The PDN Bandini Mountain and Other Things I Didn’t Know I Didn’t Know

In engineering, it's what you don't know you don't know that can ruin your day and keep you awake at nights. This is especially true after you get your prototypes in the lab, or worse, field returns from the customer. This is one reason why I have been going to DesignCon for the last few years, and this year has been no exception.

One of the sessions I attended was the Power Integrity Boot Camp, hosted by Heidi Barnes of Keysight Technologies, and Steve Sandler of Picotest. At this boot camp, I learned that I didn’t know I didn’t know how important it was to match the voltage regulator module (VRM) output impedance to the power distribution network (PDN) input impedance. Steve and Heidi recently presented a webcast which offered a condensed version of the DesignCon boot camp session. If you are involved in PDN design, this webcast will provide you with an introduction to power integrity and give some insight into the latest tips and techniques to achieve flat impedance designs.

Of course, I always try to attend some of Eric Bogatin's presentations, because I always come away with something I didn’t know I didn’t know. Eric is an adjunct professor at the University of Colorado and the dean of Teledyne LeCroy’s SI Academy. He was honored at this year’s DesignCon with a well-deserved Engineer of the Year award.

The speed training event he hosted with Larry Smith from Qualcomm was on the top of my list to attend. During the session, Eric explained that the most critical feature of PDN design was controlling the “Bandini Mountain.”

Bandini Mountain.JPGThe Bandini Mountain expression has often been used to describe a tall pile of manure. Originally, it referred to a 100-foot tall mound of fertilizer built by the Bandini Fertilizer Company in California prior to the 1984 Los Angeles summer Olympics for advertisement purposes. When the company went bankrupt, this large mound of smelly fertilizer was left behind and everyone wished it would go away.

A little background: Because of this little bit of trivia, the late, great Steve Weir coined this term to describe the large resonant frequency peak formed by the parallel combination of the on-die capacitance and the package lead inductance, as seen from the die looking into the PDN. This peak is inherent in all PDN networks, and it is almost impossible to get rid of. And like the Bandini Mountain, it was something PDN designers wish could go away.

Steve was a regular icon at past DesignCons until his sudden passing in August 2015. Steve was one of the smartest guys I knew, and I always looked forward to catching up with him when I visited DesignCon. If you knew Steve, like many of us did, you know that he often had very humorous analogies to describe empirical or simulated results. This example is no exception. He will be sorely missed, and we all appreciate his contributions to the engineering community. 

So, what I learned I didn’t know I didn’t know from Eric and Larry’s presentation was that every PDN design will have a Bandini Mountain, and unless you know its frequency and take steps to try to mitigate its peak, it could ruin your day! Even if the system seems to work in the lab, that doesn’t mean that it’s robust enough and won’t fail under certain field operating conditions that affect the transient currents.

Eric has made available the speed training slides and the associated video on his SI Academy website. If you look under Video Recordings, Presentations and Webinars (VRPW) and scroll down to the bottom you will find the slides titled, “VRPW-60-35 DesignCon 2016 PDN speed training.” If you watch the whole presentation you will learn all about the “PDN Bandini Mountain” and techniques to mitigate its effects. And while you are there, have a look at the many other videos and presentations available for free and by paid subscription.

Eric and Larry have also co-authored a new book scheduled for release in June 2016 titled, “Principles of Power Integrity for PDN Design.” I can’t wait to buy this book to add to my library so that I can find out more of what I don’t know I don’t know about PDN design. If it’s anything like Eric’s other books, I won’t be disappointed.

Back

2016

The PDN Bandini Mountain and Other Things I Didn’t Know I Didn’t Know

03-30-2016

Originally, Bandini Mountain referred to a mound of fertilizer built by the Bandini Fertilizer Company in California prior to the 1984 Los Angeles Olympics. When the company went bankrupt, this mound of fertilizer was left behind. Steve Weir coined this term to describe the large resonant frequency peak formed by the parallel combination of the on-die capacitance and the package lead inductance, as seen from the die looking into the PDN.

View Story
Back

2014

Accelerating the SI Learning Curve - Bogatin's SI Academy

08-06-2014

Columnist Bert Simonovich writes, "Last year, Dr. Eric Bogatin, the 'Signal Integrity Evangelist,' announced the end of his famous signal integrity classes. At the time I remember thinking to myself, 'What's next for Eric?' If you know Eric, like I do, you realize that the end of one phase of his career usually means the start of the next one. And now we know what that is."

View Story

Bert's Practical Design Notes: Accelerating the SI Learning Curve - Bogatin's SI Academy

08-06-2014

Columnist Bert Simonovich writes, "Last year, Dr. Eric Bogatin, the 'Signal Integrity Evangelist,' announced the end of his famous signal integrity classes. At the time I remember thinking to myself, 'What's next for Eric?' If you know Eric, like I do, you realize that the end of one phase of his career usually means the start of the next one. And now we know what that is."

