3

I have a creek crossing in the woods: enter image description here This photo is in the early spring. During the summer it's a bit like a jungle. We've considered the idea of putting foot lights along the edges of the crossing. They could either go on at dark or use motion sensors to go on when a person or vehicle comes along. Solar would be a nice option since it would save the effort of running a line from the barn (in the background) or the house all the way to the crossing. (I realize that solar powered lights would not last for more than a few hours after dark.)

But I have a question that came to me after reading this answer about using solar panels to provide power for lights in a an area that does not get direct light. The answer in the link provides good information, there's a frustrating issue when you're looking at doing something like this: There aren't clear numbers for what you can do or what kind of power you can generate this way. If the question is, "What do I do to provide power to this spot for lights," and someone specifies the wire gauge to use, there's clear math to figure the voltage drop and so on. But for solar, everything is unknown.

That means researching small solar panels (including, possibly, finding ones that go with the atmosphere of the place), then buying them, setting up a test install, evaluating, and possibly returning them. Repeat as needed to see if other similar products perform better. It's guesswork.

Is there a way to do this more precisely? For instance, are there light meters that can be used to figure out how much light a spot is getting and how much power that would provide for charging a battery? I've thought about the idea of getting a small test solar panel and putting a Raspberry Pi on a battery so it can monitor the output of the panel. Of course, a regular Pi can only work but so long on a battery, so it might be necessary to go with a Pi Zero or something.

While this is one example, I have other areas in the woods (paths, garden spots...) where a small light would be helpful, so I'm looking for something I can re-use in different areas to evaluate whether solar would provide enough power. I don't mind doing math (of course a lot of that can be put in a simple spreadsheet for reusability).

The goal is to find a method, without spending a lot of money (and I already have extra rPis, so that isn't a problem), that can be used to evaluate if an area gets enough light to power small LED lights, like footlights.

Tango
  • 480
  • 1
  • 6
  • 21
  • 1
    @Criggie: This question references your answer in a recent question. – Tango May 10 '22 at 17:16
  • Looks like a case for dataloggers and collecting a couple of years data. Did this for our location by finding data from two other places and creating an "average" - came out pretty good, at least our prediction was more accurate than the software insisted on by the power company. But they paid based on their numbers so worked in our favor :) – Solar Mike May 10 '22 at 17:27
  • @SolarMike: I'm not so sure it takes years of data. Wouldn't a few weeks provide a good idea, overall? I would think getting readings on cloudy and rainy days would be enough - but I'm also talking about a small install and it sounds like you were looking at data for a large install, like providing power for your house. – Tango May 10 '22 at 17:29
  • Are you assuming cloudy days are the same every year? – Solar Mike May 10 '22 at 17:38
  • Why not evaluate the energy needed for suitable lights and work from there? – Solar Mike May 10 '22 at 17:38
  • @SolarMike I can look at the energy needed for lights, but that still means I need to somehow measure what I can get in this area to see if it's enough for that. As for cloudy days, they vary, but we're not talking lighting a home. This is about something like LED footlights for a walkway. If I check on rainy days (which usually means rain all day around here), that'll give me a good idea for what I'll get on bad days. I can also compare numbers and if it's close to "not enough," then I can look into it more. – Tango May 10 '22 at 17:43
  • There are very inexpensive light meters that can be used to spot check an area , I find motion activated solar don’t have to have as large solar panels the sensors draw less, I found in Oregon at my last home it was cheaper to trench and install lights that were hard wired, over 23 years later those same fixtures are still in service, I ran them so I could also have an override switch and turn them on for parties, there normal functionality was 3 sensors 1 at each end and 1 in the middle so when something entered the property or left all the lights went on to light the road. – Ed Beal May 10 '22 at 17:47
  • If you want to test your worst case, you’ll need to wait until winter. There is a very big difference between received solar radiation in winter vs summer. Then you have the difference between a sunny day vs an overcast day vs a stormy day. – DoxyLover May 10 '22 at 17:50
  • @DoxyLover Yes, there are differences, but in winter, there are no leaves to block the sunlight. This is in Virginia, with 75-100' (or higher) trees with thick foliage and lots of leaves. As I mentioned, in the summer, it's almost like a jungle. In the winter, almost every spot gets a lot of actual sunlight and a lot of that is direct. – Tango May 10 '22 at 18:18
  • @EdBeal Trying to be sure I get what you're saying. You think, in the long run, trenching a direct line would cost less? (Maybe I need to add that as close as that barn looks, it's over 200' away and for most of the distance, the only place to trench is in the roadway, which would be a nightmare to trench. Right beside the endwalls it would be easier to handle, plus, for that distance, I'd be using very low power DC lines.) – Tango May 10 '22 at 18:20
  • 200' is an awfully long way to haul low voltage, though. Watch that voltage drop like a hawk. – Harper - Reinstate Monica May 10 '22 at 19:11
  • I did that length with 120v and still used low wattage I originally used CFL lamps but changed to led when the cost came down, DC low voltage would be a headache and may colt more in the long run, the difference would be low voltage doesn’t require the burial depth, if I would have gone low voltage that system would be on its 2nd or 3rd rebuild, the kids said the neighbor just replaced everything again he put his in after I did because it looked so nice, I don’t know the quality he used but I am sure he has spent 2x what I did by now. – Ed Beal May 10 '22 at 19:27
  • you could run an arduino sd-card logger much longer than a pi, since it's power consumption is a tiny fraction. You can use solar for sure, the question is how much to over-spec. Triggering the lights only on motion will really slash the energy needs. Solar panels still put out 25-50% of rating in shade (per googling). In that case, a 100W panel would get 150wh/day avg worst-case. If you ran 50w of light upon motion for one hour a day, that's only a third of the available. At any rate, adding a 2nd or 3rd 100W panel if needed turns the overkill into an extinction-level event; go for it. – dandavis May 10 '22 at 19:57
  • @Harper-ReinstateUkraine: Yes. 200'. That's why I'm looking into solar options and one of several reasons I don't want to run lines. – Tango May 11 '22 at 02:11
  • 1
    @Criggie Sure! No problem, especially since it'll help someone. – Tango Jul 03 '22 at 03:26

