Switcher Design Without Tears

From an expert switching regulator designer :

I personally try to develop a model that starts as an averaged model and ends as a close as possible to the actual implementation of the switcher that I'm trying to design.

While Matlab is useful for quick checks, it is not as useful as VerilogA when it comes to actually designing the controller and doing model/circuit comparisons.
State-space averaging is a neat way to include a lot of variables into your model, but for a more simple and direct way refer to The Fundamentals of Power Electronics by Robert Erickson & Dragan Maksimovic. I strongly suggest that you read this book, especially the control section. It's tough to design a switcher without first having a good handle on the different loop dynamics. If you're working with other folks that are more senior in the area of power management, ask them to share you the very many other issues and hidden obstacles that exist along the way; they will bite you. These are too many to mention on this message.

Resume Tips

froma guru

1. Conform to the Specific, Measurable, Attainable, etc rule for terminology
2. Don't list tasks
3. Start and finish on one page
4. Ditch the objective.
5. No grammar errors allowed
6. Resume means summary in Fr, so don't have a summary on a summary. Oops:)
7. Tailor it to where you want to go - so leave out irrelevant stuff.

The Analog Design Flow

Need to get this to the PDK master soon. A good example of the sort of stuff I do is on page 6 here, a good presentation anyways. One of the independent contractor guys says he uses Simucad and Tanner. We used Cadence at work and, while I miss it, I can think of a hundred ways it could be better. One of the heaviest and most inflexible tools you've ever seen. Not surprised they reported a 1.4b loss recently (q? yr?)

To do analog design:

1) Specifications - all values in the absolute maximum ratings (supply, temp) must be set in stone - they should not be TBD
2) Tool for schematic capture (meaning ability to generate a spice netlist)
3) spice - or, if working in RF, an equivalent of ADS or spectre
4) waveform viewer to view results.
5) spice models - this is the huge bottleneck - and depends heavily on a relationship with a vendor like MOSIS. But, there are other options like CMP and e-shuttle. Relationships are the key here - there are people who will let us put IP on their chips so we can test them
6) PDK - which really includes the spice models - we need to know what devices are available in the process, how to estimate die size, etc
7) corner spice models and mismatch models - design for manufacturing.
8) tool to use mismatch and statistical models to predict yield - this is important to productize a design. If you know that a parameter is guaranteed by design, then you don't need to test it - if you characterize it. But, if you know that you will have some yield loss which is acceptable, then you need to prescribe a test that can catch the bad chips

BTW, one name I've heard tossed about in the forums for cheaper analog design - compared to the Cadence suite is penzar : http://penzar.com/

Influential Leadership

Attended an IEEE Networking Dinner on Monday - Sam Palazzolo of Pathos Leadership spoke. Are you an influencer or an influenced person? If you do a favour for someone and he thanks you, the influential response is "You'd have done the same for me". Don't say you're welcome, don't mention it, etc.

There are 6 themes - reciprocity, consistency and commitment, liking, authority, social proof and scarcity.

• You should use information to differentiate yourself. If you have useful information - share it.
• People like other people who compliment them and cooperate with them - charisma.
• Authority is founded on knowledge and trustworthiness.
• When you want someone to say yes to a request, you can try to first present him with a big request that he's sure to say no to. Present in a nice way that'll make him feel bad to say no. Then, present him a smaller request that will be easier for him to say yes to. For example, girl scouts trying to sell cookies will say - would you like to chaperone one of our sleepovers. No? Then how about buying 2 boxes of cookies instead? (Really? David Lieberman presents research on the foot-in-the-door technique - get them to say yes to a small request first then present the bigger one after you've primed them.
• When you are trying to sell people something, the negative sometimes might work better when you try to get them to go from A to B. If you watch diet/slimming commercials, you'll see this in action.
• Your legacy is made up of your work, family, health, happiness and friendships.
• Ask yourself "In what area is it clear that I can't get to where I want to go with the story I've got?"

If some or all of this didn't make sense to you, you're probably not alone.

There's Art, And..

there's art that leaps out at you - like this one by

The Art of War

by David D'allesandro. You can get an idea of what he's like by searching for that name on Goog video and you'll get a Businessweek sound byte. Seems okay, but, to me, seems to lack the refinement that you expect in a CEO. But who am I to judge based on one sound byte?

