Want an interactive community experience? Join our Discord

We are in process of transcribing and editing this program. We are teaching the refreshed version in early November. Here is the (still rough) transcript if you want to see where we are going with it.


Batteries provide flexibility to the grid, and their exponential growth represents new territory for building designers seeking to better understand their value. Off-grid is no longer the only focus of this market. An increasing number of “grid-connected” batteries can increase project value, boost economic performance, and even manage “micro-grids” during a black-outs. Learn the fundamentals of evaluating battery technologies.

Learning Objectives:

Understand Energy Arbitrage, Demand Management, Grid-Assist, and Off-Grid Configurations
Explore AC vs. DC Coupling of Batteries to Solar Arrays
Examine Cost Differences between Above Configurations
Identify basic system components

Explore the product specification sheets of these technologies
Learn to calculate cost per kilowatt, cost per kilowatt hour, and total cost of ownership.
Understand how charge and discharge rates impact system efficiency
Propose battery strategies for a variety of applications and budgets

Battery Class, Part 1

What is a battery?

Welcome to the Solar Battery class.  This is part one, and where better to start than with the first question, “What is a battery?”   A battery has two basic parts: electrodes and electrolyte.  There are two electrodes in every battery. Both are made of conductive materials, but serve different roles.

One electrode, known as the cathode, connects to the positive end of the battery and is where the electrical current leaves (or electrons enter) the battery during discharge. That’s when the battery is being used to power something. The other electrode, known as the anode, connects to the negative end of the battery and is where the electrical current enters (or electrons leave) the battery during discharge. 

Between these electrodes, as well as inside them, is the electrolyte. This is a liquid or gel-like substance that contains electrically charged particles, or ions. The ions combine with the materials that make up the electrodes, producing chemical reactions that allow a battery to generate an electric current.

So inside the battery there is liquid which can be electrically charged. Like a solar panel, a charged battery has a positive and negative side. When subject to a load (including a dangerous electrical issues) electrons will flow out of the battery

When evaluating your solar battery options, you must consider a few things. First, how long the solar battery will last, and then, how much power can it provide? We will learn the criteria used to compare home energy storage options, and get the right battery for you or your client.

Types of Batteries

At this time, there are only two viable battery types for residential and commercial solar: lead acid or lithium ion.   Saltwater batteries are the new kid on the block and while the concepts are exciting, the companies that have brought them to market have struggled to remain in business. Lithium ion batteries are the best option for a solar array, but lead acid can be cheaper to the point where it becomes a necessity for larger residential off grid batteries.

The lithium ion battery market share has recently eclipsed lead acid. That doesn’t necessarily mean that it’s the best solution – but they are becoming the market leader, with substantial discussion being paid to which type of lithium ion battery is best. Since lead acid is still cheaper, you might want to go with a lower-cost lead acid today and upgrade to lithium ion when the price drops in the future. Similarly, you might want to start with a lower-cost Lithium-Ion battery rather than a top shelf. By the end of this class you should be able to perform cost analysis on batteries as well as read technical specification sheets to determine what is the best solution for your project.

We’ve seen over a ten year period an 80% drop in the cost of a solar panel.  Lithium-ion is most likely follow suit, but right now we’re in a first generation lithium ion product, and not just the battery itself. The inverters and software and system integration that goes along with it is just getting off the ground as well.  This is something to think about. Do you want to be the early adopter? Even if lithium-ion is the superior technology, is it worth its  cost? There is no easy answer or perfect solution when choosing batteries.

In order to understand batteries better, we first must understand Depth of Discharge (DoD).

The US Solar Institute defines Depth of Discharge (DOD) as, “an alternate method to indicate a battery‘s state of charge (SOC). The DOD is inversely proportional to SOC: as the one increases, the other decreases. While the SOC units are percent points (0% = empty; 100% = full), DOD percent points are (100% = empty; 0% = full). As a battery capacity can fluctuate with temperature, or sometimes be understated by the battery manufacturer, it is possible for the DOD value to exceed the full value (e.g.: 110%), but regardless of battery chemistry it is never a good idea to fully discharge a battery. Similarly we must be cautious how we fully charge a battery.

