Blog

Nov 09, 2024

Charged EVs | Renewable energy sources for off-grid EV charging - Charged EVs

Posted August 26, 2024 by Jeffrey Jenkins & filed under Features, Newswire, Tech Features, The Tech. In the previous article on off-grid EV charging, the focus was on the energy storage battery and

Posted August 26, 2024 by Jeffrey Jenkins & filed under Features, Newswire, Tech Features, The Tech.

In the previous article on off-grid EV charging, the focus was on the energy storage battery and the DC-AC inverter that supplies the EV charger (and all of the other AC loads). This time the focus will be on the renewable energy sources themselves. For reasons of practicality (and availability) the emphasis will be on solar, but wind and hydro will be briefly discussed as well for the rare cases in which they make sense.

That last statement is bound to be controversial, so we might as well get it out of the way first. The main issue with hydroelectric generation is that it will only be an option for those lucky few who either have (legal) access to a fast-moving stream/river or a decent size pond/lake at a much higher elevation on their property (and some way to dispose of the water that will be drained from said body to generate electricity). Since this is necessarily a short article, I have to do triage here and eliminate the less-practical and/or less-available options from consideration, and hydro definitely falls into the latter category, though it is otherwise an excellent option.

The main issue with wind is that it is rarely economically justifiable, and the oft-cited justification that wind can produce power when solar can’t (at night, during a storm, etc) makes this a difficult argument to make, but I’ll give it a shot anyway. Firstly, wind speed goes up with height and all types of wind turbine—whether employing drag, like the Savonius, or lift, like the classic horizontal propeller type—work best when the wind flow is non-turbulent (aka laminar)—so the turbine needs to be placed as far away from (and above) any obstructions, including trees, buildings, etc, as possible. The cost of even a bare-minimum 10 m (~33 ft) high tower will be the same as 6 or more solar panels on a ground mount, while the cleared area needed for the wind generator could just as easily support more panels—hence you’re better off getting the panels.

Secondly, most residential-scale wind generators (i.e. in the range of 200 W to 2 kW nominal rating) are spec’ed to deliver rated power at a wind speed of 8 m/s, or 18 mph (which is quite breezy!), and power output is a cubic function of wind speed, so if your average wind speed is just 4 m/s, as it is in most of Florida, then prepare to be disappointed by an 8-fold reduction in output (you can check your average wind conditions here: https://www.climate.gov/maps-data/dataset/average-wind-speeds-map-viewer).

Thirdly, high wind conditions are potentially even more of a problem, for both mechanical and electrical reasons. Turbine RPM is directly proportional to wind speed (if braking torque—that is, amperage draw—is constant) but it generally takes a really strong (and sustained) wind to cause mechanical failure. Of more potential concern is that the generated voltage is also proportional to RPM, and it is quite possible that it could rise to dangerous levels during a run-of-the-mill storm. There are purely mechanical solutions to protect against overspeed, such as governors and automatic yawing (to turn the blades away from the wind above a certain RPM), but these are generally too costly to be economical on small-scale wind turbines, so the usual solution is just to make the charge controller more tolerant of high input voltages (which itself incurs a penalty in higher electrical losses) and/or to switch on an additional load resistor to apply more braking torque. Even so, a distressing number of wind turbines fail every year from excessive speed, as a cursory search of videos will show.

Photovoltaic solar is arguably the most practical choice among renewable energy sources. It is relatively easy to trade off between efficiency, area required, complexity of the mounting system, etc, to achieve a certain amount of power output and average daily energy production.

Destruction from overproduction isn’t possible with photovoltaic (PV) solar, however, and as most of the planet receives a decent amount of sunlight per day (aka insolation), it is arguably the most practical choice among renewable energy sources. It is also relatively easy to trade off between efficiency, area required, complexity of the mounting system, etc, to achieve a certain amount of power output and average daily energy production with PV solar, and it is far easier to increase the power output of an existing PV system compared to wind or hydro.

Generally speaking, the ideal location for a PV panel array has an unobstructed view to the south (in the northern hemisphere) over as much of the day as possible (but at least during the peak generating hours of 10 am to 6 pm), and the most energy will be generated if the panels track the sun over the course of the day, while the tilt angle is varied over the course of each season. That said, these mechanically-complex sun tracking schemes provide a relatively modest increase in total energy production (10-20% is typical) compared to their costs, so mounting the panels at a fixed tilt angle (approximately the same as the latitude) pointed directly south is usually the most economical option.

The two main mounting options are on the ground or on a roof. Ground mounting is the most flexible with regards to the above considerations of tilt angle and orientation, but any obstructions that could shade the panels need to be farther away and/or shorter. Roof mounting systems tend to be a lot cheaper, and the gain in height relaxes the shading issues, but I would only consider such if the roof has a lifetime exceeding 25 years, and, of course, the roof has a slope roughly the same as the latitude and is facing south. If the bulk of the roof faces east and west then it is possible to split the panels up into two banks feeding separate charge controllers on the premise that the east-facing array will provide energy over more hours in the morning while the west-facing array will do the same in the afternoon, but the overall cost will still be higher than for a south-facing array.

Before you get too set on where to mount the panels—particularly if on a roof—you’ll need to figure out how many of them will be required to meet your average daily energy demand, and that will depend on the insolation value, which is the average number of hours per day that PV panels will produce close to their rated power (another useful search term here is peak sun hours). A good resource for such data (in tabular as well as map form) is available from the National Renewable Energy Laboratory’s website (https://www.nrel.gov/gis/solar-resource-maps.html).

