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.

LM3424: Need help for LED driver

Part Number: LM3424

Hi,

One of my customer is evaluating LM3424 for LED driver application.

Specifications

Input voltage  - 5.2V to 8.0V

Output  - 60V/250mA or 30V/500mA.

They want to develop common driver & current will be adjusted through dimming functionality depending on the LED strip, means when they connect 60V strip current will be adjusted to 250mA & when they connect 30V strip current will be adjusted to 500mA. This functionality will be achieved through microcontroller.

At present they are working on EVM to achieve the functionality but facing challenges with the design. Board is working as expected if input voltage is above 7.0V by adjusting few components like current sense, Resistor on CSH pin, added RC filter on MOSFET current sense etc....Board is working ok if voltage is above 7.0V to 12V but as soon as voltage goes below 7.0V, board is showing instability. Tried adjusting compensation, inductor values, slope resistor etc... but no luck.

Need inputs - what is the issue? & how can we resolve the same. Looks like max. duty cycle is creating the issue. Suggest the values of components if device looks suitable.

If this device is not suitable pls. help with better device suggestion.

Mitesh

  • Hi Mitesh,

    Thank you so much for reaching out. An engineer will respond to this next week.

    Best Regards,

    Caroline

  • Hello Mitesh,

    Did they go thru the calculations in the data sheet for their design?  Please forward the schematic with values that they used with the calculations.  Did they use the appropriate saturation rated inductor please give part number for that?  

    Thanks Tuan

  • Hello Tuan,

    We are following the calculations as per datasheet as well as EVM.

    I have attached the schematic with modified values using for design trials.

    let me know if you need any more details.

    like Mitesh mention above, for Input > 6.5V it is working fine. plz refer below waveform.

    for Input less than 6.5V, its become unstable or missing its regulation. as shown below.

  • Hello Saga,

    Please give me a few days to go thru your circuit to get you an answer.

    Thanks Tuan

  • Hello Sagar,

    I've asked for the part number along with the saturation current of the inductor that you use in the previous post.  Can you please provide that for us.

    Thanks Tuan

  • Hello Tuan,

    The Inductor Part no is "7447709220" and specifications are below.

    let me know if you need any more details.

  • Also like to mention few observations,

    while Increasing voltage from 5.3V to 6.8V it is unstable but once it turn on properly it is maintain the regulation till 6.V input if we slowly decrease the voltage.

    Inductor waveform at 6.4V when increasing the input voltage from 5.3V 

    Inductor waveform at 6.7V when increasing the input voltage from 5.3V, its stable at this point.

     

    Now Inductor waveform at 6.0V when decreasing the input voltage from 7V, As per this, it is try to maintain the shape till 6.0V but below this its again unstable or reducing the output current.

    all these trial i have taken with R6=300mR. Rest everything is same.

    one more thing, i have replace the IC also to check the variation but no change in behavior.

    also try to change the switching frequency but no change in behavior.

    kindly guide me.

    I insist you if you can check this at your side practically on EVM that will help us more to understand the issue.

  • Hello Sagar,

    We believe that you are getting too narrow of a pulse on the gate drive and it's approaching our ton-min time below.  Please increase Rt to 30k to reduce the switching frequency to see if it helps.  You will also need to change your Rslope accordingly to the Rt that you settle at along with the Vo.

    Thanks Tuan

  • Hello Tuan,

    In my above comment, there is typo error of R6 value, I used R6 = 30mR.

    Also as per above waveform, the CRO is setting is for high side duty cycle hence it is showing 16%.

    but actually it is off time, So duty cycle is 84% in our case. so there is no ton-min time issue here.

    I adjusted the slope values as per RT only. 

    but slope changing has not help me out here.

  • Hello Sagar,

    I did a simulation and based on the 0.03 Ohm and the values you have above the gate drive and loop was stable.  However, I noticed that with a 0,03 Ohm, VIS feeding back is really small.  The peak is about 90mV and turn on is about 75mV so that is about 15mV of delta sensing on the current sense.  This can be a noise issue here with such low signal to noise.  As a test can you please change RLIM to 0.06 Ohm to see if it solves the issue.  You will also need to adjust Rslope accordingly to this also.  Note that this is just a test here to figure out the real problem.  

    Thanks Tuan

  • Hello Tuan,

    Two Observations need to discussed.

    Observation 1:

    Modification on schematic: R6 change to 50mR. R9 = 250mR & Slope (R14)=10K.

    It is working fine for below volt range.

    Input voltage  - 5.2V to 8.0V

    Output  - 30V/500mA.

    But when I started Dimming through nDIM, then there is noise from Inductor.

    I try to change the Dimming Frequency (from 100Hz to 10KHz) but no luck, above 10KHz its being unstable so not tested.

    Kindly Guide.

    Observation 2:

    with Same Values when I change the Load to 50V/250mA & do the 50% Diming for adjusting the current value to 250mA then

    Current is not constant for full Input voltage range.

    Means, When I start increase the voltage from 5.3V, output current increases by 5mA for every 100mV increase in input voltage.

  • Hello Sagar,

    The noise from the inductor is due to the construction of the inductor and dimming in the audible range.  You will nee to find a molded inductor if you want to reduce this audible noise.  As far as you observation 2 you will have to trouble shoot more to figure out what it causing you the problem. Your design might not be able to cover all the range that you need for it to cover,   These are fundamental of designs that you will have to go thru yourself.

    Thanks Tuan

  • Hello Tuan, 

    Thanks for your valuable inputs.

    For Observation1: will it be any impact of audible noise on Emission part ?

    For Observation 2: I will look into design again.

    One more issue need to understand, when i do the dimming with Load of 50V/250mA, I have seen the noise in the output current, this will be persistent till the input voltage range of 6.5V, above this noise has not observed.

    why this is happening?

    Ch1- Switch

    Ch2- Input Voltage

    Ch3- Output Current

  • Sugar,

    As I mentioned you will have to go thru the design yourself to see if it satisfy all the conditions of over-current on VIS ect.  All the design equations are in the data sheet.  Again, you will need to trouble shoot things yourself here with your design.  

    Thanks Tuan

  • Hello Tuan,

    will it be possible for you to share the Simulation file of LM 3424, I have the TI Tina simulation software. it will be helpful for me to try the different combinations.

  • Hello Sagar,

    We only have PSpice on the following link under Design tools & simulation.

    LM3424 data sheet, product information and support | TI.com

    However,  you have the hardware and it's best to do your design verification first then use the hardware.

    Thanks Tuan