Rutgers Logo

Rutgers Chevy VoltRutgers Solar-to-Vehicle (S2V) Project


This website reports on a demonstration project aimed at testing electric-vehicle day-time work-place charging in conjunction with solar canopy parking structures.

Part of our emphasis is on quantifying the fraction of typical work-a-day commuter driving that can be powered on electricity that is sourced from solar generation. This has economic and green-house-gas impacts as well as providing information relevant to the design of future electric and hybrid vehicles, their battery units, and the furtherance of infrastructure to support plug-in vehicles.

This study started in December of 2012 and now more than one year of usage data has been compiled. Some recent data are discussed just below. The earlier weeks can be found at the end of this page. We hope that this information about charging availability, electric usage models, green energy, and vehicle characteristics will be useful.

Comments can be posted here in blog mode at: Solar2Vehicle.blogspot.com.

Specific questions or inquiries can be sent to: dunbar.birnie@rutgers.edu



Engineering Complex - Year 1 - Data Digestion.

 One of the more difficult aspects of this study has been logging the use of the parking spaces by non-plug-in internal combustion engine (ICE) vehicles. This has been done manually upon arrival and departure and provides at least glimpses of the actual usage of the locations -- and their potential availability for EV charging. Using the logs from the first year of our study (12/12 - 12/13) we notice a gradual increase in EV usage over time but given the placement and the other users in the same parking lot then we are pretty well saturated and looking for ways to increase the infrastructure installations on campus (keep fingers crossed!). 

Five vehicles queued up for charging

As a testament to our state of congestion -- here is a picture of 5 plug-in vehicles recently at our station -- four plugged in and one waiting. (Thanks to Robert for snapping this!).


Utilization of Parking - by hour of dayTaking the usage data for the entire year, the occupancy data were processed to provide an hour-by-hour overview of the usage of the four spots. To avoid confounding affects caused by measuring my own utilization, the data reported below are based on observations of the remaining parking spots subject to the understanding that I was typically occupying one of the spots already. The figure at right shows how the parking space utilization was as a function of time of day, where the data were grouped as: “Available”, “Blocked” (meaning occupied by a non-plug-in vehicle), and “Other EV”. This chart shows a pattern that would be typical for a university location: basically empty in the early morning, then with people leaving substantially by 6PM. And, we see pretty constant occupancy throughout the day which might be expected for a work-place location where most drivers stay for the majority of the day, though clearly there is some turnover. This shows a pretty steady usage, but the “blocked” fraction is quite significant at around 30-40% for most of the day. On average this is at least one full parking spot prevented from access for most of the day.


Chart showing usage changes through one year of  studyIt is interesting to see how the usage might have changed during the progress of the one year of study. So taking the samples from the bins covering the most usage, but then grouping them by date instead (into 2-week bins), we get the plot shown at Right. It is interesting to see that the usage by EV’s has increased significantly during this year (which was also clear by the appearance of new vehicles that hadn’t been system users when this study commenced). And, the bigger change with time is the reduction of blocking by non-EV parked cars. 

While the signage is clear that preference should be given to EV’s, there is not specific prohibition and there is no ticketing mandate in effect (and present no warnings are even being issued). However, it seems that the population of non-EV drivers at least has gradually improved at how they preserve these spots for EVs. And, this has happened in tandem with the arrival of more EV’s and so these may be helping assist in emphasizing that the EV stations are being used.



Because the start and end times are contained in the full session charging/connection reports then it is possible to identify times when more than one EV was charging – up to the maximum of four that are possible with our installation. Peak simultaneous users through one yearThe figure at right shows the daily peak simultaneous usage of the available plug-in spots as a function of time through the year. The fitted line shows a general trend toward more usage consistent with the trend shown above. Also, several days were logged when all four spaces were being used simultaneously – both Level 2’s and both Level 1’s. This happened later in the test year when more vehicles were becoming regular users of the site. Sadly there was a two-month period when one of the Level-2 units was broken (see light-blue shaded block). Still in that part of the year there were many days when all three remaining functional connection spots were simultaneously in-use for some part of the day.



