Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
Kai Klaas
12/05/04 10:18
Read: 459 times
Germany


 
Msg Score: +1
 +1 Informative
#82560 - Highcap ceramics against paralleling
Responding to: Joseph Hebert's previous message
Joseph said:
For really broadband protection, it is a good idea to use two or three caps in parallel, each performing best in a different range.

Unfortunately, this paralleling of decoupling capacitors is very likely to result in unwanted resonances, which can cause relevant impedance peaks: The equivalent series inductance of a larger capacitor reacts with the equivalent series capacitance of a smaller capacitor. The resonance is the more sharp the more the equivalent series inductance of involved capacitors differ and the lower the equivalent series resistance of the larger capacitor is. Unsuited is, for instance, a combination of big fat tantal (10µF, leaded) in combination with tiny SMD ceramic capacitor (10nF, X7R, 0603). A better combination, which prevents resonance is 10µF/16V electrolytic in combination with 100nF, X7R, 0805. Here the rather high equivalent series resistance of electrolytic very effectively dampens any resonance. Unfortunately, at the same moment this rather high series resistance of 10µF/16V electrolytic degrades the effect of decoupling. So, this combination might be good enough for an OPamp circuit, but for a microcontroller circuit something better is needed, which shows very low impedance at higher and at lower frequencies, free of impedance peaks.

Since some years highcap ceramics are available, which offer an equivalent series resistance of about 0.01Ohm in a very broad range of frequencies. I would recommend the use of 4.7µF/10V/X5R in 0805 case or 1µF/10V/X5R in 0603 case. By this you have both, the much larger capacitance compared to 10...100nF ceramics and the very low equivalent series inductance of SMD case. And this combined with an equivalent series resistance of 0.01Ohm and the absence of any impedance peak. These highcap ceramics provide an outstanding decoupling performance!

Kai

List of 22 messages in thread
TopicAuthorDate
EMI/RFI noise      Daniel Hertanu      11/28/04 15:52      
   Keep looking      Steve M. Taylor      11/28/04 16:19      
   Brownout?      Andy Neil      11/28/04 19:00      
      It is Brownout      Prahlad J. Purohit      11/28/04 19:56      
         brownout protection        Erik Malund      11/29/04 15:40      
   EMI/ RFI solutions.      Prahlad J. Purohit      11/28/04 20:02      
   Archives      James Krushlucki      11/28/04 22:08      
      0.1uF      Steve M. Taylor      11/29/04 03:56      
         Caps      Joseph Hebert      11/29/04 14:50      
            Highcap ceramics against paralleling        Kai Klaas      12/05/04 10:18      
               Highcap Ceramic.      Prahlad J. Purohit      12/05/04 11:24      
                  Thanks!      Kai Klaas      12/05/04 11:28      
         .01uf      James Krushlucki      11/30/04 23:07      
      decoupling      Erik Malund      11/29/04 15:42      
   EMI problem solution.      Alberto Geraci      11/29/04 15:55      
      is it ?      Erik Malund      11/29/04 16:00      
         That's all      Alberto Geraci      11/29/04 18:27      
            to Alberto      Erik Malund      11/30/04 07:06      
               To Erik      Alberto Geraci      11/30/04 19:41      
   Electrically Noisy Environments      Andy Neil      11/29/04 17:02      
   brownout - no, everything else - maybe      Daniel Hertanu      12/01/04 11:30      
   Problem solved      Daniel Hertanu      12/04/04 05:21      

Back to Subject List