Wednesday Feb 08, 2006

Punking the boss with a square wave

So, if you've read any of my blog, you will know that we here in Area 51 of the Labs have a penchant for pulling pranks on our boss. (See this prank for more information.) Maybe we have too much time on our hands, or maybe we're just too creative, I don't know, but the possibilities, and the imaginations around here, are just too limitless!

So anyway, last week we made a field trip to Halted Electronics which, if you haven't been, is a veritable candy store for us geek-types. Every piece of used motor, circuit, wire, etc. you can imagine. But I digress (as usual). On the $1.00 table I found a switch-plate with a single black button and a "Press for Service" sign (picture to come later), and for $0.75 a small piezo buzzer. After some work designing and building a small circuit to go on a SPOT, and some time writing some software, the pieze makes noise.

This is where the square-wave comes in. Turns out that in order to make a piezo buzz, you have to supply it with a 5+V square-wave. Lucky for us, the SPOTs have a number of GPIO (General Purpose I/O) pins. We can control the Piezo from them, but how to generate a square wave? Well, Pulse Width Modulation (PWM), of course! So the following bit of Java code will generate the requisite square-wave:

     PWMOutput output8 = new MAX6966PWMOutput(DemoSensorBoard.getIOPort2(), 8);
     for (int y = 0;y < 5;y++){
            for (int x = 0;x < 500;x++){
            } catch(Exception e){

This produces 5 'beeps' on the piezo 1/4 second apart. Perfect.

So I mounted a Sun SPOT in the switch-plate, hooked the switch to the GPIO to catch switch-presses, and then hid another SPOT in the Boss' office with the piezo attached. Walk up to the door, press the button, and the piezo squawks 5 times. Just enough to get his attention, but not long enough that he can actually locate the offending noise-maker.

So far the "Press for Service" button has come in quite handy, and has been a complete success.

There were other parts included with the switch plate which we will use for part II of the Press For Service Prank™. Details to follow ...

[!07/11 PDP a ni deppart m'I !pleH]

Sunday Jan 15, 2006

More Mythbusting

To follow up on my post about the Mythbuster Jamie Hyneman, I thought I'd mention that, while we tried to keep the 'star struck' factor to a dull roar -- something that is no small feat, to be sure -- and for the most part we did pretty well.

That is, of course, until we got to talking about the specifics of the show itself. One of the best things about the Mythbusters is that they don't just grab a myth, and test it. They actually walk through the background of the myth, and then they go through the mathematics and sometimes the physics of the myth.

The example I like best is the 4th episode about the penny dropping from the Empire State building. Will it really kill a person? Well, rather than doing the obvious and just dropping one off the top of the building, they actually took the time to calculate the terminal velocity of a penny, and the force that the mass of the penny would be capable of delivering at terminal velocity. Cool! Way better than just dropping a penny!

So you see, my son loves the show. He loves that they blow stuff up, sure, but he really loves that they do real math on the show to prove what they are doing (see the penny example above). He loves math. He's a whiz with math.

So back to the 'star-struck' part of the Mythbusters meeting (remember, that's where we started). Finally, at the end, I just had to ask Jamie for an autograph for my son. Randy would never have forgiven me had I not gotten him him one, for certain. And being the great guy that he is, Jamie agreed to have a picture taken with me, which I then printed out and had him sign to Randy. Check it out:


Neat, huh? But the neatest thing was that he actually first invited me and Randy up to his shop some weekend to get a complete tour and meet everybody. Too bad Randy still lives in North Carolina and I can't take him up on the offer. Now that would blow Randy's mind!

Friday Jan 13, 2006

SPOTs in the wild

So, as promised ... I'm going to blog more.

We have been quietly letting a few folks outside of the labs -- and even outside of Sun -- play with Sun SPOTs for a while now. Actually quite a while. The first real outside deployment, of sorts, was a Sun-sponsored class at The Art Center College of Design called "The New Ecology of Things."

It was a graduate class centered around designing retail, domestic and social futures. All using Sun SPOTs. The students were amazing in the ingeniousness of what they envisioned, and what they executed, in this class. There is a class blog that chronicled the class, and their achievements. I went down there on at least 3 occasions to view presentations by the students.

But there are others as well. We have been seeding a few select researchers and customers with small development kits (3 SPOTs, the software SDK, and a USB cable) to let them play with Sun SPOTs. The response to these limited availability programs has been, shall we say, overwhelming. I am completely buried with requests from hundreds of individuals in dozens of companies -- as well as hobbyists -- who want these things.

We've even started a public discussion forum for Sun SPOTs (complete with an RSS feed). Some of our early adopters have been posting questions there (most of which we actually answer), and I have even been posting some sample code there as well.

So think about what you might want to do with Sun SPOTs. Tell me where you think they might go. What are sensors good for? Where will they end up? What glaring hole in the world will Sun SPOTs -- or, more generally, sensors -- fill?

