Shop OBEX P1 Docs P2 Docs Learn Events
Choosing and using bypass Capacitors — Parallax Forums

Choosing and using bypass Capacitors

john_sjohn_s Posts: 369
edited 2013-04-21 14:19 in General Discussion
I found this worth to share with all http://www.intersil.com/content/dam/Intersil/documents/an13/an1325.pdf
a practical and well documented!

Comments

  • davejamesdavejames Posts: 4,047
    edited 2013-04-19 10:17
    ...I agree - have used that reference many times!
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2013-04-19 15:28
    I recently came by a rigol dsa815 with tracking generator, and whipped up a test board to try some of my bypass caps. I was surprised to see that the .1uf and 100pf I had paralleled actually caused a dead spot, rather than complementing each other as some other combinations did.
  • davejamesdavejames Posts: 4,047
    edited 2013-04-19 21:30
    ...please 'splain "dead spot"? :confused:
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2013-04-20 18:29
    Test setup is tracking generator connected straight through stripline with caps soldered onto stripline. Obviously the lower the db the more effective.

    Email notifications don't seem to be working.

    0.1 uf
    Cap01uf.png


    0.1uf + 100pf , note 'dead' spot at 380mhz
    cap01uf100pf.png


    100pf
    cap100pf.png


    4700pf
    cap4700pf.png


    0.1uf + 4700pf
    cap4700pf01uf.png
    Cap01uf.png
    800 x 480 - 33K
    cap01uf100pf.png
    800 x 480 - 33K
    cap100pf.png
    800 x 480 - 32K
    cap4700pf.png
    800 x 480 - 33K
    cap4700pf01uf.png
    800 x 480 - 33K
  • davejamesdavejames Posts: 4,047
    edited 2013-04-20 19:50
    ...another question - what were you expecting to occur with the aforementioned capacitor combinations?

    And yes, the email notification issue has been noted by a bunch o' people.
  • ercoerco Posts: 20,256
    edited 2013-04-20 20:10
    davejames wrote: »
    And yes, the email notification issue has been noted by a bunch o' people.

    As others have also noted, I get the "Are you sure you want to leave this page" dialog box at every post.

    Edit: Including this one! :)
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2013-04-21 13:08
    I was expecting the 100pf to complement the .1uf, but not to hurt the curve in part where the .1uf is strong. What exactly would you expect in this situation? Can you explain the rise at 382 mhz as shown? I would say its unwanted resonances perhaps similar to this http://licn.typepad.com/my_weblog/2011/01/bypass-capacitor-boobytrap-john-dunn-consultant-ambertec-pe-pc.html ?? Because neither cap on its own is as bad at 382mhz as paralleled.

    I had a older engineer tell me something like this to watch for about paralleled caps, and it went in one ear and out the other, till I decided to try my cap set just to see what would happen.
  • davejamesdavejames Posts: 4,047
    edited 2013-04-21 14:19
    ...good article - thanks.

    I like the last line; sums things up nicely:

    "There is no escape from this effect. Just check to be sure that your design's unwanted resonance doesn't occur at a frequency critical to your circuit(s) operation."
Sign In or Register to comment.