Archive for September, 2013

h1

An email to a coworker regarding instant messaging

September 27, 2013

I asked why he was never online. He replied:

The [group IM program's] notifications have trained me to ignore it since the notification are from someone who has sent an ALL message. For me, [it] is as effective closed as it is open. Is there a better way to use it? The thing I cannot have is interruptions every few minutes.

I left it alone for nearly a week, it wasn't important. But it mattered so I decided to write up the following.

If we worked in the same office, I would occasionally come up to your desk and ask if you had a minute. Then, if you said yes, we'd have a conversation that was back and forth (and often more friendly) than via email.

We don't work in the same office but IM gives us the opportunity to have conversations. I don't mind sending you emails but sometimes it is easier to have a faster flow of back and forth for little issues.

For example, if you were on IM today, I'd have asked the questions about schematic revisions I sent over email (and probably spent less time wondering if it was something broken on my end). Actually, if you were on IM, I probably would have asked that two days ago since I put it off until I was sure I couldn't figure it out on my own.

Another example: last week, I sent you two lists of schematic issues. If you'd been on IM, I would have asked if each one made sense, about an hour after sending them… since you didn't see the second one until I asked why the modifications were absent, in a group meeting, that would have been useful. Actually, [another EE at this company] and I have done the whole review via IM so he could clarify each point as I was making them (and fix them while I went to look for others). That was probably the best not-in-person review I've ever been part of.

Sure, it provides interruptions but you can always ignore or turn it off for a few hours when you are working on something hard. Having it on and available is like having an open office door. I won't interrupt you for stupid crap (unless I'm in a “let's not work, let's play nerf wars” sort of mood but that is pretty seldom since I'm a contractor, and even then you can close your virtual door).

This IM program is a part of this company's culture. I understand not liking the interruptions but urge you to give it a try for a few days, see if the benefits outweigh the annoyances.

I'm done with my soapbox. ūüôā

My coworker appeared on IM today. I hope it works for him.

 

h1

Forecasting from the Tesla

September 15, 2013

C and I were driving to the beach yesterday and started talking about the future of technology. We don’t normally do this, while we will talk about clients in our off time, we don’t just gab about tech so this was special.

Here are some of the things we decided while driving about:

  1. Virtual reality is neat and it is really, truly, finally coming. I’ll be able to put on a helmet and see a different world, be able to put virtual objects in places and move through them instead of working in a two dimensional environment. Our brains are¬†wired for this: we are supposed to live in a 3D world. When this happens, the whole world will accelerate (again) as we share our mindscapes with other people.¬†Motion sensitive people beware (that would be me, I get carsick just by thinking about being in a car).
  2. Augmented reality, where you see the real world with information overlaid on it, is also coming. The physical world will become a playground, mashed up with our mental models. Google glass is dorktastic but only the start.
  3. To get AR and VR to really work, we need to interact with the items that are not really there. Haptic feedback must come soon, driven by the virtual environments. C wants a glove that will make him believe he is holding a tennis ball when he is wearing an Occulus Rift VR headset. We talked about balance muscle wires, small solenoids (to put pressure on the fingertips), and those bed of nails things you see at museum stores with actuators.
  4. The iphone’s new 64 bit architecture is strange and interesting. Why does it need that much computing power? What purpose does it serve?
    1. First, the oomph will make it more usable as an enterprise (and government) device. It has three factor authentication now (something you have: the phone; something you know: a password; something you are: your fingerprint… and probably your voiceprint).
    2. The new iphone 64 bit architecture is on the path to the phone replacing computers. Now all they need is a generic dock (one for work, one for home) that has a keyboard and display. Then all your computer information comes and goes with you, all the time. No more laptops. No more computers. Your phone is everything, including your ID.
    3. Finally, the additional iphone computing power comes into play as a game console (like Wii, Playstation, even PCs). Add a game controller, plug the iphone into your TV, and have a serious physics engine simulate a system with intelligent agents.
  5. Energy harvesting is neat. We used to have it with solar powered calculators but now our devices are too powerful (and power hungry). The harvesting technology is coming, probably over the next five years. I think it will hit consumers in augmentation of battery devices (making them last longer), not eliminating batteries entirely. We agree that energy harvesting is more interesting in small, pervasive devices; it is less interesting in large scale energy generation where more traditional sources will continue to improve (e.g. wind, solar).