1 Answers1

4

Maybe explore Power over Ethernet (POE) as an option. If the run is under 100 metres (328 feet) from the barn then you can get up to 15 watts per run of cable. All you need is a POE injector per run OR a POE switch.

Equip each run of wire with a POE splitter at the bridge, and you can have 5V or 12V at the far end, then hard-wire that to a low-powered LED string on either side of the bridge. Use waterproof boxes to protect all the joints, of course.

It would be more of a "being seen" light than an illumination or security light. Another advantage of POE runs is that you could run CCTV camera/s in the area and monitor from the house.

As for logging the light levels, it may be more cost effective to simply go overboard on the solar panel size, and expect to replace them every 3-10 years. Cheaper home-grade solar panels will work adequately even in the shaded areas.

In your photo, I'd put a solar panel laying almost flat on the stonework on either end+side of the bridge, for four in total. Not completely flat so that rain and condensation can drain off. There are "in-road" solar panels that you walk or drive over, but they'll be pricey and directly over a creek is likely to be a harsh environment especially in the winter. I would plan on the roadway being overtopped by the stream at some point.

If you're a coder, then consider a Raspberry Pi has many kinds of sensor available, and can also be powered by POE. This could be located at the bridge in a waterproof housing.

Criggie
  • 9,262
  • 1
  • 21
  • 58
  • 1
    That last sentence is the most helpful in this case - yes, if I can run a line that long for PoE (about 220' in this case), and power a Pi, then I can find a few light sensors. (I'm thinking I could use 2-3 different types of sensors, depending on what they each measure.) Also, the issue about the roadway being topped - it's 1" above the 100 year floodplain, but it's been topped during several rainfall events. The number of 100 year events around here is surprising! (I hear they're revising the numbers because of climate change.) – Tango May 11 '22 at 02:44
  • 1
    While I didn't want to go into it for the question, I would be careful and creative in mounting any solar panels. We have gone out of our way to give the house, the barn, and the areas in between, a bit of a "fairy tale" look - or to make them feel as if maybe they're out of a story about Middle Earth. So the technology gets disguised. – Tango May 11 '22 at 02:48
  • 1
    These aren't security lights. They would basically mark the edges of the crossing so, at night, it's easy to not wander off and trip over the chain along the edge by accident. – Tango May 11 '22 at 02:49
  • 2
    @Tango The solar panels don't have to be at exactly the same spot. They obviously need a line of sight with the sun for best performance, but you could find a nearby clearing and obscure them from the road with carefully placed bushes (as long as those don't take light away from the panels of course). – MiG May 11 '22 at 06:14
  • 1
    Climate change increases flood problems in some places and droughts in others. I'd waterproof these structures just in case. – MiG May 11 '22 at 06:15
  • @MiG Concur - anything outside will suffer from condensation and rain intrusion over time, so waterproofing and venting are both required. – Criggie May 11 '22 at 10:25
  • 1
    @Tango Bird houses make great "disguises" for cameras and panels, and might be a good home for a bird as well. Squirrels etc may be worth considering though, depends where you are in the world and what wildllife is nearby. Rock piles are good too, either covering something, or just having one fake-hollow rock in a heap of normal ones. – Criggie May 11 '22 at 10:27
  • 1
