Thursday, July 17, 2014

Solar Power 101

Over the past day or so I made a number of postings on as part of a discussion about solar power on boats.  It occurred to me that it might be interesting to our blog readers, so here is a reconstruction of what I wrote over there....

As a little background, we have a house in VT that is 100% off the grid, and solar powered.  It has an inverter, batteries, solar panels, and a backup generator.  For all intents and purposes, it's a boat that ran aground.  I have now designed and built three generations of power systems for this house, which is over 200 years old and was NEVER electrified.

The first generation system was about 15 years ago and was mostly a feasibility experiment.  My brothers and I would go to the house for weekends, drink too much, and I was sure it was just a matter of time until someone knocked over a candle, oil lamp, or lantern and burned the place down.  So I set out to install electric lights.  A few years earlier as part of some renovations, we wired the house for the first time, even though there was no utility power.  But it allowed us to plug in a generator and have normal power in the house for a few hours here and there.

The first inverter system, ironically, was all based on marine equipment.  Anyone remember the Heart Interface Freedom 10?  Well, that was the inverter/charger, coupled with 4 deep cycle batteries that I bought at Walmart and a Blue Seas DC breaker.  Battery recharge was only via the generator.  It worked great and proved viability, but was under-sized and needed solar panels.

Generation 2 was a more elaborate and more integrated system.  The goal was a system that would allow the house to be used like a "normal" house, and be sufficiently automatic that anyone could use it, not just me.  The new inverter/charger was a Trace SW4024, an main-stay device in the boating world, and the off-grid gold standard of the time.  We added a modest solar array totaling only about 800W, and an Onan 4KW propane RV generator that I built into a small shed away from the house.  Batteries were 8 Trojan L16 flooded lead acid.  The Trace inverter provided automatic start/stop control over the generator, so the system would tend to itself.  It worked great, and ran the house for about 8 years.  By year 7, one of the batteries had to be replaced, and by year 8 others were going too.  Plus, we were using the house much more and needed more solar capacity.

Generation 3 has been running now for coming up on 7 years (wow, how time flies), and consists of 3200W of solar, a Schneider XW6048 6KW 240/120V inverter, and 1300 Ah of giant Surrette batteries at 48V.  The backup generator is now a 14kw Kubota diesel commercial set.  This system also works great, with essentially zero generator run time between April and October.

But enough of that.  Let's get back to the questions about solar on boats.

Should I install solar on my boat?

Lots of people wonder about the merits of installing solar on a cruising boat, and more particularly a power cruising boat.  There are a variety of arguments against solar, all of which are valid in some way.  The arguments usually are one or more of the following:

Argument 1) My boat has a large enough power load that I run a generator 24x7, so why bother with solar.  If your boat and its power requirements are like this, then I can't argue with you.  Let that generator rip.

Argument 2) I don't run my generator all the time, but my power load is still too large for solar to make any real difference.  This might be true, but might not HAVE to be true.  Lots of boats are built with a seemly total disregard for power conservation.  Stylish halogen lights versus LED lights is a good example.  If your boat is like this, but you are still interested in reducing your generator run time, then it's well worth looking carefully at ways to reduce you power loads to the point where solar can make a meaningful difference.

Argument 3) There is very little space where I can reasonably locate solar panels on my boat.  Small space means little generated power, and no real benefit.  This is very real on most boats, and goes hand in hand with (2) above.  Most boats have very limited space that is flat and free from traffic, and this inherently constrains how much solar can be installed.  Between load reductions and space dedicated to panels you need to be able to generate enough power to make a difference in your cruising or it's not worth it.  Typically "making a difference" translates into reducing the amount of generator time while at anchor.

Argument 4) I already have a generator and it's cheaper to run it than to buy solar.  This generally isn't true.  As an example, I'm putting 750W of solar on my Nordhavn. The parts cost for the panels, mounting rails, clamps, and feet, plus a top end MPPT charge controller was $1500. It will produce about 3KWh per day, accounting for weather etc. Some days more, some days less. A diesel generator produces about 10KW/h for every GPH, so at $4/ gal of diesel that power costs $0.40/KWh. And that's the fuel cost alone without figuring in the costs of maintenance and amortizing the cost of equipment, so that number heavily favors the generator. I recently heard about a Nigel Calder article that figured the fully loaded cost of generator power at more like $1.00/KWh. Regardless, with the solar displacing generator power at $0.40/KWh, payback will be in about 3 years.  That's a pretty good payback for any form of capital investment.

But arguing in favor of solar is more than just the hard financial numbers.  There are a number of other factors to consider.   For example, I don't like listening to a generator running if I can help it. If solar can reduce gen time while in a peaceful anchorage, that alone is worth it to me.

Then there is the whole issue of top-off charge for your batteries. Nobody wants to run their generator long enough to fully charge the batteries, so we all typically charge to 80% or so. With solar working away all day, full charge will be reached much more frequently, thereby extending the life of the batteries.

So there are lots of factors that play into one's decision about going solar or not, not just hard economic numbers.

How do I figure out how much solar I should install?

Chances are real good that the answer is "as much as you can fit". One of the challenges with a boat is that there is usually far less space available for solar than there is demand for power. In other words, physical space will be your limiting factor, not how much power you need. On really power hungry boats, the amount of solar that you can fit isn't enough to make a meaningful dent in the boat's power consumption. Other Nordhavn owners have warned me about this, but I have decided to do it anyway, determined to lower my electric loads enough to make the solar significantly helpful.  Hopefully I'll succeed.

