[SI-LIST] Re: Resonance Frequency

Wael - It is generally true that chip capacitance (Vdd to Gnd) keeps
going up and packaging inductance keeps going down for every micro
processor generation.  At the same time, the power consumption of the
processor keeps going up while the voltage keeps going down, which
implies that the current consumption is going up drastically.  This
forces all micro processor manufacturers and system designers to pay
close attention to the power distribution system.  Many processor
packages today have capacitors mounted on them to effectively put
another capacitance in the path from the PCB to the chip.  We are
effectively designing a large power filter.  The goal of the power
filter is for the chip to look out and see a reasonably low impedance
over a broad frequency range in it's power distribution system.

It is also true that the chip/package resonance frequency has remained
about the same over the years, for the reasons discussed in the
previous email.  Fifteen years ago, the clock frequency was often well
below the resonance but today it is well above the resonance.  That is
OK as long as we don't have significant energy right at the resonant
frequency.  Also, I believe we are managing our "Q's" better today so
that if the processor does draw current at the resonant frequency, the
consequences are not so bad.  

Customer code is capable of causing the processor to draw current at
virtually any subharmonic of the clock.  This is the assumption that
should be made when designing a power distribution system for a micro
processor.  If we design our power filter such that the impedance
is flat over the important frequency range, problems will be avoided.
The chip/package resonance is effectively damped out.

regards,
Larry Smith
Sun Microsystems

Wael El Essawy wrote:
> 
> 
> 
> Larry,
> 
> in a previous email you said:
> "For the last 15 years, the chip/package resonant frequency has been somewhat
> under 100 MHz. I worked on a product in 1988 where it was about 80 MHz. The
> inductance was about 1 nH, so by my favorite formula, f0 = 1/{2pi*sqrt(LC)},
> the capacitance in the chip must have been about 4 nF. Since then, the
> capacitance keeps going up and the inductance keeps going down, leaving the
> product almost constant. For the product analyzed in the paper, the
> capacitance is about 100X what it was in 1988. We tried very hard to reduce
> the inductance, but it is not 1/100th of a nH. I guess we have some more
> workto do.. That is why the resonant frequency was just 35 MHz."

> Is this
> generally true for high performance microprocessors? In other words can I
> assume that chip/package resonance frequency has changed from a fraction of
> a microprocessor cycle to tens of cycles. Can I assume that in future
> microprocessors resonance will occur at repeated high current swings over
> hundreds of cycles period only?
> 
> Thanks
> 
> Wael El Essawy
> Ph.D. student
> ECE Dept.
> University of rochester
> 

------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: