Location: Blogs North Pacific Coast Garden Railroad - Tom FarinRailroad Power and Solar |
 | | Posted by: Tom Farin | 4/11/2008 10:13 AM | This entry relates a solar yard light to an off-grid solar system. Then it discusses conversion of a string of low voltage yard lights using incandescent lighting to individual solar power and LED lighting.
Here's an example of one of the simplest and cheapest off-grid solar systems available.
It is a stainless steel solar powered yard light available from Harbor Freight for around $5. I stuck two of these in beds last summer. Every night (almost without fail) the two lights turned on at dusk and supplied power well into the evening. We had a sunny day yesterday. They were still shining as the sun came up this morning.
The self contined power center of the unit is a disk about 5" in diameter and about 1" thick. Visible in the top of the unit are the light sensor (small dot in the bottom left of the top), and the Photovolactic unit (square in the center of the top).
Unscrew the four screws in the bottom of the unit and open it up and you will see this.
The large disk in the left of the photo is the top half of the power unit. The smaller disk to the right is the lower half of the power unit. The left disk contains the printed circuit board that is the brains of this device. The PC board takes incoming sun-generated power from the PV unit, converts it to the 1.2 volts needed by the rechargable battery in the right side, storing the sun's energy in the battery. It also contains a charge controller that cuts off charging to the battery when it is fully charged. This portion of the circuitry operates best when the sun is out, not quite as well in cloudy conditions, storing the electricity in the battery.
When evening comes, assuming the switch on the right disk is turned on and the light sensor detects a low light condition, the brain will begin pulling current from the 1.2 volt battery rectifying the voltage to that needed by the LED and turn on the light. The LED will continue to produce light until either the battery's voltage falls below the minimum allowed by the brain, or the light sensor indicates that it is daylight.
The LED would not normally be visible in this photo. But you can see it laying just to the left of the center in the left hand disk. Its two silver leads originally protruded through the two holes dead center in the right disk. The two leads were attached to the green and blue wires laying loose in the right disk. To remove the LED, I merely shipped these wires and pulled the silver leads through the holes by tugging on the LED located on the non-visible side of the right disk.
Why did I destroy the electrical continuity of this neat little unit? Because I wanted to adapt it top an alternative use.
You don't need to know anything earlier in this entry to use a solar powered yard light. Turn on the switch, stick it in the ground where it will receive light and use it to illuminate pathways around your railroad. But this cheap $5 self-contained off-grid PV system can be adapted to a variety of other uses.
But before doing so it is useful to understand that LEDs are not light bulbs. LED stands for Light Emitting Diode. The diode is the electrical equivalent of the plumbing check valve. A diode allows current to move through the device in one direction, but not the other. A Light Emitting Diode produces light when the current moves through the diode in one direction, and is completely dark when current attempts to move in the wrong direction. So it makes a difference how a LED is wired into a circuit.
The two leads coming out of a LED are the LED's cathode and anode. Generally the cathode is marked in some way. With the Harbor Freight LED there is a flat spot on the diode's bottom next to the cathode. The cathode was connected to the green wire before the leads were cut. You don't need to know the difference between a cathode and anode. You just need to know that when you reconnect the spliced wires, the green wire needs to be reconnected to the cathode and the blue wire to the anode.
The most common reasons for separating the LED from its incoming wiring is that we want to increase the distance between the power unit and its LED or change its orientation.
One other thing about LEDs. They produce more light per unit of power than any other lighting device priced at less than a king's ransome. And they last nearly forever.
Powering Your Garden RR Structures Off Grid
Say you spent 20 hours putting together your steam engine house. To be prototypical, you want it to light up at night. You could run wires from the grid and light a grain of wheat bulb in the engine house. But what a bother. First of all you would need to run wiring. Then you would need a switch to turn the lights on and off. Then if you want to take it inside out of inclement weather, you would need to disconnect it from the grid. If it's raining and you're in a hurry, you'll probably forget, yanking wires out of the ground. As Winnie would say, "Oh Bother."
Of course, you could power that nasty power consuming incandescent grain of wheat light with a locally contained battery. You could use a light sensor to turn the lights on and off. But then you'd need to change batteries as those power hungry grain of wheat bulbs run them down. We build railroads to run trains, not to perform maintenance. Have you wondered why most traffic signals have been converted to LED lighting?
