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.

LM3488: Strange behaviour

Part Number: LM3488

Hello,

I have built and run an LED driver as shown in the picture below. It consists of a few blocks:

- over-voltage protection and noise filtering,

- anti-surge protection,

- 16V regulator (that works as a voltage limiter for the LM3488 chip),

- 5V regulator for PIC chip,

- PIC chip that controls LED brightness using PWM.

When the driver (and the LED load) is supplied from an autotransformer it works fine - no problems whatsoever (figures 2, 3). But when the supply voltage is taken from a traffic light controller (that uses triacs for switching the driver on and off) it quite often shows abnormal behaviour (figures 4, 5).

The heart of the driver is a SEPIC converter. I found, when I ramp the voltage up from 0V, the current is very high before the converter "kicks in". Then after everything is back to normal. Figure 5 indicates that the input voltage (coming from that traffic light controller) has its peaks cut off. I suppose this happens because each cycle (or half-cycle) the SEPIC converter requires high current to start and the source (traffic light controller) limits the current to prevent overloading.

Is there any chance to prevent the keying MOSFET Q4 from being on before the converter starts?

Figure 1   Driver diagram

Figure 2   Normal function - input current

Figure 3   Normal function - input voltage

Figure 4   Abnormal function - input current

Figure 5   Abnormal function - input voltage

Thanks.

Kind Regards,

Piotr Rozycki

  • Hi Pitor,

    The LM3488 will start operating when the VIN pin is around 3V. Until then Q4 should be turned off and there should be no switching.

    It is possible to increase the soft-start time of the device, which will help reduce the initial current need to start the converter. This application note show how to increase the soft-start time when the device does not have an external soft-start pin. Please refer to this document.

    Let me know if you have any other questions.

    Thanks,

    Garrett

  • Hi Garrett,

    Thanks for your reply. The "mystery" of the asymmetrical current waveform has been solved now - the load was too low and this caused the triac to open intermittently. Since I added a 10k resistor in parallel to the driver input this phenomenon has disappeared.
    Regarding the soft start, I'm intending to use another I/O of the PIC chip to wake up the converter (SD pin) when the input voltage is 20V AC.


    Kind Regards,
    Piotr Rozycki