Go Down

Topic: Another decoupling question...? (Read 5290 times) previous topic - next topic

xl97

I got to thinking about one comes to the 'values' of caps used for decoupling their power lines/rails..

read some threads.. and came across this:
"That is why you have to use both a small cap 10 to 100 nF in parallel with a big one 10 to 100uF."

I went back and looked at a schematic I made (and posted for review).. and noone seemed to point that out to me on this project??
How crucial is it to have TWO caps in parallel on the regulated lines from the vReg?

(and Im not thinking that 22uF is WAY to big to have a cap on the input of the vReg lines?)

So for this diagram/schematic.. 

1.) Is only having the 1 cap (100uF) after the regulator 'ok'? ...single cap? (doesnt match the statement made above though?) (I just ordered the PCB's too!)  LOL
2.) Im thinking 22uF is way to big of a value to use?  (C1/C2)  should I knock it down?







Next.. I have this schematic..(and already made boards..etc..working..etc)



**it 'does' have 2 x caps in parallel..(like the statement above says to do)...

**and it uses a WAY lower value for the cap in front of the vRegulator

will the first project suffer much?


thanks

MarkT

Depends on what the datasheet for the actual regulator says.  Most will work with 1uF on input and 10uF on the output, but best to check what the datasheet says, _especially_ for low-drop-out regulators (much more fussy).
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

fungus


How crucial is it to have TWO caps in parallel on the regulated lines from the vReg?


Large capacity (eg. electrolytic) capacitors are slow to react. You need to add a small/fast one (eg. ceramic) to take the strain while the big one is starting up.

That's why you need two...

Advanced Arduino

MarkT

The regulator normally doesn't care, the fast decoupling needs to be near the logic chips themselves.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

oric_dan

#4
Nov 06, 2012, 08:06 pm Last Edit: Nov 06, 2012, 08:13 pm by oric_dan(333) Reason: 1
Actually, bypass and electrolytic capacitors filter noise in different frequency ranges, high and low,
respectively.

Although the capacitive reactance formula, Xc = 1/(2*pi*f*C), indicates that the impedance of a
cap should be even smaller at high-frequencies than at low-frequencies, for real-world electrolytics,
this is not the case [real-world always varies from the ideal]. See curve here,

http://www.passivecomponentmagazine.com/conductive-polymer-aluminum-for-electrolytic-capacitor-technology/
http://electronics.stackexchange.com/questions/7993/caps-on-h-bridge-seem-a-bit-redundant

The bypass cap works at those higher frequencies where the electrolytic cap impedance goes
back up.

dhenry

Quote
Actually, bypass and electrolytic capacitors filter noise in different frequency ranges, high and low,
respectively.


Those are the same reasons.

oric_dan

Quote
Those are the same reasons.


If you're gonna comment, why don't you explain what you mean, so we don't have to read your
mind.

1. same as what others have said [I don't think so].
2. bypass same as electrolytic [not hardly].
3. high frequency same as low frequency [doh!].

????

dhenry

Quote
I don't think so


Think harder then.

xl97

thanks for the replies..   

Im going to try and dumb this down.. so "I" can understand it!  lol.. (since Im the OP and the one looking for help here)

so in the first diagram..

that is 'faulty'?... and will not work as is? or just be 'slower'  (slower for what? to power up?)

(yikes.. like I said I already ordered the pcbs)  =(

but if I need to make changes and re-order...I'd guess I'll have to do that.

Should I just mimic the layout/schematic like in the second image then?...


thanks!


oric_dan

Quote
Think harder then.


Aha. Well, the critical thing about the 2 types of caps is the Xc equation and the frequency
curve I pointed to.

oric_dan

Quote
so in the first diagram..

that is 'faulty'?... and will not work as is? or just be 'slower'  (slower for what? to power up?)

but if I need to make changes and re-order...I'd guess I'll have to do that.

Should I just mimic the layout/schematic like in the second image then?...


This forum is totally screwing up lately - I wrote a long post, and it ate it. Short form,

1. the 2nd image is what you normally see.
2. you always need both electrolytic and bypass caps on the Vcc buss, although a bypass on the uC
    Vcc pins will probably be adequate if not too far from the v.reg.
3. if the electrolytics are too large, Vcc may come up slowly, and the uC may not boot properly,
    but you'll probably not have that problem with values of 1..22 uF.


JoeN


This forum is totally screwing up lately - I wrote a long post, and it ate it. Short form,


On EVERY forum I post on I copy my message before I post it just in case.  I have gotten screwed too many times in too many places not to take the extra third second to hit ctrl-a, ctrl-c
I will never ask you to do anything that I wouldn't do myself.

xl97

hmm..good to know.. (still learning it all I guess)

on this project.. (which is based/built form the SECOND image)..

there is no Electrolytic caps even being used  (guess that is in incorrect).. I didnt see any +/- in the schematics..where others 'did' have it

here is the board made from it:



I havent had 'any' problems with anything so far with it.. as far is it not working/flashing bootloader/loading a sketch/playing audio..etc..

and again.. no electrolytic caps were used?

(elec caps are SO big!.....how to keep small footprint down when I see boards like/similar to this then)?

I am basically still learning.... I looking at open source schematics.. read, copy..trying to understand and implement into my own boards..etc

so thanks for the help/feedback..

Im sorta wondering why when I posted the first (full) schematic for review here, everyone said it was 'ok'??

take care

thanks!


dhenry

Quote
there is no Electrolytic caps even being used


I have had numerous systems running without decoupling caps. It is one of those things that you don't need it until you need it.

Grumpy_Mike


I have had numerous systems running without decoupling caps. It is one of those things that you don't need it until you need it.

So your systems are as good as your answers here, I did wonder.
The thing is that a seemingly functioning system can be just on the edge of functioning and things like ageing of components can push it over the edge. So to use no decoupling capacitor is just asking for it.

It is rare that there is too much decoupling. The only case I have ever encounters is that some DC to DC converters will not start with too much of a capacitave load, and the odd case of low dropout regulators, most are not that fussy.

Go Up