×

SparkFun will be closed May 25, 2015 for Memorial Day. Orders placed after 2pm on Friday the 22nd will ship out on Tuesday. Thanks!

Mahdi Olfat

Member Since: December 17, 2009

Country: United States

  • Now where was that “How to Free Yourself From Your Master’s Leash” tutorial!? (Open-Source FTW)

  • And that kids, is how we do it at SparkFun!

  • It’s all good.

    I was really addressing whoever was reading the comment. By “you”, I didn’t really mean you :). Sorry for not checking my post before submitting it.

  • 1uF is fine. But chances are, if you put it there and take it out, you won’t see a difference. The link @trophosphere recommended is fine.

    The easiest way I can explain to you how the values are chosen is this:

    By characterizing the “system” (your circuit, filters, decoupling caps) at hand, you’ll end up with a capacitor impedance vs. frequency plot. You may then look at that plot, and point to a frequency range (bandwidth), and determine that a particular capacitor with certain capacitance would block (not pass) those frequencies.

    The reason that you were “recommended” decoupling capacitor values instead of being asked to characterize you circuit, is that at low frequencies (low speed designs or DC power lines), those general values (100nF, 1uF, 4.7uF, and 10uF) work relatively fine.

    If you really really want to learn what’s going on, check this out (no spaces): www . speedingedge . com / rtft_book . htm

  • @Bunk, do not use a decoupling capacitor on you signal line. What are you trying to achieve with decoupling caps here? is your circuit not working? Does it have a lot of noise on it that you are observing?

    Let me know, perhaps I can help.

  • Hmm. @TimCole, sorry but your post is very misleading. You are describing high/low pass filters. Ideal capacitors alone are NOT high/low pass filters. There are other parameters that let the decoupling caps to act as such (i.e. their form factors or packages - which is somewhat useless if you are using through-hole packages).

    Also, DO NOT USE CAPACITORS ON SIGNAL LINES (i.e. capacitor from signal to GND) unless EXPLICITLY demanded by the datasheet. Why would you wanna voluntarily introduce capacitance on a signal line? Capacitance will most likely kill the digital logic your trying to transmit (mess up the falling/rising edges). For example, capacitance on I2C signal line (SDA) would kill communication. And guess what limits the number of devices on I2C? Capacitance.

    100nF, 1uF, 4.7uF, and 10uF decoupling capacitors are used frequently between VDD (IC power) and VSS (IC GND) and are placed as close to the power pins as possible (if you’re designing a PCB, USE SMD caps, the smaller the better, 0402 or 0603 packages are recommended). But, if your circuit is working, and the datasheet has not explicitly recommended any decoupling caps, just do not use them (especially on breadboards and simple low HZ 2-ly PCBs).

    Lastly, I can tell you from experience that these “rules of thumb” are almost always wrong. If you want to use decoupling caps, follow the datasheet. If you want to know exactly what’s happening with decoupling caps, take some time to study the underlying principles.

  • I won after 3 tries :)

No public wish lists :(