Solar Panel Installation by Paul Andron

Solar Panel installation

thought process:

The first thing I need to do is buy some aluminum bar stock and drill some holes in it. Then go out and measure… or… wait… no, first buy a solar panel then measure… no that’s not it either. Oh, I know! Install a controller FIRST, THEN buy aluminum bar stock, then drill holes, then measure, and then buy a solar panel!

Finally I settled on this process:

The first step is to approximate what your daily power consumption is. Sounds pretty straight forward, but it turns out that it depends much more on what kind of person you are than what kind of stuff you have that runs on electricity. An organized person would take their boat out for a week and every time they turned a light bulb on, transmitted on the radio, turned the radar on, ran the engine, or did anything else that would in any way affect the charge state of the battery. Unfortunately, even after 4 years of engineering school, and 4 years in the military, I’m still not an organized person.

First a checklist (because I like lists):

1) Figure out what you will be using the panels for. Trickle charging your starting battery while you’re away from your boat, or providing power to watch TV and run the microwave and dishwasher while cruising?

2) Budget? (In my experience any project on a boat costs exactly $1000)

3) Do an energy budget. See below for a poor example of one.

4) Where and how will you mount it? It needs to “see” the sun, the sun moves (so does the boat and the earth) and the more direct the sun the better, and yes, apparently this really matters.

5) What is your battery capacity?

6) How many battery banks do you have? Some charge controllers independently charge multiple banks. Also, it makes no sense to have 250 watts of solar panels and the only batteries you have are two starting batteries. You need somewhere to store the energy.

7) Do you really need solar panels? They’re expensive and take up space. If an engine can burn 1 gallon of dinosaur in an hour and completely charge your batteries that’s $2.80 per charge. You could charge your batteries 300 times for the cost of a solar installation! That’s once a day for 10 MONTHS! On the other hand, they are silent, elegant (in my humble opinion), don’t smell, and in theory are maintenance free and will last (assuming my 25 year warrantee is about right) 25 years.

8) Is it sunny where you live? We live in Juneau, AK. It is sunny about 30 days a year (but we’re leaving).

So I did a little “guesstimation” and came up with this as an energy budget:

Here are a couple of little formulas for a DC circuit:

amps x volts = watts

watts x hrs = watt-hrs

.

Cabin Lights: 6 watts x 3 lights x 4 hrs = 72 watt-hrs

Sidelights (LED): 1 watt x 2 lights x 8 hrs = 16 watt-hrs

Stern light: 6 watt x 8 hrs = 48 watt-hrs

Radio: 0.6 amps x 12 V x 24 hrs = 172 watt-hrs

+ 5.8 amps x 12 V x .5 hrs = 35 watt-hrs

Laptop/GPS: 30 watts x 24 hrs = 720 watt-hrs (Power adaptor says max of 70 Watts, but I’m sure it rarely runs at maximum because the screen is only on maybe half the time, and sometimes the computer is “sleeping” so I put down 30 watts just for fun. I’ll have to hook a ammeter up to it to find out for sure)

Radar: 8 watts * 4 hrs = 32 watt-hrs

Cockpit instruments: 2 amps * 12 volts * 24 hrs = 576

Stuff we don’t use much or at all (but somebody might):

Refrigerator: 35 watts * 24 hrs = 840 watt-hrs! (we usually just stick to block-ice while underway)

Stereo: 35 watts * 4 hrs = 140 watt-hrs (my Ukulele uses much less power)

So, adding up the power consumption (just the top half) comes to 1671 watt-hrs. This really is thumb and eyeball engineering here, but 1671 watt-hrs divided by an 8 hours of sunny day comes to 208 watts. That would mean we would need a 208 watt solar panel that was in direct sunlight for 8 hours a day to make up for a typical day of battery usage. By the way, you also need (1671 watt-hrs / 12 volts) = 139 amp-hrs minimum battery capacity (but a good rule of thumb is three times that) to hold that power that the panels create.

So then it was time to shop for a solar panel. There are hundreds to choose from. Big ones, little ones, rigid ones, floppy ones, some with round cells, and some with square cells. Then there are three basic types of solar panel available in decreasing order of efficiency and cost: mono-crystalline, polycrystalline, and amorphous. The way I understand it is solar panels are made up of silicone crystals (isn’t everything these days). Mono-crystalline means one-crystal. So each cell is made of a single crystal that is “grown”. So like all crystals, the bigger the crystal the more expensive. The advantage is that mono-crystalline panels are much more efficient, but of course you pay for it. Apparently, besides cost, mono-crystalline are also very sensitive to shadows. If part of the crystal is hidden in a shadow that part will actually drain power rather than provide it drastically affecting performance of the whole array. Polycrystalline is the next step down. They are made up of a bunch of smaller crystals that are somehow mashed together. Less expensive, less efficient, and less sensitive to shadowing problems. Then there’s amorphous. These are the least efficient, but have a couple of huge advantages: they’re cheap, are not affected by shadows, and can be made flexible! So you can lay them out on deck on sunny days, and roll them up and stow them on foul weather days, but they don’t last as long, and produce half as much power for the same surface area.