 

We’ll see how these turn out. And what are we missing?

 

h1

Why I podcast

September 13, 2013

A couple times, after we stop recording, people have asked why I have a podcast. I need a better answer than I currently have. This is all going to be stream-of-conscious as I sort out the reason and come up with an answer.

The podcast started as a way to learn something new. I was going to do a half dozen shows so I could understand how recording worked. It was a replacement for my normal community-center classes on stained glass, soldering, clay, hula hooping, laser cutting, woodshop, bookbinding, etc.

We’re way beyond six shows and I’m starting to make plans for people three months out: I’m going to be doing this for awhile. I need a better answer to the why question.

Let’s see… It isn’t “to give back”, I hear that reason from other people but I don’t feel that motivation myself. ¬†“To share my passion with other people” is closer. I really like engineering and building things. I want other people to come and play in this lovely sandbox.

I suppose that dovetails into my other reason, “I like people to share their passion with me”. Wow, that kinda came out wrong. And now that I’ve seen that, I look back to the other and snicker. Oh, go ahead and snicker, might as well laugh with me instead of making me go it alone.

Do you ever have those moments when everything is snicker worthy?

Back to topic!

Ehn, let’s circle around and see if I can put this another way.

I like Sports Night, the TV sitcom about running a sports TV show. It is smart and funny. But what I liked best is that the people on the show loved their jobs. Their jobs were annoying, sometimes dumb, sometimes hard. They were jobs. But the characters loved doing them. A common thread with my TV habits is that I like watching happy people do neat things.

Do you see where I’m headed?

I like doing the show because people share their passion, enthusiasm, amusement, happiness, spark of ingenuity, whatever-it-is, with me.¬†I think this is a solid reason and accounts for about half of “why”.

I’m a little shy so I need a way to engage with people, especially in this way.¬†I need the cover of the show to be able to go up to the head of content at EETimes and say “talk to me?”. Or to have a serious (semi-serious) talk about going to Mars with a woman I’ve never met. Or to have a good friend talk about his views on engineering without it seeming like makework.

There is another, subtler reason. I’m a proponent of women in technology (and science). I’ve heard from many sources that a lack of role models causes women to give up too easily. And I’ve seen how never having met a female engineer has warped some folks sense of the possible. I make sure to get a big cast of female engineers on my show. If we’ve have two males guests in a row, I start mining my women-in-tech contacts to make sure the trend doesn’t continue. It hasn’t been that difficult. Really.

To sum up, why I do a podcast, in order of importance:

  • To talk to people who love their jobs
  • To share my passion for engineering
  • To promote the visibility of women in engineering

Huh, ok. Those are good reasons. I don’t need to introspect any deeper. Oh, and in case our accountant reads my blog:

 

h1

Trip report: Shaping the Future of MEMS and Sensors

September 10, 2013

I went to STM's one day conference on MEMS and sensors. When C asked what I was expecting, I said I didn't know. I remember going to an Atmel mini-conf and got some neat technical information about their MCUs and a dev kit or two… but that was many years ago. Since this was less technical, less hands on, I figured I'd hear more about theory and future plans. I was still hoping for a dev kit. (I always hope for a dev kit. Even when I go to the grocery store.)

It was good that I didn't have a particular goal because it was nothing like what my unformed expectations were.

Taking a step backward, Karen Lightman and I got to talking at DesignWest this year. We've chatted a few times (once on my podcast) and she invited me to the MEMS Industry Group Executive Congress. I suppose I thought this event would be somewhat similar, a little less business focus but still a very “where is MEMS headed?” approach.

One thing I should mention is that this mini-conference was free to attend. And they fed us several times. It is at the Santa Clara Marriott which is a nice location (though I'm going to have a nightmare that involves the garish pattern of the rug). Cherche la femme. Except with money.

And yet, unlike the Atmel processor event, this has not been an overt advertisement. I'm having trouble describing what the conference has been like, hence the roundaboutness. Let me continue with what the day has not been:

  • It has not been very technical. A few sessions got into technical details but most were high level information, more like an introduction why wireless sensors are neat or how the MEMS industry has changed over time, particularly in the morning.
  • It has not been particularly future looking. (It is called “Shaping the Future”.) While there are some microfluidics actuators that I was unaware of, I haven't learned much about sensors that exist now (or that will exist in 3 months, 6 months, or 2 years.