On the other hand, we could splice about a foot or so of wire between the the LED leads and the blue and green wires in our Harbor Freight power unit, mount the LED inside your structure, and place the control unit in a sunny location, maybe on the back side of the structure's roof. But there would be this big power thing mounted to the back side of the roof of your otherwise-pristine engine house. But with a bit more creativity you might be able to remove the light sensor and PV unit from the power unit, splice some additional length into the electrical lines feeding those devices and mount the electronics under the roof and just the PV and sensor on the roof.
Doing so would substantilly cut maintenance and allow you to move the whole thing inside in inclement weather.
Best yet, you could brag to your friends that your engine house has a "complete self-contained, off-grid Photovoltaic system". Then you could launch into an explanation of how an off-grid PV system works. You know, the PV unit, the electronic brain, the battery storage, and the controls over your LED light. Those bragging rights would come at the princely cost of $5 and a few hours of your time. And in a tiny little way, you'd be reducing our dependence on non-renewable energy sources and making use of our vastest renewable source, the sun.
I bought 13 of these Malibu low voltage lights at pre-winter close out at Home Depot around 6 years ago. They are all metal powder coated lights. My recollection is they listed for around $40 apiece and I paid around $20. they are as close to a railroad lantern as anything I had seen. Their boxes have been decorating my shed until this spring.
In a few weeks, the 13 light string will illuminate the walkway running through my garden railroad headed down to my pond. They represent a great little case study in the issues of using solar power in a Garden RR. I'm going to look at three alternative approaches to powering these lights.
GRID POWER
The approach I had planned to take is to pick up 100' of gauge 12 low voltage cable ($50), string the lights down the cable and power the lights with a 300 watt Malibu transformer ($100) plugged into 120V AC courtesy of the grid. Assuming my cost in the lights themselves is sunk, my total up front cost is $150 plus my labor.
But that's not the end of the costs. 13 lights with 18 watt bulbs consumes 234 watts of power. Assuming they are lit 4 hours per night, they will consume 1 kilowatt hour of grid electricity per day or 365 kilowatt hours per year. At a cost of 10 cents per kilowatt hour, my electrical cost is $36.50 per year.
Assuming I also replace six 18 watt bulbs per year, my maintenance costs (exclusive of my time) will be another $16.50 per year.
SUBSTITUTE SOLAR POWER
Assume that I power the same lights using the same low voltage power cable with solar power. I will save the cost of the transformer ($100) and the annual cost of the electricity ($36.50).
But a self-installed complete off-grid PV system is estimated to cost about $7 a watt. In Wisconsin, daily full sun solar radiation averages 4.5 hours per day. Given that I gather power 4.5 hours per day and light the lights 4 hours per night, the solar generating capacity of the PV system would need to be roughly 234 watts. At $7 per watt, the system would cost $1,638. Net of the savings from eliminating the Malibu transformer, I'd be out about $1,500 up front. Based on the $36.50 annual electrical savings, my payback on my $1,500 net investment would be OVER 40 YEARS.
You could validly argue that my payback calculation fails to consider the increasing cost of electricity over time. I could validly counter that it also fails to consider the time value of $1,500 spent now against future savings in energy costs. But our arguments would be moot. At age 62, I'm never going to see a return on my $1,500 investment.
The above example illustrates why most people feel solar power is not a cost effective solution. But this next example will prove that assumption wrong.
REDUCE POWER CONSUMPTION THEN SUBSTITUTE SOLAR POWER
Pick up any book on installing a PV system and you'll find the first recommendation in the book is to focus on reducing power consumption BEFORE sizing a PV system. That's because steps taken to reduce power consumption often have very short payback periods. A good example is replacing incandescent light bulbs with compact fluoresent bulbs. The case is so compelling that some states have enacted statutes REQUIRING the replacement of incandescent bulbs with CF bulbs.
But beyond that, a movement to PV solar power forces you to change the way you think about power consumption. You are moving from a situation where power feels like it is unlimited as long as you are willing to pay a nominal cost (10 cents a kilowatt hour where I live) for your additional power consumption. You are moving to a situation where additional power is extremely expensive ($1,500 for 236 watts of solar PV power).
I found a good example by accident in following the TOC/GR/Bachmann controversy. By the way, in spite of the fact I've crossed swords with TOC once or twice in the past, I'd defend to the death his right to free speech. But that's not my point.
