• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 200 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.

TM4C123GH6PM: Current draw conditions for nHIB pin on TM4C123GH6PM

can altineller
Expert 1470 points
Part Number: TM4C123GH6PM

Hello,

When using the nHIB pin on the TM4C123GH6PM to switch on/off an external power supply, if current is drawn from the nHIB pin, even in range of microAmps, the voltage on the nHIB pin drops, naturally.

In the event of an arbitary power disconnection, nHIB stays high, and since the switching mechanism still drains a little bit of current, and this drains the backup battery.

Let me clarify about the switching mechanism: it is a high side MOSFET switch, triggered by a BJT transistor, schematics below:

Basically, current drawn from nHIB pin by Q3: 1. causes a voltage drop on the voltage provided by nHIB pin, 2. causes the battery to drain when power is removed arbitarily.

I have tried many things, increasing the bias resistors, using a reverse diode to feed the gate of Q3, all with various results. Even if I use a 500kOhm for the base, it still consumes few microamps.

I also understand that if I was using a LDO like TPS73633 to switch the voltage to the circuit using the nENABLE pin, this would not happen.

I also tried changing Q3 with a discrete n-channel mosfet, (BSS138) schematics below:

Which works quite well, does not cause a voltage drop, and when power is disconnected does not consume more than the ~1.5microamps, which is the current drawn by tm4c123 in hibernate mode. It just is a little trigger happy, i.e. if you touch the gate, it will turn on. Or if the power is connected, without the vbat being connected, it will turn on. This does not happen when the battery is connected because nHIB will assert a low.

So this is my question: How can I switch on/off an external mechanism using the nHIB pin, without upsetting the hibernate module, i.e. causing voltage drop? What are the limitations on this pin current-wise? Or can a mechanism be employed to switch something on/off without spending current?

Best Regards,

C.A.

  • 0
    TI__Guru**** 159085 points
    can altineller said:

    I also understand that if I was using a LDO like TPS73633 to switch the voltage to the circuit using the nENABLE pin, this would not happen.

    I also tried changing Q3 with a discrete n-channel mosfet, (BSS138) schematics below:

    Which works quite well, does not cause a voltage drop, and when power is disconnected does not consume more than the ~1.5microamps, which is the current drawn by tm4c123 in hibernate mode. It just is a little trigger happy, i.e. if you touch the gate, it will turn on. Or if the power is connected, without the vbat being connected, it will turn on. This does not happen when the battery is connected because nHIB will assert a low.

    So this is my question: How can I switch on/off an external mechanism using the nHIB pin, without upsetting the hibernate module, i.e. causing voltage drop? What are the limitations on this pin current-wise? Or can a mechanism be employed to switch something on/off without spending current?

    Hi Can,

      I really don't know what to suggest as to why the BJT transistor would draw current on the nHIB high enough to drain the battery. Can you try another BJT transistor. Just wanted to make sure there is no issue with the component. Since you said that either a LDO implementation or a BSS138 would solve the current drawn issue on nHIB pin, then it is something specific with the high side MOSFET implementation for which I don't have enough experience to comment. I wonder if any other pin like a GPIO connected to the base of the BJT transistor, will the voltage also drop like nHIB? I'm sorry that your question is beyond the scope of MCU that I have not  enough knowledge to help. Would you please also try places like Stackoverflow or the vendor of the MOSFET for consultation on your external circuit. If you do solve the problem on your own, please do let us know and hopefully it will benefit the community.

  • 0 in reply to Charles Tsai
    Expert 1470 points

    Hello Charles,

    I removed the connection between nHIB and the mosfet circuitry. I placed various resistors to ground from the nHIB pin, to observe the voltage drop, even if it is couple of microamps.

    Could I have damaged my tm4c123gh6pm? I soldered and desoldered it couple of times, and it has been subjected to heat, and also, the hibernate pin might have been exposed to different voltages.

    The problem with the this setup is that it is a default-on device, i.e. if battery finishes, it will turn it self on.

    I will probably use a 10-30Mohm resistor, pulling hibernate to ground, and just maybe 100nanoAmps more spent, [that is if power is arbitarily disconnected] - which is totally acceptable, to fix this problem.

    One coin battery of cr1220, will last around 2.5-3 years @ 1.7microamps.

  • 0 in reply to can altineller
    TI__Guru**** 159085 points
    can altineller said:

    I removed the connection between nHIB and the mosfet circuitry. I placed various resistors to ground from the nHIB pin, to observe the voltage drop, even if it is couple of microamps.

    Could I have damaged my tm4c123gh6pm? I soldered and desoldered it couple of times, and it has been subjected to heat, and also, the hibernate pin might have been exposed to different voltages.

    Hi Can,

      Do you have a new unit that you can swap in to redo the measurement? How will it compare with the old unit? I don't know if the device is somehow damaged or not. Please also be aware of the below errata. 

    can altineller said:

    The problem with the this setup is that it is a default-on device, i.e. if battery finishes, it will turn it self on.

    Would you mind to elaborate this a little bit more? When you first power on, the main supply will supply the power. You will then arbitrarily remove power and rely on the battery until you reconnect main power again. You don't really use the nHIB pin. Is that the correct understanding?

    can altineller said:

    I will probably use a 10-30Mohm resistor, pulling hibernate to ground, and just maybe 100nanoAmps more spent, [that is if power is arbitarily disconnected] - which is totally acceptable, to fix this problem.

    One coin battery of cr1220, will last around 2.5-3 years @ 1.7microamps.

    Glad that you find a solution. 

  • 0 in reply to Charles Tsai
    Expert 1470 points

    Hello Charles,

    Charles Tsai said:
    Would you mind to elaborate this a little bit more? When you first power on, the main supply will supply the power. You will then arbitrarily remove power and rely on the battery until you reconnect main power again. You don't really use the nHIB pin. Is that the correct understanding?

    I use the nHIB pin, but if the battery is not connected, and if power is connected, because there is no path to ground on the mosfet gate pin, the circuit turns itself on.

    After many testing the final circuit is:

    It keeps the mosfet gate high, drawing as little as current as possible, while providing a 10M ground escape path for charge on the mosfets gate.

    It only costs 300nA more @ 3V with 10M resistor. And that is only and only if main power is disconnected arbitrarily. I could increase the resistor to 30M to limit the wasted current to 100nA.

    When the battery is NOT connected, and sicne the mosfets gate is grounded by r14, it will not default turn on,

    Notice the mosfet in this circuit, turns on other mosfets,so it is only passing 10microamps or so.

    Best Regards,

    C.