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.

  • TI Thinks Resolved

WEBENCH® Tools/LM3410XSDSEPEV: LM3410XSD

Intellectual 450 points

Replies: 14

Views: 223

Part Number: LM3410XSDSEPEV

Tool/software: WEBENCH® Design Tools

Hello

Pleas help me, I am bought your EVB LM3410XSD in order to implement that solution in my circle.

I made several adjustment on EVB,  i assembled two my LEDs - Omron, GW CSHPM1.PM-LSLU-XX51-1

Disconnect R2 (in final design i will use internal pull down in MCU).

My problem, once i connect DIMM (yellow wire) pin to GND (Black wire)  the LED is still light (weak, but still on) What's the problem?

while i connect DIMM (Yelow wire) pin to VCC (red wire) , LED light with wanted power, this is OK.

Please advise, what I mike wrong?

Thank you

Ilya

  • In reply to Ilya Feldman:

    Hi Ilya,

    I have moved your query to boost LED forum. It was mistakenly posted in Webench. Kindly note that as this a holiday season replies to your query might get delayed.

    Best Regards,

    Bhushan 

     

  • In reply to Ilya Feldman:

    Hello Ilya,

    Can you look at pin 6 with an oscilloscope?  Can you also measure pin 4 with an oscilloscope?  How long is the ground wire pulling dim low?  If this was a boost I can understand why the LEDs may do this but a SEPIC should not.

    Best Regards,

  • In reply to Irwin Nederbragt:

    Hello

    I know what the issue is, but I can explain why he behaves like this.

    When Vin = 5V, the LEDs light and I_input = 48mA.

    When Vin = 3.8V, the LEDs turned off and I_input = 3.5uA.

    Please find SCOPE plots in attached doc . (for that test i used external power supply)

    if Vin=5V, pin6 (SW) = ~5.4V

    if Vin=3.8V, pin6 (SW) = ~3.7V

    Vfb, pin 4 at Vin 5V and 3.8V = ~0V.

    During all that measurements pin DIM all time connected to GND.

    --> At first stage of development, work with EVB- LM3410XSEPIC and connect it to NUCLEO (MCU) and supply only Vin-5V.

    at final configuration, Vin will connect 3.8V as on my schema below.

    This is my circle:

    Questions:

    1. Why exist this situation - with Vin-3.8V led no light and Vin-5V led is light?

    According to data sheet, LM3410, Input voltage can be 2.7-5.5V.

    2. why under vin-3.8V and pin DIM, Iq=I_input = 3.5uA and not 80nA? what i need do in order to get 80nA?

    Thank you

    IlyaPIN.7z

  • In reply to Ilya Feldman:

    Hello Ilya,

    Your initial post references the SEPIC design, the schematic in your latest post is a boost.  A boost converter cannot have the output be zero volts, it will be a diode drop below the input voltage at minimum.  This will cause the LEDs to glow if the LED string voltage is near Vin.  You have two LEDs, The LED string voltage is too low for 5V input.  The LM3410 is not running (as seen in the oscilloscope pictures).

    If you removed the LM3410 from the board it would behave the same way.  Vin goes through the inductor (a DC short) and through the diode to the LED string.

    Best Regards,

  • In reply to Irwin Nederbragt:

    Hello

    Unfortunately i not clear understand your answer, sorry. I would like learn from professional, please help me. I would like deep understand what my problems. I am added notes in your answer:

    Your initial post references the SEPIC design, the schematic in your latest post is a boost.  A boost converter cannot have the output be zero volts, it will be a diode drop below the input voltage at minimum.

    Ilya - you want say that AN-1775 LM3410X SEPIC 6-Pin LLP Demo Board, designed to work not as boost?
    what is mean "converter cannot have the output be zero volts", you want say, that my design designed as boost and Vout should be high then Vin?
    Unfortunately i not found what a V drop of LM3410? If my Vout min 5.7V (two LEDS 2.85V) what Vin Max can be for correct work?

    This will cause the LEDs to glow if the LED string voltage is near Vin.  You have two LEDs, The LED string voltage is too low for 5V input.  The LM3410 is not running (as seen in the oscilloscope pictures).

    Ilya - what is mean "The LM3410 is not running (as seen in the oscilloscope pictures)." ?
    In order for the component to work properly? what my input requirements, i have two LEDs with Vled=2.85V.
    Why under vin-3.8V and pin DIM=0, Iq=I_input = 3.5uA and not 80nA? what i need do in order to get 80nA?

    If you removed the LM3410 from the board it would behave the same way.  Vin goes through the inductor (a DC short) and through the diode to the LED string.

    Ilya - you write that note because if Vin =5V and Vled drop 5.7V (2.58v+2.85v), this is mean boost not will work and Vin goes through the inductor (a DC short)?