Too many options!

The one common thread is that it turns out 208 watts of solar panel is BIG. Not to mention VERY, VERY, VERY, EXPENSIVE. A single solar panel costs anywhere from $50 to $2000. I boiled it down a bit and a pretty good bet is $5 to $6 per watt (very rough calculation, but gives a ballpark). 208 watts worth will cost about $1100, and that’s just for the panels!

So when the rubber met the check book I decided that anything over 100 watts would be a good start and I’d get a controller (did I mention you need a controller?) that would allow for future expansion if we so wanted.

Because solar panels put out an amount of electricity proportional to the amount of sun they receive (rather than say, the amount you want, or the amount you need) you need some sort of device to keep them from frying your batteries. These cost from $30 to $300 but typically you can figure $150 or so (for the power I’m talking about here).

Here’s what we decided on (from e-Marine):

Kyocera 125 watt solar panel (this one is polycrystalline)

Morningstar 30 Amp Charge controller (way more capacity than we needed, a 15 Amp controller would have been fine).

Some other things we needed:

20 feet of 10AWG dual conductor wire (thicker wire is better because it gets more juice from the panel to the batteries)

Some other things it turned out later that we also needed:

3 pieces of bar stock marine grade aluminum

assorted nuts and bolts

assorted connectors

shrink-wrap

zip-ties

e-Marine is a great resource. Low prices, great selection. They ship FedEx though, which when something weighs in at 40 lbs may not be the most economical method. The customer service is kinda crappy too. I would send an email with ten questions in it and get an answer back that read something like “yes”, “I don’t think so.”. Meanwhile I was busy trying to spend $1000 at their store. You’d expect them to be a little more forthcoming with some ANSWERS!!! I’m off my soap box.

The Kyocera solar panel arrived and Saesha and I just stared at the thing for about 10 minutes in wonder. The construction is beautiful (I’m sorry, I’m an engineer, I get off on stuff like this), we’d shine a light at it at different angles and watch the panels glisten and shimmer. Very cool, not to mention BIG.

I then went out on deck and measured the space between posts on the Radar Arch (click on Radar Arch Installation for info on this debacle). Measuring the a-frame probably would have been better done PRIOR to ordering the solar panel, but alas that would be too sensible. Luckily Poseidon graced me with forgiveness that day and the panel would indeed fit! So now how to mount it? This was another step that would have been better performed PRIOR to ordering the solar panel, but alas that would be too sensible indeed! e-Marine conveniently makes extremely expensive mounts for the solar panels they sell, but because the crossbars on our arch are 2 in aluminum as apposed to 1 in stainless steel (like the stanchions) for which ALL of e-Marine’s mounts are designed, I had to work something “custom” out.

I called down to T&S Welding and asked for three pieces of 2 in x 3/8 in x 25.7 in (notice that I measured) marine grade (I hear this is important) aluminum bar stock to use as cross beams. I settled on 3/8 in thickness here because, well, damn-it, I’m an engineer and therefore am fully qualified to make such determinations! I settled on three cross beams because two wouldn’t be enough, and four is obviously too many.

Skip the next paragraph if you’re reading this to find some useful information…

I swung by T&S after work to pick up my shiny new aluminum cross bars! I walked in and started poking around the store. I had some time because the guy at the front desk, a guy with no front teeth, was busy trying to run a customers ATM card through his credit card machine. Industrial stores are great. They always have unique items that you’d never find in a commercial wasteland like Fred Myer or Ace Hardware or even Sears. Things like silvered gloves that would allow you to stick your hand behind a jet engine without feeling a thing. Did you know that an acetylene welding torch costs like $200? It’s two pieces of 8 in brass tubing held together with a clamp with a valve (about $5 worth of parts), but I digress. I casually mentioned something to clowns at the front desk about the card being an ATM card and that it really would matter weather they cleaned it off, put it in a plastic bag, put a piece of scotch tape on it, licked it, rubbed it, or did anything other than stick it in ATM machine. It just wasn’t gonna work (I’m sorry, I was getting impatient and was going to be late to pick up Saesha!). So with that cleared up I asked the guy with no front teeth if I could pick up my aluminum that I had called about earlier. I got a blank look for a few seconds, then saw a glimmer of remembrance (give the guy a break, it had been almost an hour since I had called him!). He said it turned out that they didn’t have any 2 in aluminum, only 2 1/2 in (I looked at the order form he was holding and saw my phone number and wondered for a second why it was that I had given that too him…). I then called Saesha and told her I might be late. Yes, I said, 2 1/2 in should be fine, and handed him my ATM card.

