• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 62 subscribers
  • Views 442 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LM3478: Start up problems

Julian Worskett
Intellectual 255 points
Part Number: LM3478

I am sorry if this has already been asked, but I cant find it.

I am converting 5V USB to 10Vdc at 1.5 Amps for charging 2 lithium cells, but i am using a resistive load for testing at present. All is fine if I turn on the LM3478 with no load. However, if I power on with a load, the lab power supply goes into current limit (3 Amps). If I remove the load, then reconnect, all is fine, about 2.2 Amps.

Circuit as web bench with 220uF AVX TPS series tants at input and output, all other components as on the BOM.

I used a p channel mosfet on the output as a delayed switch, 3478 ramps up, wait 100 ms, connect load. That works most of the time, but it depends on how the USB is plugged in, so any noise due to the action of pushing in the plug sometimes causes it to draw too much current. If the USB is unplugged, then plugged in quickly, overcurrent can happen.

I also read somewhere about using an RC filter on the pin8 supply to the chip. I used 47 Ohm and 10uF but this doesn't help, any more than 47 Ohm and i can see the supply failing.

Any ideas please?

  • 0
    TI__Guru 65450 points

    Hello Julian,

    Would you please check the input voltage when connecting to the 5V supply? It is possible that the high current during startup causes a voltage drop on the input. This increases the input current even further.

  • 0 in reply to Brigitte
    Intellectual 255 points

    Thanks Brigitte,

    I had quite long (2ft) 7/0.2 wires to my power supply. Reducing the length and increasing the wire thickness to 32/0.2 has definitely made a difference, so thanks for that.

    My worry now is that the surge capability of USB hubs etc may vary and they may not supply the transient current. If I reduce the load to 1 Amp, that also increases the reliability at power up, so I may reduce the charge current on my LT4002-8.4 charger and the customer will have to accept that charging time on USB will be increased over a 12V wall block.

    It's a shame that we can't externally control the soft start time of this chip. I was wondering whether i could put a capacitor on the output voltage feedback pot to start it at a low output voltage and ramp up. However, there is still the problem of a noisy USB plug i.e that the input voltage drops in spikes as the connector is pushed in, too fast for my load switch to disconnect.

    Anyway, thanks for your help on this. I think it is resolved.

    Julian.