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.

TPS61181AEVM-259: TPS61181A

Part Number: TPS61181AEVM-259
Other Parts Discussed in Thread: TPS61181EVM-259, TPS61181A

Hi,

We are doing tests on a TPS61181EVM-259 board from which we have removed the jumpers, so that the IFB pins are not connected to Gnd.

The supply voltage is 13.30V and we used series of 6 white leds (18.90V when on).

In first test we used one series of leds on IFB1, and one series of leds on IFB2 with IFB3 to IFB6 unconnected.

We tested concretely what is referred to in the manual in

After set Enable pin, Vo reaches the maximum voltage and the leds turn on, the voltage starts to decrease until it gets close to 0V, and the leds turn off definitively.

The leds stay on only for a few seconds.


We tested several configurations with jumpers, but it is enough that at least one IFB pin is disconnected from Gnd for what was described to happen.

As we cannot use jumpers on our board, and we have a problem that we don't know how to solve in order to use a variable number of IFB pins with leds.

Any comments would be appreciated.

regards

Jose

  • Hi,

    I am out of office and will reply to your questions tomorrow

    BR

    Patrick

  • Hi Jose,

    Because TPS61181A also has an IFB overvoltage threshold as typical 17V. So when boost is ramping up the output voltage to OVP threshold (typical 39V), the IFB voltage may over 17V and as a result, device shuts down all IFB pins.

    If you don't have jumpers on your board, a 0ohm resistor connect IFB to GND is also okay.

    BR

    Patrick

  • Hi Patrick,

    In our system the number of led strips is variable, between 2 and 6, so it is not feasible to be adding / removing resistors on each board at the time the system is assembled.

    In the datasheet it says 

    8.3.5 IFB Pin Unused
    If the application requires less than 6 WLED strings, one can easily disable unused IFB pins. The TPS61181A
    simply requires leaving the unused IFB pin open or shorting it to ground. If the IFB pin is open, the boost output
    voltage ramps up to VO overvoltage threshold during start up. The device then detects the zero current string,
    and removes it from the feedback loop.

    What we checked on the demo board is different from what the manual says, that is, it is enough to have at least one IFB output unconnected for all IFB lines to be turned off.

    Besides wanting to understand why this happens, we are looking for a solution other than putting wires in the unused IFB lines connecting to Gnd, as this becomes expensive and complex in the final assembly of the system.

    We have tried using a Mosfet, controlled by a GPIO of the microcontroller, but that Mosfet when active, and connecting IFB to Gnd also creates a direct connection from Vo to Gnd. If someone makes a mistake and connects the series of leds on that IFB line before disconnecting it from Gnd via the microcontroller, the leds will be connected from Vo to Gnd without any current limit.

    So I ask what options we have to overcome the problem.

    regards

    Jose

  • Hi Jose,

    Let me check this and feedback later.

    BR

    Patrick

  • Hello Patrick,

    Do you have news regarding the pending issue ?

    regards

  • Hi Jose,

    Sorry for the late. I have checked this on EVM. That is strange because I got the result as datasheet mentioned. If I didn't connect IFB2~6 and only connect IFB1, then only LED1 will be turned on. And the LEDs won't be turned off after seconds.

    So could you please send me a picture of how you connect the LED strings with the EVM? I think there might be some setup issues.

    BR

    Patrick

  • Hi Patrick,

    Do you have an e-mail address or other way I can send you the requested information ?

    regards

    Jose

  • Hi Jose,

    I have sent you my email through friend request.

    BR

    Patrick

  • Hi Patrick,

    Please check your e-mail.

    regards

    Jose

  • Hi Jose,

    Since we have contacted by email, I will close this thread now.

    BR

    Patrick