An issue with LT3762-based boost converter. The .asc file included

Dear colleagues,

Is it possible to use for LT3762 VIN a voltage supply that is lower than the main input voltage of the boost converter?

I have the successful LTSpice simulation but the live circuit stood working for a few seconds only, from an external PWM of 40%.

After that is happened, PWMTG signal do not starting permanently even after the power OFF then ON. 

Nothing was heated noticeably. BG and TG signals are continuing work normally.

May different power sources for VIN and rest of the circuit be the cause for the issue?

LT3762_9ABoost.zip

Parents
  • +1
    •  Analog Employees 
    on Feb 4, 2019 4:35 PM

    Hello MPY,

    There shouldn't be a real issue using a 12V supply to power the Vin pin of LT3762 while using a separate 15V supply to power the boost stage, although you are likely to get more current capability out of the INTVCC regulator from powering VIN from 15V as well. Given the number of MOSFET devices used in this design, there is always the chance that you are overloading the internal regulator, leading to other failures in a high power system. I don't know if the NMOS devices listed in your simulation are the ones being used in the actual circuit, so I can't really comment on if you are overloading the INTVCC regulator.

    From your description, it sounds like something could have damaged the PWMTG driver or PWM PMOS device. I would double check the part used for the PMOS to make sure it has an adequate current handling capability with an appropriate gate drive voltage (the one listed in the LTSpice sim file is NOT one that should be used for this design). Another thing that could kill the PMOS / TG driver would be if you are connecting the output of your converter through long lengths of wire, adding inductance to the output which can cause voltage spiking during PWM dimming. If there is enough ESL caused by long connecting wires the LED current will ring during PMW dimming, which can pop the PMOS or TG driver from an overvoltage if severe enough.

    Kind of hard to speculate as to what went wrong given a simulation which appears to be functioning somewhat normally. 

  • Hello Klawrence,

    have you read my message from Feb 5?

    Please look, I made some changes in the circuit but it still not working. 

    Now BG driver keeps NMOS opened permanently.

    I did everything ( I hope ) according to the datasheet and the PCB reference design.

    Please can you tell where my mistake could be?

    Thank you for your help

  • +1
    •  Analog Employees 
    on Feb 21, 2019 4:37 PM in reply to MPY

    Hello MPY,

    I did not realize that your PWM dimming frequency was as high as you stated in your previous post. 20kHz dimming of a 450W boost system is not going to be easy to do and will really stress the IC. You already have quite a lot of switching losses from the doubled up NMOS devices in the power stage, but now you also have a high frequency switching PMOS. On top of that, the PMOS that you are using for the dimming is out of spec for this parts PWMTG driver. You need a -6Vgs PMOS to guarantee that the FET is fully enhanced when driven.

    I'm afraid that what you are trying to accomplish isn't going to be possible with a synchronous boost controller. You may want to try to use a sync buck controller like LT3744 if you need that combination of high current / high dimming frequency. You would need to break your LED string up into smaller segments and run from a higher input voltage as a buck, but the dimming and high current delivery that you are trying to accomplish is much more feasible with a buck than a boost. 

Reply
  • +1
    •  Analog Employees 
    on Feb 21, 2019 4:37 PM in reply to MPY

    Hello MPY,

    I did not realize that your PWM dimming frequency was as high as you stated in your previous post. 20kHz dimming of a 450W boost system is not going to be easy to do and will really stress the IC. You already have quite a lot of switching losses from the doubled up NMOS devices in the power stage, but now you also have a high frequency switching PMOS. On top of that, the PMOS that you are using for the dimming is out of spec for this parts PWMTG driver. You need a -6Vgs PMOS to guarantee that the FET is fully enhanced when driven.

    I'm afraid that what you are trying to accomplish isn't going to be possible with a synchronous boost controller. You may want to try to use a sync buck controller like LT3744 if you need that combination of high current / high dimming frequency. You would need to break your LED string up into smaller segments and run from a higher input voltage as a buck, but the dimming and high current delivery that you are trying to accomplish is much more feasible with a buck than a boost. 

Children
No Data