OK folks, this is going to be a looong post, so bear with me. We visited Sandpoint, Idaho yesterday evening to view the SolarRoadways demonstration. For anyone who wishes to visit, the City of Sandpoint has a webpage with a webcam on the demo site here:
http://cityofsandpoint.com/visiting-sandpoint/solar-roadways#ad-image-11The system as of 10:20 PDT is now installed and operating. Watching people on the webcam is interesting because all exhibit the same behavior. First, they place a foot on the panels to see if they are secure, then they look. Next, they take a tentative step onto the panels and finally decide to walk across them.
The installation commits the sin of partial panel shading. Even at 10:00 am, the panels are shaded by the building they are installed adjacent to.
Here are some details about the panels themselves. We now no longer need to guess at values. Each panel is 44W and there are 30 panels in total. So, the theoretical maximum power generation is 1320W. Based on the datasheets for the microinverters, I am surmising that the panels operate at 24VDC, since the lowest input voltage is 16VDC.
Solar Roadways experienced some manufacturing problems and, if I recall correctly, only 14 of the panels are operational. The defective ones will be replaced as they get new PCBs. Apparently the PCBs were delaminated during the vacuum curing process.
Each panel consists of two pieces of tempered glass with a hollow cavity to hold the cells and electronics packages. The electronics packages are potted in some type of clear resin (silicone?) between the two pieces of glass. There have obviously been some problems with evacuating the air out of the sandwich – as evidenced by the bubbles in the photos.
The panels are connected together by 4-pin flat trailer lighting connectors. The wires on these appear to be 16AWG, while the DC wiring back to the panel is 14AWG. Each panel is connected by two wires – and this is where things get interesting – by looking at the panels closely it clearly appears that each was originally intended to have an eight wire interface. Six of the wires have been cut off at the edge of the panel. More on this in a bit.
The installation is a bit strange in that the rated wattage of the panels exceeds the ampacity of the cabling that should be carrying the power to the panel. The installation should generate 55A at 24VDC at peak, yet only three 14AWG cables are provided back to the panel with a rated ampacity of 15A each. Even stranger, there are 5 microinverters for the three 12VDC cables. Maybe two more cables were pulled during the night, but it looked like the conduit was close to full to me.
I need to say that while I did not meet the Brusaws, the team members on site were incredibly nice and helpful. I was given full access to photograph what I wanted and even invited to stand on a panel – which I did – and it survived my immense bulk admirably.
I would urge people to be kind when commenting. The SolarRoadways employees have been absolutely working their assess off with little to no sleep getting this demonstration ready. Whatever I have to criticize technically, I certainly would not criticize the herculean efforts that good people have put forth in bringing their idea to fruition. So, good work, SolarRoadways crew!
These particular panels are installed on a sand base in nearly the same manner one would install concrete pavers. The panels are locked together with metal strips top and bottom which are screwed together. The edges that border the concrete have aluminum strips that are attached to the concrete to lock the panels into place.
Here is the third panel being laid down:
I also managed to look inside the control panel / kiosk, and this is where things get interesting. Solar Roadways has kept its topology of using EnPhase micro inverters. These inverters are the M215-IG models. I was told that the inverters connect directly to the local utility. I’m not sure that this is correct, since these are 120/240V inverters, but perhaps the utility has enabled some type of net metering.
Where things get really interesting is in the upper right hand side of the panel. One finds a total of six MeanWell 500W power supplies. This does not mean they are supplying 3000W constantly, but the fact that the power supply is more than double the rated wattage of the power generation side, should lead one to some obvious conclusions.
What I do not know at this time is what wattage the heaters are rated at.
Here are some calcs with what I know about the installation:
SolarRoadways has conducted demonstrations of the panels using a power supply to light the LEDs – even in daylight. This is surprising to me because one would think that each solar panel, at 44W, would be able to provide its own power for the micro and a couple watts of LEDs. I saw several panels that were ostensibly functional, but saw none operate under natural or artificial lighting. I have a suspicion that these panels operate as only powered devices at the present time. I will know more next week once the meter goes on line. I will be contacting the City of Sandpoint on this as well.
I honestly think SR was overly ambitious with this installation. If they had simply confined themselves to solar generation only, without worrying about microcontrollers and blinky LEDs, that this may have gone better. What the city of Sandpoint has at this time appears to be a piece of artwork, more than a power generation system.