• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 65 subscribers
  • Views 257 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.

UCC28780

Jung-hui Cheng
Prodigy 150 points
Part Number: UCC28780EVM-021
Other Parts Discussed in Thread: UCC28780

Above screenshot is the startup test waveform on Q2' Source and Gate in UCC28780EVM-02 . The ramp-up time is around 50ms.

We prototyped the circuit with Q2 replaced by BSP135 and the same capacitance 40uF on VDD. The new circuit uses extra 2.5mA from VDD for other circuits. We found  the time on ramp-up waveform is increased to 4.5 seconds. In  Fig. 16 of datasheet SLUSD12A,  the statement is described below

RDDS is set at 12 kΩ when VVDD is below 1 V to limit the maximum fault current under VDD pin short events. RDDS is reduced to 1 kΩ when VVDD rises above 1 V to allow VVDD to charge faster. The maximum charge current (ISWS) is affected by RDDS, the external series resistance (RSWS) from SWS pin to QS, and the threshold voltage of QS (VTH(Qs)).

Question: The 4.5-second ramp-up time is due to damaged  RDDS  remaining at 12 kΩ even VDD is above 1V ? Or due to QS (VTH(Qs)) is reduced  under higher  initial charge current of Isws which is limited by the equation of VTH(Qs)/(RDDS+RSWS)?

 

  • 0
    TI__Mastermind 31930 points

    Hello Jung-Hui, 

    Thank you for your interest in the UCC28780 ACF controller and its EVM. 

    I don't think that Rdds is damaged and stuck at 12kR.  If it were so, I think there would be other problems as well. 
    Looking at the BSP135 datasheet, I see that its Vgs threshold voltage is about -1.5V on average.  This voltage would be impressed across the internal 1kR (Rdds) plus the external 121R (R22) which results in about 1.34mA of charging current to the VDD node (through SWS). 

    The UCC28780 controller draws only ~70uA before start-up, so that leaves 1.27mA remaining to charge VDD. 
    Since 40uF charges about 7.5V in 4.5s, this calculates to ~67uA charge rate, so I estimate your other circuits are drawing about 1.2mA during this long charge-up time.  

    Compared to the BSS126, the lower threshold voltage of the BSP135 reduces the total available current for start-up and the extra circuits on VDD use up most of that.  Very little is left to charge Cvdd. 
    Graph 10 of the BSP135 datasheet indicates that Vgs(th) increases when it is heated.  A simple experiment of temporarily directing a hot-air stream at the BSP135 should increase the Vgs threshold by ~0.4~0.5V.  This additional voltage should increase the starting current by ~360uA or so, which is small but still about 6X higher current than calculated above.  If start-up ramp time reduces to <1 second, this proves that your extra circuits on VDD are "starving" the VDD capacitor.  In some cases, VDD may not reach the 17.5V start-up threshold at all.  

    If the functions of the extra circuits are not needed until after the ACF controller starts up, then these circuits may be powered by a separate auxiliary bias source from the AUX winding, with it's own diode and filter capacitor, separate from that on the UCC28780 VDD net.  This would relieve the VDD net from the burden of supporting the extra circuits before start-up.

    Regards,
    Ulrich

  • 0 in reply to Ulrich Goerke
    TI__Expert 7080 points

    Hi Jung-hui,

    Ulrich's analysis make sense to me , I total agree with him.

    Since I see you have created another thread, I'll close this post and let's discuss in that one.

    Thanks.