• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 62 subscribers
  • Views 1648 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.

LM5060 and OVP behaviour

Simon Edwards80
Prodigy 120 points
Other Parts Discussed in Thread: LM5060, LM5069

Hi,

I’m using the LM5060 as a power switch and want to use the OVP feature to protect the device I am controlling the power supply to.

I have a question on the behaviour of the OVP feature as follows:

  • If the supply to VIN is set so that OVP will be no higher than 1.76V, when the EN pin is set high the FET is enhanced and turns on.
  •  If the supply to VIN is set so that OVP is higher than 1.76V but below 2V, when the EN pin is set high the FET does not turn on.

If I am reading the datasheet correctly then the OVP feature should only work when the OVP pin is higher than 2V. So why am I seeing this behaviour?

Interestingly if I place an oscilloscope probe on the OVP net then the FET does turn on in the second instance. If the oscilloscope probe is already on the OVP net and the power is applied, the FET does not turn on.

Once the FET is on, OVP behaves as it should i.e. bring OVP past 2V and the FET turns off, bring it below 1.76V and it turns on again.

I have tried this with a few different LM5060 devices and the effect is the same.

  • 0
    TI__Genius 12621 points
    Hi Simon,

    There is hysteresis on the OVP pin (240mV typical). You can see the equations on page 20 and 21 in the datasheet.

    So essentially, the OVP should trip at 2.0V (typ) However after it trips, the voltage will need to drop below 2.0V - 240mV (typ) = 1.76V typ before the FET will turn back on.

    This is to have hysteresis so that the circuit down not turn off and on sporadically.

    Thanks!
    Alex
  • 0 in reply to Alex Triano
    TI__Genius 12621 points
    Hi Simon,

    After reading your post a second time, I see that you understand how the OVP function should work.

    Based on your test with the oscilloscope probe, you mentioned that the FET turns back on once you connect the probe. It may be that the circuit was very borderline from turning on, and adding the FET probe with its small level of capacitance was enough to have OVP dip below the 1.76V threshold and turn on.

    For the other test of turning it on to 1.76V with the probe already connected, could it be that there was a small overshoot greater than 2V on the OVP pin? For example, if wires were hot plugged in to a live power source, perhaps this created a voltage transient greater than the OVP setting when applying power? This would cause it to be off, until OVP dipped below 1.76V (typ) to turn on.

    Thanks!
    Alex
  • 0 in reply to Alex Triano
    Prodigy 120 points
    Hi Alex,
    This doesn't explain why the LM5060 will not turn the FET on when the power is applied and OVP is below 2.0V but above 1.76V. The trip point is 2.0V for a rising voltage and 1.76V for a falling voltage (that's the hysteresis).
    I've tried the TINA model of this device and it behaves like the datasheet explains but not how the device on my board behaves!

    Simon
  • 0 in reply to Alex Triano
    Prodigy 120 points
    Hi Alex,
    I was typing my response before your second comment was made. I will try and look at the OVP pin during power on and see what is there. I thought there was a 9.6us filter on the OVP before it turned the FET gate charge pump off so I'm looking for a transient that is greater than 2.0V for longer than 9.6us, correct?
  • 0 in reply to Simon Edwards80
    TI__Genius 12621 points
    Hi Simon,

    You're correct.

    1) FET turned on once oscilloscope probe was applied
    2) FET remained off if oscilloscope probe was applied prior to turn on.

    #1 seems like the probe's capacitance helped put it below the threshold. I can't think of another simple explanation for this.

    #2 yes, this behavior would be apparent if there was a transient greater than roughly 2.0V for roughly 9.6uS.
    - Note that the min spec is 1.88V for OVP, and the OVPdel is 9.6us typ, but not spec'd min and max, nor over temp.

    Thanks,
    Alex
  • 0 in reply to Alex Triano
    Prodigy 120 points
    I've verified your answer but it seems that if you get even close to the 2V threshold at power on for any amount of time, the OVP takes over until you transition below the hysteresis point. Using the oscilloscope I couldn't see any transients on the OVP pin during turn on but there must be something there to trip the comparator.
    Is there another FET controller that I can use with OVP protection?
    My design is nominally 24V powered and is protecting an expensive piece of equipment that can accept no more than 28V. The supply may exceed 28V due to back EMF generated by freewheeling motors etc. In my opinion, transient suppressors are not an option as the duration of the transient is likely to be long compared to what the suppressors are designed to withstand.
  • 0 in reply to Simon Edwards80
    TI__Genius 12621 points
    Hi Simon,

    You can try adding a small capacitor (RC) to the OV pin of the LM5060 to help filter this on the front end of the internal comparator (if that is the issue). If using a cap, it may be worth using an independent, 4-resistor option for UV/OV (option B in the datasheet, figure 30, page 21).

    Otherwise, the LM5069 offers additional protection features over the LM5060. It has an active, adjustable current limit, a power limit to keep the FET's operating within their SOA, and short circuit protection (fast shutdown for severe overcurrent event).

    In addtion to this, it has UV/OV protection with hysteresis, but the amount of hysteresis voltage is adjustable based on the resistor selection (there is a hysteresis current source which turns on and off).

    www.ti.com/.../lm5069.pdf

    Thanks!
    Alex
  • 0 in reply to Alex Triano
    TI__Genius 12621 points
    ++ Also, if using the LM5069, we have a design tool which can be downloaded at www.ti.com/hotswap --> "Tools & Software".

    This tool can help you with component selection, such as determining the UV/OV resistors and hysteresis using the automated equations rather than by hand.

    Thanks,
    Alex
  • 0 in reply to Alex Triano
    Prodigy 120 points
    OK thanks Alex, I'll try the cap first then look at the LM5069.