Re: [SI-LIST] : A Question About Power Noise.

Stephen Peters ([email protected])
Wed, 17 Mar 1999 15:05:11 -0800

Hello Mark:

A question. I assume that in RF design the purpose of power plane
'decoupling' is to keep RF energy out of a devices power leads -- i.e.,
preserve a quiet DC reference. This is done by providing a low impedance
path between power and ground at the frequency of interest. Therefore,
it seems to me that you would want a decoupling capacitor to be
operating right near its SRF, but if I read your comments correctly, you
imply you do not want it to operate it at that point. What am I missing?

(Note: In the digital world, the rule is to use as big a decoupling
capacitor as possible (for a given package inductance). The reason being, of
course, is that in digital design the purpose of a 'decoupling capacitor' is
to supply transient current to a devices when it switches. In other words,
the decoupling capacitor is acting as a charge storage device, and
the bigger the capacitor the better. It's interesting that while we both
refer to those capacitors as "decoupling capacitors", they are used for
very different reasons.)

Stephen Peters
Intel Corp.

Mark Radol wrote:

> Vinu Arumugham wrote:
> >
> > Is a low value of capacitance really important? At 1GHz+, the package inductance dominates the characteristics and so the impedance of a 0.1uF and a 100pF capacitor in an 0603 package will not be very different. As you point out,
> > for better performance at higher frequencies one would need a lower inductance 0402.
> I'm not as concerned really about the exact number of effective pF as I
> am the series reonant frequency (SRF). If the inductance remains the
> same, the SRF increases when you decrease the capacitance.
> SRF~1/[6.28*(Leff*C)^0.5]
> If the SRF is outside my range of interest, then the capacitor will
> still look sufficiently like a capacitor (low Z actually) so that I
> don't have to worry about the resonances effecting the circuit. In
> other words, I'll get the low Z supply and not get the oscillations and
> suckouts in the frequency response that I'd get from a larger valued cap
> with a lower SRF.
> So yes, while I agree a 0.1uF cap will be an inductor @ 1GHz, a 100pF
> will still look largely capacitive.
> Ray brings up some points that perhaps explain the differences between
> what we are seeing. Typically my designs are 2 layer RF boards with
> coplanar lines. The "ground" is beside the signal/power line. That can
> yield a MUCH higher mounted SRF than if you have a 4 layer board with
> the vias to GND/VCC. I still see a reduction between quoted SRF's and
> board mounted SRF's, but the difference is not as severe as when vias
> are used. The difference in SRF's is only a couple X at most instead of
> a couple orders of magnitude. So how much does using an 0402 help when
> using > 2 layers since the PCB dominates the inductive parasitic?

**** To unsubscribe from si-list: send e-mail to [email protected] In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at ****