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.

LM5007: Convertor goes in to Current limit mode or Shutdown mode with nominal inductor value change

Part Number: LM5007

Dear E2E Expert Team,

I have been using LM5007 design from last 6 years and has zero failure till date.

How ever recently have done changes in inductor design and changed inductor value 100uH/0.6A to 120uH/1A.

With 120uH,my buck design working and gives the output 11.65 for 49VDC input but coil gets heated nearly 45 Centigrade.

However if i switch ON/OFF the input voltage several time then LM5007 suddenly going in to shutdown mode or current limit mode and didnt get recoverd. During this condition output of converter is nearly 2 volt and its oscillating.

Convertor working fine with 100uH/0.6A coil and no problem observed till.

I want to know why LM5007 going in to shutdown mode or current limit mode for 120uH/1A.Please refer the below our regular convertor design for your reference.

  • first_name last_name,

    This device is rated for 500mA. Operating in 1A is hitting current limit. The oscillation is probably the frequency set by the RCL resistor.

    -Sam

  • Dear Sir,

    Thanks.

    I will check for current limit resistor.

    One more thing i want to tell you, I have using this convertor design for 200mA load current which is a basically MCU and relay circuit. Even I am using 120uH/1A inductor but my load current is only 200mA.

    Is this inductor value causes problem or i have to check other things. Please suggest.

  • Arvind,

    Sorry, I misunderstood as 120uH at 1A. My mistake. I wouldn't expect this inductor change to cause an issue like this.

    Inductor heating: Caused by IL^2*DCR and by core losses from switching. Have you double checked that the device is switching at 400kHz (approximate CCM frequency from your conditions)?

    Device latching: Please check VCC during the issue. And please also share scope shots of the SW waveform. Also try placing a 2.2uF (or greater) 100V X7R (or similar) capacitor from VIN to RTN.

    -Sam

  • Sir,

    Sorry for delayed reply, I was on long holiday and access to emails was very restricted.

    Today I took the observation as below,

    A)Switching Frequency:

    I have checked the switching frequency which comes 400 to 436 KHz and its vary with load.

    B)Device Latch :

    1)With Device latch condition voltages at each pin of LM5007 is as below,

    SW=2.46 V ,BST=6.59V, RCL=0V ,RTN=0V ,FB= 0.524V, RON=2.18V, VCC=7.04V, VIN=49V

    2)In latch condition at SW pin no frequency observed ,got only steady 2.46 V.

    3)I have tried placing a 2.2uF to 100uF 100V X7R capacitor from VIN to RTN but no effect of such capacitor on device latch.

    C)The problem comes on several power ON/OFF cycles.

    D)Please check below SW pin and VCC capture as below

    SW Pin Capture

    VCC Capture

    Capture at Working Condition 

    Sir ,Problem seems to be related to current limiting resistor.

    When I use the our regular 100uH/0.6A Inductor with RCL changed to 150K,i observed same problem. It might be related to volt-second balance on inductor.

    The comparison between 100uH/0.6A and problematic 120uH/1A coil is as below,

    Measurement Taken with LCR Meter
    Regular
    Drum Shaped
    100uH
    Frequency 100Hz 120Hz 1 Khz 10 Khz 100 Khz DCR
    Inductance 0.101mH 0.100mH 101.4 uH 96.84uH 93.68uH 0.25 Ohm
    Q 0.272 0.325 2.86 14.45 48.2
    Problematic
    Toroidal Shaped
    120UH
    Frequency 100Hz 120Hz 1 Khz 10 Khz 100 Khz DCR
    Inductance 0.120mH 0.119 mH 120.7uH 119.71uH 118.49uH 0.13 Ohm
    Q 0.685 0.814 8.07 31.8 21.6

    Please suggest what to do next.

    Regards,
    Arvind Wagh | Design Support
    PVR Controls | A-160 TTC Ind. Area, MIDC Khairne, Thane-Belapur Road, | New Bombay 400705. India
    T +91 22 2778 2005 | F +91 22 27782008 |
    www.pvrcontrols.com
  • Arvind,

    VCC = 7V so the device should be on and sensing.

    If SW is 2.46V, I assume VOUT is also 2.46V with no load and no switching. Please confirm.

    What is the voltage on BST during this condition. Perhaps the pre-bias from on-off-on is causing an issue with BST voltage.

    I will speak with the team as well to see if we have seen something similar to this before.

