• TI Thinks Resolved

UCC3802: Implementing Short Circuit Protection in Flyback

Prodigy 50 points

Replies: 9

Views: 88

Part Number: UCC3802

Hello Sir,

we have designed a Flyback(40W) using IC UCC3802, but short circuit protection is not working,

in our design voltage across R_sense Ramped to 1V, but IC doesn't go to fault mode even after output gets shorted.

Please Guide us on how to implement Short Circuit protection in the Same design.

Thanks & Regards

Shivam Singh

  • Hi Shivam,

    The CS pin feeds two comparators, the PWM and Overcurrent. The CS pin voltage needs to rise to 1.5V to trip the overcurrent comparator and trigger a restart.

    If the CS pin voltage is is clamped to 1V then it sounds like the PWM comparator is still in control, what is the voltage on the COMP pin during the short circuit?

    please also post your schematic so we can review the circuit.

    Thanks

    Peter

  • In reply to Peter Meaney:

    Hello,

              Thank you for your response. Please note that the cycle by cycle power on the primary side is limited to 1V as sensed by the MOSFET current  resistor. Thus the voltage never reaches 1.5V required by the overcurrent comparator. Anything above 1.1V  tends to fry the secondary side components. A circuit which can kill the PWM comparator when the MOSFET current sense voltage rises to 1.1V or so would be appreciated. The primary side schematic is attached with this response for your ready reference.

    Thanks & Regards

    Shivam

     

  • In reply to SHIVAM SINGH:

    Hi Shivam

    How is the Vcc derived? Normally Vcc is supplied from a auxiliary bias rail from the transformer which collapses under short circuit condition and results in a hiccup mode of operation as the power supply alternates between restart and shutdown.

    Regards

    Peter

  • In reply to Peter Meaney:

    Hi Peter,

                   The VCC for the controller is derived from the bias winding of the flyback transformer. The bias winding voltage does not collapse during a short circuit as the primary side current is limited to 1V across the MOSFET current sense resistor. It's not clear why the short circuit limit has been internally set to 1.5V which I think is causing the problem. Is there a way to circumvent this ?

    Thx and regards

    Shivam

  • In reply to SHIVAM SINGH:

    Hi Shivam,

    What are the values of the RC filter on the CS pin, it looks like R24 is 1MR and C33 has no value. Is it possible that the CS signal is been filtered.

    The IC has leading edge blanking so the external filtering should be small, maybe 1k and 100pF.

    The Aux winding is probably peak detecting during the short, maybe adding a resistor in series with the aux winding to its bulk cap will decrease its ability to peak detect and may allow the winding to collapse during short circuit.

    Do you have waveform images of the voltage across the CS resistor R11 and the waveform on the CS pin.

    Thanks

    Peter

  • In reply to Peter Meaney:

    Hi Peter,

                   Though the circuit diagram does have it, we did not put any filtering components on the CS pin when connecting up the supply. What value would you recommend for the auxiliary supply resistor ? It is specified to give 15V and the bulk cap is 100uf.  The peak value of the wave form across the CS resistor during short easily exceeds 1V thus ensuring that the primary power limiting takes place. 

    Thx and regards

    Shivam

  • In reply to SHIVAM SINGH:

    Hi Shivam,

    The series resistor value is usually in the order for 10R to 100R as it depends on the current been drawn by the controller which will give you a constant voltage drop across the resistor and at worse case operating condition where the Vcc voltage is a minimum, I suspect no-load on the output, you will want enough voltage on the Vcc rail to keep the IC operational.

    Is the aux winding more tightly coupled to to the primary or secondary ?

    Regards

    Peter

  • In reply to Peter Meaney:

    Hi Peter,

                  I am sorry but I forgot to mention that we have a 150R resistor connected in series with the bias winding for the internal zener of the flyback controller. The primary winding is interleaved and the bias winding is closely coupled to it. You probably might have some short circuit protection circuit for the UC3842 which I think with some modification could work. Please let me know.

    Thx and regards

    Shivam

  • In reply to SHIVAM SINGH:

    Hi Shivam,

    The difficulty is how to recognize the short circuit, the only information you have that the device is under overload/short circuit is that the voltage on the CS resistor exceeds 1V.

    If the aux was wound closer to the secondary its voltage would follow the secondary winding, instead it closely coupled to the primary which means it can still regulate when the secondary is shorted.

    In the past we have used a simple transistor pull down which is uses a NPN who's base is charged through an RC circuit from the CS pin voltage. If the RC is selected such that is takes some time say 10ms to cause the NPN to turn on the this time can be used as a filter to prevent nuisance tripping during turn on. The collector of the NPN can be used to pull COMP pin low and disable the PWM gate drive. If more precision is require a comparator can be used to flip high when the CS pin exceeds the desired threshold and drive the RC circuit which controls a switch to pull COMP low.

    Regards

    Peter