• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 4 answers
  • Subscribers 62 subscribers
  • Views 1111 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.

LM3414: LED current difference

Satoshi
Guru 19585 points
Part Number: LM3414

LM3414 tested below condition, LED current was differ from design(calculation) value.

If there any reason and improvement, please let me know.

Schematic and information are attached below;

■UV: NVSU233B ×5ch

・Design value: 500mA

・Measured value: 518mA

■Green LED: NCSG119B ×2ch

・Design value: 350mA

・Measured value: 342.5mA

■IRL: SMBB780D ×6ch

・Design value: 450mA

・Measured value: 470mA

Best regards,

Satoshi

  • 0
    TI__Mastermind 28655 points
    Hello Satoshi,

    How are you measuring current? The part has tolerance and these are close to the tolerance limits. If you include resistor tolerances, the current operating point (half or less than 1A) and measurement tolerances these are probably close. The accuracy of the LM3414 is at 1A regulation, it will become worse as the regulation point is lower, 100 mA is the minimum and it must stay in continuous mode.

    Best Regards,
  • 0 in reply to Irwin Nederbragt
    Guru 19585 points

    Irwin-san

    Thank you for reply,

    Please see measured condition below, I think that 450mA (vs 470mA) differ is a little larger.

    I want to clear reason of above differ.

    【Measured condition】

    ・Oscilloscope： TDS3034C (TDS3054B) / Tektronix

    ・Current probe： TCP202 / Tektronix

    ・Digital multi-meter ： 34401A / hp

    ・Test image is attached below, ILED line cut and measure digital multi-meter by way of universal PCB.

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Mastermind 28655 points

    Hello Satoshi,

    I'm not sure I understand, larger than what?

    If the set-point is 450 mA and the accuracy is at 1A and if the error is caused by internal offsets the accuracy at 450 mA would be:

    (1A/0.450) times the EC table percent error:  or approximately 93% to 107%.  (470 mA/ 450 mA = 1.044 or 4.4%.  Note there are other errors associated with current set-point such as tolerance of the current setting resistor, layout and other tolerances.

    From the EC table:

    "RIADJ = 3.125 kΩ      0.97      1      1.03 A      TA = 25°C
    ILED Average LED Current
    RIADJ = 3.125 kΩ      0.95       1       1.05 A"  (over temperature)

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Guru 19585 points
    Irwin-san

    Thank you for advice,
    Customer condition is Ta: 25℃ and 24V±10%, ILED accuracy is ±3% (and RIADJ, layout, test condition)
    But 470mA is more than 4% and larger current.

    Is there possibility of cause larger difference?

    Best regards,
    Satoshi
  • 0 in reply to Satoshi
    TI__Mastermind 28655 points
    Hello Satoshi,

    Of course there are other possibilities, that is why they need to look at all the signals and the inductor current etc to see where the error is coming from. Remember that the part accuracy is stated at 1A, it will be worse at lower currents. For example would you expect it to be 3% accurate if the current was set to 1 mA? The internal offsets would not allow it to be accurate. If they need to be more accurate they could set up a power on test to adjust one way or the other to make the set-point more accurate.

    Best Regards,
  • 0 in reply to Irwin Nederbragt
    Guru 19585 points
    Irwin-san

    Thank you for advice,

    Is there specification or graph for relation of ILED and accuracy?
    (For example, graph of ILED vs accuracy)

    Best regards,
    Satoshi
  • 0 in reply to Satoshi
    TI__Mastermind 28655 points
    Hello Satoshi,

    Unfortunately no. This part, like other LED drivers, has a current accuracy specification at full current. Some part will state input offset voltages which help in calculating error at different levels. I would look at the 1A information of 0.97A minimum and 1.03A maximum and scale from there. For example, at 500 mA I would assume the error is all offset and state the 0.47 mA minimum and 0.53 mA maximum. This doesn't include other errors such as the current setting resistor.

    Best Regards,