For a quick and dirty ballpark estimate, divide your average daily energy use by the insolation hours to get the bare minimum of panel power required (noting that this does not account for panel aging, exceptional uses, extended periods of cloudiness, etc). For example, to supply 20 kWh per day on average at a location that receives 4 sun hours of insolation you would need a minimum of 5 kW PV power capacity, which could be from 16 panels rated for 313 W each, or 12 panels rated for 417 W each, etc. You can install more panels than that, of course, but there are diminishing returns beyond about 3 times the above-calculated number unless you have the storage battery capacity to absorb the excess energy and need to handle extended periods of cloudiness without resorting to a backup generator.

The typical solar panel available these days will deliver 36-44 V open circuit and 8-11 A short circuit, while virtually all charge controllers (whether internal to the inverter or standalone) require the voltage from the PV array to be higher than the battery voltage, so any practical off-grid array will consist of panels wired in series.

The typical solar panel available these days will deliver 36-44 V open circuit and 8-11 A short circuit, while virtually all charge controllers (whether internal to the inverter or standalone) require the voltage from the PV array to be higher than the battery voltage, so any practical off-grid array will consist of panels wired in series. Maximizing string voltage will minimize conduction (I2R) losses, and given that pretty much all PV panels made today have internal bypass diodes, the issue of the cells—or entire panel—being forced into reverse conduction when shaded is eliminated.

Wiring panels in parallel trades the big swings in voltage that result from shading of one or more panels in the string for a reduction in total string current. However, the bypass diodes are typically axial-leaded Schottky types without any heatsinking besides said leads, so they can’t really handle more than about 10 A or so, especially when baking in the hot sun. Consequently, two—or maybe three—panels in parallel are the practical upper limit. If you need more power than you can get from about 300 VDC open circuit and 20 A short circuit, then simply break up the PV array into multiple strings that each feed their own charge controller (that can all feed a common storage battery). Most of the higher-power AIO inverters have two PV array inputs, anyway, so that gives a practical power handling capacity of 12 kW right there.

The charge controller that goes in between the PV array and the battery is the final key piece of equipment to consider. As discussed in the previous article, hybrid solar/all-in-one inverters have a PV-input charge controller built into them (and likely even two), and while that is certainly convenient, it might not be the most flexible solution, and it also might not do the best job of MPPT, or Maximum Power Point Tracking. There is also a compelling argument that a separate inverter, AC-input battery charger and PV-input charge controller will be less expensive to maintain if (or when) something fails. The other concern—that it might not do a good job of MPPT—is harder to quantify in the real world unless you have two identical PV strings, with one feeding the charge controller in the AIO and the other feeding a standalone charge controller. I did just that, and while a dataset of one is hardly authoritative, I did notice that the standalone charge controller consistently extracted about 10% more power from its string compared to the AIO, and it also responded more quickly to intermittent shading from passing clouds.

That last observation points to a better/faster MPPT algorithm, which basically hunts for the best combination of loaded voltage and output current from the panels to deliver the most power, as above a certain current (which is less than the short-circuit current) the output power starts to decline, and this current is proportional to the light intensity striking the panels, hence the need to hunt for it on a frequent basis.

Finally, there are numerous ancillary items that will be required for a PV energy system, including DC-rated fuses and/or circuit breakers, combiners and/or Y-cables, surge arrestors, conduit and wire, and a whole bunch of PV connectors.

Finally, there are numerous ancillary items that will be required for a PV energy system, including DC-rated fuses and/or circuit breakers, combiners (to parallel panels at the string level) and/or Y-cables (to parallel panels at the panel level—but make sure they have inline fuses on each branch of the Y), surge arrestors, conduit and wire, and a whole bunch of PV connectors (usually—but not always!—the MC4 type).

Off-the-shelf PV combiner boxes feature a two-pole circuit breaker for each string in the array with a two-pole surge protection device on the output side (i.e. from each side of the array to the usual two or more ground rods pounded into the earth) along with knockouts along the perimeter of the box for half-inch or three-quarter-inch conduit, making them a convenient way to wire everything up. The wire that joins the panels together and leads back to the combiner needs to be specified for PV use (e.g. UL 4703 rating) but the wire that goes from the combiner to the charge controller can be of any type if the conduit is continuous and buried at least 24 inches deep, as long as it is of an appropriate gauge for the current (#12 is typically used for 10-15 A and runs of 100 feet or less; #10 is recommended for longer runs and/or if bumping up against the 20 A practical limit for a single array).

However, given the price of PVC conduit these days, it might very well be more economical to go with cable rated for direct burial (such as type UF-B or mobile home feeder) so that only short stubs of conduit (that extend 18 inches below grade) are needed at either end of the run.

When the cost of everything involved in an off-grid energy system is tallied up, the economics might seem questionable, but that will almost certainly fade into insignificance the first time you recharge your EV with it.

This article first appeared in Issue 68: April-June 2024 – Subscribe now.

The free webinar sessions from our Fall Virtual Conference are now available to view on-demand. Register for a session below to watch the recording and download the presentation.

Register

Register

Register

Register

Register

Register

Register

Register

Register

Register

LOAD MORE SESSIONS