 The full one-year report can be loaded here.

Specific further questions about the on-going Solar-2-Vehicle study should be directed to my email: dunbar.birnie@rutgers.edu



General Background about this Study

This study was initiated in reference to an academic article from 2009 that performed a simple estimate of the power needed for commuter transit in relation to the area of a solar panel associated with a parking-canopy solar structure. At that time most PHEV’s were converted from normal hybrid – ie., they were not production model vehicles. Now, with the advance of battery and vehicle technology there are more choices and parameters for different real vehicles. The source article analysis of this concept appears here: Journal of Power Sources, “Solar-to-Vehicle (S2V) Systems for Powering Commuters of the Future”, Volume 186, 539-542 (2009), by D. P. Birnie, III.

 

In relation to transportation electrification and mobility then:

1)      We have a very large solar parking structure (8 MegaWatts!!) on the Livingston campus (shown below left), and one ground-based array already generating about 1.4 MW.

2)      We have 2 Plug-in Vehicle Charging nodes at the School of Engineering beside the CAIT Building – one node behind Martin Hall on Cook Campus – and four more in the process of being installed in conjunction with the solar parking structure over on Livingston (#1).

3)      We have 2 Chevy Volts – bought by the Rutgers EcoComplex, through a grant provided by the New Jersey Board of Public Utilities.

4)      And, we have faculty in several departments who are interested in solar, batteries, information systems, algorithm optimization, vehicle design, alternative fuel public policy, and many other topics. Many of these faculty are networked through the Rutgers Energy Institute.


 

Rutgers Infrastructure

showing the solar array to students

Coulomb Technologies charger

Rutgers Chevy Volt

The new solar parking canopy array on Livingston Campus covers 32 acres and provides 8 MW peak power.

Plug-In charging units located at three different locations on campus.

Test vehicle is a Chevy Volt (from the RutgersEcoComplex). This vehicle has an all-electric range of around 35 miles. Longer distances are provided by gasoline.

This seed project is supported by:

1)      Rutgers Energy Institute

2)      Rutgers EcoComplex

For more information contact:

Dunbar Birnie (dunbar.birnie@rutgers.edu)



Compendium of earlier weeks’ commentary:

 

Parking and Plugging Utilization - Engineering Complex Recent Experience.

 The two charging nodes here by the Engineering complex have been the major source for day-time plug-in charging for the Solar-2-Vehicle project for the last 6 months. During this time the energy used by the vehicle and distances travelled have been logged and analyzed (see many comments below). One additional category of data that is being logged is the actual parking space occupation of the four spots here along with the duration of when EV vehicles have been plugged in. These data come from visual identification when arriving to work or leaving from work -- and recently also at times when intentionally moving the vehicle away from the EV charging spot once notification has been received that the battery is full. For the present vehicle and commuting energy usage then typical battery charging has required four hours or a bit more.

Two weeks of parking space utilization

The figure at right gives a graphical representation of the parking spot usage data for the four EV-enabled parking spaces by the CAIT building. Two weeks of data for 9:00AM to 4:59PM (16:59, miltary time) are given. Each tiny square represents a 5 minute time interval. The blue lines are for times when the Rutgers Chevy Volt was plugged in. Yellow lines are for times when other EV's were plugged in and charging. Green squares are for empty parking spots and black squares are for spots occupied by internal-combustion engine (ICE) vehicles. The green and black are only logged in a 5-minute time interval when I've been moving the car to free up the space, or when I've been leaving work for that day. 

The two week interval started May 6th, 2013 and ended May 17th, so the data covered the last couple days of the regular semester as well as all of finals period and up to the Friday before graduation.

The Results?.....:

 Based on these data and integrating over the full 9-5 workday for those 2 work weeks, then 36% of the time both of the charging stations were in use simultaneously. And, 91% of the time at least one of the charging stations was in use. There was even a short time when THREE vehicles were plugged in (two with the "level 2" plugs and a third with the regular "level 1" plug)! ---> These charging stations both have the capability of 240V and 120V simultaneous charging, though most users prefer the faster charging rate level 2 plug. This is why they were installed with proximity to four parking spaces. 

