• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 63 subscribers
  • Views 1682 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS40211 Boost LED driver constantly restarting

Rifat Sahiner
Prodigy 20 points

Hello,

I have designed a boost led driver which drives 12 power LEDs in series with 1.2A constant current. LED string and power supply voltages varies 39 to 44V and 23 - 26V respectively. I made the calculations for %20 p-p ripple with 700kHz switching frequency from the information given in product datasheet. I attached the schematic to post.

When i powered the system LEDs started flashing. I first thought soft start cap should be bigger value than i put a 1uF but this only changed flashing speed. This meant that driver was sensing an overcurrent state so i lowered Isense resistor but didn't work. Than i tried to decrease LED current and flashing stopped at 350mA. Since than i tried a few things too but couldn't stop the flashing above 350mA current.

I hope i could have explained the situation well. Waiting for possible reasons and solutions for the problem. Thanks in advance.

 

 

  • 0
    TI__Guru 56925 points

    I suspect that your issue is the over-voltage protection circuit pulling up on the current limit pin.  D1 is a zener diode that pulls up on the ISNS pin if the output voltage exceeds the expected maximum voltage, but it is feeding a 1.15kOhm resistor with a 100mV threshold, so the converter will trigger OCP and restart if the zener diode conducts 100uA of current.  Your 47V 1W zener diode might be conducting 100uA of current a little below 47V and tripping this OVP circuit a little early.

    You could try 2 different ideas on this:

    Decrease the ISNS impedance by reducing R7 and increasing C8 by the same factor (try 2x) which will require 200uA of zener current.

    Try a 51V or lower current 47V zener diode.  This zener diode is never going to see a lot of power dissipation.  As soon as it conducts 100uA of current, the converter will shut-down and restart.  At most, it's going to see 100uA continuous current if it's on the very edge of triggering OCP.  That's a worst case power dissipation of 5mW.  It doesn't need to be a 1W zener.  A 100mW zener diode will more than suffice, be smaller, less expensive and have lower zener current below its zener voltage.

    If you don't have a smaller zener diode or higher voltage diode, try adding 40kOhm series resistance.  This will increase the zener voltage 4V @ 100uA forward current, the equivelant of going from a 47V to 51V zener in this application.  In the end, I would recommend a 100mW 51V zener for that location in this application.

    If that doesn't work, here are some additional things you can try:

    1) As a check, try reducing the bandwidth of the control loop.  The control loop needs to adjust for changes in LED forward voltage, which is typically a slow process.  Try changing C3 to 220pF, C4 to 10nF and R4 to 3.16kOhms.  This will reduce the loop-gain by 20dB and provide a more stable control loop.  If the issue is stability resulting in higher peak inductor currents, this should address it.

    2) Right now you are using a 54ns time constant filter on the current sense.  This is following the datasheet recommendation, but it offers very little filtering of the gate-drive current pulse.  The first thing I would try would be to increase this.  Increase C8 from 47pF to 150pf.  This should filter much more of the initial gate drive pulse from the ISNS signal and provide a more accurate current limit function.

    3) R6 provides a series resistance to the gate drive, reducing the gate drive current and slowing the turn-on.  Increasing this to 10Ohms will increase switching losses slightly, but reduce the initial "spike" on the current sense by reducing the gate drive current.