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.

UCC28880: Failure to output voltage upon POR (Occasionally)

Part Number: UCC28880
Other Parts Discussed in Thread: UCC28881

The UCC28880 part is being used to supply power to a control circuit. The current design outputs 11V. Recently, the product this part is in has had several failures when the power is turn off then back on and the switcher stops operating. From what was scoped, the mosfet is no longer switching to provide output voltage. This failure can be repeated but does not occur every time. When the device is removed from AC for longer than 15 seconds then reapplied to AC, the IC will begin operating again. The measured values during failure: VDD=5V, FB=0V, HVIN to GND of the IC =160V with 120VAC input.

  • David

    Do you have a schematic of your design that you can provide?

    Did you use our excel calculator http://www.ti.com/lit/zip/sluc625 

    Best Regards,

    Eric

  • Hi Eric,

    I used the WEBENCH designer tool. 

    The schematic is shown below. One of the components that I am wondering may be a problem is the Cvdd capacitor. The excel sheet you provided shows a 10nF cap and the webench had a 100nF. Currently, there is a 1uF cap there. Could that cause an issue with voltage slew rate on power up and power down that would lockout the IC? To test this, we are putting a 100nF cap on and will conduct test to see if we can replicate the issue. 

    Regards,

    David Strong

  • David

    In section 8.2.1.2.3 (page 20) of the datasheet, it's listed that 100nF is recommended for Cvdd (C17).  We agree that the test you are running where you reduce this capacitance is a good idea.

    I also noticed that there is no pre-load resistor or zener clamp.  If there is a minimum load always on the 3.3VDC output then a component to provide a minimum load on the output isn't needed, but if 3.3VDC can have no load, you may need additional pre-load to insure proper operation.  Section 8.2.2.2.2(Page 26) of the datasheet has details on how this minimum load (RL) impacts the regulation feedback.

    Best Regards,

    Eric

  • Hi Eric,

    The pre-load resistor was omitted because there is always a load on 3.3VDC rail as well as the 11VDC rail.

    Through testing, if we aren't able to replicate this with the capacitor change how would we know it would work in the field for its lifetime? Is there more to the answer than just what's recommended in the datasheet?

    Regards,

    David 

  • David

    What we recommend is that you experimentally verify that the change you make to the design will eliminate the undesired behavior you are observing.  I highlight sections in the datasheet to help explain why a specific change will resolve the issue.

    If reducing VCC capacitor does not resolve this issue, the next steps we recommend are to measure what exact voltages/currents UCC28880 sees to make sure its the same for what's expected.  Since UCC28880 is floating and has a high voltage swing every time the FET turns on or off this can be difficult to do, so having a good understanding on how UCC28880 should operate is recommended.

    Before you finalize your design I do recommend that you verify that the current thru the internal FET does not exceed the abs max limit of 320mA.  If you find that this is happening, we recommend you use the pin-to-pin equivalent UCC28881.

    Best Regards,

    Eric