• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 55 subscribers
  • Views 8219 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.

Low-Leakage Capacitors for Coin Cell Products

Charles Gervasi
Expert 2280 points

This TI White Paper (SWRA349) deals with coin cells and peak current draw.  The upshot of the article is bursts of high current draw reduce coin cell life but not by all that much and a capacitor across the battery mitigates this.

What is a good cap type to use that has reasonably low leakage?  Capacitor leakage undoes some of the benefit of protecting the battery from current spikes.

  • 0
    TI__Expert 4060 points

    Hi Charles,

    We have been using SMD ceramic capacitors of X5R or X7R type (also in order to get the maximum capacitance in a minimum area), and these seem to be quite OK when it comes to leakage. One specific type I looked at specified an insulation resistance of 500 MOhm*uF, which for a 47 uF cap is equivalent to a leakage current of a few tenths of a uA at 3V (in other words, it shouldn't really affect the total sleep current on its own). Of course, if you use a huge one, you might have problems, but we have generally been looking at values between 10-100 uF, also because of physical size concerns, and these should still have quite low leakage currents.

    The temperature variation of capacitance value that is more pronounced in X5R/X7R compared to NP0 or other dielectrics doesn't really matter here, neither does microphony or other effects, as we are using the capacitor for energy storage rather than as a tuned element. Basically, this is a classic decoupling use-case, so the important thing is to have a big enough capacitance, not so much what the exact value is.

     Best regards,

    Karl

  • 0 in reply to Karl T.
    Expert 2280 points

    Karl T. said:

    One specific type I looked at specified an insulation resistance of 500 MOhm*uF, which for a 47 uF cap is equivalent to a leakage current of a few tenths of a uA at 3V (in other words, it shouldn't really affect the total sleep current on its own).

    Can you post a link to one or more of the caps you considered with low insulation resistance?  Thanks, CJ

  • 0 in reply to Charles Gervasi
    TI__Expert 4060 points

    Hi,

    Here's one example: http://www.avx.com/docs/catalogs/cx5r.pdf. They quote 500 MOhm*uF, and have a 47 uF cap with a voltage rating of 6.3V in 1210 package. I guess the isolation resistance should be high, not low :-)

    Best regards,

    Karl

  • 0 in reply to Karl T.
    Expert 2180 points

    Karl,

    We are also facing button cell question as Charles. The AVX cx5r part you recommended is of 1210 package, which is actually quite large (1210 = 3.2mm x 2.5mm) which is comparable to half the footprint of a CCxxxx chip.

    After doing some online search we collected the minimum packaging for MLCC (Multi-Layer Ceramic Capacitor) SMD capacitors. The left-most column showing increasing capacitance, and the corresponding row shows smallest size capacitor that could offer the capacitance.

    µF

    MLCC SMD Size (X5R or X7R)

     

     

     

    × 10-6

     

     

     

     

    × 10-3

     

     

     

     

    0.1

    0402

     

     

     

    0.47

    0402

     

     

     

    1.0

    0402

     

     

     

    2.2

    0402

     

     

     

    4.7

    0402

     

     

     

    10

     

    0603

     

     

    15

     

    0603

     

     

    22

     

    0603

     

    33

     

     

    0805

     

    47

     

     

    0805

     

    68

     

     

     

     

    100

     

     

     

    1206

     

    In your first post you mentioned capacitance between 10 and 100uF. For the 47uF AVX cx5r in your second post it seems that 0805 size is sufficient. More aggressively if a 22uF capacitance is used, then 0603 capacitor would be enough.

    Have you ever tested such small size capacitors on any product, and how would the performance be?

     

     

    Matt

  • 0 in reply to Matt Shira
    Prodigy 80 points

    Did anyone actually use any capacitor and conclude. I am facing the same issue. Need some help on deciding the capacitance value across battery (CR2032).

    Thank you.

  • 0 in reply to archana jabade
    Expert 2280 points

    I gave up and went to a bigger battery.  I found I couldn't get nearly the rated charge out of a 2032 b/c the resistance rose during its life causing the voltage to drop during high-current operations.  This is obviously a known problem, but I had the option of going to a larger battery, so I abandoned button cells.

  • 0 in reply to Charles Gervasi
    Prodigy 80 points

    I have a size limitation and can not go beyond 2032. Anyway I will post a new query for this. Thank you Charles for your feedback.

     

    Regards,

    Archana.

  • 0 in reply to Matt Shira
    Intellectual 310 points
    MLCCs do indeed have good leakage, but you should keep in mind capacitance degradation due to DC bias. This is a spec not widely advertised by manufacturers - only Murata provides full curves for their line AFAIK. Capacitance dependence on DC bias is directly related to package size, and to a lesser extent voltage rating. A 22UF 6.3V 0603 will give you around 8UF with 3V across it; an 0805 will yield 12UF, and a 10V 1210 package 20UF.
  • 0 in reply to archana jabade
    Mastermind 9135 points
    It is also important to place the cap at the right position.