Solar batteries must retain some charge at all times because of their chemical composition. If you use 100% of a battery’s charge, its useful life will be drastically reduced.  The depth of discharge (DoD) of a battery refers to the percentage of a battery’s capacity used. (Most manufacturers will specify a maximum DoD for optimal performance). If a 10 kWh battery has a DoD of 90%, using more than 9 kWh of the battery before recharging it will reduce the longevity of the battery considerably. Basically, a higher DoD means higher utilization of your battery’s capacity. A lower depth of discharge means longer utilization of the battery. This means the error of oversizing the battery simply means it will have a longer life, but undersizing the battery could dramatically reduce its life.

So, lead acid batteries are old school technology that have been used in off-grid energy systems forever. But a larger market for lead acid have been and mobile equipment such as heavy machinery and forklifts. These mobile application markets are being eroded by lithium ion, but that doesn’t mean the stationary off-grid market should switch technologies just yet.  Lead acid has a shorter life and less depth of discharge than lithium ion, but they are very good when price is an issue. Clients who want to go off-grid and require lots of storage can be very happy with lead acid. But smaller, high power options should be done with lithium ion. We’ll learn how to calculate that determination later.

Energysage describes lithium ion as, “used in the majority of new home energy storage technologies. Lithium ion batteries are smaller and lighter than lead acid batteries. They also have a higher DoD and longer lifespan than lead acid batteries.”  Not so fast…that’s what this class is about.  It really all depends on several factors and at the end of this class you will know what to chose in a variety of situations.

A final word about saltwater batteries: what is all the buzz about? Well, a saltwater battery doesn’t uses saltwater as the electrolyte to improve environmental friendliness. The environmental impact of a global battery industry is not only an environmental concern, but also an environmental constraint in terms of manufacturing scale. Saltwater batteries in theory address both these concerns but the one company that makes salt water batteries for home use, Aquion, filed for bankruptcy in 2017 despite massive funding from the likes of Bill and Melinda Gates. Lithium ion continues to gain market share.

Capacity & power

Capacity is the total amount of electricity a battery can store, and it is measured in kilowatt-hours (kWh). Most batteries are “stackable” to some degree, which means batteries can be added to the system later. But that feature requires electronic controls which can manage the additions to the system – similar to micro-inverters or DC-optimizers in solar which manage independent solar panels. In other words, adding more batteries to the system later may often require buying more inverter capacity as well. It is more difficult to just add more batteries to the literal circuit. You do not want multiple uncontrolled circuits of parallel batteries because of resistance which will result in performance degradation. Likewise, you want all the batteries on a circuit to share the same characteristics, which is a function of usage. So snapping on a few batteries to the circuit is not typically what is done when adding batteries to a battery system – there are more material costs to consider! More on that later.

While capacity tells you how big your battery is, it doesn’t tell you how much electricity a battery can provide at a given moment. To get the full picture, you also need to consider the battery’s power rating. In the context of solar batteries, a power rating is the amount of electricity that a battery can deliver at one time. It is measured in kilowatts (kW).  The nameplate power rating of the inverter is only one part of battery capacity. Of course the inverter needs to be large enough to power the load, but likewise, almost all batteries cannot deliver all of their power over the course of one hour. Such a rapid discharge rate would cause the battery to lose much, if not most of its stored electricity as heat. So in addition to the inverter, the battery size needs to give some consideration to the amount of instantaneous load it will be subjected to. This is not a concern for a battery with a high capacity and a low power rating, such as an offgrid home with multiple days worth of battery storage. It is more of an issue when you have a relatively small battery with high power rating which might only run for a few hours.