And, based on the eyewitness records (mostly during noonish to midafternoon times - see figure), then all four parking spots were simultaneously physically occupied 67% of the time (meaning no easy access to charging for any other drivers arriving during typical afternoon times). Also, 47% of the time spaces were occupied by internal-combustion vehicles during that time period.

Clearly, the usage patterns and population of potential users are continuously changing, but for now this confirms that these chargers are located in a relatively busy lot and that a relatively high level of EV usage is being experienced.


Plug and Play!

 
Rutgers day (Saturday April 27th, 2013 was a great success. A beautiful sunny day! The system organizers counted around 83,000 participants (and many of them walked by and checked out the Rutgers Chevy Volt) along with other great electric vehicles that were on display!

 Example EV and PHEVs from Rutgers Day

Here, left to right: Tesla Roadster, Tesla Model S, Tesla Roadster, Nissan Leaf, Mitsubishi MiEV, and the Chevy Volt.


Welcome to Rutgers Day!!! – General Interest Information about Rutgers

 

This is the website for just one of many activities at Rutgers aimed at understanding energy, improving energy technology, and changing policy or societal practices. Here are links to some of the other projects that you might be interested in:

         Rutgers Energy Institute

         The Volt parked near CAIT - not plugged inRutgers EcoComplex

         Center for Energy, Economic, and Environmental Policy (CEEEP)

         Rutgers Climate and Society Initiative

         Center for Advanced Energy Systems

         Rutgers Facilities and Green Purchasing

         Graduate Training Program (IGERT) inNanotechnology for Clean Energy

         Graduate Training Program (IGERT) inRenewable and Sustainable Fuels


Solar-Commuter Data Digestion (After 3 1/2 Months)

 

Now that the weather is warmer the daily commuting efficiency is much better. Less energy is expended for windshield defogging or for seat warming! Also, the battery is warmer and able to charge and discharge more efficiently (though the hot summer might be a different story!). For example, during the last week, which included four normal round-trip commute days then each day’s travel was completed using only electrical power, as charged the day before… on campus, tied in to the Rutgers grid and solar generation. For these transits the average electric fuel economy was 4.3 miles per kWhr.

 

Still, the usage patterns and congestion at the plugs is a problem and there are more users than spaces. So now, since the refill charging time is in the range of four hours, then I will be attempting to free up the space when the battery is full (as pictured at right – full and moved from the station). I’m hoping to encourage other plug-in users to swap-out when they are full – and to also encourage normal vehicles to leave those spots free for plug-ins.

Finally, I am interested in receiving comments from EV drivers and other enthusiasts at the project blog here:Solar2Vehicle.blogspot.com.

Solar-Commuter Data Digestion (After 2 1/2 Months)

 

In earlier reports (see below) I have commented about difficulties with traffic congestion, ambient temperature, charging unit availability and other EV-related issues. Now, I have collected enough commuting round-trips to start to be statistical and quantitative about the data and correlations.

 

Looking closely at the many round trips and recognizing that many factors are always at play in determining the electric usage efficiency then I can make these quantitative observations, as discussed in the following categories. But before that, as a point of pride, through careful driving and attention to energy usage I have so-far logged 5 individual full-round-trip commutes that have been powered ONLY by electricity for the ~36.5 mile trip. This reinforces the emphasis going into this project that full-solar commuting is possible for a reasonably large zone encircling the daytime solar charging destination. Now with spring arriving then I expect that many more regular commute days will fulfill this all-electric goal.

 

Ambient Temperature: After noticing that when the weather was cold the vehicle’s power management system would select the gas-powered electric generator instead of using nominally available battery power, then I started logging temperature as well as energy use and mileage. When considering only the miles travelled under electric power and performing a regression against ambient temperature then there is a moderate positive correlation. For my normal route the average electric efficiency has been running at 3.4 Miles-per-kiloWattHour (MpKWh). While there is a lot of scatter, it looks like for every degree colder, the system operates about 1% less efficient. Since this data is mostly winter-based then some of this cold-weather effect must be due to using the window defroster and the interior heating and seat warmers (though I try to “rough it” as much as I can).

