No Script, No Fear, All Opinion
RSS icon Home icon
  • EEVblog #626 – Ceramic Capacitor Voltage Dependency

    Posted on June 7th, 2014 EEVblog 19 comments


    Dave explains, shows, and measures a potentially big trap with using high value ceramic capacitors.
    If your 10uF capacitor really 10uF in your circuit?
    Those humble X7R caps you think are a “stable” dielectric? think again…

    Class II and above ceramic capacitors can vary their capacitance drastically with DC bias voltage level and also the applied AC voltage.
    Links:
    http://www.murataamericas.com/murata/murata.nsf/promo_dcbias.pdf
    http://www.avx.com/docs/masterpubs/mccc.pdf
    http://www.ece.ucdavis.edu/vcl/asap/asap_v1/docs/X7R_C.pdf
    http://psearch.murata.com/capacitor/product/GRM21BR60J106KE19%23.html
    http://www.murata.com/products/design_support/simsurfing/outline.html
    http://www.avx.com/SpiApps
    Capacitor Tutorial – Ceramics & Impedance
    The piezoelectric effect demonstrated

    Forum HERE

    Be Sociable, Share!
    • ac

      Need part 2 with some standard vs “hi-fi/audio” caps. Gigabyte is making pc atx motherboards with audio-electrolytics, they used to use those polymer or whatever earlier (and all-electrolytics before going all polymer).

    • Raphael

      Wow, that was deeply impressive! I always thougt that the X5R/X7R were halfway decent caps compared to the crappy Z5U/Y5V types in terms of cap. stability over voltage, but now…. Guess I have to think about them in a different way now.

      Thanks a lot for this video Dave, keep up the great work!

      PS: One thing that’s been bugging me for a long time: Why don’t they print the value on ceramic smd caps???

    • Paul

      Now this was really interesting, and seeing you work the digital scope with ease, I’d have to have the manual out a few times to get that far.

    • Michael

      You forgot the 50 Ohm of the Gen.

      So 10.6ms/1.05kOhm = 10.095uF

      Bye

      Michael

      • Chris

        Thats a good observation actually. Dave, whats the explanation here?

        • poot36

          That 50 ohms was for channel 3 of the scope I believe not channel 1 and 2 that he was using.

          • Chris

            This is true, so the 50 ohms should not effect it. Better observation.

            • Michael

              Not the scope is the problem.
              The problem is the output impedanze of the function generator.

    • Jay

      Dave, thanks so much for this. I started using MLCC capacitors just recently, and was unaware of the issue.

      MLCC capacitor technology is relatively new and developing rapidly, and the methods used to put 10 uF or more in such a small package are very high-tech as compared to simple ceramic disc capacitors!

    • http://www.geppettoelectronics.com Nick

      Is the solution (or at least one) to simply dramatically over-spec the voltage rating of your caps?

      • Niklas

        Yes, uprating the voltage can help. The capacitance drop increases as the applied voltage is approaching the rated voltage. Also remember to do the thermal derating if the capacitors are operated at 60′C or higher as pointed out on page 2 in this catalog from Murata:
        http://www.murata.com/products/catalog/pdf/c02e.pdf

    • Marco

      I did see you “outtake” and maybe you are right, maybe you have to clean the whole lab, say 1x a year. Cleaning, dusting and use a vacuumcleaner, who knows how many “lost” objects you wil find.

    • Bearman

      Great vid. Very interesting results.

      Thumbs up.

      B

    • Neganur

      just a side note, at 07:13 min you meant to say 1 kHz +/- 0.1 kHz :)

    • pinkman

      This is good info, keep in mind that you all should already know this from physics class and any EE classes you have taken if you’re in an EE program. It is sad that most people nowadays are taking less and less away from these programs, I have encountered engineers who have been in the field for a few years or more and are still somehow unaware of even basic concepts like this. Thanks to Dave for shining the spotlight on this important topic, you can not design with ceramic caps without this basic knowledge permanently programmed into your brain. Do not forget it!

    • Nic

      You can’t blame the education.. these fancy dialectrics and chip capacitors didn’t exist 20 years ago when I got my degree.
      As Dave says, it’s a trap for older as well as younger players..
      Hmm.. I also want to check back for facts about the lemonade diet.. :-)

    • Jimmy

      Constant current charge would of illustrated the response better then naff RC

    • euzer

      To aid measurements on the Agilent scope change the cursor mode to “track measuremnt” so you don’t have to change both the X & Y. Alternatively increase the mark time of the sig gen to let the capacitor charge some more and then use Quick Measurement Rise Time, which defaults to 10% 90% thresholds which is a rise time of 2.2RC. You could manually set the thresholds to give you a value of tau directly.

    • caspencer

      I wonder, is this voltage dependency more of any issue with higher value (> 1~10uF) caps or do all caps exhibit this behavior?