This is a nuanced issue becomes less important by choosing a higher end battery. But for an idea of scale, flooded lead acid is pushed to its limits around a six hour discharge rate, with substantial degradation for anything faster than a 20 hour discharge rate. And so flooded lead acid works best when you have a day’s worth of storage capacity available, assuming a relatively even load. Lithium ion is less impacted by discharge rate, but low-end lithium ion has a maximum discharge limit for about two hours.

The nuance comes from the actual load itself too. There is a big difference between a 4kW load that is sustained for 24 hours vs. a 16kW load sustained for six hours. We’ll cover these concepts in greater detail later.

Round-trip efficiency

A battery’s round-trip efficiency represents the amount of energy that can be used as a percentage of the amount of energy that it took to store it. For example, if you feed five kWh of electricity into your battery and can only get four kWh of useful electricity back, the battery has 80% round-trip efficiency (4 kWh / 5 kWh = 80%). Generally speaking, a higher round-trip efficiency means you will get more economic value out of your battery.

Battery life & warranty

For most uses of home energy storage, your battery will “cycle” (charge and drain) daily. The battery’s ability to hold a charge will gradually decrease the more you use it. In this way, solar batteries are like your cell phone – you charge it each night to use it during the day, and as your phone gets older you’ll start to notice that the battery isn’t holding as much of a charge as it did when it was new.

Your solar battery will have a warranty defined by a guarantee that the battery will cycle X number of times. Because battery performance naturally degrades over time, most manufacturers will also guarantee that the battery keeps a certain amount of its capacity over the course of the warranty. Therefore, the simple answer to the question “how long will my solar battery last?” is that it depends on the brand of battery you buy and how much capacity it will lose over time.

For example, a battery might be warrantied for 5,000 cycles or 10 years at 70 percent of its original capacity. This means that at the end of the warranty, the battery will have lost no more than 30 percent of its original ability to store energy.  

Because lithium ion batteries have better performance than lead acid, there is less emphasis placed on depth-of-discharge cycle sheets and warranty expectations. Instead, the warranties are based around simple benchmarks like total battery output, with a single depth-of-discharge point occasionally mentioned. Here are some examples.


Many different types of organizations are developing and manufacturing solar battery products, from automotive companies to tech startups. While a major automotive company entering the energy storage market likely has a longer history of product manufacturing, they may not offer the most revolutionary technology. By contrast, a tech startup might have a brand-new high-performing technology, but less of a track record to prove the battery’s long-term functionality. 

Whether you choose a battery manufactured by a cutting-edge startup or a manufacturer with a long history depends on your priorities. Evaluating the warranties associated with each product can give you additional guidance as you make your decision.

Lead acid needs to be discharged over a period of about five hours or longer.  This means the lead acid battery must be at least five times larger than the greatest sustained peak load. I recommend taking the highest sustained 15 minute load measured in kW, multiply by five, and have that be the smallest possible flooded lead acid battery bank size to match the load. If using lithium cobalt, such as the Tesla Powerwall, I would multiply by two instead of five. Just as a rule of thumb.

The efficiency drops dramatically if you’re trying to fully discharge a lead acid battery bank in a matter of minutes. Not that you can’t do it, but you shouldn’t.

Lithium ion can be discharged in a matter of minutes and still maintain its efficiency in about 15 minutes to an hour.

Flywheels and super capacitors

There are other forms of storage, like super capacitors and flywheels.   I do see super capacitors coming to market in Australia.  If you get on LinkedIn and follow solar in Australia there are companies out there putting in 7 kwh super capacitors. They seem exorbitantly expensive but what these super capacitors can do is smooth out the second-by-second volatility in your electric inflow and outflow.

A small super capacitor can help maintain some system efficiency particularly on an unreliable or unstable electric grid.  I have a grid-assist site running a hot rod of a solar array (large but cheap inverters, small flooded lead acid battery) and the signal quality out of the inverter is not the greatest. I’d love for a supercapacitor to simply plug into the system for voltage stablization. It’s not impossible, but I don’t know a market-tested consumer product for this and haven’t experimented with it on my own.