View Story
Back

2013

Are Guard Traces Worth It?

08-14-2013

Some claim that a guard trace should be shorted to ground at regular intervals along its length using stitching vias spaced at 1/10th of a wavelength of the highest frequency component of the aggressor's signal. But others believe separating the victim trace to at least three times the line width from the aggressor is good enough. Bert Simonovich addresses both arguments.

View Story

Bert's Practical Design Notes: Are Guard Traces Worth It?

08-14-2013

Some claim that a guard trace should be shorted to ground at regular intervals along its length using stitching vias spaced at 1/10th of a wavelength of the highest frequency component of the aggressor's signal. But others believe separating the victim trace to at least three times the line width from the aggressor is good enough. Bert Simonovich addresses both arguments.

View Story
Back

2012

Bert's Practical Design Notes: Perils of Lumped Via Modeling

05-30-2012

Popular opinion has held that PCB vias were mainly capacitive in nature, and therefore could be modeled with lumped capacitors. Although this might be true when the rise time of the signal is greater than or equal to 3x the delay of the via discontinuity, I'll show you why it is no longer appropriate to think this way.

View Story

Bert's Practical Design Notes: The "Stubinator" vs. Back-Drilling

04-11-2012

I was intrigued by a DesignCon 2010 paper presented by Dr. Nicholas Biunno on a new matched terminated stub technology developed by Sanmina-SCI Corporation. The company calls this technology MTSvia, and it allows the embedding of metal thin-film or polymer thick-film resistors within a PCB stackup during fabrication. Personally, I like to call this technology the "Stubinator."

View Story

Backplane High-Level Design: The Secret to Success

01-04-2012

In my previous column, I touched briefly on the concept of backplane high-level design (HLD). For any new backplane design, I always recommend starting with a HLD. It helps you capture your thoughts in an organized manner, and later provides the roadmap to follow for detailed design of the backplane. This week, I will touch on key aspects that go into this process.

View Story
Back

2011

Bert's Practical Design Notes: Why Backplane Architecture is Crucial

11-04-2011

I am often asked what I do for a living. When I say high-speed signal integrity and backplane architect, the next question is usually, "What is a backplane architect?" By my definition, a backplane architect is any person who plans, devises or contrives the achievement of a backplane design. And the earlier you consider the backplane's physical architecture, the more successful the project will be.

View Story

Bert's Practical Design Notes: Fiber Weave-Induced Timing Skew

09-17-2011

A couple of times this year, fiber weave effect timing skew came up for discussion on the SI-List that many of us subscribe to. This is becoming more of an issue for many designers as bit rates continue to soar upwards. For signaling rates of 5GB/s and beyond, it can actually ruin your day. So what is fiber weave effect anyway, and why should we be concerned about it?

View Story

Bert's Practical Design Notes: Cross-Sectional Geometries

08-10-2011

So, why do we need to study PCB cross-sectional geometries? Because they describe the details of the dielectric substrates, traces and reference planes in a PCB stackup. Their relationship to each other is used to predict the characteristic impedance and interaction of the respective traces. Understanding these geometries can help you determine odd-mode and even-mode impedance, average and differential impedance, crosstalk and more.

View Story

Practical Differential Via Modeling Made Easy

07-14-2011

In my last column, I discussed the twin-rod model. Now I'll explain how a twin-rod transmission line model can be the basis for a practical differential via circuit modeling technique and a simple alternative to a 3D field solver. This method can yield an approximation much faster than a field solver, and when you need a rough estimate, this via modeling method may be just what you need!

View Story

The Three Amigos: Twin-Rod, Rod-Over-Plane and Coax

06-22-2011

In almost all cases, equations used to calculate the loop inductance and capacitance of transmission lines are approximations. However, there are three unique cross-sectional geometries that have exact equations: Twin-rod, rod-over-plane and coaxial. I've dubbed them "The Three Amigos." Get to know them -- they can be your friends.

View Story

IBIS Output Impedance Made Easy

05-03-2011

I subscribe to the SI-List forum on signal integrity. People often pose the question, "How do you find the driver impedance information from the IBIS file?" Most of the time we want this information so that we can explore mitigation techniques to control reflections caused by impedance discontinuities of the transmission path. Let's look into IBIS output impedance.

View Story

New Column: Bert's Practical Design Notes

04-13-2011

Over the years, I have held a variety of hardware design engineering positions and pioneered several advanced technologies into products. After all this time as an engineer, I still have the passion I had as kid to learn and understand new things. This column is about sharing some of that passion.

View Story
Copyright © 2017 I-Connect007. All rights reserved.