Continuing my randomwalk to getting to the point, I misread the email and went to the Santa Clara Convention center this morning. I saw many tech people streaming in the building: backpacks, jeans, a few suit jackets, all men, mostly with bad posture. I wondered very briefly if I was in the right spot but it seemed like my normal sort of crowd so I followed them in.

 

I was wrong. The people here are more like VP of Engineering level or managers. There are application engineers and some development engineers (and consultants, ahem). There are more suites than jeans at this conference and very few backpacks (oops, though my blue backpack really is a lovely specimen of backpackdom). Shoes are mainly leather with very few sneakers, lots of nice heels. Oh, and lots of women. I don't know how they got such a good ratio but they did. Maybe MEMS has more women?

 

Anyway, these are the grownups. So why are they here?

 

The speakers seem to be trying to convince me that wireless sensors, sensor fusion, and wearable technology are important (and coming soon). Ummm… Some of that is already here, I know, I've shipped those products. For the rest, we've got all the pieces, we just need to figure out how to put it together, make it cheap and easy to use, and then ship it. I suppose I'm too tactical given that I already accept these proposed strategies.

I suppose if I wasn't already in this industry, maybe this would be new information to me. But I like MEMS sensors and I've been working with them for two different clients. And I do tend to try to be industry current and this is, I suppose, relatively current.

Still, the level of material is different from my normal conferences, not technical enough to satisfy me as an engineer and not future-looking enough to make me excited as an architect looking at far-off products. Maybe if I was more used to be only in the role of manager, this might satisfy a need to know what's new and likely to be on the schedule in the next year.

For the most part, the people I met were more interesting than the talks. But there were some good talks so let me start there.

My favorite talk was from Jon Kindred from Starkley. He spoke on hearing aids, the current technology and the future plans. He was not only a good speaker, I was jaw-droppingly interested in the technology contained in the tiny, tiny super power efficient, more smarts than my 1992 computer, in-ear-canal hearing aid. I meant to leave to go see another session but I stuck around after (even ambushed him later). So, that was good. I do so like looking at other peoples' technical underwear. And I'd love to work on that technology and application. And have him on my podcast.

I think my second favorite was from Alissa Fitzgerald. She designs custom MEMS sensors, particularly for medical applications. She talked about the risks of putting things in human bodies. I mean, the risks to the chips, a different perspective. I met Alissa doing the DesignWest sensors in health panel so seeing her talk was one of my main reasons to attend. Not disappoointed.

Finally, I enjoyed Michael Emerson's talk about Preventice, an Android phone connected EKG monitor. It made me think of the podcast with Dr. Edward White about medical technology. I wish I could get them to talk to each other (and let me listen).

As for the people, let me start with people I already knew. Mike Perkins was my boss' boss at HP (omg, that was so long ago) and then again at LeapFrog (that was still more than a decade ago). I wish he'd had his own session, I'd like to hear more about his opinions on the industry and his role at Neato Robot.

I also met Eric Wilson, my boss at the six-week, seldom-spoken-of stint at Steve Wozniak's Wheels of Zeus. I sat next to him at lunch, 75% confident it was him. He didn't recognize me. He seemed awkward about it though I felt it was funny.

Other people I met, I wonder if I'll meet them again at the MEMS Executive Council. Some of them, I think. Though that will be very business related, not technology related. I suspect that will be a very strange conference for me, outside my normal realm. I'm pleased to be going as an expert in shipping embedded products, I'm comfortable in that role though it isn't one I tend to introduce myself as.

I handed out my card, both book cards and Logical Elegance business cards. I'd love to work on hearing aids or something neat like that. But I didn't hear about a lot of super whizzy product using little-heard-of sensors. Maybe during the ending networking session.

 

 

h1

MOSFETs, games, and a demonstration of shiny object syndrome

September 9, 2013

Conversation this morning:

Me: You named one of your gadgets MOSFET... 
Him: Yes...
Me: Why? Is it something to do with Star Wars? Boba Fett?
Him: (Uncomfortable silence.) No. They are used in music amplifiers.

I keep coming back to MOSFETs. I keep seeing them in different contexts where they end up being the “oh you just put a MOSFET in there and it will work the way you want” component.

Maybe it is time I learned what this magical component does.

As a pretest, though, I’ll splat out everything I can think of.

