Walt at Random

You pay the big bucks to install a photovoltaic system on your South-facing roof–paying up front, not on a mysterious contract basis. So you’ve essentially prepaid most of your electric bills for the next 20 years or so (let’s say 80%, since most systems are sized to replace 60% to 80% of the power needs).

Now you can relax and ignore your electricity use, right? After all, you’re almost never going to use more than Tier 1 (assuming tiered pricing by usage), so whatever power you do buy will be cheap.

Why do I say “photovoltaic” rather than “solar”? Well, it’s a much neater word–but also, quite a few rooftop solar installations don’t generate electricity; they’re solar water-heating installations, frequently used for swimming pools. An entirely different technology, with the only real commonalities being the use of the sun as a power source and the use of dark roof-mounted panels.

The paradox

We had good electricity habits before our photovoltaic system went live–we both grew up turning off lights when we left rooms, now that we have air conditioning we set it at a moderate temperature, we were already using CFLs in some appropriate spots, etc. In fact, that made us a poor candidate for some photovoltaic companies–while the American household average use is apparently more than 900kWH/month (900 kilowatt-hours, that is), we ran between 360 and 480 the first few months we were here (and using AC for the first time).

But here’s the thing:

• When PG&E installs the new meter, you see exactly how much energy you’re using (rounded to the nearest 10 watts, when read in conjunction with your inverter’s output reading), and total usage is in big bold numbers.
• At least for a while, it’s a bit of a game: You know your system’s rated maximum output, the inverter shows current output whenever it’s operating, you want to see how close you’re coming at various times of day and weather conditions.
• Even more of a game, actually: “Hmm. We’re generating 1,900 watts and the power meter says the utility is receiving 1,500 watts. What’s using 400 watts in the household?” You get very interested in what the house uses “at idle”–during the daytime with no lights on and no appliances or computers running.
• Somehow, even though you know you’re actually generating better than 80% of the power you’re using, even in mid-November with shorter days and possibly lower efficiency, you want–or at least we want–to see what would lower overall usage even further, without deprivation or anything.

That’s the paradox: Making a change that should make electricity use almost irrelevant, even on an environmental basis, has made us more aware of electricity use. We’re paying attention more–and I’m pretty sure we’re not alone.

The contractor we used, Solar City, makes it even easier to obsess just a little: The systems include SolarGuard, at least for the first five years–a wifi-based monitoring system that reports the system’s current and overall generation every fifteen minutes. Solar City uses that to monitor for signs of trouble–but each customer also has a web page. With a daily graph–showing, half-hour by half-hour, how much you’ve generated (and the total for the day, the total to date, and previous daily, weekly, monthly charts). Oh, and also how many dollars worth of electricity you’ve generated and how many pounds of CO2 haven’t been emitted because you’re using less utility electricity (making some assumptions about the source of that electricity, to be sure).

It’s an interesting setup. They assume you have broadband, and provide a tiny little wifi router that plugs into an available Ethernet port on your system (in our case, on our own wifi router); it communicates with a transmitter in the inverter. (I don’t know who actually builds the wifi router, but it’s “designed and manufactured in the USA”–as were our thin-film panels.)

Favorable unintended consequences

We’re a little more conscious of how and where we use electricity than we were, and we were pretty good before. Is that good, bad or indifferent? I don’t really know. (We figured out that our “idle rate” is between 50 and 80 watts–after realizing that turning on the garage lights while checking the idle rate was skewing it, since those old-fashioned fluorescents themselves use about 80 watts.)

There is, as it turns out, another unintended consequence, one that’s specific to SolarGuard, and that one’s absolutely favorable.

I’d been noticing that on some, maybe most, mornings when I first turn on my notebook PC, it could take anywhere from three to ten minutes of futzing around before I could get a stable DSL connection–sometimes, I’d have to unplug the modem or reboot it.

My wife, who uses wifi for her notebook, had noticed that some afternoons or evenings, after I’d been off my computer (or at least off the internet) for a couple of hours, she’d have trouble getting a connection, and sometimes couldn’t get one at all without rebooting the modem or the router.

I was wondering whether AT&T DSL (or the modem) was simply “forgetful”–that, if not used every so often, it would just drop the connection.

Well… SolarGuard apparently sends info every 15 minutes, whether the inverter’s actually generating power or not.

Since SolarGuard went live, neither of us has had problems connecting. At all. (Cross fingers.)

Maybe being “reminded” once every fifteen minutes is just enough to keep DSL live. If so, it’s a nice little extra.

Hmm… Lemme see here… We just passed 10kWh generation for the day. Not bad for November.

This entry was posted on Saturday, November 14th, 2009 at 4:07 pm and is filed under Stuff. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.