mode and Level trigger?
The desired operation is to output the LED current only when the Strobe signal is High in Flash mode.

Q2) The datasheet states that the strobe is 1ms or longer, but is it correct that the LED output does not come out when it is shorter than that?

2) IR LED Mode – Abnormal operation
When setting to initial IR LED and giving pulse to Strobe, LED current is synchronized to Strobe and outputs normally.

After a few seconds, as shown below, Rising starts with delay from Strobe signal and does not reach the desired current.

Regarding the above mentioned problem, I would like to request review to see if there is a design error.

Please check. Thank you.

  • Hi Grady, 

    Thank you for reaching out! I am glad to help. 

    I have gone though your questions. I will have a summary and reply to you tomorrow. 

    Thanks! 

    Best, 

    John. 

  • Hi Grady, 

    1)

    Q1): According to datasheet table 7.5.1, if you need to use strobe to control the output, for Enable Register (0x01), Strobe Enable bit should be set to 1, but Mode Bits could be set to 00 or 01. 

    • If set to 00, current rising/falling speed will be slow ramp. 
    • If set to 01, current rising/falling speed will be fast. 

    Q2): >1ms is just a recommended strobe pulse width to output a well-shaped current. <1ms strobe pulse width are also acceptable. 

    2)  

    This is weird.

    From the waveform, it seems that the strobe behavior is changed from IR mode to ext Flash mode.

    Also, it is possible that thermal scale-back may happen. 

    IVFM may also lead to this, but it is disabled now. 

    To verify this, could you please help provide read-back value of register from 0x01~0x05? 

    Thank you! 

    Best, 

    John. 

  • Read data from registers(0x01~0x05) after initial setting (normal operation)
    (0x01)LM36011_ENABLE : 0x25
    (0x02)LM36011_CONFIGURATION : 0x1E
    (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0xFF
    (0x04)LM36011_LED_TORCH_BRIGHTNESS : 0x00
    (0x05)LM36011_FLAG : 0x00


    Read data from registers (0x01~0x05) after Abnormal operation
    (0x01)LM36011_ENABLE : 0x24
    (0x02)LM36011_CONFIGURATION : 0x1E
    (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0xFF
    (0x04)LM36011_LED_TORCH_BRIGHTNESS : 0x00
    (0x05)LM36011_FLAG : 0x00

    The Mode Bits of the ENABLE(0x01) register are arbitrarily changing from 01b (IR mode) to 00b (Standby mode).
    What could be causing this problem?

    Please check. Thank you.

  • Hi Henry,

    Our expert is currently out of office for Chinese New Year Holiday. He will reply to you after Feb.5 when back to office. Thanks for your understanding.

    Best Regards,

    Steven

  • Hi Henry, 

    I will doubt if thermal shutdown happened. But it is weird to see 0x05 is all 0. 

    Does the read after abnormal operation the first time read or the second time read after abnormal operation happened? 

    Thanks! 

    Best, 

    John. 

  • Hi John

    This is the first data read after the abnormal operation.

    When the high temperature condition was intentionally induced by making the STROBE signal turn ON more frequently, the "Thermal Current Scale-back (TSB) Flag" was set and the LED current was confirmed to change to the Torch Current level.

    However, under the conditions I normally operate, the FLAG value was always kept at 0, and only the Mode Bits value was changed arbitrarily.

    Even the I2C lines (SDA, SCL) have always been silent after initialization.

    What could be causing this problem?


    Thank you.

    Henry.

  • Hi Henry, 

    Could you please try flash brightness conditions below in initial setup to see if the same thing happen or not? 

    • (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0x7F
    • (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0x0F
    • (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0x07

    Thank you! 

    Best, 

    John. 

  • Hi John

    I tried all initial setups, but the same abnormal operation was confirmed in all of them as follows.

    (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0x7F

    [2025-02-07 15:39:49.976] LM36011_ENABLE : 0x25
    [2025-02-07 15:39:49.976] LM36011_CONFIGURATION : 0x1E
    [2025-02-07 15:39:49.976] LM36011_LED_FLASH_BRIGHTNESS : 0x7F
    [2025-02-07 15:39:49.976] LM36011_LED_TORCH_BRIGHTNESS : 0x00
    [2025-02-07 15:39:49.976] LM36011_FLAG : 0x00
    [2025-02-07 15:39:50.985]
    [2025-02-07 15:39:50.985] LM36011_ENABLE : 0x24
    [2025-02-07 15:39:50.985] LM36011_CONFIGURATION : 0x1E
    [2025-02-07 15:39:50.985] LM36011_LED_FLASH_BRIGHTNESS : 0x7F
    [2025-02-07 15:39:50.985] LM36011_LED_TORCH_BRIGHTNESS : 0x00
    [2025-02-07 15:39:50.985] LM36011_FLAG : 0x00


    (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0x0F

    [2025-02-07 15:30:05.621] LM36011_ENABLE : 0x25
    [2025-02-07 15:30:05.621] LM36011_CONFIGURATION : 0x1E
    [2025-02-07 15:30:05.621] LM36011_LED_FLASH_BRIGHTNESS : 0x0F
    [2025-02-07 15:30:05.621] LM36011_LED_TORCH_BRIGHTNESS : 0x00
    [2025-02-07 15:30:05.621] LM36011_FLAG : 0x00
    [2025-02-07 15:30:06.630]
    [2025-02-07 15:30:06.630] LM36011_ENABLE : 0x24
    [2025-02-07 15:30:06.630] LM36011_CONFIGURATION : 0x1E
    [2025-02-07 15:30:06.630] LM36011_LED_FLASH_BRIGHTNESS : 0x0F
    [2025-02-07 15:30:06.630] LM36011_LED_TORCH_BRIGHTNESS : 0x00
    [2025-02-07 15:30:06.630] LM36011_FLAG : 0x00


