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.

LM5006 output capacitor

Other Parts Discussed in Thread: LM5006

I see on the LM5006 eval board, Vout is set to 5V, and the output capacitor (C2) is 15uF, 16V, X7R.  I have used this exact part in my design and it is working well.  However, I am hoping to replace this capacitor (which is 2.8mm high) with one that is a bit shorter.  So, a couple questions -

Is the 16V rating required for any reason?  Given the 5V output, I would expect plenty of margin with a 10V capacitor?

Any concerns about a X5R, X6S, or X7S capacitor?  I seem to be able to get a much shorter capacitor in these dielectrics instead of X7R.

Thanks,

Dave

  • There are a couple things to consider.  First is dc bias voltage.  Nominal capacitance is based on 0 V operation.  As dc voltage is applied to a capacitor the actual capacitance will vary.  In most cases it will decrease rapidly as the applied voltage approaches the rated voltage.  So sometimes we will use a higher voltage rating to maintain capacitance, such as a 16 V capacitor in a 5 V circuit.

    Dielectric also plays a part in capacitance over temperature.  So dielectric choice will depend on your temperature requirements.  First letter is lower temp range, number is higher temp range, last letter is tolerance over temperature.  For example X7R is -55 to +125, +/-15%...X6S is -55 to +105, +/- 22%.  So you can pick the dielectric nbased on your needs.  I would not use any Y5V for example.

  • Hi John,

      Very interesting.  I've also seen that package size can make a difference in how the part reacts to a DC bias.

      So, I took the recommended capacitor (15uF, 16V, X7R, 1210) and plugged it into the TDK capacitor tool.  I see that at 5V it loses 15% of its nominal value, making it a 12.75uF capacitor. 

      Next, I found another capacitor - 22uF, 16V, X5R, 1206, and the TDK tool showed that at 5V, it loses 40% of its value, but that still leaves me with 13.2uF.

      There are many other combinations I could try that will give me different capacitances around my operating point.  I also found another part that was even closer to the original - 22uF, 16V, X5R, 1210 - but it's not from TDK so I can't use their tool to compare.

      I'm not terribly concerned about moving from X7R to X5R as my application doesn't go above 85C. 

      I suppose the big question is - if I select a different capacitor, as long as the resulting capacitance is about the same at my operating point, can I safely use it as a replacement?  And, once I've selected a replacement, what is the best way to test it?  The output capacitor's main function is to reduce ripple voltage - so if the ripple voltage is acceptable with the new part, is there anything else to worry about?

    Thanks,

    Dave

  • My comments were in general terms about ceramic capacitor usage.  I do not directly support LM5006 and have never used it personally.  But the LM5006 uses constant on time control.  It requires a certain amount of ripple voltage at teh feedback pin.  Usually that ripple voltage is generated by the ac ripple current in the inductor flowing thru the ESR of the output capacitor.  So you will need to pay attention to that in your design.  I think the datasheet has enough detail to allow you to select the proper output filter components for your design.