• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 277 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.

UCD9090A: Trouble with Delay Settings...Please Advise

Darren (FAE)
Expert 5483 points
Part Number: UCD9090A

I have a customer having some difficulty with the delay timings.

They are using the Fusion Digital Power Designer. I have attached some screenshots of the setup helping outline the below outline:

Basically, under the Hardware Configuration Tab -> Logic Controlled GPOs -> Configure -> Delay Time, they set this to 2ms

Then, under the Rail Configuration Tab -> Voltage Fault Max Glitch Time = 20ms -> They set this for both OV and UV faults on the rails

---

With this setup, they are expecting a delay of around 22ms from when the voltage rail goes past the OV/UV setting, to when the GPO output changes.

But they are only seeing a delay of around 2ms. Your help is much appreciated.

Regards,

Darren

image003.zip

  • 0
    TI__Guru 65750 points

    Hello

    The glitch filter is only valid for the fault response. device will report UV or OV fault as soon as detecting it, but the action of the fault would be delayed by the glitch filter time. The GPO output is not affected by the glitch filter.

    Regards

    Yihe

  • 0 in reply to yihe
    TI__Expert 5483 points

    Hi Yihe,

    Thanks for the quick reply. I am still a little lost, sorry...I will use Figure 16 from the Datasheet to help.

    1. From the above figure, it looks like there is this "MAX_GLITCH_TIME" which, according to the settings attached in my previous post, should be set to 20ms for both OV and UV. This setting means the OV / UV condition needs to continue for 20ms for it to be detected as a "fault". Correct?

    2. After the "fault" has been detected, there is a delay of TOFF_DELAY[1], and the RAIL 1 EN (GPO) goes low, and the RAIL 1 VOLTAGE falls. Is this the DelayTime which was set as 2ms in my previous post?

    3. The customer was measuring the RAIL 1 VOLTAGE and GPO/EN with an Oscilloscope, and was expecting the EN pin to go LOW after at least the MAX_GLITCH_TIME, which was 20ms. But instead, the EN pin went LOW after only 2ms of the OV/UV condition...
      In your post, you said the GPO is not affected by the glitch filter...But the above figure shows it should be affected by this setting...how can the software be configured so the GPO doesn't flip state until the glitch filter time has passed?

    I really appreciate you helping me understand this for the customer.

     

    Regards,

    Darren

  • 0 in reply to Darren (FAE)
    TI__Guru 65750 points

    HI

    Please see my comments below.

    #1. the fault condition is detected as soon as the voltage is out of valid range, but the action of the fault(de-asserted the EN signal) is taken if the fault last over the glitches time. The image shown in the data sheet is right. but the EN signal listed in the data sheet is not the Logic GPO. They are different functions.

    #2. no, what you set is the delay for the logical GPO, which is not the same as TOFF DELAY

    #3, the EN pin is different with LGPO. this is the root cause of the issue.

    Please refer the attached slides. It shall help to understand how to configure the device properly.

    0167.UCD90xxx Quick Guide.pptx

    Regards

    Yihe

  • 0 in reply to yihe
    TI__Expert 5483 points

    Yihe,

    I think I get it now, but just want to confirm.

    In the Fusion Digital Power Designer software, as shown in your slides (slide 8), there is an "enable" option for the "Rails & Enables" box. The pin associated with this will have the fault delay (glitch filter) applied.

    Whereas the Logic Controlled GPOs section sets a different pin, and has a different delay (just a response delay from the moment in time the fault was detected)

    This understanding is correct?

    Regards,

    Darren

  • 0 in reply to Darren (FAE)
    TI__Guru 65750 points

    HI Darren

    Yes, you are right.

    Regards

    Yihe