An elevator goes up to the top floor and then it drops back down to the bottom floor in a matter of seconds.  And so, they have to pick a storage that fits the application, whereas storing that power and lithium-ion is not quite as efficient as a flywheel.  Large heavy loads instantly click on ramping up in seconds rather than quarter hours. So there are applications where instantaneous power deployment or even storage are needed. Lithium ion does not have an instant response time, and lead acid is even worse, so there are applications for supercapacitors and flywheels. But these applications are more common in commercial or industrial settings. 

That said, you commonly you won’t need the responsiveness of supercapactitors or flywheels for grid-tied building demand management, and off-grid storage quality can be resolved by upgrading from lead acid to lithium ion and careful load management and appliance selection. A building maximum load period is billed based on the maximum 15 minute peak load but that actual load may occur over a two or three hour window, or have nearby peaks approaching the maximum for that duration.  So even when you have a spiky load, you will want to have storage capacity that is a multiple of the duration of that load, to ensure that you can manage it successfully. A two-hour peak load might be managed with a five-hour battery, just to be on the safe side. The last thing you want to do is fully deplete your battery bank and then not have any reserve capacity available when your demand is still high. The larger the battery compared to the load, the better its responsiveness independent of technology. So even for a commercial battery, it may be a better strategy to go with a larger, less responsive battery rather than a smaller, more responsive battery.  

I think it is a great idea to combine multiple battery chemistries, having a small expensive battery reserved to peak demand management, and a larger cheaper battery for supplying a more stable baseload. While such an arrangement can be programmed using existing components, inverter manufacturers do not warranty their components under that kind of configuration. Even battery manufacturers which advertise multiple battery chemistries are simply indicating that their products will work with multiple battery chemistries, but not at the same time. 

That said, when looking at your system costs, the cost of that inverter warranty is not the only consideration. When powering an entire building off-grid, there are other motivations, such as the cost of grid expansion or the building owner’s personal goals. The inverter itself is only one component, and while battery inverters cost more per watt than solar-only inverters, and can cost more per watt than the solar panel itself, the actual batteries are by far and away the most expensive component in an off-grid system, such that the inverter warranty may not be the driving the project decision making. 

  So if an extreme early adopter wanted to piece together an off-grid system that is responsive to their facility needs, so they could use a high end lithium ion battery when the demand spikes momentarily, they discharge the lithium ion, and then have a separate inverter and a separate battery system for a lead acid base load. 

My clients always ask if they should choose lead acid or lithium ion.  The answer is, “It depends.”  There is the capital cost consideration, since lithium ion at least twice the price of lead acid. That’s the primary reason residential off-gridders go with lead acid. But smaller batteries, such as those used for significant commercial demand savings or for homeowners seeking to run the whole house for only part of the day should give lithium ion a closer look.

There’s also the system maintenance consideration.  Lithium ion batteries and some lead acid batteries are sealed batteries so when it operates it doesn’t emit water vapor or gases. That sounds like a good thing, but the problem is that battery charging and discharging will produce chemical reactions, including gas, and letting the gas vent to the outside will result in less degradation than sealing it inside the battery. This is particularly true for lead acid. Sealing the battery also adds cost, and so a sealed lead acid battery costs more and does not last as long as an unsealed lead acid. So when I use lead acid for off-grid, I always use an unsealed battery, which means the battery there’s a lot of evaporation in the battery due to use, and so one of the big maintenance tasks of a lead acid battery is to water the battery.  

