• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 19 subscribers
  • Views 814 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.

Logic devices: output short-circuit protection design?

Daniel Ka
Intellectual 497 points
Other Parts Discussed in Thread: SN75123, INA381, LM334

Hello, 

I have a question regarding implementing an overcurrent/short-circuit protection for outputs of the logic family ICs. I could not find this question in the FAQ. I understand that some logic devices have outputs that are short-circuit protected (e.g., SN75123), but most don't.

What is a recommended practice to protect logic chips from damage in configurations where the output current limits could be exceeded?

For example, if the output signals of a 74LVC3G34 (triple buffer) or 74AHCT541 (3state octal driver) are available externally to the user (e.g, to attach own devices), I think they should be designed to withstand shorting to GND. 

A simple solution would be to use a series output resistor to limit the current. For example a logic device operating at 5V, a resistor of 500 Ohms would limit the current to 10mA (i.e., below the ABSOLUTE MAXIMUM RATINGS). However, the voltage drop at the resistor would prevent a specific HIGH level for different output currents. If the customer application would sink a current of 5mA, the external voltage would drop to 2.5V. 

What would be a better design? 

Thanks.

Daniel

  • 0
    Guru 244270 points

    There is no easy, general solution.

    If you want to limit the output current of your existing logic gates without much of a voltage drop, you have to use a small resistor and add an opamp or current sense amplifier (e.g., INA381) that switches the output off.

    There are self-protected MOSFETs with an internal current limit (typically 1 A or more); you could use them to build your own output drivers.

    If you have a higher supply voltage, you could add series current limiters (LM334 or discrete) and then regulate the voltage down to the desired output voltage.

  • 0 in reply to Clemens Ladisch
    Intellectual 497 points

    Thank you, Clemens, for your suggestions. 

    While looking at the different solutions, I came across "PTC resettable fuses". Without preferring one brand over another, this is an example for these devices:

    MF-FSMF010X (Datasheet: https://www.bourns.com/docs/product-datasheets/mffsmf.pdf?sfvrsn=a3c99b25_6)

    This model will hold 10 mA of current and trip at 30 mA. However, due to the heating effect these fuses are based on, their reaction time is quite long. According to the datasheet, the fuse trips after 1 second of a current load of 500 mA.

    Will this suffice for the discussed overcurrent protection?

    If not, what reaction time (e.g., to switch off) in case of an overcurrent event is required for efficient protection?

    Thanks.

  • 0 in reply to Daniel Ka
    Guru 244270 points

    The absolute maximum ratings must not ever be exceeded, not even for a short time. If there were a higher rating for short pulses, it would be mentioned in the datasheet.

    Even the opamp-based protection might be too slow for certain cases.

    Consider rethinking the overall system design.

    Typically, plain logic signals are accessible on development boards, and in extension devices like PCI cards. However, in those cases, the users are supposed to know not to abuse them.

    Most signals that are externally accessible (e.g., USB, RS-485) are designed to be short-circuit resistant and have different electrical characteristics.