Re: [SI-LIST] : Inductance vs. frequency
Fred Balistreri (email@example.com)
Fri, 24 Apr 1998 15:52:34 -0700
> I don't know chips and fine lines, but I would guess that at 10MHz+ you
> would be in the LC regime rather than in the RC regime.
> Thank you for your response. My guess is that if we measure R at DC, then
> the whole RC delay can be off 10X at 10MHz, or even more at 100MHz. Which I
> haven't seen on chips.
> May be someone in this group can comment on it for us.
> -----Original Message-----
> From: firstname.lastname@example.org <email@example.com>
> To: Chip Chip <firstname.lastname@example.org>; email@example.com
> Date: Friday, April 24, 1998 10:08 AM
> Subject: Re: [SI-LIST] : Inductance vs. frequency
> >I can't give you an absolute answer, but it looks constant from 0 to 10KHz
> >and rises approximately as the square root of frequency from 100KHz to
> >That's about what I would expect from skin effect.
> >Hi, interesting data.
> >One quesiton, the resistance goes up 10X from DC to 10MHz, isn't that too
> >high ?
> >-----Original Message-----
> >From: firstname.lastname@example.org <email@example.com>
> >To: si-list(a)silab.eng.sun.com <si-list@silab.Eng.Sun.COM>
> >Date: Thursday, April 23, 1998 10:25 AM
> >Subject: RE: [SI-LIST] : Inductance vs. frequency
> >I have some inductance data on 22 gauge high capacity twisted pair
> >parameters are Hz, ohm/mile, mH/mile, umho/mile, uF/mile
> > f R L G C
> > 0 166.8 .802 --- .0825
> > 1K 166.9 .798 1.2 .0820
> >10K 168.6 .792 30.2 .0817
> >100K 209 .773 603 .0810
> > 1M 545 .654 8900 .0808
> >10M 1725 .593 78400 .0808
> >The inductance in this data is at least qualitatively
> >similar to your data.
> >I believe this data was determined by measuring characteristic
> >impedence, propagation attenuation, and propagation velocity
> >and then calculating R,L,G,C to match the data.
The reason the skin effect does not play a more major part in IC is the
geometry. Look at a skin depth chart over frequency for aluminum. An
IC interconnect thickness is rather thin. That means most of the
current is still flowing across the whole cross section rather than
on the edges at even 100Mhz. Therefore the DC resistance still
dominates. The concept of inner and outer inductance only applies
when the skin depth at the frequency of interest is a small portion of
the overall conductor cross section. Not so in digital IC's of today.
The delay in IC's is still dominated by RC although L plays a major
role in the packaging. That is changing but mostly still true at the