[SI-LIST] Re: VRM Modeling of commercial DC-DC convetors
- From: "Smith, Larry" <larrys@xxxxxxxxxxxxxxxx>
- To: "sij99@xxxxxxxxx" <sij99@xxxxxxxxx>, "istvan.novak@xxxxxxxxxxx" <istvan.novak@xxxxxxxxxxx>
- Date: Sun, 14 Jul 2013 22:14:12 +0000
Jack - The paper you mentioned is most likely the origin of the component
parameters required by Allegro because the parameter names are identical. When
Ray Anderson, I and some other engineers at Sun Microsystems wrote this paper
back in the 90's, the typical VRM was in a silver box mounted in the computer
chassis and connected to the mother board with wire cables about a foot long.
VRMs that mounted to the PCB were just becoming popular. The 4 element circuit
topology made a lot of sense back in those days. But now, I just use a series
L and R for the switching portion of the VRM similar to the topology that Dr
Johnson has shown. Here is a link to our paper buried in some old SI-list
archives:
http://www.si-list.net/files/published/sun/cpmt_1999.pdf
The paper gives some guidance on how to calculate L_slew from knowledge of the
transient current (dI), the allowable voltage droop (dV) and the amount of time
that it takes the switching regulator to slew that amount of transient current
(dt). The best you can hope for from simulation of this model is an estimate
of the bulk capacitance needed to go with the VRM with this kind of transient
response.
Let me make a comment on frequency domain analysis and switching power
supplies. Buried in frequency domain analysis are assumptions about linear and
time invariant circuits. Switching power supplies are neither.. The circuit
topology varies with time ( the switches..) and they are very non-linear
(doubling voltage does not double current). We must be very cautious when
attempting to use frequency domain models for switching power supplies. The
transient response to a rising current waveform is often quite different
(asymmetrical) compared to a falling current waveform. Our simple model would
require different L_slew values for rising and falling transient current
waveforms. A complete analysis requires time domain simulation and more
complex models.
The same comment can be made for step function PDN loads where on-die circuits
demand a square wave of current. The damping is very different on the rising
current edge compared to the falling current edge. The load for our PDN
circuit is not time invariant. Power integrity analysis is usually done in the
frequency domain with linear and time invariant assumptions but a complete
analysis should involve both frequency domain and time domain simulation.
Regards,
Larry Smith
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of Jack Si
Sent: Saturday, July 13, 2013 6:15 AM
To: istvan.novak@xxxxxxxxxxx
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: VRM Modeling of commercial DC-DC convetors
Thanks Istvan, Howie and Faraydon for the useful info to model the VRM.
Thanks a lot,
Jack
________________________________
From: Istvan Novak <istvan.novak@xxxxxxxxxxx>
To: sij99@xxxxxxxxx
Cc: "si-list@xxxxxxxxxxxxx" <si-list@xxxxxxxxxxxxx>
Sent: Saturday, July 13, 2013 9:59 AM
Subject: [SI-LIST] Re: VRM Modeling of commercial DC-DC convetors
Dear Jack,
For these converters you can get good models from the vendor and you can also
get a generic SPICE simulator that you can use to simulate the rest of your
circuit. Or, if you want to use your simulator instead for the rest of the
power distribution network, you can take the output data from the vendor
simulator and fit a simple model and use that model in your simulator.
Best regards,
Istvan Novak
Oracle
On 7/12/2013 10:15 AM, Jack Si wrote:
> Dear Istvan,
> I am using Dc/DC uModule LTM4616 and LDO LT1763 in my design. When i set the
> PDN analysis in Allegro, it requires slew inductance(Lskew), Flat
> Resistance(Rflat), Output inductance (Lout), Output Resistance(Ro). I was not
> able to get the info directly from the data sheet. I referred "Power
> Distribution System Design Methodology and Capacitor Selection for Modern
> CMOS Technology", but it gives typical values.
> My questions are,
>
> How can i derive these values?
> Whether the typical values will acceptable result?
>
> Thanks and Regards,
>
> Jack.
>
>
>
>
>
>
>
> ________________________________
> From: Istvan Novak <istvan.novak@xxxxxxxxxxx>
> To: sij99@xxxxxxxxx
> Cc: "si-list@xxxxxxxxxxxxx" <si-list@xxxxxxxxxxxxx>
> Sent: Friday, July 12, 2013 6:50 PM
> Subject: [SI-LIST] Re: VRM Modeling of commercial DC-DC convetors
>
>
> Jack,
>
> The question is too broad, because what model you need also depends on
> what simulator you use. First check with the vendor to see what
> models they may have. If they have none, you can get a sample,
> measure, and create a behavioral model yourself.
>
> Best regards,
>
> Istvan Novak
> Oracle
>
>
> On 7/12/2013 3:01 AM, Jack Si wrote:
>> Hi Experts,
>> As the part of the PDN Analysis, i need VRM model for the commercial Buck
>> DC-DC power module. Please suggest the accurate way to model it and use.
>>
>>
>> Thanks,
>>
>> Jack
>>
------------------------------------------------------------------
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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