Watering the battery is relatively simple. There are consumer-grade and industrial-grade watering kits to connect the individual cells in the battery together. I’ll look at watering systems in a little bit. But if you’re going for low upfront cost and better performance and cost-effectiveness, go with the unsealed lead acid batteries and locate them either outside the building, or in a room along the edge of the building where the gas from the battery can be vented to the outside. Whereas if you’re going for no maintenance or no off gassing, lithium ion is a great choice.  If you’re in an apartment building and you want to put your battery in an equipment closet, you’ll have to go with lithium ion just due to the ventilation requirements that come with sealed batteries. Likewise if the site location is such that maintenance is an issue, such as a partially used off-grid cabin in the middle of the woods, lithium ion is the way to go. But if that same cabin lacks the budget for lithium ion, starting out with lead acid can get the job done.   

When I look at the performance characteristics between lead acid and lithium-ion, the main problem with lead acid is when the load becomes to large that it wants to discharge the battery in less than a full day – let’s go ahead and say under six hours. This is more reflective of the instant load over this time frame, rather than the average load over this time frame. When beginning to evaluate the discharge rates of batteries, it is good to ask yourself “at this instant load, how fast will my battery discharge?” and then to understand that number would double if say, the battery is at a lower state of charge. 

Some of that can be rectified by smart appliance selection.  

If I’m doing an off-grid home design I am NOT going to specify a tankless water heater, because it can be a 20 kilowatt load that is instantaneously applied.  So, tankless water heaters are not appropriate for an off-grid lead acid cabin. As we talk about in our off-grid class and our upcoming smart home class, there is the potential for demand management to help reduce this load. In periods of heavy load and no offsetting solar charge, a thermostat on an air conditioner could be dialed back, a refrigerator or dehumidifier clicked off, the ceiling fans turned off. Sometimes it is easier to do such controls with dumb appliances and smart switches, and other times it is a more serious programming job.

Now I’m getting into specification sheets and there’s a there’s a couple of terms for a lead acid battery that I need to understand.  There’s “float”,  “absorb” and “bulk”.  The float of the lead acid battery refers to the very top of the battery capacity and at the very top of the battery capacity right here at the 20% – 30% range. Then I can see this is an asymptote that’s getting us all the way to the top. So the float is the top range of the battery, the absorb is the mid range, and the bulk is the bottom range.

So, back to the DoD.  This chart indicates the number of cycles a battery will go through, based on how deeply I discharge the battery.  In a sense, if a battery has double the capacity, it’s going to be discharged half as much. So I can right-size our battery by finding an ideal cycle level to our battery that I should not exceed.

 People say a lead acid battery shouldn’t discharge more than 80%, or if you fully drain a lead-acid battery it’s not going to charge back up again. Those are very fast and loose rules of thumb that can give a wrong impression about lead acid.  These two charts are actually two different kinds of lead acid batteries.  

Lead acid batteries are available from low to high quality and everything in between. These two curves are different shapes and what I like to look at is the difference between absorb and bulk is somewhat indicated by where this straight line starts to deflect.

So that point of deflection in this curve is somewhat of a quality indicator. If I look at these two batteries in the manufacturers spec it shows on this lower end battery, you don’t want to cycle it down more than about 35% on a regular basis, whereas on the this battery you’re going get elastic stress and strain curves of metal.

This straight-line range is the elastic range of the battery. This bottom range is where the battery material gets strained and you see more degradation in that and it’s not exactly the same as stress and strain.  It’s really more of at what level of chemical interaction do you get more degradation in the battery anode.

But what I can see is that within the flooded lead-acid category you can find flooded lead-acid that can take all the way down to a 65% DoD on a regular basis, and then you can also get flooded lead-acid that can only really take a 35% DoD.

The reason why I want to point this out is because lithium-ion guys will say you get 100% DoD, so you can get away with a smaller battery bank. That is true, but you can also apply the same argument to staying with top-shelf flooded lead-acid batteries. Then you could get away with a smaller battery bank too, and so might not really an issue.

What I commonly do in my off grid designs is specify a top shelf flooded lead-acid battery and then program the control system so before I get to the point where I’m operating in the bulk range of the battery, meaning before I get to the point where the battery is being strained, I’m going to turn the generator on and charge it all the way back up to the top.

