Does adding PV reduce a building's EUI?

PV reduces net EUI.   Net EUI should be viewed as energy use from the grid. Net EUI may not be a holistic matrix.   The embodied carbon of PV is high, I am not aware of whole life carbon study of onsite PV versus offsite renewable, e.g. solar thermal etc., regarding which one is better when it comes to whole life carbon.  Please share if you have studies regarding this?

Agree that PV reduces net EUI. But it depends on the measuring stick you are using…AIA 2030, ILFI certifications, Zero Code, LEED, etc. Time of use is becoming more important since renewable electricity is not generally dispatchable – you can’t just turn on the wind or sun to meet increasing loads, so the utilities need to add not only more renewable energy but also storage of some kind. As a thought experiment, if we installed enough solar to get the entire US to net zero EUI, we’d have major problems with the grid. But we’re in no danger of that now since PV is a small percentage of national energy use. To Luke’s point about embodied carbon, all energy generation has some embodied carbon associated with it. Putting PV on-site may make it slightly less efficient in terms of orientation, but it may be slightly more efficient in terms of reduced grid losses. https://www.carbonbrief.org/solar-wind-nuclear-amazingly-low-carbon-footprints According to this article/study: PV, Wind, CSP, and Nuclear have lower energy embodied in infrastructure and fuel than other sources of electricity. (Note that the metric is energy in this case, not carbon) -Kjell From: Luke

Kjell, Interesting!   All things the same, the data from the article suggest offsite site wind has less emboided energy over solar in whole life ENERGY.  Since: "...Wind and solar..., at 2% and 4% respectively, equivalent to EROIs of 44:1 and 26:1" The embodied energy of wind is HALF of solar and has much higher EROIs.  That questions, at least in terms on ENERGY, whether onsite solar is better choice than offsite wind.

In our Paths to Carbon Zero study (case studies of 8 projects), our research fellow Kelsey Wotila found (p19) that rooftop 'sawtooth' PV using simple aluminum frames laying on a low-slope roof had a carbon payback of less than 3 years in the south Louisiana climate.  More heroic support frames (say, the solar carports one sees with all their steel, aluminum, concrete footings) would involve more carbon up-front and presumably take longer to pay back their carbon (more on that below).

Other studies from NREL have concluded thatthe energy payback for PV modules being typically in the 2-3 year timeframe as well, but the carbon payback depends on the source pf energy used to make the modules--as noted in this study from the journal Solar Energy Materials.

That study cited in the Carbonbrief article Kjell referenced starts from energy payback but does spread the carbon emitted making and installing PV and spread them over the service life of the systems to get to an amortized carbon impact of solar.

As to Luke's question about rooftop vs ground installations:  There's a nice study by Erik Alsema (2012) "Energy Payback Time and CO2
Emissions of PV Systems" that addresses that question; Figure 1 in that paper shows that because roof-mounted systems are (literally?) piggy-backing on existing structure, they have a lower embodied energy / carbon cost.  Unsurprisingly, the payback time depends on how sunny the climate is.   Of course, the ground mounted arrays at utility scale are coming in significantly cheaper than building-mounted PV due to economies of scale and engineering costs being spread over a larger installation.

It's important to note that due to the massive scale-up in manufacturing of PVs, studies done more than a few years ago are probably out of date, and carbon intensities continue to drop.  The Log Law of Learning in action.

Love you all!   Thank you for the quick and wonderful information. One take away can be, the question that whether onsite net zero is "better" (embodied + operational energy/carbon) than offsite renewable, the answer may "depend"...   Obviously, transmission losses, how much sun, wind, etc.will need to be account for...There are data suggest PV onsite CAN be better than offsite, but wind offsite CAN potentially be less whole life energy than solar on site.  Hydroelectric is worse than concentrate solar power and nuclear offsite, and the above 3 are worse than PV onsite...also geothermal...other?  Not one size fits all?