One thing he mentions more than once is the "room-changer" thing: Who is the room-changer? Which one creates a dramatically different and more positive aura whenever he or she walks into the room? (tell me about that one - I've seen a few types who cast a spell of doom on the room!) Even if your back is to the door, you don't need trumpets or 'Hail to the Chief' to know that a person of presence has come in.

How?

1. Offer something substantial that's not just self-importance. Energy to listen, solve problems, make things happen. 
2. Be true to yourself and the things you believe. Don't craft a personality. If you choose to take a position on an organizational issue, how will it look in a 30 second political attack ad? 
3. Have a sense of perspective. The people whose paychecks you sign will only say good things about you and tell you positive stuff. You have to have friends who don't care about what your position is. (to me, this also means : "don't try to win all the time - in every conversation")

The things about Nike being a place to learn marketing but not practice it is another big thing I got out of this.

Then, about bosses - they may tell you they're interested in helping your career. What they mean is, "Wise up. I'm interested in making my own career better. If that works for you, well and good." Basically, you want to help your boss's career in the hope that you can rise with her.

I learnt some things I didn't know about GE. I thought GE was a great place where the best people were democratically chosen to be leaders because everyone agreed they were the best. In Jack: Straight from the Gut, the a.b. of Welch, he describes the politics surrounding his own contest for the chairmanship of GE as "thick."

Then, how do you find the mole? Use the CIA trick. David Lieberman would love this one. Say you have three suspects, give each one a different piece of false, damaging information about you and see which one surfaces. Get scared if all three do: You are really unpopular!

One funny part of the book is where he talks about one bad hire he made: Not only was her work shabby and inaccurate, her behaviour was so bizarre that she soon lost all credibility with the people who worked for her. Here is just one example among many: She had this strange relationship with M&Ms. She would buy a giant bag of M&Ms, and instead of having lunch, she would take them into a conference room and lay them all out, by colour, battalions of M&Ms, all with the letter "M" facing her. Never mind the sanitary issues of eating them right off the conference table, she'd talk to her candies before she ate them: "Okay, yellow battalion, your turn!"

Some of these outfits can pay a headhunter a $200k fee for hiring someone in a senior management position. 

In the old days - before the 90's about 1 in 10 CEO's were sacked by the boards. In the new era, it's about 1 in 3.

Socially, once you reach senior management, don't socialize with your employees! You just increase the odds of something nasty happening to undermine your career! Damn - what this must do to the legions of young riffraff aiming for the high posts just for the social life that should accompany it.

One thing he talked about which struck very close to home was a loan shark culture at work: When I was a kid, my dad borrowed from loan sharks. The interest was five percent a week, and it had to be paid every week. If the loan shark's collector walked in, and my dad said, "Look, my kid was in the hospital this week. I can't pay you until next week," the collector would say, "I'm really sorry to hear about your kid in the hospital. Pay me my money."

At some organizations, this attitude is endemic.

Now, those who know I worked for a quality outfit might wonder about this. Was it like that in my place of work? What I can say is that the businessscape is extremely competitive. To survive, you have to compete against a lot of shady outfits and cost is king. So is delivery. If you can't deliver, you're in trouble. Which is worse, to know that you're dead if you don't get across the tracks before the train gets there or to dread nasty meetings on a daily basis? My take.. it trickles down from the top. If you're careful about how you hire, you'll never get into such problems. I've seen entire teams dragged down because one or two resources don't care about pulling their weight. And they'll wrestle - they're pigs, and they love it. Hiring is one of the toughest decisions that involves so little work. Really, what can you do? You can ask questions, but how hard is that? This is where experience and judgement come in. Even the smart ones who can deliver can become tyrants and terrorize the other members of their team and make other worthwhile contributors walk up and leave. I've seen that happen too.

Based on what happened to me, the next time you contemplate changing groups internally, you're looking at getting a whole bunch of answers. Which business group are you really part of? What is the history? Why do you guys sit here when the rest of that business group sits there? Who is really responsible for this project? Which business group is going to pay for new hires to staff this project? What if things don't go well? Who cracks the whip? If I do deliver, what can I expect in one year? In two years? In three? And the thing you might find really hard - telling your present boss about this opportunity and getting his take on it. Sure, things aren't going well. But, he has a stake in seeing you depart gracefully. It makes your boss look bad if you just up and go.

