The US Department of Energy (DOE) and its Commercial Building Integration (CBI) program commissioned Pacific Northwest National Laboratory (PNNL) to evaluate the potential energy-efficiency benefits of a move to a DC grid in commercial buildings. PNNL has published a whitepaper on its research reporting possible energy savings of 10–18% while also revealing some of the challenges to a DC grid. LED-based lighting was a key element in the research and many companies now have DC power as a standard option in their luminaire portfolios.
We’ve covered the potential of DC grids going back almost a decade and of course the AC versus DC argument dates to Edison and Tesla. Some commercial building loads such as large fans and motors require AC input while an increasing majority of the load percentage has swung to the DC side with the transition to solid-state lighting (SSL) and the prevalence of PCs and other electronics. Moreover, renewable energy sources such as wind and solar and storage technologies (read batteries) are more efficient on a DC grid.
We wrote about one DC-grid project in a Florida bank branch that went online in 2013 using solar power, and a DC grid to supply LED lighting, computer power, and more. Even then, the technology enabled a net-zero commercial building, although it lacked some of the challenges a large corporate facility might face.
The PNNL research included direct interviews with identified industry experts, a request for information that companies in the space responded to, and an exhaustive review of literature on the subject. Most state-of-the-art, energy-efficient buildings today still use an AC grid as the PNNL research reveals. So renewable wind and solar sources that produce DC energy must be inverted for use on the AC grid and therefore immediately suffer a 5–10% efficiency loss in the conversion. Then an AC-DC conversion happens at every LED luminaire, every PC, every phone charger, and more each adding another 5–10% efficiency loss.
If instead a conversion to DC happened where AC enters the building, there would be fewer conversions and simplify the concept of a building operating on a DC micro grid and batteries with the ability to operate normally even in the absence of AC from the grid. That transition would also mean fewer solar panels and batteries would be required ideally because of fewer power conversions and associated losses.
Moving to a DC grid sounds like a no-brainer win for industry and society. But nothing is ever quite so simple. We lack standards for DC distribution voltage levels. Generally, the grid has to operate at a relatively high voltage of, say, 380V, so current levels remain relatively low and minimal losses happen over the wiring infrastructure. Meanwhile, that grid voltage has to be dropped into the 24–60V range for LED lighting and to 12V for electronics.
The one technology for DC-powered lighting that has achieved some demonstrable success is Power-over-Ethernet (PoE). In PoE, computer network cables (Cat5, 6, 7) both power LED lighting and also provide connectivity to each fixture. A PoE-enabled network switch delivers the DC power. But to date, the PNNL notes that no Ethernet switches are on the market that operate from 380V DC.
The PNNL report also notes that DC grid technology for the eMerge Alliance also sets standards for a DC grid. The first two linked articles up above cover eMerge technology. But it seems to have lost some momentum in SSL, or rather eMerge proponents see a more immediate challenge and opportunity in micro grid technology.
The PNNL does note from a positive perspective that 22 different LED lighting manufacturers now have DC-powered luminaires as a standard part of their portfolio, and that list includes startups to stalwarts. The majority at 17 manufacturers are focused on PoE.
Evidently, there has been lesser buy-in on DC from the building power supplier community. The PNNL report says that only seven companies, mostly smaller organizations, make the micro grid controllers or power servers that connect renewable-energy systems to a DC grid.
There is one further positive that PNNL identified specific to LED lighting in a DC grid environment. PNNL had previously published research on AC-DC LED drivers with that component being the most likely point of failure in SSL products. Most of the failures in drivers identified by PNNL happened in the first stage of the driver where the AC-DC conversion takes place. A DC grid would eliminate that failure node and allow the LEDs to deliver their inherent long-life potential.
The report concludes with making recommendations for next steps in a DC movement including specific instructions to building owners, utilities, lighting manufacturers, and more. You can download the full white paper from the PNNL website. Gabe Arnold and Grace Pennell authored the report.
For up-to-the-minute LED and SSL updates, why not follow us on Twitter? You’ll find curated content and commentary, as well as information on industry events, webcasts, and surveys on our LinkedIn Company Page and our Facebook page.
