• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 64 subscribers
  • Views 517 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.

TPS62A01A: External Sink Capability

Greg
Expert 2031 points
Part Number: TPS62A01A

I would like to know if the TPS62A01A has the ability to sink current when an external source raises the voltage of the power supply rail.  In particular, I have this question in relation to ESD events that get clamped to the power supply rail through the steering diodes of the IC getting zapped with ESD.  In looking at the block diagram of the TPS62A01A, it looks like there is a low-side MOSFET that can be turned on to sink current from the inductor.  Here is the block diagram:

If this is true, it seems the TPS62A01A could react to the rising voltage and keep it in check.  But is the speed of the ESD event too fast for the TPS62A01A to compensate?  Or does all of the capacitance on the power supply rail slow down the rise, allowing enough time for the TPS62A01A to keep the voltage from rising?

Regards,

Greg

  • 0
    TI__Mastermind 25430 points

    Hi Greg,

    I will get back to you on Monday.

    Thanks,

    Febin

  • 0 in reply to Febin
    TI__Mastermind 25430 points

    Hi Greg,

    Thank you for your patience.

    I think you touch two different topics - Current Sink capability & ESD protection.

    If I rephrase your question, you want to know whether the device would be protected if there is a very fast ESD event on output? Is my understanding correct?

    Regards,

    Febin

  • 0 in reply to Febin
    Expert 2031 points

    Hi Febin,

    Thank you for helping out with this.  The question isn't about protecting a device.  It is about what a protection mechanism might do to a power supply rail when an EOS or ESD event occurs.  In a video on the TI website about overvoltage, they show what can happen if current from an external source is dumped into a power rail.  Here is a screenshot from the video:

    The voltage could be from an ESD event or a continuous overvoltage as shown in the picture.  The regulators in this example are likely linear regulators which cannot sink current, which is why the voltage rises until the TVS diode clamps it at 19.2V.  But a switching regulator could have the ability to sink current as well as source it.  My question is whether the TPS62A01A has the ability to sink current, and whether or not it can do it quickly enough to counteract a step input of current into its output.  On one hand, the 2.4 MHz switching frequency might be slower than the event, but on the other hand, it could be that all of the capacitance on the rail will serve to slow down the surge, and a 2.4 MHz switching frequency will allow the TPS62A01A to keep the voltage from rising because it can either stop injecting current, or even sink it to keep it from rising.

    Please let me know if you need more information.

    Regards,

    Greg

  • 0 in reply to Greg
    TI__Expert 5115 points

    Hi Greg,

    Today is a holiday in Germany. Febin will get back to you by tomorrow.

    Thanks

    Yann

  • 0 in reply to Yann Ducommun
    TI__Mastermind 25430 points

    Hi Greg,

    - TPS62A01A is ESD protected as per the specifications in Section 7.2 in the datasheet. 

    - If an over voltage condition occurs on any pin, as long as the AMR of the device is not exceeded, the device will be protected.

    Greg said:
    But a switching regulator could have the ability to sink current as well as source it

    -  Not all switching regulators.

    Greg said:
    My question is whether the TPS62A01A has the ability to sink current

    - TPS62A01A can sink a few mA of current. 

    Greg said:
    whether or not it can do it quickly enough to counteract a step input of current into its output. 

    - What is the output current step you expect?

    Greg said:
    On one hand, the 2.4 MHz switching frequency might be slower than the event, but on the other hand, it could be that all of the capacitance on the rail will serve to slow down the surge, and a 2.4 MHz switching frequency will allow the TPS62A01A to keep the voltage from rising because it can either stop injecting current, or even sink it to keep it from rising.

    - Which event are you referring to? ESD or over current event or over voltage event?

    If you have a specific application challenge, please specify that. It looks like several topics like ESD, current sink capability, over voltage event are all put together and it is difficult to understand your exact concern.

    Regards,

    Febin

  • 0 in reply to Febin
    Expert 2031 points

    Hi Febin,

    Thank you for helping me with this.  I'm sorry if I caused some confusion by asking the question about the low-side MOSFET and how it would sink current.  I did some research on how synchronous buck converters work, and now I understand the role of the lower MOSFET in the block diagram.  It actually functions as a diode, circulating the current back to the inductor when the upper MOSFET is off.  Therefore it doesn't sink any current as I originally thought it could.

    I looked for the ESD/EOS video that I grabbed a screenshot from, but I couldn't find it.  But I found one very similar.  Here is the link:

    https://www.ti.com/video/4080367001001

    I think it will help you understand the question I am asking if you watch that video.  An example is shown of a linear regulator, and how it is unable to sink current from an EOS event.  The solution is TVS diodes on the power rails.  Is the TPS62A01A any different than an LDO in its ability to sink current?  You told me that it is only able to sink a few mA, and that makes sense because I'm realizing it wasn't designed to have that capability.

    My question doesn't have anything to do with the TPS62A01A's ability to protect itself.

    Ultimately, my question is whether or not I need to include TVS diodes on the power supply.  If the supply can absorb the current, I will not need them.

    An LDO has a pass transistor that does not conduct in the reverse direction, so the output of the regulator rises due to the injected current.  This is more of a problem on circuits that are lightly-loaded and with little capacitance.  The upper MOSFET in the TPS62A01A is similar to the pass transistor of an LDO, but the circuit differs when the upper MOSFET is off because current flows back to the inductor through the lower MOSFET and the output voltage falls.  If the output doesn't drop (or even rises a little) because of the injected current, the TPS62A01A should sense this and the upper MOSFET won't be on for very long each cycle until the output voltage drops.  So in this sense does the TPS62A01A sink current injected in its output?  Adding extra capacitance on the output should help with absorbing injected currents.  And of course it depends upon the load.  If there is a lot of load on the output, the injected current will be minimal in comparison and it won't matter.

    Maybe this problem is only typical in lightly-loaded analog circuits that are fed by LDOs?

    Regards,

    Greg

  • 0 in reply to Greg
    TI__Mastermind 25430 points

    Hi Greg,

    Now I understand better. EOS and ESD is different.

    To my knowledge, TPS62A01A does not have EOS protection circuit internally. Anyways, I will double check with the designer and let you know in one or two days.

    Regards,

    Febin

  • +1 in reply to Febin
    TI__Mastermind 25430 points

    Hi Greg,

    Greg said:
    If the output doesn't drop (or even rises a little) because of the injected current, the TPS62A01A should sense this and the upper MOSFET won't be on for very long each cycle until the output voltage drops.  So in this sense does the TPS62A01A sink current injected in its output? 

    I confirmed with the designer, TPS62A01A can sink a few mA range of current (~100mA).

    I suggest to add some diodes as protection mechanism externally.

    Regards,

    Febin

  • 0 in reply to Febin
    Expert 2031 points

    Hi Febin,

    Sorry for the delay in replying.  Thank you for getting this information from the designer.  100 mA seems like a lot of current, which is good to hear.  And of course this is very dependent on the load and the amount of capacitance on the output of the regulator.  Thank you for digging into this for me and getting the answer.  I will probably use signal TVS diodes to prevent the current from entering the protection diodes of the IC.

    Regards,

    Greg

  • 0 in reply to Greg
    TI__Mastermind 25430 points

    Hi Greg,

    That sounds great... Thank you for the update!

    Thanks,

    Febin