I stumbled across an article TOC wrote at the GR Web site explaining a potential modification to the Bachmann K-27. It involved removing the LED from the headlight and substituting a grain of wheat bulb. That recommendation comes from a world where asthetics are important and additional energy has been really cheap. The esthetics of a more prototypical color warmth produced by the grain of wheat trumps the minor additional energy cost in TOCs mind. I understand TOCs viewpoint as I've lived there all of my life.
But were TOC to find it necessary to pay for the additional energy capacity needed to PV solar power the electrically inefficient incandescent grain of wheat bulb as compared to the very energy effecient LED, he might be forced to make a different decision. This very small example represents the change in priorities we all may be forced to make as we deal with global warming and our dependence on non-renewable energy sources.
The TOC grain of wheat/electrical cost concept directly applies to my string of yard lights. The problem with powering the yard lights with PV solar isn't the cost of PV solar power. The problem is the highly electrically inefficient 18 volt incandescent bulbs.
What if I were to substitute a white LED for the 18 volt incandescent bulb in each of my 13 yard lights? What if I was to power each LED with the self-contained off-grid solar PV unit provided as part of the $5 Harbor Freight Yard lights? How would that change the economics of my decision?
I would save the cost of the Malibu Power pack and the 100' of low voltage cable ($150). I would save the annual cost of grid power ($36.50). I would save the annual cost of replacing the 18 volt bulbs ($16.50) -- the LEDs are unlikely to burn out in my lifetime.
And it would cost me $65 for 13 Harbor Freight lights. I come out $85 ahead up front and save about $50 per year.
"But what about your labor, Farin?" I'm betting that my up front labor investment won't be significantly greater in converting my 13 lights to solar than the initial time needed to string the cable, attach the lights, debug the system and set the timer on the grid based system. Annual time savings could be significant as I won't be squatting on old legs to change bulbs and won't have to deal with interruptions in my electrical line cause by critters and shovels.
It's time to get the soldering iron and solder out. Step one will be to solder the LED to the two terminals shown in this photo.
Once I've done my first conversion, I'll post some conversion photos.
_________________________________________________________________________________
Comments by Tony Walsham
I am reading your thoughts with great interest.
Have you considered that the incandescent bulbs actually radiate light very well from the side, whereas LED's generally are very directional and don't radiate light from the side very well at all.
You may have to construct a special pcb insiide the lantern holding multiple LED's pointing sideways to radiate light through the frosted lens.
_________________________________________________________________________________
Comments by Cale Nelson
to add to what Tony said...well not really, but another thought...I use the 12v GOW bulbs in my engines, very little draw...and a nice light?
____________________________________________________________________________________
My Response
Tony,
Thank you for your comments. You certainly raise a valid issue relating the the relative light dispursion of GOW vs. LED. In my yard light application, my primary need is to illuminate the pathway through the bottom of the light rather than through the sides. So the LED will probably work in an acceptable manner in this application. I'd have to put up with the lack of significant glow through the sides of the light unliss I dispurse the light in some manner.
Lack of dispursion would be a much greater issue in lighting a structure. Some kind of diffgusion method would be needed in that application.
Cale,
I plead guilty of trying to stretch an example a little too far. What you are saying to me is that the draw from the GOW is relatively small in relation to the draw from the engine, sound board and other accessories. Fair enough.
But that raises a much larger issue in powering a garden railroad with solar. One of the criteria we may be forced to elevate in evaluating engines and their accessories in the future is the overall current draw of the eengine. In a today's world where the supply of power seems to be infiniite and the marginal cost of power is very low, they are not huge factors. Not as long as we are willing to put up with shorter run times on battery powered locos and purchase higher capacity transformers for track powered engines.
But in the future when the marginal cost of additional power is much higher than today ($7 a watt in a solar PV application), we may need to reevaluate our priorities. I wonder how the solar powered railroad linked to earlier, dealt with the issue of power usage in their model railroad.
Solar Railroad
In their site they talk a lot about the system they use to supply the power. But I don't see any discussion as to the choices they made to limit power usage before designing the system. Maybe they had enough funding that they didn't need to address the power usage issue. As you and I already stated, most of us are not in that position.
| | | Permalink | Trackback |
|
|