 

Traffic Congestion: Using the departure and arrival times for each of the commute legs it is possible to make allowance for the level of road congestion experienced during travel. However, a plot/correlation of the electric efficiency versus travel time shows only a very slight negative slope and lots of scatter. While more congestion certainly wastes energy from stop-and-go, the present data suggest this is a relatively small effect. The morning and evening commute times are very similar as well. It is possible that less congestion leads to faster driving which counteracts the expected correlation.

 

Weather: If we average the electric efficiency for days when it was either raining or snowing, then we find a significantly lower value. For this handful of bad weather days the average mileage was about 2.7 MpKWh. Rain certainly increases the rolling resistance and there is usually increased congestion on these days as drivers have lower visibility and less vehicle control.

 

Time of Day: It is notable that the average MpKWh for the morning commute direction is about 3.25, while the value for the evening commute direction is averaging 3.62 – substantially better. Part of this can probably be sourced back to the change in temperature (the data show that the evening has been on-average about 5 degrees warmer than the morning) however the difference from AM to PM is larger than due to only this effect.

a warning ticket left for ICE vehicle in a charging spot

Type of Driving: Only a few trips have involved commute variations along the congested Route 1 corridor. Averaging these data shows that I got only about 2.6 Miles-per-KWh during these trips. Route 1 is very congested, yet it has traffic that is trying to maintain higher speeds. This usually entails more accelerating and more braking, which is wasteful of energy and results in this lower efficiency value. Another way of looking at this is that the normal back-roads routes are more suited to a passive driving style that saves energy.

 

As a final note, just this week I noticed that the Rutgers parking controllers have started to encourage people to leave the EV charging spots open for plug-in vehicles. The picture at right shows an official “Warning Only” that was left for one car. I don’t know how or if there has been a real policy change (the signs are still the same), but this is the first time that I’ve seen warnings being issued.

 

Finally, I am still wanting to gather information about WHERE there would be solar generation in conjunction with EV charging spots.….If you do know of similar locations then please share through the project blog here: Solar2Vehicle.blogspot.com.

Solar-Commuter Data Digestion (weeks 6/7)

 

My reporting has been interrupted because of a computer hard-drive failure (ouch!), but I continue to log commuting mileage, electrical and gasoline usage and to monitor congestion and/or availability of the plug-in charging spaces near Rutgers Engineering. I do occasionally run into days when the spots are all taken and I don’t get good on-campus charging. Likewise (as noted in week 5) there have been many cold days and slower commuter traffic which has made it difficult to reach full solar-electric-only commuting round trips…..but the data quest continues!

 

Row of plugged in vehicles at Trenton stationOne notable event happened last, week, though: I made the trek down to Philadelphia to attend a meeting of the Greater Philadelphia Clean Cities Coalition (one of nearly 90 regional groups promoting alternative fuel vehicles). For this trip I knew that it would be out of range for electric only travel so took the opportunity to do park-n-ride from the Trenton train station – and in the process use one of the two solar-array-outfitted parking garages at the Trenton Amtrak station. These arrays aren’t new, but when the parking garages were built in 2008 they included 6 plug-in vehicle charging spots in each of the two garages. These are all level one chargers, but are still quite nice for commuters who might leave their car during a full work-day commute into NYC or Philly. When I arrived in the morning there was already one Chevy Volt parked there and charging – but the best part was when I returned from the GPCC meeting in the afternoon I was one of 5 vehicles all parked in a row and charging! Here we are… 4 Volts and 1 Leaf taking in electricity parked under the large solar canopy!

 

So, it makes me wonder…. How many places are there where it is possible to charge electric vehicles in conjunction with solar generation? Do you have a workplace with solar canopies and plug-in? Are there other commuting hubs that facilitate this? If you do know of similar locations then please share through the project blog here: Solar2Vehicle.blogspot.com.

