• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 64 subscribers
  • Views 1177 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.

TPS63000 Output Oscillation Problem

Mark Williams
Intellectual 510 points
Other Parts Discussed in Thread: TPS63000

Hi,

I have a design that uses a couple of TPS63000 devices to create a 3.3V and 5V power supply from a Lithium-Ion battery (and a TI charger/power management IC). I initially experimented with the evaluation board and built several prototypes with great success.

Our first batch of production product arrived the other day but I instantly noticed a problem. After a bit of investigation the 5V power supply output was oscillating - I had approx 2V peak to peak signal sitting around 4.5V at about 96kHz. Three boards were tested at random from the production batch with the same issue.

While the power supply was oscillating the input current was significant (even with no output load) at around 200mA and the device got warm to the touch.

The normal circuit load on the output of this power supply is about 100mA. I tried disconnecting the load, but the oscillation was still present, just slightly less ripple.

The layout follows the development board (I can upload if necessary) and uses a 10uF 0805 ceramic on the input (10V, X5R), a 3.3uF inductor, an identical 10uF 0805 ceramic capacitor on the output with 100n 0603 in parallel.

I slowly migrated a component at a time over to the development board starting with the feedback network, then the inductor, then the output capacitors then suddenly the oscillation occurred on the output of the dev board (with no load or 100mA load present, at 3-4V input).

The 0805 10uF 10V ceramic output capacitor (I can get the exact part details when I am back in the office tomorrow) appears to be causing the problem. What is interesting is that there is no oscillation at 2V and 5V input, and as you ramp the input voltage over this range the oscillation builds up then diminishes.

I have a solution, I tried two 10uF 0603 6.3V capacitors on the output (as per the dev board) and have not seen the oscillation since, over the full input range and with no load or 400mA load, and with load transients.

However, I would like to know what could have caused this issue so I can avoid this in future? I used a 10uF ceramic and regardless of load I got nasty oscillation on the ouput - even on the development board supplied by TI?!

For the record, I also tried adding the feedforward cap in parallel with R1 on the feeback network but this had very little effect.

Just on the point of the output capacitor selection, could someone explain the equation in the datasheet - what values do I use for uF/uH?

Thanks in advance,

Mark

  • 0
    TI__Expert 4525 points

    Hi Mark,

    Equation 7 in the TPS63000 datasheet is meant to say that output capacitance should be at least 5 times the value of the inductor.  The "uF/uH" is simply a unit conversion.

    So, in your example:

    5  x  3.3uH  x  uF/uH  =  16.5uF

    The minimum recommended output capacitance is 16.5 uF (for a 3.3uH inductor).  There are two solutions:

    1) Use a smaller inductor (the 2.2uH on the EVM will work fine)

    and/or

    2) Increase the output capacitance to greater than 16.5 uF

    Also, in the capacitor datasheet you should see a curve that shows effective capacitance vs. DC Bias Voltage.  The effective capacitance will decrease as the voltage across the capacitor increases.  A 10uF capacitor is only 10uF at 0 Volts.  Depending on the size and type of capacitor, there will be some decrease in effective capacitance as you increase voltage.

    Please let me know if this helps.

    --Stephen

  • 0
    TI__Expert 4525 points

    Here is an 0805, X5R, 10uF, 10V capacitor.  You can see that at 5V the capacitance is down to ~60% of the original value.

  • 0 in reply to Stephen Nortman
    Intellectual 510 points

    Stephen,

    Thank you very much for such a quick and detailed response! I think if I had understood the equation in the datasheet I would have selected the correct capacitor! I suppose the following statement from the datasheet sums it up well:

    "A capacitor with a value in the range of the calculated minimum should be used. This is required to maintain control loop stability. There are no additional requirements regarding minimum ESR."

    Out of interest I added another of the same capacitor in parallel to the output but it only halved the ripple. It is very interesting to see how the capacitance is affected by DC bias, I am going to look further into this to try and understand why (this link has a scientific explanation if anyone is interested http://www.murata.com/cap/faq/faq04.pdf).

    The capacitor I was using was a Taiyo Yuden part (LMK212BJ106KG-T). Interestingly they have a plot for the capacitance change with frequency, but not applied DC voltage:

    http://www.yuden.co.jp/ut/product/capacitor/LMK212BJ106KG-T.html

    I am now using two 0603 6.3V 10uF X5R capacitors in parallel, as detailed in my original post, with success. This is an easy and cheap change to make to the boards we have in stock.

    Mark

     

     

  • 0 in reply to Mark Williams
    Prodigy 10 points

    Hi,

    What's the part number of your new capacitor please ?

    Mathieu