Re: caps (was: 8250LP....)

From: silverdr_at_wfmh.org.pl
Date: Tue, 22 Aug 2017 00:05:00 +0200
Message-Id: <838C2994-0DC8-4964-91C8-CECB0B7DC5FB@wfmh.org.pl>
> On 2017-08-21, at 23:07, Mike Stein <mhs.stein@gmail.com> wrote:
> 
> How is any cap across the IC's power pins not "filtering the power rail(s)"?

The main, desired effect of the typical for digital ICs 47nF/100nF bypass cap is limited to a relatively small distance. That's why you put them close to the component rather than at the power connector common for the whole board. That's what I call bypass cap. You can have other, higher capacitance cap, which bypass/filter other frequencies... 

> If you look along the IC's supply rail you will probably also come across some 10 - 100uf caps across the power rails, maybe even right beside a ceramic; are they filters or bypasses? As I said, the only difference is the frequency they're intended to 'bypass' which will determine the apppropriate type and location.

... and you can have (usually even more bigger) ones that "filter" much lower frequencies for the whole board or section and are therefore placed close to the supply points of the board / section rather than the small component.

> But if you want to call electrolytics and tantalums "filters" and ceramics "bypasses" that's OK; we'll know what you mean... ;-)

:-) No, not exactly. My arguably bad old habit is to call the small ones, placed out of necessity next to the component / IC and filtering mostly the respective component "bypasses" and those which are used to provide filtering for larger sections "filters".  It doesn't have much to do with the type/manufacturing process of the capacitor.

-- 
SD! - http://e4aws.silverdr.com/


       Message was sent through the cbm-hackers mailing list
Received on 2017-08-21 23:00:03

Archive generated by hypermail 2.2.0.