• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 64 subscribers
  • Views 1182 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.

LM5069: Problem using LM5069 as a load switch / e-fuse

Nicholas Jones
Prodigy 20 points
Part Number: LM5069

Hi,

I'm struggling a bit with a design based on the LM5069 as a high side load switch controller. I'm controlling the LM5069 by applying voltage to the UVLO pin. If I switch the output on with no load, the voltage comes up and I can then apply load with no problem. If, instead, the load is connected to the circuit output and I apply voltage to the UVLO pin, the circuit does not bring the output up.

I have designed the circuit using the design calculator and simulated it in TI-TINA but my built circuit doesn't match the simulated behaviour. I've tried with and without the dv/dt additional circuit and got the same behaviour.

Using an LM5069MM-2, the voltage on the timing capacitor rises at about 80 µA, which I understand to mean an over-current or over-power fault is detected. However, the voltage across my 5 mΩ current shunt never seems to go above 10 mV and my supply voltage is ~28 V. My power resistor is 25 kΩ and I measure ~500 mV across it when the controller is trying to switch the output on. It seems like the LM5069 is regulating the gate voltage down to limit maximum power but to a point much lower than what I've set with the RPWR. 

As shown in the waveforms, the controller seems to try the output once, then continues to automatically restart detecting a fault. After the first attempt, the restarts are not accompanied by gate voltage or current flow, so I'm not sure what is making the controller think there's a fault. 

I appreciate that the controller is really designed for protecting a system when the input power is connected and disconnected, rather than as a load switch with constant input power but I didn't think I'd be too far outside the intended design use with my application. I'm essentially trying to replace relays and fuses with semiconductor switches in an auxiliary 24 V automotive system. I'd like to switch up to 55 V, and different loads I have require current "fusing" at 3 A, 7.5 A, 11 A and 15 A. Is the LM5069 suitable for this, or if not, is there another product you'd recomend?

1033.LM5069_Design_Calculator_REV_C.xlsx

  • 0
    TI__Guru** 105211 points

    Hi Jones,

    I will review and get back by tomorrow.

    BR,  Rakesh

  • 0 in reply to Rakesh
    TI__Guru** 105211 points

    Hi Jones,

    What is the type of load? Can't we enable the load using PGOOD output of LM5069 to reduce the power stress across the MOSFET. ?

    If you want to start the load during LM5069 startup, you need to increase the power limit for successful startup. Please try as below and share me the test waveforms along with input current.

    Use Rpwr=75k ohm, Ctimer =22nF; use dvdt method. I have attached modified design sheet for reference. Also go through the design examples at www.ti.com/lit/an/slva673a/slva673a.pdf 

    BR, Rakesh

    1104.1033.LM5069_Design_Calculator_REV_C.xlsx

  • 0 in reply to Rakesh
    Prodigy 20 points

    Cheers Rakesh,

    The load will be an electronic clutch, pump motor or solenoid valve. (The design has multiple output channels.) These load can obviously have a bit of inductive kick. I've been testing with resistors or a Rigol DL3031 Load bank. Unfortunately the loads cannot be controlled with the PGOOD signal.

    I will try the modifications suggested - thank you. I'd like to understand the reasoning. Increasing Rpwr seems to set the power limit to 240 W. My MOSFET (SIR626DP) has a rated power dissipation of 104 W at 25 °. Do I also need to usea larger MOSFET? I originally had RPWR = 25 kΩ for power limit of 80 W but the LM5069 appeared to limit to approximately (VRSENSE = 10 mV / 5 mΩ) * (VSUPPLY = 28 V) = 56 W.

    I'd also like to understand why the LM5069 appears to detect a fault when there is no gate voltage and no current flowing (after initial fault) in the waveforms.

    I will be able to provide new waveforms when new parts have arrived - shipping to my location unfortunately takes a while.

  • 0 in reply to Nicholas Jones
    TI__Guru** 105211 points

    Hi Jones,

    104 W at 25 ° is steady state condition. We need to consider SOA graphs in the datasheet as we are allowing high power only for short interval set by timer cap. 

    I will wait for more test results from you!

    BR, Rakesh