Thursday, October 27, 2011

Signal Integrity (SI) Education in North America

Last year I went to a career fair at a well known university in Southern California to help my company scout out some possible new grads to join the team. I must have talked to many students, and though many were very bright and certainly worth hiring them, I was surprised that only one was familiar with SI. Specially surprising since that university has a great engineering department.
Russ Dill, licensed under CC

The following month, very surprisingly, the story was repeated at another, prestigious university. I was a bit shocked.What is just a coincidence? Maybe the students I talked to missed that class because they were busy trying to cure the hangover after a frat party (no, the stereotype of the engineer as a nerd who does not party is not completely accurate).

I thought, why SI wouldn't be by now a better known area of electrical engineering? Why wouldn't it be a subject introduced in Universities?  After all, learning SI is just like a recipe. All seniors have the basic ingredients to learn it, but it is just a matter of combining all those concepts, and just like a recipe, some proper guidance makes a great chef.

In all fairness, I still need to do proper research. I've found some good SI-related topics on UCLA's page. Maybe universities have some formal SI programs or directed studies, but maybe at the grad level. Besides, SI is a relatively young discipline, and I think universities are naturally slow to keep up with technological advances.

It is important more students know about it as early as possible, since good SI engineers are hard to find, and demand for them should be strong for several years. To the student, that also means job security. Though this is not necessarily everywhere, as I'm finding it here in Vancouver (this deserves it's own post).

I know there were, are and will be more graduates like me, who would prefer knowing early on of as many career paths the electrical engineering profession offers as possible, so they can hopefully figure out which one they'll like to pursue.

Creative Commons LicenseGD

With all this in mind however, when I look back, for me not knowing at the moment what I wanted to do with my degree is what got me to learn about SI and discover my career. Like Steve Jobs said in his very famous Standford commencement speech, for me, now I'm just connecting the dots..

Monday, October 17, 2011

The "Why"

In previous post I talked about my background and how I came to learn about Signal Integrity. Now I want to talk about the why I liked it so much.

Electronics can be broadly divided into the Analog and Digital worlds. During junior year, most students start finding their area of concentration, and it seemed like most went for either the purely digital or the analog tracks. For me, I liked them both equally, and couldn't see myself just going into one single route. By the time I graduated, the best compromise I could find was DSP, and I even took some graduate courses about it. On the downside, DSP may require a lot of coding for which I was not particularly fond of. At that same time I had started with my first engineering job and found Signal Integrity.

Signal integrity was just what I was looking for as it deals with digital signals but with an understanding that those signals need to interact with the real world. The real world - for the most part - is analog.
This was no mystery before, however, before it didn't really matter.

What changed?

In 1999, the state-of-the-art Pentium II processor had a max clock speed of 450 MHz and the front side bus speed was 100 MHz. Compared to today's processors... I'd say not impressive.

Ten years later, the first iPhone had very similar specs. The number of electronic devices with an embedded computer increased dramatically, in addition, the consumer's appetite for data transfer and faster devices seems to have no end.

“Any sufficiently advanced technology is indistinguishable from magic.”Arthur C. Clarke

Photo by katerha
Just as a car going down the road at 5 mph is more maneuverable than the same going at 100 mph, slower signals are easier to send from one computer to another. At 5 mph, there's no need for a car with ABS, ESC, etc., but if you're driving at 100 mph, you're more likely to make it in one piece at your destination with all those extras.

After my first signal integrity class, I realized something that is best expressed in an anonymous quote Eric Bogatin worte in his book:

"There are two types of engineers: those who have signal integrity issues, and those who will."

This is a nice video showcasing the new Thunderbolt technology that provides 10 Gbps speed! I was fortunate to witness all the signal integrity work needed to achieve this incredible speeds.

Monday, October 3, 2011

It all started when..


As this is my first entry into this blog, I thought I'd give some of my background and how I became a “signal integrity enthusiast”, and how for me it changed from this somewhat obscure term to what it is, in my opinion, one of the most exciting and challenging disciplines and electrical / electronics engineer has to deal with now and many years to come.

So it all started when..

.. I destroyed my cousin's doll because I wanted to know how, when you pushed the back of it, the doll would say some words or start crying. OK, maybe it was too because the cries were extremely annoying. Yes, she was not amused, and yes, I did apologize.

Fast forward to highschool. All students were required to take a shop class and for me the most intriguing was electronics. I built my own variable voltage power source, a sound amplifier, made my own PCB, among other cool stuff. Unlike most of my classmates, by my junior year I knew I wanted to be an engineer. I had become a nerd!

After 5 years in college, it was getting close to graduation and like most of my classmates, it was time for job search. Like most of my classmates, Cal Poly Pomona's School of Engineering provided me with a broad set of skills which would make me ready for all of the possible paths this profession can lead you to; and like most of them, I had no good idea what I really wanted to do with my degree.

The more time one spends on a particular field, the more difficult becomes to change to a different one. My fear was that I would start working on one field which I might not enjoy doing for a long time, and then go through job hunting and take a pay cut for changing into a new field. But there was also the pressure to get a job because I didn't know if I was going to get any offers soon. I figure I could try something for about a year and if I didn't like it, the change would not be that bad.

Two Choices

I was able to line up 2 offers, and it was decision time.

Job Choice #1

Small company, working on digital design for computer hardware, a very common (or popular) type of job (at least in most Cal Poly student's minds).
 The interviews went well with standard questions on digital design. Flip-flops, counters, programming PLDs, microcontrollers, etc. Can't say I remember much detail about it.

Job Choice #2
Berg Electronics, manufactured cable assemblies for computers and telecomm equipment. The job required performing cable testing, some circuit design and assist manufacturing.
It was in that interview when I first heard of "signal integrity". I remember well that the engineering manager had asked me if I heard of it, and I said "No". Then he asked me if I knew what an "Eye Pattern" was, that one I knew because for some reason it made an impression on me in my digital communications class though it was mentioned as a side note without much detailed explanation. He said I was the only candidate that new that one.

Although the money was a factor in my decision, I also thought about the opportunity to try something different from what most of my classmates were doing, plus it was "just for a year".

My first year in that company was one of the most exciting times as I learned so much beyond electrical engineering. My boss put me in charge of a small cable assembly line for a new product and had to hit the ground running. I never imagined how much work and detailed design goes into a "simple" computer cable.

Then came the time to learn about signal integrity. I went to a class sponsored by the University of Milwaukee that was taught by one of the foremost authorities in the field: Mr Eric Bogatin. The class was an "eye opener" and I was hooked.

So it has been way more than just a year.