What d'Allesandro says is to walk quickly if you find you're in a loan shark culture - at least start looking now!

And the part of the book I really liked :

At John H, we had a simple cultural goal: We wanted to make the company a great place to be at and a great place to be from.

A great place to be at, in my opinion, is one where there is a free exchange of ideas, where there is some humor and fun, and where people's contributions are rewarded. Why did we want to make it a great place to be from? Because no organization can hang onto all of its good people. There is only one head of marketing slot, only one CFO's job. If nobody leaves, you wind up with many highly talented but extremely frustrated men and women.

Great one! Reed Hastings, in "No Rules Rules" talks about building a team with high talent density. Part of that is - adequate performance earns a generous severance package 😊

Soft Skills Aren't Soft

Just ask Peggy Klaus

self-awareness
trustworthiness
conscientiousness
adaptability
critical thinking
attitude
initiative
empathy
confidence
integrity
self-control
organizational awareness
likability
influence
risk taking
problem solving (that's definitely not a soft-skill)
leadership
time management

Now, which of these do you learn in school? And which ones can you learn playing on a sports team?

Thinking about an MBA? Consider this : a survey by the GMAC found that MBAs were strong in analytical aptitude, quantitative expertise, info gathering, but lacking in other vital areas - strategic thinking, comm skills, leadership, adaptability.

What soft-skills do you need to work around politics - org awareness, spreading your influence, having a mentor, managing up, deflecting gossip.

Other skills you need to succeed - being able to ask for help and ideas, listening to your gut, self-belief, being able to change game-plan and priorities, and getting back on your horse after being knocked off.

Know your bosses communication style - how often does he need updates, emails, face-to-face chats, etc.

Think you're ready for leadership? Think again - know your strengths and w's, take the initiative to address your shortcomings, influence and motive people to think and act the way you want, let others express their ideas and trust them to do the job, express your needs and goals clearly, swallow humble pie occasionally, treat everyone with r irrespective of rank. Do you enjoy working with people, helping them grow and be successful? handle uncertainty well? make decisions without all the facts? need everyone to like you? want immediate and constant reinforcement? feel nervous about having legal and financial responsibilities for others? balk at the idea of evaluating or firing someone? have the time for this?

Career tips.

Jack on Leaders

1) Relentlessly upgrade their team, using every encounter as an opp to evaluate, coach and build self-confidence
2) Make sure people not only see the vision, but live and breathe it
3) Get under peoples' skin, exuding positive energy and optimism
4) Establish trust and candour, transparency and credit
5) Have the courage to make unpopular decisions and gut calls
6) Probe and push with a curiosity that borders on skepticism making sure their questions are answered with action
7) Inspire risk taking and learning by setting an example
8) Celebrate

Curent Reading - Winning - Jack Welch

Audio book by playaway. Nice.

Mission statement is important. Must come from those who are ultimately responsible.

Focus on the behaviours that will complement your mission. Values is a bit vague. Behaviours is understandable.

"People development should be a daily event integrated into every aspect of your goings on"

Book Reviews

Managed to get through a couple during this 2 week vacation. The best part of the vacation was the early morning walk through the Theosophical society in Adyar - all the way to the beach.

Now, Discover your Strengths - Cunningham and someone else. Okay book. Might pick up one or two things. You want to focus on using and developing your strengths and find ways to manage around your weaknesses. Do not dwell on your weaknesses. Analysts are good at finding patterns in data.

How to Work a Room - Susan Roane - good tips - read one newspaper a day - there's nothing small about small talk - it's essential. Be approachable. Practise host behaviour instead of guest behaviour. Common sense + kindness = manners. Use business cards only after a rapport has been established. To join a conversation, if it's two people, use caution - they might want privacy. If it's more, they'll probably be inclusive - get to the edge and give facial feedback and see if they open up to you, etc.

Making Work Work - Julie Morgenstern - still finishing this one - but, some good points - multitasking doesn't work - it takes 20 minutes to get back to the level of concentration needed - use bread-crumb trails to help you get back faster.