Author Topic: Parellel Capacitors Question  (Read 3842 times)

0 Members and 1 Guest are viewing this topic.

Offline crucialfixTopic starter

  • Newbie
  • Posts: 2
Parellel Capacitors Question
« on: October 16, 2014, 11:04:52 am »
Quick question regarding my first power supply build. I am following this schematic here:


There are 2 parallel capacitors on the input and output of the LM317. My question is why the pairs?

I understand the purpose of the capacitors in those particular positions and that capacitors in parallel effectively increase the area of the plates, but why does the designer opt for 2 in parallel vs 1 large cap? In his tutorial all he mentions is that C2 and C5 is optional.

Thanks
 

Offline Whales

  • Super Contributor
  • ***
  • Posts: 1899
  • Country: au
    • Halestrom
Re: Parellel Capacitors Question
« Reply #1 on: October 16, 2014, 11:33:40 am »
Capacitors are not perfect -- they contain an internal resistance amongst many other things.  Placing two in parallel tends to provide better performance than one large equivalent value.  You can easily get away with just one in almost all cases, but it's good practice to use more if you can.

EDIT: There's also the possibility multiple capacitors exhibit less oscillator-like symptoms with other R and L in the circuit, but I'm assuming that's tiny.  Can anyone else comment here?
« Last Edit: October 16, 2014, 11:36:42 am by Whales »
 

Offline RobertHolcombe

  • Frequent Contributor
  • **
  • Posts: 275
  • Country: au
Re: Parellel Capacitors Question
« Reply #2 on: October 16, 2014, 11:39:06 am »
There will be both a large and small value cap on both the input and output. They are serving the same purpose however the larger value cap will be better at filtering low frequency ripple/mains noise/changes in load, where the smaller value cap will filter out higher frequency noise.
 

Offline crucialfixTopic starter

  • Newbie
  • Posts: 2
Re: Parellel Capacitors Question
« Reply #3 on: October 16, 2014, 12:15:01 pm »
Awesome info guys. Thanks!
 

Offline dentaku

  • Frequent Contributor
  • **
  • Posts: 881
  • Country: ca
Re: Parellel Capacitors Question
« Reply #4 on: October 16, 2014, 04:22:45 pm »
By the way, did the person who drew that schematic specify values for C1 and C2?

I recently built a power supply and C1 is 4700uF electrolytic and C2 is a little ceramic 0.1uF X7R.
 

Offline wraper

  • Supporter
  • ****
  • Posts: 16864
  • Country: lv
Re: Parellel Capacitors Question
« Reply #5 on: October 16, 2014, 04:29:28 pm »
By the way, Not connected pot terminal should be connected to gnd for better reliability. As potentiometer contact is not ideal, when it becomes open, regulator output will shoot up to almost the input voltage. Especially if the pot is an old one. Connecting this terminal to gnd won't affect normal operation at all, but will limit that negative effect to some extent.
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: Parellel Capacitors Question
« Reply #6 on: October 16, 2014, 04:34:45 pm »
Yes, I was just about to post the same exact warning. Connect the other end of the pot to ground.

By the way, Not connected pot terminal should be connected to gnd for better reliability. As potentiometer contact is not ideal, when it becomes open, regulator output will shoot up to almost the input voltage. Especially if the pot is an old one. Connecting this terminal to gnd won't affect normal operation at all, but will limit that negative effect to some extent.
 

Offline mrkev

  • Regular Contributor
  • *
  • Posts: 225
  • Country: cz
Re: Parellel Capacitors Question
« Reply #7 on: October 16, 2014, 05:44:39 pm »
As it was already said. The large cap is better for low frequencies (duh, because it is large). But electrolythic caps have some inductance and also the actual dielectric inside is not any good for frequencies above few kHz (physical properties).
So when you have that elect. cap at frequency of some MHz, you could even find that it's not a cap anymore and it behaves as a coil... On the other hand, f.e. small ceramic cap is pretty good far beond that point. So the combination of two (one large, second small but effective at at high freq.) can cover much bigger bandwidth.
Just a hint - Always learn the actuall flaws of components you are using. In the real world, many things won't behave as you would expect them to, because of their parasitic properties...
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16615
  • Country: us
  • DavidH
Re: Parellel Capacitors Question
« Reply #8 on: October 17, 2014, 06:26:20 am »
For a regulator like the LM317, the small capacitors are needed if the input and output capacitors are removed some distance from the regulator and this is especially the case on the input side.  In most applications they may be left out because the bulk capacitors are close enough but there is nothing wrong with adding them even if they are not needed and they will improve the high frequency performance.  Small 10uF aluminum electrolytic capacitors or smaller solid tantalum capacitors may be used in place of the 0.1uF ceramics but the ceramics are probably the least expensive.

As pointed out above with how different types of capacitors perform, when they show a 100uF in parallel with a 0.1uF, it is not to make a 100.1uF capacitor.  Sometimes multiple capacitors of the same value are placed in paralleled to lower their ESR, ESL, and/or to achieve a higher ripple current rating but that is a different situation.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf