• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 61 subscribers
  • Views 364 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.

LM5118: Not producing 24V on startup & Q1 MOSFET Overheating

Ritchie Hua
Prodigy 30 points
Part Number: LM5118

Hi,  

I am trying to design a Power Board Supply with the LM5118 that will output 24V 2A that is then further connected to a 12V (OKI-78SR-12/1.0-W36-C) & 6V regulator (L7806CV-DG).  

I have two problems: 

1. When I start up the from an adjustable DC power supply and supply  >10V, the LM5118 output drops below 5V. Even when the either of the regulators are not connected. However, when I dial the power supply back to 9.8V it resets and the Vout is the expected 24V. Is there a way to ensure that the output will 24V when it boots up with input voltages up to 24V? 

2.  When the lm5118 outputs 24V and it is connected to the 12V & 6V regulator are connected, the mosfet (Q1, KIA50N06B) that is connected to Vin will heat up very quickly. Is there a way to fix this? Does this have anything to do with the gate driver pin H0? 

Thank you for your time. 

Please note:

That I have removed R10 to ensure the EN pin has a high enough voltage supply.

Here is the schematic diagram.

 2570.PowerSupplyBoard.pdf

  • 0
    TI__Guru 59845 points

    Hello Ritchie,

    1. So the output voltage is proper when the input voltage is below 10V, but when you increase the input voltage above 10V, the output voltage is falling out of regulation. Is my understanding correct?

    Does this happen at startup only or as well when you start with 9V and then increase the input voltage to 10V and above?

    Is there a current or power limit in the DC power supply you are using that changes the available current at 10V?

    The LM5118 does not change operation modes at 10V input if the output is set to 24V, so I do not think that the problem is caused by the circuit itself.

    2. How much load is drawn by the regulators? Please check the load that is connected to them. If they discharge the output faster than it is charged by the LM5118, this can cause high power losses on the transistor.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 30 points

    Hello Brigitte,

    Sorry for the late reply and thank you for responding. Hopefully this helps narrow it down. 

     So, the output voltage is proper when the input voltage is below 10V, but when you increase the input voltage above 10V, the output voltage is falling out of regulation. Is my understanding correct?

    When starting up above 10V, Voltage out falls out of regulation, and the current draw (11mA) and is below expected values (>=33mA without load). While the supply is still on, increasing the voltage does not change the voltage output. 

    However, while the supply is still on, decreasing the voltage down to 10V, the current draw goes to ~33mA, and the circuit output voltage maintains the correct voltage regulation. I can then adjust supply voltage to above 10V without affecting the voltage regulation.

    Does this happen at startup only or as well when you start with 9V and then increase the input voltage to 10V and above?

    When the voltage is at 9.8V - 10.2V the voltage regulation is as expected. Could this have something to do with the UVLO specification of 9.8V? and that decreasing the voltage could reset the chip itself? 

    Is there a current or power limit in the DC power supply you are using that changes the available current at 10V?

    The current limit set for the DC power supply is 1.5A. I don’t think this is an issue, as when it voltage regulation is working correctly it will only draw a current of minimum of 33mA and max of about 100mA before attaching a load to it. 

    2. How much load is drawn by the regulators? Please check the load that is connected to them. If they discharge the output faster than it is charged by the LM5118, this can cause high power losses on the transistor.

    While the lm5118 chip is outputting the correct voltage regulation and larger voltage and current is being,

    Reviewing the Q1 mosfet signal for HO and HS pins of the design, the signal appears to be extremely noisy. It looks like there is not enough capacitance between the pins and it might be overheating due to transient losses? Please correct me if I am wrong. 

    Please note that the probe is 10x measurement and the oscilloscope is set to 10x.  

    I have attached the signals appearing at the HS/HO pin at when the current draw is higher and the mosfet overheats.  

     HS Pin.zip

    HO Pin.zip

    Regards, 

    Ritchie

  • 0 in reply to Ritchie Hua
    TI__Guru 59845 points

    Hello Ritchie,

    Please check if the system is able to start if you increase the power limit of your supply to 3A.

    The startup current is dependent on the input voltage and the higher the input voltage, the higher is the startup current. You can increase the softstart capacitor to reduce the inrush current.

    Where exactly did you measure the HS and HO signals. On the pins or on the transistor gate pin?

    As the inductor current is getting to 0A in every cycle, the transistors will show some ringing.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 30 points

    Hello Brigitte, 

    Thank you so much for helping me.  

    Just to follow up,  increasing the current limit did help. 

    But I am not quite sure that I understand why this occurs. 

    • Why would the DC power supply cause it to trip the IC? 
    • Should this occur even when there is no load to the system? 
    • As this application is for a battery-powered power supply,  would attaching one to my board cause the IC to trip every time it is connected?  
    • Are there any other parameters besides the soft starter capacitor that could adjust the IC handling of the inrush current?

    Best regards, 

    Ritchie 

  • 0 in reply to Ritchie Hua
    TI__Guru 59845 points

    Hello Ritchie,

    It is normal that a power supply needs more current during startup than in normal operation. At startup, the output capacitors are not yet charged and this needs to be done in the time that you allow the power supply to reach the correct voltage. So increasing the softstart time will reduce the inrush current as then you give the converter more time to reach the right output voltage.

    Inrush current is mainly defined by the time the converter has to charge the output capacitors, so changing the SS cap is the best way to reduce the inrush current.

    For your battery, it depends on the current capability of the battery where you need to set the inrush current limit. This needs to be checked with the real battery, I think.

    Softstart current is for sure as well dependent on the amount of output capacitance connected, but this should not be the driver for defining the output capacitance. Stability and transient response should drive the needed output capacitance value.

    Best regards,
    Brigitte