  • When I wanted to connect a IO to a motor, the IO pin didn’t have enough current. So the EE (Phil) put in a MOSFET so I could toggle the IO line to get the motor to turn on and off without pulling current through the processor.
  • When I asked Phil about my resistor divider sucking too much power, he said I should put in ¬†a MOSFET so the divider was only on when I needed to measure the battery. I didn’t carefully read the ensuing EEspeak because I don’t have an extra IO pin.
  • A FET is a form of MOSFET. Probably one with less moss… Actually, it is probably that a MOSFET is a form of FET. Probably with more moss. Really though, there is probably no moss, it tends to be damp and that’s bad for electronics.
  • It seems to be a switch that is controlled electronically.
  • They come in n and p varieties. I don’t know the difference.
  • (And this just in!) They are used in musical amplifiers.

So, let’s be clear: I’ve used MOSFETs before. But as I try to design my own hardware, I keep getting smacked upside the head with them, like getting hit with a fish tail as the fish escapes the boat and swims away. That’s always been ok with me: I never really wanted the fish. But now I want to understand MOSFETs. And not just for a few minutes until some shiny other thought comes along.

Let’s go back to what Phil said in my resistor divider email exchange.

To prevent that leakage current you could just put a FET to turn off that voltage divider when you’re not checking the battery voltage. You could use a high-side P-Channel FET, (between VBat and R1) but turning it off solidly would require a voltage equal to (or higher than) the battery, or you could use a low-side N-Channel FET, but then you’d still have the leakage current through the ADC. ¬†Although that 50KOhm is probably only while it is sampling, not when it is not in use. ¬†So that is what I would try for a minimal-part, minimal cost solution, if you have a spare digital I/O to turn that FET on and off…

Ok. Well, I guess I retained enough of that though I suspect Phil copied the paragraph from the last five times I asked him something that required a MOSFET. I suppose what I need to do is use MOSFET in a circuit I design myself.

First, a little more information, courtesy of Wikipedia. Wow, that article is long. Let’s just come back to it, ok?

And since my attention span is tiny this morning, let me tell you about this game I’ve been playing. It is Circuit Coder for the ipad. It gives me little challenges, like build a NAND gate using only NOR gates, then I build what it wants, trying to think through the problem. Here is my half-adder.
Circuit Coder Half Adder solution

I had a decent amount of logic in college, in the CS courses, so this NAND and NOR gates are buried deep in my brain but have a solid foundation. I like the puzzle aspect. I’m a little stuck on SR latches but I have a plan to go read about them and I suspect they’ll fall pretty quickly.

Since some of the puzzles are tricky, there is a walkthrough. I was afraid to look at it for fear of taking away the puzzle aspect. However, the walkthrough is only for the first 3.5 minutes of the game so I am more likely to cheat using a computer engineering text book. And a game that can get me to look up how components work and demonstrate logic gates so effectively… this is more what I want from learning. Sneaky learning. Though, I wish the game had a little more help and could be a little more competitive (what is the minimum component solution for each?). But completing puzzles is very satisfactory.

I was hoping one of the components I need to make is a MOSFET. But looking around now, I don’t see that in Circuit Coder. (Though I do see a review¬†that rates this game highly and suggests Codea¬†as a good learn-to-code game. Whew, expensive though! And then my appolearning “trial” expired and I am considering whether to buy their (too expensive) app so I could learn about other instructional apps.)

And this is why my quest for MOSFET intuition has, to date, been for naught. I keep looking at the Wikipedia article and then finding something else to do. It is too theoretical and not tactical enough: what should I use a MOSFET for and why?

Getting away from the shiny distractions available in Wiki, I switched to looking at Charles Platt’s Encyclopedia of Components in Safari. (If you write an O’Reilly book, you get a free lifetime subscription to Safari online. Happy perk!)

There was nothing about MOSFETs in the table of contents. That seemed so unlikely for a book titled thusly. I searched for MOSFET, found some entries under Chapter 29. field effect transistor.

I bet “field effect transistor” has been used to sound very sci-fi. If I had a band (made up entirely of light theramin), it would be called “field effect transistor”.

And apparently, I should read chapters 26-28 before proceeding so I understand diodes and transistors because those are related.

I had hoped to share my new-found understanding with you in this post. Instead, I feel like I’ve wandered around in circles until exhausted. I’m going to go sit in a corner with this book and see what I can learn.

Or maybe I’ll do some paying work.

Naw, I think I’ll go run errands.

Or maybe push reload on twitter.