[All diplomacy is a continuation of war by other means. -- Chou En Lai]


Ok, so this is too cool. At least I think so. And no, I'm not one of those 'star-struck' types. I rarely even notice a 'celebrity' when I see one, have been friends with several for a years before I realized who they really were. So there. :-P

But I was lying on the couch one night surfing the web and watching Mythbusters as they were doing something with robotics (I think they were disproving the myth that when Bugs Bunny stuck his finger in a shotgun it 'banana-peeled' on Yosemite Sam. Turns out that myth is false -- who would've guessed!! -- and it just takes the hand off at about the wrist, but I digress) and I thought "hmmm ... robotics ... we do something with robotics with SPOTs ... " So I Googled Jamie Hyneman, one of the hosts, and dropped him an email. A non fan email and asked him if he'd be interested in talking about SPOTs with me.

He's a great guy, very personable and reasonable, and we exchanged a few emails before having a conference call. A few emails later, and a few months later, and I got an email from him saying he had some time and would stop by the Labs yesterday after lunch. And so he did! We spent about 2 hours talking about all sorts of projects he's working on -- turns out he gets contacted by a very diverse group of folks to talk about a very wide range of projects, as you might suspect.

Looks like we just might be collaborating with Jamie on a number of aspects of Sun SPOTs and robotics! How cool is that?

[I feel better about world problems now!]

Monday Dec 12, 2005

Sun™ SPOT Blogging

Well, it seems that since others are blogging about Sun™ Small Programmable Object Technology (Sun SPOTs), maybe I should do more of it. Especially since it would be best if the information that is out there is actually (and factually) correct.

So, to start with, Sun™ SPOTs do not actually have Bluetooth. Sorry. What they do have is 802.15.4 radios. Don't confuse that with Zigbee since they are not really the same thing.

Got all that? Confusing enough for you?

Ok, I promise (and this time I mean it!) that I'll blog more about Sun™ SPOTs from now on. If I don't, ping me and tell me to get off my butt and say more! They really are way cool!

Monday Jun 20, 2005

More Battery Woes

So I thought we had the whole battery issue all ironed out. But that's what I get for thinking. I had requested samples from a couple of vendors, and one even provided them. But $8.00 per battery is a pretty prohibitive cost, so I was waiting for the samples from the other vendor. That was 3 weeks ago, and still no samples. Here we sit spinning our wheels, waiting to move forward. When I contacted the vendor, I find out that they have not even shipped them yet. They are "having trouble delivering samples." That does not leave me with a warm fuzzy feeling about this vendor. They say that they are not having trouble delivering orders, just samples, but that doesn't make a lot of sense to me. So I'm going back to the drawing board. Starting over on the whole battery issue is not something I really wanted to do, but I guess I have no choice. With any luck I should have samples from a couple of other vendors by the end of this week.

Monday May 23, 2005

Battery Woes

You'd think that laying out a tiny little processor board, or a board covered in sensors, or getting the Java VM running on these sensors would be the hard part. But if you thought that, you'd be dead wrong. Turns out the hardest part -- at least for me -- is getting the right battery! We could stick with the Double-A Alkaline batteries we used in the RevA and RevB boards, but we would go through batteries so fast we'd fill up the landfills. Plus they look huge and clunky and they just aren't sexy enough! After all, we're building these really cool and small devices! Why would we want to hook them to 20th century batteries? So we came up with a RevC board that uses a Lithium-Ion battery, but it's soldered to the board, and its a separate battery board still. We can make it smaller, I just know we can! Randysensorreflect3.5 72 See how small it is? See how cool it looks? But, as I said, we can do better. So I've been trying to track down an appropriate Li+ prismatic battery. Those are the ones that are in your cell phone. They are square, not round, and can be very small and powerful. But it turns out that small and powerful are mutually exclusive (the science of 'Duh'). We have some minimum power requirements, and wedging those into a package that is not larger than our maximum size requirements turns out to be quite difficult. I have spent the last 3 weeks talking to major battery manufacturers, boutique battery manufacturers, OEM suppliers, you name it. They are all almost right. But not quite. But I think we've finally found one. I have to wait for the final word back from the board designer on the final size of the new board (RevD?) and then try to find just the right battery for it, but I think that I have finally found the right stuff. Time will tell.

Tuesday May 17, 2005

Too much of a good thing

So the spots are going great guns. We have written demos (in a very short time) and showed them at the Sun Labs open house. We have begun talking to Universities and other potential early adopters and .. well, we have a problem. Everyone wants them, and they want them now. But this is an early-stage research project, and we're just beginning to produce boards in large enough numbers for our own research team to play around with (and believe me, we don't even have enough for that right now. So what are we going to do? Of course, we'd love to be able to make enough to give them to everyone who wants them. We'd love to be able to sell them to anyone who wants them, but we're just not there yet. So, where, exactly, are we? Well, let's see. We are redesigning the boards to make them smaller and less bulky (if you can call a 1.5" x 2" x .5" system 'bulky'). If we can find the right sized battery, we'll be doing well. It turns out, though, that batteries are incredibly hard to deal with. We found this great little Maxell battery that was just the right size, but retail price is astronomical, and I can't find a wholesale distributor that will sell them. You would think that with the ubiquity of devices that need 3.7 volt Lithium Ion batteries you would be able to find one in just about any size of capacity, but you'd be wrong. We need a battery that has some specific characteristics. And so far, I can get 4 out of 5 of them in any one package, or fewer. But with a device that is this small, the last thing I want is for the battery size to drive the device size -- we did that with the first version where we had a giant Double-A battery pack that was bigger than any other part of the system. Not a very elegant solution.



« July 2016