So the bulk range of the battery will actually have a different charge setting than the absorb range of the battery.  The bulk range is the danger level of the battery. Get down into the bulk range of your battery you’re going to want to turn that generator on and crank it really high and charge to get out of the bulk range as fast as you can.

During the absorb range, the generator setting is generally going to run for a shorter time and be less aggressive. Here, I might just run the battery for three hours and just get a little bit further up and down, and then in this absorb range, you can just yo-yo your energy levels back up and down without worrying about battery damage.

The float of the battery and when the battery is fully charged at the top of the range is similar to the absorb range of the battery.  You’re not getting too much degradation even though the electricity is flowing through the battery if your battery is at full charge and your solar array is tied into the same bus as the battery bank before you land on the inverter.

 You have the voltage flowing through the top of those battery leads.  It’s just it’s not going to be cycling or damaging the battery because you’re not draining it further into it’s strained range. A 48-volt battery is just the nickname of the battery.  You may be wiring 4 12-volt batteries to create a 48-volt battery. Read the spec sheet and it’ll say the 48-volt battery bank is actually rated for a maximum voltage capacity of 60 volts.

Why is my battery when it’s at full charge, closer to 60 volts than 48 volts?  You’ll never see your battery get all the way up to a 60-volt charge. What’s happening in this cycle and DoD charge?  You can see it forms a NASA taupe?? that goes further and further up. What happens when you charge it all the way up to it’s full 100% no DoD capacity is that it immediately starts to self discharge and drop back down.

 You really can’t maintain a flooded lead-acid battery at 100% state of charge and so I simply don’t ever charge them all the way up to 60 volts.  I might charge them up to 54-55 volts and then when they get down into this lower range is actually when they start getting into their 48 volt category.  Once they drop below 48 volts they really don’t produce any power for your 48 volt applications but do get into diminishing returns in terms of efficiency.

If I try to charge our batteries all the way up to 100%, that high voltage is going to want to press the electricity back down into the appliances and not be able to store it in the battery very efficiently.  There’s actually only one circumstance when you charge your lead-acid batteries up to approaching that full voltage.  That’s during the maintenance cycle of the battery.

The maintenance cycle of the battery you charge it to the top of the float range.  That’s not what always you’re going to do if you’re if you’re in the absorb range and you click on your generator. You’re really only trying to charge it up to the top of the absorb range or into the float range, but what you’re not trying to do is charge it all the way to the top.

Charge to the top of the float during maintenance. Do this because it actually is going to bust off plaque that will cover the anodes.  During chemical reactions you have a solid that is sticking down into some chemicals and electrons are flowing through the solid into the liquid. That reaction is causing a little bit of corrosion in the solid anode and it “plates” the anode.  That’s not something to do all the time, but you really do it more often when you’re operating the battery down into its full DoD.

 So, maintain your battery, but it is not something to do all the time.  You also don’t need to maintain the battery as much when operating in this absorbed range of the battery. It’s really only when you start cycling your battery down below what it can take that more regular maintenance tasks occur.  

You will need to maintain your battery more frequently during extreme months: in summer with its high air conditioning loads, and in winter with less sunlight, but less in the spring and fall, when electric loads are low and production is high.

When I think about the Tesla battery, it sounds like a very high-end battery. In actuality, it is the standard lithium ion battery. It is the lower end of lithium ion technology. Why? For one, it has fewer cycles, which I know speeds degradation more quickly than the top end.  

The Tesla battery is, generally speaking, a technology that is referred to as lithium ion phosphate and the main difference is the lithium iron phosphate anodes experience less degradation than anodes without iron phosphate. The lithium ion batteries have a shorter life, but in order to get a longer life span, there’s also a high production cost.

