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.

TPS61045 noise / spikes

Other Parts Discussed in Thread: TPS61045

Hi,

I do use the TPS61045 in a circuit.

Input from 3 to 4 V aprox. Output 8 or 15V.

When the output is set to 15V, the output is good.

When using 8V, some devices have a correct signal and some of them have some spikes that generate problems in some measurements.

Can anyone give a clue of what could produce this behavior is some devices and not in others? Any clue of which part could introduce this noise?


Thanks

  • Hi,

    What is the maximum load @ Vo=8V condition?  What is the main inductor value and the output capacitance ? Could you please paste your schematic here?

    BR,

    Helen

  • Hi Elen,

    See the schematic. It is based on the sample circuit.

    We use the inductor model: (4,7uH)
    LQH32CN4R7M33.

    The load current is constant around 55 uA.

    Revising the circuit, I saw that the calculation of the frequency was higher than the maximum value, so I did a test with the following inductor:

    LQH32CN100K33L of 10uH.


    The results were much worse introducing a sinewave of 110 Hz aprox. See picture.

    Best regards

    Guillem

     

  • Hi Elen,

    Sorry, I did a mistake in the data I gave to you. I was calculating the current at 3V input. The output current at 15V is around 10mA

    Guillem
  • Guillem,

    4.7uH inductor is OK in your usage.

    Is the total output ripple =160mV in your application?

    I suggest you to add one more 2.2uF or 4.7uF ceramic cap at the output side. It will be helpful.

    BR, Helen

  • Hi Helen,


    I attach the complete view of the circuit. I have been testing and if I remove L3 (I shortcircuit it) then the circuit works properly. Changing the value L3 to 4,7uH or 10uH doesn't help.

    I have also tested to increase the value of C42 to 2uF but the results are not good. Do you think that the inductor at the output is a problem?


    Regarding your question, yes, the ripple when the circuit works bad is 160mV. When it works without spikes is around 10mV.

    I also noticed that the error occurs when the battery is fully charged (4,2V input), after 1hour (voltage < 4V aprox) it disappears.


    Do you have any ideas?


    Thanks!

  • Hi Mitja,

    If remove the L3, let C8 directly in parallel with C42, then the total output cap is about 48uF. The cross over frequency will getting  much lower, and phase margin will be higher in this condition. So it should be more stable. 


    The highest cross over frequency happended at maximum input voltage, lowest output voltage condition, this is why at 4.2Vin, 8Vo, you found it unstable.


    So the easiest solution is to short the L3 in your application.

    BR,Helen

  • Hi Helen,


    Thank you for your answer.

    Just a last question: the L3 was there for the filtering of the signal. Removing L3 makes the supply a bit noisier than before.

    If I want to put a similar filter, shall I put 47uF at the boost output, then L3 and then again the 47uF?


    Best regards

    Guillem

  • I think you can put 2pcs 22uF/25V - 10uF/35V ceramic cap at the output side, these ceramic should be very close to the output side, the 47uF Aluminum electrical cap should be at the load side, then the ripple will be much better.

    Only the capacitance of the ceramic can help for the ripple. If the stibility is OK with 2pcs 25uF ceramic cap , then I think the 47uF cap can be also deleted.

    BR,Helen