Thursday, August 20, 2009

One year ago today, I handed in my resignation at GD. Well, handed isn't really the right word. More like marched down to my program manager's office and told him flat out that I was quitting. It's normally kind of a rough conversation to get started. Fortunately, my manager's manager also was leaving, and gave me a good segue: "Hey, I heard Barry gave his notice today. I've decided to do the same." I had to resist the urge to skip down the hall and do a jumping heel click as I walked out of his office. Then I was giddy for about a week. Then the panic set in. The "oh god what did I do" thoughts started swirling. I knew something had to change at GD, and that meant either the workload and company needed to change (fat chance), or I needed to leave. So leave I did. But I was concerned that I wouldn't be able to hack it at my new company, going from a hotshot to bottom rung experience-wise. Almost a year later, I still have days where I spend all day struggling with something that ends up being really stupid. But I take comfort in having eventually solved it on my own. In any case, it was definitely a change for the better to join team Sedna. And though there are stressful days, I'm certainly much happier here. I've gotten the increase in responsibility and pay that I wanted but without all the GD baggage!

Monday, August 17, 2009

The heat wave that hit DC last week and is still in full effect, plus the power outage we had last night got me thinking about a couple projects I've been mulling over for a while. Both of which involve some kind of backup power solution and/or house monitoring. The main thing I'm concerned about is the temperature sensitivity of our chinchillas, so keeping their room nice and cool at all times requires a separate air conditioner. But when the power goes out, so does their source of cool dry air, and during the summer, it gets steamy in their room quickly. So I'm looking at some way to make sure their A/C stays on and also a way to monitor the room remotely.

Project 1 - Backup power source
Option 1 - UPS - This is the "simplest" solution. Plug a UPS into the wall outlet, plug the A/C into the UPS and you're good to go. Any UPS large enough to support an A/C should also handle a florescent lamp and whatever monitoring device I come up with in Project 2 without breaking a sweat. However, UPS's aren't really designed to handle large inductive loads (like motors), so I have some doubts about how well this would actually work. They also are generally only meant to provide enough juice for a few minutes to shut down all attached loads when the power fails, so one large enough to run for say, an hour, is going to mean a huge battery and that means huge $$ as well - something in the range of $2500 by my estimation. They do have the nice feature of providing a near-seamless transition from wall power to battery power that other devices can't match, though power being off for 15 minutes honestly isn't going to hurt anything. We're more talking about outages that last longer than an hour.
Option 2 - Gasoline/LP/NG Generator - This is certainly more complicated no matter how you do the installation, but also provides a lot more flexibility. This requires getting some kind of generator, portable or permanent, finding a place for it in the back yard, possibly building a shelter for it, keeping it in working order, and ensuring that it has plenty of usable fuel. There are a lot of sizes and options out there, but if we're going to go this route, rather than having to run an extension cord across the house (bad idea) and up 3 stories (worse idea), we might as well get one we can patch into the house wiring, which means more complexity, but higher reliability and more flexibility. We could use it to provide backup power to other appliances as well, such as the fridge/freezer, internet router (we could still access internet via laptops), maybe some basic lighting, maybe even the central A/C if the genny's rated high enough. Obviously, cost and size go up as capacity increases, so we'd have to find a good balance between necessity and convenience. Between generator and installation costs, this is likely to be in the $2k-4k range. Generators can often be setup to start automatically when the power goes out, and can provide power directly into the existing wiring system, so you can pretty much whatever you want depending on which circuits have the backup wired in. Long-term, a permanent system is less hassle free, but those generally require a gas line connection, and since our house is 100% electric, we'd have to have a gas line run in from whereever, if that's even possible in our neighborhood.

Project 2 - Room monitoring
This one I'm much more excited about. Here's kind of what I had in mind: A device that has temperature and humidity sensors and potentially a camera. It would have a configurable timer and warning/critical threshold for each sensor type. And then some kind of network connectivity, preferably wifi just to make installation easy. Every 10 minutes/hour/whatever (depending on timer setting), it would poll the sensors and, if available, do a camera capture. Then it would upload this information and its timestamp to a website where I could monitor this data from anywhere. This could be as simple as leveraging the Twitter API to post the info as the amount of test should be very small. Depending on the complexity of the camera capture software, it could also do an overlay of the image with temp/humidity/time and just dump that to Flickr or Twitpic or whatever. Then the second piece to this is an alert system. Whenever the temp/humidity approaches the set threshold, it would set a warning and when it exceeds it would send a critical alert. These alerts could be the same posts, but with higher frequency and perhaps a prefix like WARN or CRIT, or they could be emails or text messages to a location I specify. All of this would ideally be configurable through a web interface, though I'm not personally above having some separate configuration software that downloads to the module via USB or something.
I know there exist standalone cameras with their own webservers built in that you just put on your network and they do their thing. And there are modules for several home-security products that have cameras and various other environmental sensors. But I'm not aware of a product that does all the things I've just described. I've been scoping out information on the Arduino and it's extension boards, which seems like it would meet most of my needs in this particular area. I think I would have to write most of the code/interface myself (which would be fun), but I think pricewise this Frankenstein is going to get out of control pretty quickly (I'm thinking in the $400-500 range when all is said and done). Anyone know of any other product that does anything even close to what I described that I might be able to re-purpose?
Temporary solution might be to go with an existing webserver camera and just point it at the Oregon Scientific weather station sitting in the room. This still has the limitation of not being able to post updates when the power goes out due to power loss in the camera and router. But hopefully that will be resolved in some manner by Project 1.

Anyway, some more things for me to start considering and researching. By the time I come up with something, summer will almost certainly be over and then I'll start worrying about how to keep the house warm instead of cool when the ice storms roll around in about 6 months.

About me

  • I'm Rev. Adam
  • From Oakton, Virginia, United States
My profile

Twitter Updates

eXTReMe Tracker