    Criggie: Good idea about bird houses! I could use small solar panels as roofs for them! I also have some 12x12x8" j-boxes that are ground level, on another project, that I have to disguise. I have a 3D printer and a CNC. I'd like to make some nice statuettes and pedestals. The pedestals would fit around the j-boxes. On them I want to put some statuettes of some Tolkien characters, like Sam and Frodo on one and Beren and Luthien on the other. I'd use the CNC to carve them from wood. I think having those in the middle of the woods would look cool. Rock piles are a good idea, too. – Tango May 12 '22 at 02:52
  • 1
    @MiG (and Criggie): This is in Virginia. Lots of rain. A lot! High humidity, too. Plus, as Criggie pointed out, the possibility of the stream topping that spot in a 100 year flood event. (I've seen something like eight 100 year floods in my lifetime - and two 500 year or greater ones.) Wildlife around here? A lot. Birds, squirrels, raccoons, fox, supposedly coyotes and bear (never seen them), deer, groundhogs, lizards, frogs - those are the things I'd have to protect against. – Tango May 12 '22 at 02:55
  • 1
    One more note on PoE: This got me excited - then I remembered that within the barn, there's a concrete tunnel and I had ethernet cables running through the tunnel, connecting the barn to the LAN in the house. I have had multiple occasions where I've lost equipment connected to those cables. So if I use PoE around there, it'd be a short term thing only for an expendable rPi. (I'm thinking of using a Pi Zero - have to check, but I hear batteries last a good while for them.) – Tango May 12 '22 at 02:57
  • @Tango if the barn is on a different power supply to the house, it may be on a different electrical phase and causing issues there with a low-voltage differential. This can result in ~50V AC on the data lines, bad for many things. You might need a fibre backhaul to the house to avoid this. – Criggie May 12 '22 at 03:08
  • 1
    There is about 400' of 4/0 wire going from the house to the barn - underground. I posted some questions dealing with that in March when one of the underground lines blew. It turned out it was at a point between the line I buried in 2017 and the rest, which I buried in 2018. I figure I either hit it with a shovel when cleaning up at that point to run the 2nd half of the coil or it got seriously kinked up. Anyway, it's all part of one system again. The data lines from house to barn are fiber optic - I just thought the 30' or so in the barn, under the barn floor, would be okay for ethernet. – Tango May 12 '22 at 05:07
  • 1
    PoE is a great idea. I would make sure the ethernet cable is well protected (sheath or duct) from both weather and wildlife. However small, there IS a current in there, so animals and insects may be drawn to it, and locating a line break over a long distance is a pain. – MiG May 12 '22 at 05:50
  • 1
    Do note the maximum distance of 100m for both PoE and data (either might still work beyond that, but they might also not). Not too familiar with PoE, but I assume any repeater would just add to the power draw so that won't be possible. – MiG May 12 '22 at 05:53
  • 1
    @MiG See my comment 5 comments up about PoE. I got excited at first, but it'd be a real risk around here. It's a shame because I was already thinking I could do a bit of digging to run it on the surface, but just under the railroad ties that line the roadway there. The conduit for that would only have to go into the ground as deep as its OD. While Pis aren't expensive, I really don't want to deal with them or the PoE brick blowing out on me multiple times. – Tango May 13 '22 at 05:18
  • 1
    Marking this as the answer - mainly due to the last part about using a Pi. I won't be using PoE, since lightning is such an issue around here, but I'll be using a Pi - still working on adapting a tractor battery with the right kind of connection. – Tango May 26 '22 at 04:07