RE: insuring the stability of power delivery
In general I agree to the concept described by Brian.
I am not sure that the method of how to specify it is the right one.
The reason is that we need to check that adding the 3-5db to the PSE output
impedance allows low cost PD dc/dc converter designs with its EMI filter.
In other words, we need to check what is the typical PD input impedance with
EMI filter of standard low cost implementation and than compare it to a
standard PSE output port impedance. When we will have these results, we can
evaluate what should be the best definitions or should we add special tricks
to handle it.
Remember that we have two scenarios:
A: A Single big PSE power supply is connected to multiport system and the
PSE output impedance is changed due to the status of each port
(connected or disconnected, full power or low power etc.)
B: Each port has its own 15W power supply.
We can always say, well we are defining only one port. However we cant
ignore how multiport system change the 1 port parameters...
As Roger said.. more fun.
> -----Original Message-----
> From: Lynch, Brian [SMTP:brian_lynch@xxxxxx]
> Sent: ה, מאי 31, 2001 7:45 PM
> To: 'R karam'; 'Yair Darshan'; 'Rick Brooks';
> Subject: RE: insuring the stability of power delivery
> for continuity, I'll reply at the bottom
> >-----Original Message-----
> >From: R karam [mailto:rkaram@xxxxxxxxx]
> >Sent: Thursday, May 31, 2001 12:15 PM
> >To: Lynch, Brian; 'Yair Darshan'; 'Rick Brooks';
> >Subject: RE: insuring the stability of power delivery
> >Comments on Brian's Input
> >Adding cable to the mix: The closed loop output impedance of the PSE
> >plus the maximum cable impedance (R and L) must be lower than the
> >closed loop input impedance of the PD plus the maximum cable impedance
> >(L and R)=> Since we do not know the cable length to insure system
> >stability, the closed loop output impedance of the PSE plus the cable
> >impedance must be less than the PD's closed loop input impedance.
> >Notice that I have said "closed loop" impedance. This means that the
> >feedback loops of the PSE power supply and PD DC/DC converter plus the
> >load of the PD are included in the calculations. Not an easy task for
> >an open standard.
> >[rkaram] this is either too simple or a mess. my guess is
> >that in Front of the
> >DC/DC we have some Choke and ceramic caps, these are meant for
> >emi control.
> >so looking into the PD the AC impedance can be inductive or
> >capacitive based
> >on where the choke's resonance is (as its parasitic
> >capacitance and the caps on
> >each side come into play with the inductance) so if we are
> >after some really low
> >Z at very low frequencies, then we would need a huge cap at
> >the pd input?
> >of course now to make this more fun, if what Yair mentioned
> >(negative Z)
> >looking into the DC there goes another twist now...
> >a lot to munch on...
> [Brian Lynch] You are right that there is probably an LC filter
> on the input of the PD for switching frequency as well as high
> frequency filtering. There is also a low frequency aspect of the
> input impedance which is partially a function of the closed loop
> frequency response of the DC/DC converter and the load. As such,
> you cannot state the impedance as "capacitive" or "inductive". You
> simply have to plot the impedance as a function of frequency while
> the equipment is operating. You can calculate the impedance, or
> measure it.
> Since PDs can be so varied, I recommend we concentrate on
> determining the closed loop output impedance of a typical
> range of PSE power supplies, and then add the cable (R&L)
> and 3 to 5dB of margin and call that the lower limit for the PD.
> I agree that this is a lot to comprehend, but this is how
> distributed power systems are designed. Any time a DC/DC converter
> is powered by a power supply, the same stability criteria has to be met.