• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 62 subscribers
  • Views 1083 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.

LM3481 UVLO max. hysteresis

Manuel Reinoso
Prodigy 50 points
Other Parts Discussed in Thread: LM3481


Hello,

We have two questions about UVLO in the LM3481 part.

Design info:

R7 = 100K
R8 = 69.8K

Vsh = 3V (calculated)
Ven = 3.5V (calculated)
Hyst = 500mV (calculated)

Additional info:

Fsw = 500Khz
Topology = SEPIC with coupled inductor
Cin = 100uF
Cout = 660uF

Iout = 2A max.

Questions:

1- Maximum hysteresis that can be configured? Could be above 500mV? The UVLO is not working as expected when powered from lithium battery, when the Vin reaches 3V the IC stops switching but quickly returns to normal state. The battery voltage rising is only about 0.35V when the UVLO triggers at 3V.

2- Maximum R7 & R8 values?

Thank you.

  • 0
    TI__Intellectual 2225 points
    hi Manuel,
    The hysteresis should be able to be 500mV. When I plug Vsh=3V, and Ven=3.5V into equations 17 and 18 in the datasheet, I get R8=69.8k, but R7=75k. That should help increase your hysteresis.
    Again, I don't know if there is a maximum value. The main thing to keep in mind is that the absolute maximum voltage on the UVLO pin needs to stay below 6V even at the maximum VIN.
  • 0 in reply to Perry Tsao96839
    Prodigy 50 points

    Hello, thank you for the answer, but the equation I'm using results in R7=100K

    R7 = R8 x ( Ven / 1.43 -1) = 69.8K x (3.5V / 1.43 - 1) = 101K

    Maybe the equation 18 is wrong in the datasheet Im using and the correct parameter is Vsh ? If I use 3V instead 3.5V I'm getting 75K for R7.

    Regards

  • 0 in reply to Manuel Reinoso
    TI__Intellectual 2225 points
    Hi Manuel,
    Equation 18 is correct. I made a mistake, your calculation was correct. Sorry for the confusion.
    I will test this on the bench tomorrow to see if I get the same results you did.
    Perry
  • 0 in reply to Perry Tsao96839
    TI__Intellectual 2225 points

    Hi Manuel,

    Sorry for the delay.  I re-checked the equations (they are correct), and checked it on the bench.  With the R7=100k and R8=69.8k, I measured VEN=3.38V and Vsh=3.04V. This hysteresis is similar to what you measured.  In choosing this hysteresis, you should keep in mind that the hysteresis will be decreased a bit by any ripple or random noise that might be seen on the UVLO line.  Therefore, if your VIN supply feeding the UVLO has 50mV of ripple, then the hysteresis will also be decreased by that amount on both the rising and falling edges because it will get tripped at the peak/valley of the ripple and not the average value you might seen on the multimeter.  In particular, the VIN/UVLO input ripple will be higher when the switcher is switching, so this will affect the falling VEN threshold more.

    To increase the hysteresis, I would just scale the R7 and R8 values.  For example, to increase the hysteresis by 50% from your measured 0.35V, I would just increase R7 by 1.5 to 150k and R8 by 1.5 to 105k.

    I hope this helps. Let me know if you have any further questions.

    Perry

  • 0 in reply to Perry Tsao96839
    Prodigy 50 points
    Hello Perry,

    I have tried with R7=150K and R8=100K and the hysteresis does not increase. In fact, I think that the UVLO does not work at all with resistor values above 100K. When the li-ion battery reaches a value next to 3V , the IC starts a loop of ON/OFF state.

    So my conclusion is that the max hysteresis achievable with the LM3481 is about 400mV

    Maybe i'm missing something?

    Thanks.
  • 0 in reply to Manuel Reinoso
    TI__Intellectual 2225 points

    Hi Manuel,

    I took some measurements with R7=150k and R8=105k, and got Vsh = 2.9V, and Ven=3.3V (0.4V hysteresis).  To further increase the hysteresis, I used R7=226k and R8=136k and got Vsh=2.95 and Ven=3.55, which is pretty close to what you were looking for.  There should be no issue with using higher value resistors. 

    When doing your measurements, I would recommend testing with a adjustable lab power supply first rather than directly with the battery.  It is possible the higher impedance of the battery is causing voltage droop which can cause the on/off cycling that you mentioned. 

    As an alternative way to select the resistors, you can increase R8 to increase hysteresis until you get the amount you want (0.5V).  Then adjust R7 accordingly to get the turn-off point that you want.

    Let me know if this helps.

    Perry