Solar-Commuter Data Digestion (week 5)

 

This has been a particularly cold week here in New Jersey and I’ve been experiencing poorer round-trip electrical efficiency rates – even possibly worse electrical usage in the morning when the car is colder. Part of this could be creature comforts, such as running the windshield defroster and the internal heat and seat warmer, though I really try to minimize my usage of these energy hogs. But my worry about the cold weather was amplified this morning (1/23/13) when the temperature was in the teens: The Volt’s power management system started to cycle part of the time on gasoline as well as part on electric – giving a dashboard notice of “Engine Running Due to Temperature” – also reinforcing the known worry that battery performance will be reduced in cold weather. I will try to gather more detailed morning and evening numbers, including outside temperature, for next week’s comparison.

Solar-Commuter Data Digestion (week 4)

 

Here are the data for the newest week. Total distance driven: 126 miles. Cumulative charging amount at the meter: 38.8 kWh. Of this travel, 94.6% was carried out using electricity accessed on the Rutgers campus during daylight hours. From the energy usage provided by the Volt’s user screen, I calculate this weekly electrical fuel efficiency value: 3.57 miles per kWh.

driving distance map from campusHowever, for ONE of these days the entire round trip was covered with only solar-sourced electricity – no gas!! (36.5 miles with 9.9 kWh of electrical use. This comes to a bit more than 3.68 miles per kWh. …. I’m getting better at cautious driving and being patient in traffic congestion.)

 

So, it is no surprise that for a given battery capacity it is possible to live within a range where full round-trip commuter travel can be carried out within a single battery charge. It happens that I live just at the edge of this car’s round-trip battery range and to an extent where I have to concentrate on fuel conservation during driving to reach that 100% goal. I live in Princeton and work at Rutgers Piscataway. The map at right is draw with an as-the-crow-flies radius of 15 miles around the charging spot that I’ve been using – which shows generally the set of homes around Rutgers where completely solar-sourced electric commuting might be possible with this vehicle.

 

And, as with gasoline powered driving, there are many factors that impact the overall fuel efficiency/usage. For electric vehicles we have to consider the ambient temperature (cold temperatures lower the battery capacity). Certainly our driving temperament is important, as noted above. Also, we need to use electricity from the battery for all the car’s functions: in winter it includes warming up the interior and in summer it may include operating the air conditioning – again, depending on the ambient temperature and our comfort demands.

 



Solar-Commuter Data Digestion (week 3)

Happy New Year! This third week heralded the beginning of 2013, but since the first was on the Tuesday it meant there were only three commute cycles of data that were be gathered. And, while Rutgers is back open-for-business, there are no classes in session yet and so no incidences of parking congestion were encountered.

 

I did make one adjustment in my commute: finding a shorter route. This shaved off about 2 miles in each direction! This has helped me get much closer to my aspiration of fully solar-electric commuting. I have learned from reading other drivers’ experiences that the battery range may improve when the weather is warmer.

 

Here are the data for the newest week. Total distance driven: 112 miles. Cumulative charging amount at the meter: 35.5 kWh. Of this travel, 81.9% was carried out using electricity accessed on the Rutgers campus during daylight hours. From the energy usage provided by the Volt’s user screen, I calculate this weekly electrical fuel efficiency value: 3.09 miles per kWh and for the gasoline-powered travel I found 32 MPG. (Note that the Volt’s display quantifies the electricity coming out of the battery in comparison to the electricity that was put during charging. Comparing three recent charge/discharge cycles then it looks like there is about a 17% loss of electrical energy during storage and retrieval – at least during the freezing temperatures we have during winter.)

Solar-Commuter Data Digestion (week 2)

This second week has been more of a “conventional” commuting week. With the end of the semester arriving and finals week study times then there has been less congestion in the parking lot at the times when I’ve arrived to work. So for this week I have collected energy totals for 5 round trips (plus one short mid-day cross-campus trip for a meeting), for a weekly total distance of 229 miles and cumulative charging amount at the meter of 62.68 kWh. Of this travel, 73.4% was carried out using electricity accessed on the Rutgers campusduring daylight hours.

