Interview with Olin Lathrop

Featured Engineer

Olin Lathrop

Olin Lathrop - Principal at Embed, Inc.

How did you get into electronics/ engineering and when did you start?

I have been interested in electronics and technical things since grade school. When I was in first grade my father got a crystal radio kit which we built together. I didn’t undestand how it worked then, but was fascinated by it and had a great time building it. The best thing I could get for birthday or Christmas was a Heathkit. Over time they got more complicated and I learned more how they worked. I was given some “experimenter” kits and acquired parts from other places, like unsoldering them from equipment others had thrown out. Back then it was all point to point parts between eyelets on the bottom of tube sockets, so it wasn’t too hard to harvest parts from old TVs and the like.

In 7th or 8th grade I got a book from the library about how transistors worked. A lot of stuff was over my head, but I kept going over sections and was able to pick up some things. I didn’t really know what I was doing, but tried to make a audio amplifier with harvested transistors. I could get some gain, but whenver I added more gain it would make all kinds of horrible noises on its own. After much frustration and poking around I found that putting a capacitor accross the power helped a lot. It’s obvious now what was going on, but back then that was totally unexpected.

As I learned more and more about electronics thru high school and college, I built various things on the side. One time I made a small AM transmitter that injected the carrier onto the power line in the dorm. We’d play radio station on Sunday evenings. I almost got into trouble when one of our audience participation events was to have everyone flush their toilets at the same time. It shook 100 years of crud from the pipes in Troy NY, and it took 20 minutes for the water pressure to get back to normal.

What are your favorite hardware tools that you use?

Whatever is needed for the job. Obviously a scope gets used a lot. We do microcontroller circuits, so in-circuit emulators and debuggers get used frequently too. A variable benchtop power supply can also be useful.

What are your favorite software tools that you use?

Again, whatever is needed for the job. I have created a lot of my own software tools over time, so of course I use those when they are appropriate since I’m very familiar with them and can modify them as needed. For example, I have created a rather extensive preprocessor for PIC assembler code.

What is the hardest/trickiest bug you have ever fixed?

Hmm. There have been a number of tricky ones over the years, each one different. A few times I’ve had revelations about particularly sticky problems while taking a shower in the morning. One such time in college we had a computer system that would randomly fail once a day or so. We set up triggers for power supply glitches, added routines to the kernel to try to find what was going on at the time, checked the disk (300 Mbyte removable platters back then), etc. This went on for weeks, and we only knew a bunch of things it wasn’t, which didn’t seem to leave anything. Then one morning I realized it was a bad spot on the paging disk, which wasn’t checked with the regular disk check utility.

Nowadays the toughest bugs are ones the customer reports that we can’t reproduce here. Fortunately this is rare.

What is on your bookshelf?

Looks like a tape dispenser, stapler, inflatable globe, 3-hole punch, a box of business cards, and two pads of post-it notes of different sizes. Why do you ask?

If you’re wondering what I read or use as reference, the answer is mostly datasheets. Nowadays if you want a introduction to some topic, it’s easy to find on the internet. When you want details, general texts don’t help much anyway. I’ve got little respect for books that just regurgitate the datasheet for you in a supposedly easier to digest fashion. At best they don’t garble anything.

Do you have any tricks up your sleeve? (special way to analyze circuits, special process you use to make something, etc.)

I don’t think so. The best thing is to really understand how a circuit is supposed to work or the science behind what it’s connected to. Everything else follows from that. To diagnose a problem, the strategy is to do experiments that divide down where the problem could be. Eventually you find the single thing that isn’t as it’s supposed to be. The two biggest mistakes I see others making in chasing bugs is to not assume it’s their fault, and to jump around randomly chasing whims about what might be wrong.

As for analyzing circuits, I use my trusty HP 11C. I can usually get a pretty good idea what’s going on with some rather simple computations long before the young whipper-snappers have finished entering the circuit into a simulator.

What has been your favorite project?

There have been so many, it’s hard to say which one was my favorite. Current things I’m doing on the side include making a flying wing glider stable with active feedback in a microcontroller, developing a new computer language called “M” for embedded systems, a comprehensive home energy management system, and some smaller things that come and go.

Do you have any note-worthy engineering experiences? (blowing up things, getting shocked, etc.)

I’ve gotten shocked and blown things up a few times, but nothing spectacular. That’s probably because mostly I don’t work on high power or high voltage stuff. About 10 years ago I built up some quick and dirty circuit on a breadboard. Something wasn’t quite right, and as I was leaning over it to take a good look the electrolytic cap accross the power supply blew up with the usual bang and bad smell, but also ejected something straight up. I was really lucky it didn’t get into my eyes. I had a welt on my forehead for a week.

What are you currently working on?

We have about a half dozen projects going, although I’m not going to identify the customers or get into details. Since electronics and microcontrollers get into lots of places, we work with varied end applications. We’re working on things now that include systems for plating very thin layers of stuff onto silicon wafers, active gas pressure controllers, a system to encourage hand washing in hospitals, active RF ID tags and receivers, high tech flashlights for the military, a x-ray tube controller, and even a high tech toilet.

What challenges do you foresee in our industry?

A problem I see coming is not enough new people entering the field. When I grew up there was the space program and people doing cutting edge technology stuff that were featured as heros you could aspire to. John Glenn’s flight and Apollo 8 made huge impressions on me. Who is there today? We make a big deal of actors and atheletes, but they aren’t going to sustain the economy of the future. We need to make it not only OK to be interested in technology, but it should be cool and there should be a few people singled out for exceptional accomplishments in a public way that kids can aspire to. Putting a teacher in space to give lessons from the space shuttle is a good idea, but we’re not doing anywhere near enough of that sort of thing.