Solar Energy Metrics: Good Bye $/Watt – Hello $/Kilowatt-Hour!

Today TI announced that it is sponsoring this year’s Solar Decathlon, to be held in Washington D.C. October 9-18.  Similarly to 2007, 20 university teams are competing in the 2009 event and I’m sure each and every team is hacking their way through home construction, product acquisition and a myriad of design changes as problems arise, are dealt with and dispatched ahead of tomorrow’s batch of “issues”…  For these undergraduate and graduate students, they are getting an early dose of real-world, private sector reality checks.

One item I’m hopeful the teams are dealing with has to do with a topic we’ve been working through at TI for some time now.  Traditionally, the solar industry measures an installation in terms of dollars per watt ($ / W).  For instance, a 200 Watt panel (otherwise known as a module) is specified to deliver 200 Watts at its peak output.  So if I put 10 of these panels on my roof, I just installed a 2,000 Watt system (10 x 200 W).  And of course, 2,000 Watts is otherwise known as 2 kilowatts (kW).  At its theoretical peak performance (with “theoretical” of course meaning this will never happen), I should expect 2kW of electricity to flow into my house on a cloudless, shadeless, sun shiny day, theoretically speaking… (realistically it’s less than 1.6kW, but that’s a subject for a different day)  Let’s say the panels cost me $500 each, the 2kW solar inverter (the box that converts the solar power from DC to AC, among other things) costs me $2,000 and the installation contractors charged me $10,000 for labor, wire, mounting systems, license fees, utility hook-ups, and all the other stuff considered to be “balance of system costs” in solar installation lingo.  Add this up all up ($500 x 10 + $2,000 + $10,000) and I have a total bill of $17,000, before any tax incentive credits, rebates or any other cost savings measures are considered.  If I take my $17,000 figure, and divide by my 2,000 watt system, I get the traditional solar industry measurement, in this case, $8.50 / watt. 

You hear the $ / Watt ratio talked about quite a bit amongst press and analysts.  The number is regionally dependent, and sits between $5 and $9 per watt in most places right now, before federal, state or local rebates and incentives.  Many in the industry are pointing to $1 / watt and below as the magical number the industry needs to hit in order to hit grid parity and wide-spread consumer and commercial adoption of solar energy.  The problem with a $ / Watt metric is that it fails to take into account the amount of usable energy that is generated for a given solar installation.  Buying and judging a solar installation based on $ / Watt is akin to buying a computer today based solely on its posted processor speed.  A 2 GHz computer for $1,000 is $500 per GHz.  It’s an interesting number, but it doesn’t tell me anything about boot time, web page download speeds, red-eye picture editing rate, DVD burn time, etc…

You’ve probably noticed that you are charged for electricity in terms of kilowatt-hours (kWh), an aptly named unit that assigns a price for using 1000 Watts over a period of 1 hour.  By definition, grid parity occurs when the amortized price of solar electricity generated ($x.xx / kWh) equals the amortized price of say coal powered electricity ($x.xx / kWh).  In my neck of the woods, coal power averages somewhere around $0.10 / kwh at my electric meter, and again this can vary widely by region.  I need to know how my solar electricity stacks up against $0.10 / kWh, thus, I have my first issue with the industry standard $ per Watt metric for my solar installation – it gives me an apples versus oranges number in terms of electricity rates.  So I need to convert my $8.50 / Watt solar installation cost into $x.xx / kWh to see how I’m doing.  This is where it gets very muddy, as not all solar installations are created equal.  Check out this scenario of three different solar installations:

Installation A is my system, with quality solar cells and inverter control and efficiency, electronics, reliability, etc… and it yields $0.28 / kWh.  Installation B is my installation with worse performing solar cells and inverter control and efficiency, etc…  Installation C includes potential technology advances that we hope to see in the next few years at both the solar cell and inverter electronics level, resulting in a $0.20/kWh electricity rate.  Notice that all installations have the exact same $ / W, but their average daily power output varies greatly, leading to widely varying $ / kWh.  At the end of the day, $ / kWh is what we all care about. 
It’s the metric our wallets live by today, and it will be the ultimate judge for the economic competitiveness of solar energy in the years to come.  Varying installation techniques, inverter topology architectures, solar cell efficiency, control electronics and overall system installation cost will determine the return on investment for solar energy.  We are dependent on a balance of economies of scale and engineering innovation to get our $ / kWh and solar installation ROI’s on par with grid power pricing.  Given the recent boom in industry capital and research, grid parity is getting within our reach.

Chris Schairbaum

Director of energy technology strategy