Twenty minutes later I was home with Saesha and my shiny new aluminum bars! Saesha had just cleaned up the boat (she does that sometimes so that I can start a new project!). So I pulled all my tools out and immediately went to work! I made some measurements and drilled a couple of holes.

My plan was to first mount the aluminum to the panel, then set the panel in place on the arch, measure, take it all down, drill more holes, then put it back up and bolt it down with some U-bolts. It worked out great, except that I didn’t drill the second set of holes in the right place (which explains why the U-bolts are slightly offset, no unfortunately it’s not some engineer thing that I know that nobody else does because I’m an engineer)… details!

Next step: wiring. The black junction box was sealed with a rubber gasket, except that it had a big hole in it. I figured out that the big hole was for the wire to go through, but also new that it would be best to make that hole water tight as well so I hopped in the old Honda civic and zipped off to Anchor Electric. The directions that came with the solar panel said something about making sure to get an IP65-rated wire clamp. The guy at anchor electric (who claimed to be, and acted like, an electrician) stared at me like I had three heads when with confidence I mentioned IP65. He handed me a grey plastic water tight wire clamp and said he was sure it would be fine. I looked at it and couldn’t disagree at $5. Maybe someday we’ll all know what IP65 is, but until then…

Wiring was easy, black-negative, red-positive, a little crimping, a little shrink-wrap, and viola! Marine grade electrical wire is expensive though. Actually, anything that says “marine” on or near it will automatically cost twice as much. We still haven’t figured out why, but I’m sure “they” know what “they” are talking about. I had ordered 20 ft from e-Marine because while at work I imagined how much I would need and my imagination came up with 20 ft. Turns out I was correct and had exactly 0 in to spare, as my dad would say (and no he’s not a farmer, he’s an economics professor, so I’m not really sure where he picked this one up) “sometimes even a pig finds an acorn”. I’m not sure why pigs are looking for acorns, but the point is 20 ft of wire was just enough. I ran the wire down the arch zip tied it along the way, drilled a hole in the rail of the boat and sealed it with a Blue Seas thru-hull wire clamp (this one must have been IP65-rated and marine-grade because it cost $20). Then the wire ran up inside the engine room, across and through the bulkhead into the quarter berth to the charge controller.

I put the charge controller in the quarter berth because it said right in the directions to mount it outside of the space with the batteries. Lead-Acid Batteries give off hydrogen gas when they are overcharged, known as “gassing”. Hydrogen gas is the same stuff that allows a person to light ones fart (no, that’s not methane, a common misnomer, and the smell comes from hydrogen-sulfide by the way, again, not methane), so seeing as how it’s flammable apparently the fear is two-fold. The obvious one is that you can blow your boat up (though I think this is rare), the other is that small pockets of gas can float into your controller (or any piece of electronics) and as the controller’s internal workings do what they do they can create tiny little sparks, which if in an atmosphere with some hydrogen can cause potentially damaging microscopic fires in the electronics. So anyway, I put the controller outside of the engine room (our AC battery charger is inside the engine room with the batteries, and has never blown up, sparked, caught fire, or failed in any way, so maybe this is all malarkey).

The last step was to wire the controller to the batteries. We run everything that charges (Alternator, Wind Generator, and now Solar Panel) to something we fondly call the “charging circuit”. Ok, I made that up just now, but it has a nice ring to it, sounds important. The “charging circuit” is a big 1-both-2-off switch so whenever something is charging we can choose whether it charges the house deep-cycle bank of batteries, or the engine battery, or “both” (risky because “both” also parallels the two banks and so all the house loads can drain the starting battery). We know it’s risky, so that makes it ok, right?

All the wires connect, one last look over everything, and turn the “charging circuit” switch to “house”, and viola! The controller came to life: the blinky lights blinked, the not-so-blinky lights didn’t blink, and even the little digital readout started reading out! I read the readout and it said .9 amps! That’s like 11 watts! Or enough for one cabin light bulb! Very exciting. Actually, considering it was the dead of winter, early morning, and in Alaska, I think that’s pretty good. The sun wasn’t even visible yet.

In hindsight, or foresight, or maybe blind sight, I think I may have to do some re-arranging. You see, the way the panel is mounted it looks straight up. That’s great when the sun is above you. After careful observation of our celestial neighbor I’ve noticed that it isn’t always straight above us. Actually, I’d say it almost never is. So I think at some point in the future we’ll have to raise the wind generator up a bit to allow some room to tilt the solar panel a few degrees to get it to point the right direction. It would involve a couple of additional holes in the aluminum cross bars (but seeing as how we have that extra 1/2 in we should be ok), and would also require an aluminum collar and maybe a 1 ft section of 2 in aluminum pipe. That’s for the future though. We’ll see how this works out!