    Best regards

    Ilya

  • In reply to Ilya Feldman:

    Hello Ilya,

    "Ilya - you want say that AN-1775 LM3410X SEPIC 6-Pin LLP Demo Board, designed to work not as boost?"  It is not design to work as a boost.  If the switching node is not switching the LEDs cannot light with a SEPIC, your oscilloscope pictures show the swtich node not switching, if the LEDs are lit it cannot be a SEPIC converter since the output is blocked by C3 assuming Vin is DC voltage.

    "Ilya - what is mean "The LM3410 is not running (as seen in the oscilloscope pictures)." ?"  Do you see the switch node rising and falling?  It is DC so the LM3410 is not running.  If you pull DIM high you will see it run.  (If you are looking at the switch node)

    "Ilya - you write that note because if Vin =5V and Vled drop 5.7V (2.58v+2.85v), this is mean boost not will work and Vin goes through the inductor (a DC short"  I would look at Vsw and Vin, If the LM3410 is not running the Vsw cannot be higher than Vin.  If Vsw is not switching the input voltage will go through the inductor and boost diode to the LED load.

    Vled = 2.85V at it's rated current Vled at lower current will be lower.  This is why you may see the LEDs glowing at 3.8V (3.5 uA), at 5V they are at 48 mA.  If you use three LEDs it will probably be quite a bit less.

    I don't know what you are using for your test, is it a SEPIC or Boost? the schematic shows a boost, the board appears to be a SEPIC.  It appears you are using a boost.  The LEDs are lit and there switch node is not switching.  If you look at a boost the input goes through the inductor to the boost diode to the LEDs.  The LM3410 can be removed and voltage will still be applied to the LEDs.  In a boost the output voltage needs to be higher than the input voltage to work properly.

    Also look at the LED datasheet, it will have a graph showing Vf versus current.  This can also be done with a lab supply (limiting the current to prevent LED damage).

    Best Regards,

  • In reply to Irwin Nederbragt:

    Hi Irwin

    First of all thank you for your time!!!

    Small clarification, on the AN-1775 LM3410X SEPIC i made adjustment and made it as my Schema - work like BOOST.

    I created Boost with LM3410 because i looking for a Led driver that can give me constant current 400Ma-700mA (this what i need). I took Boost because i calculate that my Vin 3.8V and Vled drop 5.7V (2.58v+2.85v) or for three Vled drop 7.74V (2.58v+2.85v+2.58v). this a reason of Boost. Maybe my design is incorrect, please explain me.
    Maybe i not understand how can i deliver to LED constant current and i need find other solution.
    Additional questions, in order to understand LED driver works. Is output voltage of the LED driver (LM3410) doesn't matter for me?
    and important to me a constant current that can supply LED driver for LEDs?
    In order to increase output power and supply high output power do i need supply duty cycle to DIM pin? Because i understand - if i will give to DIM pin constant "0" and LM3410 will supply constant current according to R96 (Rfb). it's looks i not understand how work LM3410, sorry, please advise

    Best regards

    Ilya

  • In reply to Ilya Feldman:

    Hello Ilya,

    Both the SEPIC design and the boost design deliver constant current.  The output voltage is determined by the load.  The boost is for LED string voltages higher than the input voltage.  The SEPIC can have the output voltage higher or lower than the input voltage.  If you are using two LEDs at 2.58V forward voltage and want zero current when disabled you will probably need the SEPIC.  If you look at the LED datasheet you can see the LED voltage versus current, the voltage across the LED will be lower at lower currents (this is why you see the LED glowing and input current higher than standby current).

    Try using the boost with three or four LEDs.  It will work better with the boost circuit though it may still glow at your higher input voltage.  This is will show you what is happening with two LEDs.  If you need to have two LEDs you probably will have to use the SEPIC design.

    If Dim is low it disables the output (however a boost can still feed current to the load from Vin through the inductor and diode if the load voltage is too low).  If Dim is high it will provide maximum current determined by R96 value.  If PWM is applied it will reduce the average current from the maximum.  It will provide maximum current only when Dim is high.

    Best Regards,

  • In reply to Irwin Nederbragt:

    Hi Irwin

    What i can understand  - if i need low power constipation 80nA, i need use SEPIC? but in fact on Boost and SEPIC current will be feed current to the load from Vin through the inductor and diode.

    If my Vin=3.8V and i am used two LEDs it will give me Vled drop 5.7V (2.58v+2.85v), is this not enough high voltage for Boost performant?

    What Delta between Vin and Vout=Vled should be in order to get very low consummation. I don't now what to do in order to get  80nA with Boost ?

    Thank you

    Ilya

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.