Searching \ for '[EE]: 5V Logic Bypass caps - SMD value/phy' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=logic+bypass+caps
Search entire site for: '5V Logic Bypass caps - SMD value/phy'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: 5V Logic Bypass caps - SMD value/phy'
2000\06\05@182604 by rottosen

flavicon
face
Robert:
If you look at the curves that you included you will see that the 0.1uf
capacitor has lower impedance at 10000 KHz then the 1uf capacitor. This
is because the inductance of larger value capacitors is larger than
smaller value capacitors.

Also I think that the value of 1 ohm is somewhat arbitrary and does not
apply well in many cases, especially high gain or high power circuits.

-- Rich


"Robert A. LaBudde" wrote:
{Quote hidden}

2000\06\07@052856 by Robert A. LaBudde

flavicon
face
<x-flowed>At 04:25 PM 6/5/00 -0600, Rich wrote:
>Robert:
>If you look at the curves that you included you will see that the 0.1uf
>capacitor has lower impedance at 10000 KHz then the 1uf capacitor. This
>is because the inductance of larger value capacitors is larger than
>smaller value capacitors.
>
>Also I think that the value of 1 ohm is somewhat arbitrary and does not
>apply well in many cases, especially high gain or high power circuits.

1. The curves for 1 uF, 0.1 uF and 1 nF capacitors were for different types
of capacitors.

2. Inductance is a function of lead length and spacing, not capacitance.

3. The curves show that electrolytics have poor impedance, ceramics have
good impedance at a narrow range of frequencies and tantalums have a flat
impedance at a reasonable value.

4. You will notice the impedance curves of small value, high quality
ceramics have lower impedance at high frequencies, they have high impedance
at low frequencies.

5. For a given capacitor type and package, higher capacitance will perform
better. Usually the reason smaller capacitance values are used at high
frequencies is that the low inductance packages are only available in low
capacitance values.

6. The "1 ohm" impedance corresponds to a current draw of 100 mA and a
desired maximum ripple of 0.1 V. This is a reasonable design target for a PIC.

7. My own experiments with a 4 MHz PIC showed 0.01 uF ceramic not good
enough to filter clock noise, but any value from 0.1 uF to 4.7 uF tantalum
quenched it.

8. I would be very interested in seeing some experimental results from
other people at higher frequency (e.g., 20 MHz). "Theories are great but
experiments are science."


================================================================
Robert A. LaBudde, PhD, PAS, Dpl. ACAFS  e-mail: .....ralKILLspamspam@spam@lcfltd.com
Least Cost Formulations, Ltd.                   URL: http://lcfltd.com/
824 Timberlake Drive                            Tel: 757-467-0954
Virginia Beach, VA 23464-3239                   Fax: 757-467-2947

"Vere scire est per causas scire"
================================================================

</x-flowed>

2000\06\07@125624 by David VanHorn

flavicon
face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


>2. Inductance is a function of lead length and spacing, not capacitance.

In any given capacitor type, there is an inductive component that is linear
in relation to capacitor size. Lead inductance is more or less independent.

>3. The curves show that electrolytics have poor impedance, ceramics have
>good impedance at a narrow range of frequencies and tantalums have a flat
>impedance at a reasonable value.

Some numbers would really help here.

>4. You will notice the impedance curves of small value, high quality
>ceramics have lower impedance at high frequencies, they have high impedance
>at low frequencies.

And if you go still higher, you will find the impedance rising again.
(would help if you put numbers on your assertions)

>5. For a given capacitor type and package, higher capacitance will perform
>better. Usually the reason smaller capacitance values are used at high
>frequencies is that the low inductance packages are only available in low
>capacitance values.

This is just plain not true.
Granted, you aren't going to find 1000uF in a 1206 SMD package, but even a
0.1uF has a definite range where is is a significantly better bypass than
at others. My observations held regardless of the capacitor type or value,
in that any given cap showed a range where it was most effective as a
bypass (lowest impedance) and at lower or higher frequencies, the same
capacitor was significantly less effective (higher impedance)

>6. The "1 ohm" impedance corresponds to a current draw of 100 mA and a
>desired maximum ripple of 0.1 V. This is a reasonable design target for a
PIC.