A lot of boats don't fall into the power hungry category, and I don't think their owners would regret putting as much solar up as they can. The panels are remarkably inexpensive at the moment. As an example, my 3 panels and mounts were about $1000, and the charge controller was $500. The controller can handle a lot more panels, so the incremental cost of more solar is relatively small. If only I had room.

Speaking of controllers, I would really recommend an MPPT controller. On land-based systems the argument for MPPT has always been that the $$ spent for MPPT gets you more power than spending the same $$ to adding more panels. The assumption, of course, is that you have ample room for as many panels as you might want.

On a boat you will almost always be panel space constrained, so the objective is to get as many watts of power out of those precious square feet as possible. An MPPT charger will get another 15% - 20% out of those square feet, maybe more. They also let you run higher voltage panels, and wire them in series for even higher voltage. This gives you a much greater selection of panels to choose from so you can find something that is just the right size to fill your available space.  And higher voltage means less current between the panels and charge controller, which means smaller wires which is always welcome when you have to fish them through the boat.

I keep hearing about MPPT charge controllers and PWM charge controllers.   What does the jargon stand for, and what's the difference between them?

PWM stands for Pulse Width Modulation.

A PWM charger simply connects and disconnects (switches) the panels and the battery in such a way as to maintain a desired voltage. The amount of time that the panels are connected vs disconnected is considered the Pulse Width. By varying (Modulating) the pulse width, the controller regulates the voltage and hence the charging of the batteries. The more time the panels are connected to the batteries, the higher to battery voltage gets.  The less time they are connected, the lower the voltage.

MPPT stands for Maximum Power Point Tracking.

Now we're going to dive into electrical engineering geek-land, so be warned.....

To understand MPPT, you need to understand a basic principal about solar panels.  They are what's know in the electrical engineering world as a constant current device. That means that over a range of loads, they produce a constant current, and only the voltage drops/rises.  This important characteristic is the motivation behind MPPT chargers.

Understanding that panels produce a constant current regardless of voltage, a PWM controller effectively operates the panels at the battery voltage. So if the panels generate 10A of current, and your battery voltage while charging is 14V, you will get 140W out of the panels (10A * 14V). The issue is that panels are typically rated at higher voltages. In fact, every panel has a maximum power rating which is base on the highest operating voltage just before the current starts to collapse. This is referred to as the Maximum Power Point, and is achieved at the maximum power point voltage, known as Vmp (Voltage, maximum power). If you look at the spec sheets for any panel you will find all these numbers, and they are all based an a set of standard test conditions. Getting back to a PWM charger, it operates the panel at 14V where the panel's Vmp is more likely around 20V. So a panel that is ideally able to produce 200W (20V x 10A) is only being harvested at 140W. The difference is the opportunity that an MPPT controller captures.

MPPT works a little differently than PWM, and is instead based on a variable DC to DC converter. It's called an MPPT controller because it operates the input to the DC-DC converter at the panel's maximum power point, namely 20V in out example. So the input power to the DC-DC converter is the full 200W that the panel has to give. It then operates the converter output at the desired battery voltage, say 14V while charging. Well designed DC-DC converters are pretty efficient - around 90% or more, so 90% of the 200W input power is available at the output yielding 180W to charge the batteries. Compare that you the 140W we got from an PWM controller. That's a 28% increase in charging power. Pretty nice, right?

That's the basics, but there are a lot more smarts in an MPPT controller. It needs to "search" for the max power point on whatever combinations of panels you have attached, and it needs to find that max power point as the sun exposure varies over the course of the day, and as weather changes. This is where the "Tracking" part of MPPT comes from.  Different manufacturers do the tracking differently, but all manage to find and track the maximum output from the panels over the course of the day.

You said that an MPPT controller permits a wider selection of solar panels.  Can you elaborate?

Back on the PWM description, you'll recall that regardless of the panel's rated voltage, all the power you get from the panel is that which is available at the battery's voltage.  As a result, it only makes sense to get panels that are rated a little above the max battery voltage that you expect. For a 12V system that ends up being around 18V give or take. Anything more than that is just wasted power and money.   Furthermore, you need to wire all your panels in parallel to maintain the 18V max voltage. When you go looking for panels, you will find a very limited selection rated at 18V, and the panels tend to be smaller in capacity (around 140W max), and a good bit more expensive per watt.

On the other hand, you will recall that an MPPT controller is based around a DC-DC converter, and pretty much all of them can accept an input voltage (panel voltage) up to around 140V. So all of the sudden you can pick panels of pretty much any voltage you want, including the 250-350W panels that are out there. And you will be able to find panels that cost around $1/watt, if not less. This ends up being a huge advantage when you are trying to fill the limited space you have available for panels because you have a much wider selection of shapes and sizes to choose from. I was able to find panels that almost perfectly fill the space available on my hardtop, and probably wouldn't have even come close if I were constrained on my selection.
Stay tuned for a post later on about the solar system to be installed on Tanglewood, but I'll probably wait until it's actually installed and I can also report on its performance.


  1. Wow, thanks for the great information. Have been reading your blog for some time and really enjoy all the insights you have. I hope solar on your N60 works. Can't wait to hear the impact it is on your new boat.

  2. Thanks much Peter, great stuff as always. The camp that insists solar has no place on a powerboat utilizing partial generator time often have not looked at the tech in quite some time. PV electro has come a long way and the expense has dropped dramatically. The system I will install on my home will be done without subsidy and should pencil out to 100% payback in 4 years time by my most conservative calculations.