    Please also share the layout for review.

    -Sam

  • Sir,

    Please check my reply in red for you questions,

    If SW is 2.46V, I assume VOUT is also 2.46V with no load and no switching : Yes sir VOUT comes only 2.46 Volt with load connected.

    What is the voltage on BST during this condition : Its 6.59V during this condition

    I have been using this design from last 6 years with same layout, no such problem came before.

    Problem comes only when i change the inductor values 120uH/1A with regular 100uH/0.6A.

    Please note that same problem comes if i change the RCL value to 150K with our regular 100uH/0.6A coil.

    There is some interconnection between RCL and inductor values.

    Regards 

    Arvind Wagh

  • Sir,

    Please check below layout for your reference,

  • Arvind,

    Please confirm if VOUT is coming to 2.46V with or without load and please confirm if the device is switching.

    We are thinking the device may be latching from the pre-biased VOUT caused by the on-off-on. You can test this by pulling VOUT low during this condition to see if the device starts switching again.

    -Sam

  • Dear Sir,

    Our application circuit works on 5V supply. To get 5V ,we converting LM5007 Output using IC7805 to 5V.

    During latch condition,2.46V from LM5007 acts as input to IC7805 and output of IC7805 comes nearly 0.6V.

    So during latch condition ,IC7805 acts as load for LM5007 Output.

    No switching happens at SW pins while in device in latch condition.

    You can test this by pulling VOUT low during this condition to see if the device starts switching again.

    Pulling low means should I short the output to check device start switching ?

  • Arvind,

    Yes, try quickly shorting the output.

    -Sam

  • Dear Sir,

    try quickly shorting the output

    I tried the quick short, once i short the output then LM5007gets recovered and starts switching.

    When I measure output ,it comes properly i.e 11.65V.

    If again I switch on-off-on LM5007  then problem comes and need to short the output again to get recovered.

    Please suggest what to do next.

  • Arvind,

    The issue is that the output is not discharging. So SW sits at VOUT and the device cannot pull SW low because we have a diode instead of a low-side FET. This is made worse by having a large BST capacitor. The datasheet recommends 0.01uF but your schematic shows 0.1uF. Try replacing the BST capacitor with 0.01uF and test again.

    -Sam

  • Dear Sir,

    Please have close look in schematic, we have connected 0.01uF Capacitor at BST pin as per recommendation.

    Please see C21 in schematic.

  • Arvind,

    My mistake, I must have looked at the VCC cap.

    Other recommendations:

    • Decrease L1 value. This is from the team and it aligns with your observation of no issue at 100uH.
    • Increase minimum load.
    • Provide bias to VCC. This pin can go up to 14V. Connect VOUT through a diode to VCC.

    -Sam 

  • Dear Sir,

    With respect to your below recommendation, please see my reply in red font

    • Increase minimum load. : I have checked with default 20mA and 10mA load connected across VOUT but problem remains same.Used 560 ohm and 1 K resistor cross VOUT.
    • Provide bias to VCC. This pin can go up to 14V. Connect VOUT through a diode to VCC. : I have connected the diode but have no effect. Diode connected as per below circuit diagram

    please suggest what to do next.

  • Arvind,

    Does your board have an additional voltage which we can use to bias VCC instead of using VOUT?

    -Sam

  • Dear Sir,

    We don't have any provision for additional voltage.

    LM5007 is the only power source in the board. 

  • Arvind,

    In that case we have two options:

    • Use the reduced inductance in the design which works.
    • Select a different device

    -Sam

  • Dear Sir,

    Can I understand what is exact reason we are not able to  use LM5007 with 120uH/1A coil ?

  • Arvind,

    This is a non-synchronous device which means there's a diode instead of a low-side FET.

    The BST capacitor provides the voltage to drive the high-side FET and is charged by the internal LDO when the SW node is low. By toggling EN, VOUT is not discharging entirely but the BST capacitor voltage has fallen, so the BST capacitor is not able to charge up enough to turn on the high-side FET and continue normal switching since SW is held high by VOUT. Synchronous devices avoid this by using the low-side FET to set SW low to charge the BST capacitor but this non-synchronous device cannot do that.

    Your design with 100uH is probably very close to seeing this issue. By increasing to 120uH, we see this issue. The factors I listed previously have an affect on this type of issue. A design which is very close to seeing this issue is susceptible to seeing this issue after minor changes.

    -Sam