• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 256 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.

UCC28251: Enable functionality

Martin Haberzettl1
Prodigy 60 points
Part Number: UCC28251

Hello everyone,

I would like to use a logic High or Low level to directly control wether the IC is enabled or disabled.

The Problem is, i am hesitant to use the EN-Pin for this because of its Pulse-enable-functionality.
According to Datasheet Page 16 "If the EN falling edge happens before the SS voltage reaches 0.3 V, the enable signal at EN pin is considered as a pulse."

I am worried that if the level enable signal i would feed to the EN-Pin was not perfectly clean (for example switch bouncing or electromagnetic interference), it could be detected as a pulse instead of a level, which would mean loss of control.
The input level at EN would be low but because of the pervious short pulse, the IC would still be enabled. (Like in Figure 22 in the datasheet)

I have tried the following ways to circumvent the problem:

  • Use a High signal on the OVP/OTP to disable the IC.
    This works, but when it is enabled again, it starts into Hiccup, which is undesirable.
  • Instead of using the EN-Pin, pull down the SS-Pin to ground.
    On the Datasheet page 24, the following is stated "Pulling SS pin to ground externally shuts down the controller as well."
    This works for the OUTA and OUTB signals, but sadly SRA and SRB keep running.
  • Clean up the signal by delaying the falling edge on the EN signal by a few milliseconds (~ 20% of Softstart-time).
    This is the obvious solution, but it also delays the shutdown reaction time and is not exactly an elegant solution to circumvent the pulse-enable feature of EN.

Do you have any further suggestions how to realize a fully reliable level-enable functionality that can enable or disable all four PWM outputs?

  • 0
    TI__Mastermind 28935 points

    Martin,

    The EN minimum pulse width is 3.2 μs and the condition you referenced from the data sheet is referring to ignoring the EN falling edge signal during SS - do you really need EN control during SS and is this a legitimate concern? If yes, then the best thing is as you said to make sure the EN signal is reliable and not bouncing or chattering. If your EN is chattering, then you may consider using an external edge triggered latch that monitors the first rising edge of the EN signal and sets the output to latch until the latch is reset. You would have to establish the appropriate reset control signal a cording to what is available in your system. Good luck nd thanks for connecting through E2E.

    Regards,

    Steve M

  • 0 in reply to Steven Mappus1
    Prodigy 60 points

    Hello Steve,

    thank you for fast reply.

    A spurious signal or bad switching transistions to happen during softstart is relatively unlikely in my application, but it is not impossible.
    Because i have to build a safe and resilient system (with enough output power and voltage to be lethal), i must adress this potential error condition.
    Of course i have redundant safety systems in place, but every redundant part should be reliable on its own.

    Your proposed solution is similar to my third approach: preconditioning the EN-Signal to supress any transitions that could trigger the pulse-detection feature of the UCC28251

    In my case, i would probably realize it by using a simple RC-lowpass with a diode, so the capacitor charges quickly and discharges slowly. This way, any spurious pulse would have its falling edge delayed by long enough to be detected as level-enable.
    Are there restrains on the minimum input transition rise- and fall-rates on the EN-pin?
    If so, i would need an additional schmitt-trigger speed up the slow transitions.

    So i assume there is no way without glue logic to have a completly reliable level-enable and those are all the options i have?

    Regards
    Martin Haberzettl

  • +1 in reply to Martin Haberzettl1
    TI__Mastermind 28935 points

    Martin,

    I've seen some really nasty EN signals that even an RC would not reliably tame, depending how long the EN signal chattering/bounce is happening. SR edge triggered latch with reset seems to be the most bullet proof and maybe even some amount of RCD filter as you proposed at the front end of the edge triggered latch. In any case, I believe we are arriving to at least a high level answer to the question you proposed. In a system such as what you are describing, My vote is external signal preconditioning/filtering and logic to introduce a reliable EN signal to the UCC28251. Thanks for connecting through E2E.

    Regards,

    Steve M