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.

UCC28782: UCC28782 CIRCUIT VALUES IN EVAL kit

Part Number: UCC28782

Hello:

We fixed the MP6908 circuit VG uncertainty issue by dropping the output thru a zener, then feed it to TPS7A2450DBV LDO and there is no failure of the output ICs.

But the circuit started off and went into bursts of switching packets at the drain of lower  NV6117.

Our circuit & routing mimics the UCC28782 EVAL kit. So what would cause the module to operate in bursts not steady-state? 

It is as if some capacitor is not high enough to hold charge...such as at S13 node most likely?

Ours is .033uF while the Eval kit shows .022uF

..Could it be controller VDD , which has now 64 uF, X5R caps?

Of course, at this time we are not operating at 65W, much less. Our Rcs is .22 Ohm, ...Pl see attached exl file for this operating point.

At this time, xmfr is 20:4:4, Lpri is 380uH, lkg is <1 uH, electronic load is Resistive 20 Ohms or around this value.

We do not use a third coil to power MP6908 though.

Any helping hint to troubleshoot will be highly appreciated.EQ30PILOT_EXLCALCJUL27debug.xlsx

  • Hi Robin,

    I see the full load power input line in the calculator is 15W,  I think that is maybe a typo when you input it . that is why you get  Lm=380uH . based on our experience . for 65W/20V design. Lm = 110uH is a good start point  , and current sense resistor is around 0.12ohm . turns ratio equal to 5~6 make sense .

    the controller has SBP2&1 , LPM, ABM, AAM operation modes , except for AAM mode , all other modes are running at burst mode , and we recommend to set the load  >50~60% full load entry into AAM mode . this can be achieved by setting the BUR pin resistors value , you also can adjust it during the debugging .

    Thanks.

  • Thnx much Jaden.

    The burst switching on startup I think occurs due to SOME values- I will double-check on your observations.

    After some cycles, there is not even startup pulses...so maybe some cold solder... such a circuit is very hard to debug. So we have to cut trace for S13. Driver with external bais & an external PWML  & track where signals are not as expected. 

    Let me report after that...

    r

  • Jaden: did you see my email to you last week? Would you pl.