It is interesting to see what the Volt provides as analysis of the travel. Here are two sequential days of round-trip travel reports.

 

12/19/2012 – 12/20/2012

12/20/2012 – 12/21/2012

107 MPG

227 MPG

Arriving to campus in the AM of the 20th, I had no remaining battery charge and the plug-in session that day delivered 12.075 kWh to fill the battery again. Using the Volt’s measurement of electricity consumed, I calculate 2.97 miles per kWh. The fuel efficiency for gasoline-powered travel was 33 MPG.

Arriving to campus in the AM of the 21st, I had no remaining battery charge, but had probably done a better job of driving cautiously – certainly the traffic had been relatively lighter (finals mostly over). I calculate 3.60 miles per kWh for this round-trip. The fuel efficiency for the gasoline-powered travel was 35 MPG.

 

For those who keep track of exact fuel efficiency for travel then you should understand that the “107” and “227” mpg data that were provided on these screens is (TOTAL miles) divided by (gasoline used) – Thus doesn’t really account for the electric energy used in travel. It is better to break the data into electric-miles per kWh and gasoline-miles per gallon, as I’ve done in the figure caption.

 

(Note that Rutgers campus is shut down for holiday during the week including Christmas through New Year.

Therefore, there will be no commuting and no reporting for this upcoming week.)

 

Solar-Commuter Data Digestion (week 1)

The vehicle was signed out for this project on 12/4/12. At this point, the vehicle had only 153 miles on the odometer! Because of it mostly sitting and not being driven then it immediately started up and gave this message:

Fuel is stale message
“Stale” Fuel! Obviously this was vexing since I was planning to be using mostly electric and measuring mostly kWhrs not gallons. But, as a complicated engineering system the designers have provided many hidden controls that relate to engine performance, lifetime, and safety. So, I proceeded to go with this, using it for regular commute and eventually filling the tank as a start. Hopefully this won’t reappear during the continuing effort.

 

I performed my first electric charge on Monday 12/10/12. I had filled the gasoline before that so this date really defines the starting point. After filling the battery (it took in just under 12 kWh) then the dashboard display indicated an all-electric range of 33 miles and 267 miles of gasoline range available.

 

On Tuesday I wasn’t able to charge because the parking lot here is relatively busy and although these spots are marked as plug-in spots, the university parking office does not consider them Plug-in-ONLY spots. They do get used by other cars.

 

Wednesday’s charge (9.4 kWh) was also partly limited by congestion in the lot. All the spaces were occupied when I first arrived, but I was able to swap into one at lunch time, so got a late start. Thursday and Friday I got good AM starts on charging and received 11.96 and 12.03 kWh on those two days.

 

Here is the Rutgers EcoComplex car receiving power on Thursday:

Plugged in at CAIT

 

Using the Wednesday-to-Friday commuting trips home and back (about 46 miles round-trip, depending on route), then I was able to get my first quantitative look at fuel use. If the projected range data starting those trips are accurate then I should have used 57 miles of electric-powered-travel and 36 miles of gasoline-powered-travel. Since the electricity was all accessed on the Rutgers campus during daylight hours then this amounts to ~61% of the commute being powered by the sun.

 

Clearly, the maximum possible percentage that I can achieve will be limited by my commute distance in relation to the battery size. This measurement is being made using the Chevy Volt, which has a battery size that is designed to provide perhaps 35 miles of all-electric range under the best conditions. Still my aggregate performance this week was also limited twice because of congestion in the parking lot which is limiting my overall achievable solar distance.

Readers who are interested in this study and who have questions or comments may post them on a blog that has been created at Solar2Vehicle.blogspot.com. Or to the email below.


Dunbar P. Birnie III    (dunbar.birnie@rutgers.edu)
Professor
Department of Materials Science and Engineering
Rutgers, The State University of New Jersey   
607 Taylor Rd.
Piscataway, New Jersey, 08854-8065

(c) 2013,2014   Dunbar P. Birnie, III