    (0x03)LM36011_LED_FLASH_BRIGHTNESS : 0x07

    [2025-02-07 15:31:59.215] LM36011_ENABLE : 0x25
    [2025-02-07 15:31:59.215] LM36011_CONFIGURATION : 0x1E
    [2025-02-07 15:31:59.215] LM36011_LED_FLASH_BRIGHTNESS : 0x07
    [2025-02-07 15:31:59.215] LM36011_LED_TORCH_BRIGHTNESS : 0x00
    [2025-02-07 15:31:59.215] LM36011_FLAG : 0x00
    [2025-02-07 15:32:00.232]
    [2025-02-07 15:32:00.232] LM36011_ENABLE : 0x24
    [2025-02-07 15:32:00.232] LM36011_CONFIGURATION : 0x1E
    [2025-02-07 15:32:00.232] LM36011_LED_FLASH_BRIGHTNESS : 0x07
    [2025-02-07 15:32:00.232] LM36011_LED_TORCH_BRIGHTNESS : 0x00
    [2025-02-07 15:32:00.232] LM36011_FLAG : 0x20


    Is there anything else I should check?

    Thank you.

    Henry.

  • Hi Henry, 

    Thank you for this trying! This is very weird. 

    This is on customer's board instead of EVM, right? 

    I will go to office and try the set up on EVM to try to reproduce it. 

    Thanks! 

  • Hi Henry, 

    I tried the same setting on EVM for 30min with different flash current and no abnormal happens. Is there any chance customer's soft ware mistakenly reset the register? 

    Thanks! 

  • Hi John

    Looking at the test results, I can see that an LED Short Fault occurred as follows:

     

     

    The shunt resistor (R139) of my circuit is 0 ohm, unlike the TI EVM. Could this be the cause of the abnormal operation described so far?

     

     

     

     

    Thank you.

    Henry.

  • Hi Henry,

    Our expert is currently out of office and will reply to you later. Thanks for your understanding.

    Best Regards,

    Steven

  • Hi Henry, 

    Change the resistor in serial with LED to 0ohm is OK. In your schematic, 2,3 is the anode and 1,4 is the cathode, right? 

    If short fault happens, indeed 0x01 register's mode will reset. It seems only in flash current 0x07 case, short fault happens. Could you please confirm if it happens in other abnormal situations? 

    Thank you! 

    Best,

    John. 

  • Hi John

    That's right. In my schematic, 2,3 is the anode and 1,4 is the cathode.

    LED Short Fault Flag was only checked in the flash current 0x07 case.
    However, it is not always checked and is irregularly checked every time I test. It was not checked in other cases.
    (In the test, I read LM36011 Registers every 1 sec.)

    Below is the signal capture when the abnormal operation occurred.
    (R139 was replaced from 0 ohm to 0.1 ohm resistor.)
    As you can see, the I2C Line is quiet.

    Is there a clue to solve the problem?

    Thank you.

    Henry.

  • Hi Henry, 

    If I am understanding correct, 'However, it is not always checked and is irregularly checked every time I test. It was not checked in other cases.' means the abnormal behavior is not every-time reproduceable? Or Short flag 1 is not every-time reproduceable? 

    One of my guess is abnormal short fault happening reset the mode to standby. And read register clear that fault flag bit. But I still not sure by short happens. 

    Could you please share me the schematic and layout (in altium format or gerber file)? 

    Thank you! 

    Best, 

    John. 

  • Hi John

    The abnormal behavior is always reproduced. But, short flag 1 is not always reproduced.

    I will tell you about the schematic and layout file sharing after internal discussion.

    Thank you.

    Henry.

  • Hi Henry, 

    Thank you! I will also try reproduce this by involving short condition to see any finding. 

    Thank you! 

    Best, 

    John. 

  • Hi Henry

    It is difficult to provide the entire board's Gerber files.
    Would it be okay if I provide the relevant PCB files layer by layer instead?

    best regrad

    song

  • Hi Song, 

    You could send me the layer-by-layer file and I will try my best to check that. 

    Thanks! 

    Best, 

    John. 

  • Hi john

    Please provide a secure email address for file transfer.
    Sharing on the web is not an option.

    best,

    song

  • Hi Song, 

    Sure. Please send to john-shen@ti.com

    Thank you! 

    Best, 

    John. 

  • Hi john

    I have sent the document to the personal email you provided.

    best regards

    Song

  • Hi Song, 

    Well received. Could you please help remove C4, C5 and see if the problem still exist or not? My guess is the charging of big C may trigger the short fault which reset the mode bits. 

    Thank you! 

    Best,

    John. 

  • Hi John,

    C4 and C5 have been removed, and the test was conducted in advance.

    Even after removing C4 and C5, it still switched to standby mode in IR mode.

    I will try again today.

    Best,

    Song

  • Hi Song,

    Our expert is currently out of office and will reply to you next Monday. Thanks for your understanding.

    Best Regards,

    Steven