• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 65 subscribers
  • Views 1282 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.

LM5050 Gate not working

Rob vreden
Intellectual 310 points
Other Parts Discussed in Thread: LM5050-1, LM5050-2

Hello,

Need some help with an issue I have with the LM5050. I use it in a standard configuration described in the datasheet. I have pins 1&3 left open/not connected.

The input Voltage is 20Volt. the load draws 2 amp of current. Th efet I use: AUIRLR3636 and/or also tried the older, higher RDs on: RFD16N05:LSM( 47mOhm)

what I measure is: source Voltage 20 Volt; Drain Voltage 19,4 Volts.  I measure 0,1 Volt  on the gate pin in respect to the source. and ),7 Volt in respect to the drain.

the Gate voltage measured  in respect to the ground is 19,7 Volt.

So the body diode is conducting , the Voltagedrop accross it is approx 0,7 Volt, but the gate Output is not driving at all.

I have swapped out three Lm5050 to see if the product was faulty. but no result. I bought the LM5050 from Farnell. 

Not sure what is gooing on. Need some help or ideas here.

kind regards,

rob van vreden

  • 0
    TI__Mastermind 22610 points

     

    Which version of LM5050?

    LM5050-1 or LM5050-2 ? Pin 1 function is different between them, the two versions are not interchangeable.

    For LM5050-1 : Pin 1 is the main supply pin for all internal biasing and as an auxiliary supply for  the internal gate drive charge pump. VS is typically connected to either Vout or Vin, or a separate supply can also be used.

    For LM5050-2 : Pin 1 is an open drain output for the FET Test circuit. The Status Out pin used in conjunction with the OFF test mode pin.

     

  • 0 in reply to Donald Jones
    Intellectual 310 points

    Hello,

    the one I have is: LM5050MK-1/NOPB. So model-1.

    Pin 1 is connected to the V-in.

    VS is Voltage Sense? that pin is No1 and already connected to the Source.

    I have removed the fet and measure no gate drive voltage other than 0,1 Volt. Or 19.3 Volt irt the ground.

    could I have an fake component?

    regards,

    rob

  • 0 in reply to Rob vreden
    TI__Mastermind 22610 points

    LM5050-1 top side mark is : SZHB

    Pin 1 (VS) is internal bias and logic supply pin.

    Pin 4 (IN) is the voltage sense connection to the external MOSFET Source pin.

    Pin 6 (OUT) is the voltage sense connection to the external MOSFET Drain pin.

    With the OFF pin open, 20V applied to the VS pin and the IN pin, and the OUT pin at least 5mV below the IN pin voltage, the GATE pin drive should turn on.

    Based on your description here I have no idea what has gone wrong.

    As for having a "... fake component ...", seems unlikely for two reasons : 1) Farnell is a reputable dealer, 2) I doubt that the LM5050-x has that large of a demand at this point that counterfeiters would bother.

  • 0 in reply to Donald Jones
    Intellectual 310 points

    I have pin 1 and 3 open/nc. For me confusing why you mention VS( Voltage sense)for pin 1 and pin 4.

    So I have nothing connected to pin 1. so no 20 Volt as you describe above. Should I connect this pin( max 5,5 Volt) to an external supply?

    I have tried to rotate the device 180 degrees, but that generates smoke.

    It is hard to see what pin 1 is, but it is detectable. So I am sure it is in right.

    the circuit is to connect two sources: wind power and solar power to an invertor. all works exept that the fets are in diode bypassing only. I need to fix that.

    regards,

    rob

  • 0 in reply to Rob vreden
    Intellectual 310 points

    Found the issue. I was looking at the wrong datasheet. I was looking at the -2 sheet. that is why I could not understand the VS pin.

    It works in the meantime. Thanks for the support. It would have cost me weeks to find it myself:-)

    regards,

    rob

  • 0 in reply to Rob vreden
    TI__Mastermind 22610 points

    My apologies.

    Your earlier statement ".... Pin 1 is connected to the V-in...." confused me.

    For LM5050-1, the VS pin (Pin 1) must be connectected to a voltage source (5V to 75V) for proper operation.

    I do not see any configuration in the datasheet where the VS pin is not connected. The VS PIN description (Applications Information, Page 11) states that "The ... VS supply must be between 5V and 75V."

    Your present statement "... Should I connect this pin( max 5,5 Volt) ..." also has me confused (i.e. max 5.5V) as the Operating Rating for the LM5050-1 VS pin is 5.0V to 75V.

    Yes, please connect the VS pin (Pin 1) to either the IN pin (pin 4), or the OUT pin (pin 6).

     

  • 0 in reply to Donald Jones
    Intellectual 310 points

    Hello,

    Well, I am not quite there. One part is working. it turns the gate on when it should.

    The other one is not working yet. The difference is that the one that works has a clean DC Voltage on the source. the other one is fed by the output of an step up regulator. working at 150KHz or so. could be that this interferes with the operation. I have two 68uF at the source, but maybe I have to add some ceramic capacitors to it as well.

    the one not working gives 0 Volts irt the source, but 20 volts irt the ground.

    regards,

    rob

  • 0 in reply to Rob vreden
    TI__Mastermind 22610 points

    SMPS are problematic for LM5050-x.

    As the gate starts to turn on, Rds(on) begins to fall, and the output voltage gets closer to the input voltage, at some point the SMPS ripple voltage will cause the LM5050-x to detect a reverse voltage condition and the gate is immediately discharged at a ~2A rate. Then the whole process starts over.

    The LM5050-x is especially sensitive when the output current is low, and the resulting Iout x Rds(on) does not produce enough voltage to offset the ripple voltage.

    One trick to try is to place a 0.01uF to 0.1uF ceramic capacitor across the mosfet, drain-to-source. Depending on other circuit components this usually allows the ac signal to common-mode across the reverse comparator inputs. This is not a fix, it is a cover-up, and it does not always work. The LM5050-x is working correctly, it's reverse comparator is very fast by design, and it doesn't take too much of a reverse voltage for it to respond by totally discharging the gate.

     

  • 0 in reply to Donald Jones
    Intellectual 365 points

    Is this the reason for the LM5050-1 datasheet recommendation to select the MOSFET Rds_ON so there is at least 22mV of drop at nominal load current?   Are there any other reasons for this minimum Rds_On recommendation?

    Thanks,

    -Derek