And has almost nothing to do with bypassing high frequency noise.
You're talking about averaged DC current here.
What "ripple" are you talking about? 60 Hz, 120 Hz, 4, 12, 20MHz....?

>7. My own experiments with a 4 MHz PIC showed 0.01 uF ceramic not good
>enough to filter clock noise, but any value from 0.1 uF to 4.7 uF tantalum
>quenched it.

I would have expected 0.022 to give optimum results here (third harmonic of
4 MHz), but you give no details of how the capacitors were connected.
Improper track routing or excessive lead length will render most bypasses
ineffective at RF.

You don't mention at what frequency you observed this "clock noise".
(critical to understanding what bypass value will work best.)

How did you measure this, what degree of supression did you see?
In my experiments, I used a Tek 7L5 spectrum analyzer with tracking
generator for 0.1uF and 7L12 with TR501 tracking generator for higher
frequencies.

You are pushing some very fuzzy assertions here.

>8. I would be very interested in seeing some experimental results from
>other people at higher frequency (e.g., 20 MHz). "Theories are great but
>experiments are science."

Which is why I made my original measurements years ago.
I don't remember reading that the laws of physics had changed in the interim.

- --
Are you an ISP?  Tired of spam?
http://www.spamwhack.com  A pre-emptive strike against spam!

Where's Dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.2 for non-commercial use <http://www.pgp.com>

iQA/AwUBOT6XOYFlGDz1l6VWEQJ0cwCg/APshUzhs5YLBgGXYs2k4HXVJt0Aniqi
RFI5keXdiiUDRFcaZLayBwyk
=f1xv
-----END PGP SIGNATURE-----

2000\06\07@182858 by briang

flavicon
face
In-Reply-To: <4.0.1.20000607112426.03837350spamKILLspammail.cedar.net>

David VanHorn <.....dvanhornKILLspamspam.....CEDAR.NET> wrote:

{Quote hidden}

This is as if you didn't read past the first sentence.

IMO The first sentence is sometime true and sometimes not.

For one thing you can't always get the next value up or down sometimes the
cap you';re using is the highest, or lowest value they make in that type.

But you can't just respond to whole of the paragraph by saying "it's not
true" when plainly the rest of it is true and expect to be taken seriously.

The rest of the paragraph describes exactly the reason why one would use a
100nF ceramic to decouple something rather than a 100nF aluminium
electrolytic.

The end of your reply is completely irrelevant you talk about varying the
frequency and seeing an minimum impedance. The paragraph you are responding
to is about choosing the right capacitor for a particular job. WTF are you
doing responding with an argument about a capacitor having a minimum
impedance at a particular frequency.

Brian Gregory.
EraseMEbriangspam_OUTspamTakeThisOuTcix.co.uk

2000\06\16@213256 by David VanHorn

flavicon
face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

At 04:25 PM 6/5/00 -0600, Richard Ottosen wrote:
>Robert:
>If you look at the curves that you included you will see that the 0.1uf
>capacitor has lower impedance at 10000 KHz then the 1uf capacitor. This
>is because the inductance of larger value capacitors is larger than
>smaller value capacitors.

Bingo.

>Also I think that the value of 1 ohm is somewhat arbitrary and does not
>apply well in many cases, especially high gain or high power circuits.

That too.  The Z of the cap needs to be substantially lower than the rest
of the circuit.
Of course if you blow the layout, then the best cap in the world won't do
you much good.


- --
Are you an ISP?  Tired of spam?
http://www.spamwhack.com  A pre-emptive strike against spam!

Where's Dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.2 for non-commercial use <http://www.pgp.com>

iQA/AwUBOTxDtYFlGDz1l6VWEQJghQCgkiGBAwLAGDES1Qj/BEXBbk8GlPsAnRQC
t6ewuXC+W1W0qby7CzQ+Fmmy
=5EYK
-----END PGP SIGNATURE-----

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...