There’s a top range of lithium-ion technology and you see this top range being used for very remote power applications and for clients for whom the budget is not an issue. So the client who wants or needs to do the project once and have it last for 30 years without any maintenance lithium iron phosphate.  

In battery technology, you do get what you pay for, and under no circumstances would I recommend cutting costs at the battery level.  Better the residential client wait to add a battery, or accept the cost as part and parcel of doing business.  It would be easier and smarter to shave cost off the panels than try and work with substandard batteries.

Conversely the commercial clients may think they can afford a top shelf battery from the get-go, but using a battery for a building- that’s a lot of storage capacity that you’re going to need! The cost, as I’ll see, of lithium iron phosphate batteries starts to become prohibitive.

So the major players in lithium ion batteries are Tesla, LG Chem, and Marseille DS.   LG Chem is the open distribution line battery manufacturer, whereas Tesla has a more closed installer network.  It’s very easy to become an LG Chem lithium ion installer,??? Why? Tesla wants to make their own products.  Tesla also wants to make their own inverters. ??

 If you’re using a solar inverter, you’re likely to be specking out an LG Chem battery. There are even higher-end lithium iron phosphate batteries that last a bit longer.  They will work with solar inverter manufacturers as Ill. About a year and a half ago, I selected a range of battery technologies and compared their cost and lifecycle cost. I selected a mid-range flooded lead-acid battery, a top-shelf flooded lead-acid, the Tesla Powerwall 2, and an ultra premium lithium iron phosphate that was made by Sony. I also selected a nickel iron battery technology as Ill.

 I have to get a little bit further into product selection to understand why this premium flooded lead-acid line was 105 kwh versus the industrial flooded blood acid at 122 kwh. You have to also understand how to estimate the size that needed, and you will soon. But, for now, what you’re really constrained by is the upper limit of batteries that will fit on no more than two batteries circuits. Within this particular premium flooded lead-acid product line, the largest size battery I could wire together to fit the 48 volt configuration was 105 kwh. 

There is a little bit more wiggle room within the industrial line and so I called the manufacturer and simply asked what was the most popular industrial flooded lead-acid battery, with the assumption that the most popular would likely be the best price point. That was the Tesla Powerwall. (1 or 2?)

So, I wondered if I spent ~$25,000 on a flooded lead-acid, could I instead use the Tesla Powerwall? How much capacity what I would I get? What would the user experience of a smaller battery bank look like?  The lithium iron phosphate I added for comparative analysis. This was to determine if the lithium iron phosphate for a 42 kwh battery was cost effective as well.

This is the top shelf, longest lifecycle maintenance-free battery, but the capital cost was beyond the clients means. I had a competitor to flooded acid, called nickel ion and it’s also more expensive and would have involved a higher upfront cost as well.  The difference was the DoD is rated at a much deeper DoD for its regular cycle. And so then I have to as what would the DoD be?

 So I looked at the average DoD for the Tesla Powerwall for the nickel iron battery for the industrial lead acid and for the premium lead acid? How many cycles would I get out of it? I took the upfront cost of the battery, divided by the size of the battery. I looked at the size of the battery in the cost of the battery and multiplied those two together to get to the upfront cost.

 I have 105 kwh battery so I go into our energy analysis and I say our average DoD for this 105 kwh battery is going to be a 30% DoD.  I do a 122 kwh battery. My average DoD is only going to be 20%. The Tesla Powerwall 2 is actually rated for 100% DoD at 37800 kwhs, so their warranty is not based on the DoD.It’s just based off of the total energy output.

The reason is the lithium ions don’t have that same stress-strain curve where there’s an operating range in the mid range and top range of the battery, is that the lithium-ion battery is more elastic than the lead-acid technology, and so you can fully discharge it and not worry about the operating range of the battery. 

However, once X amount of kwhs are used, the warranty is up.  Remember the Tesla provided inverter? They’re counting every single kwh that comes out of the battery and tracking it. so you look at their warranty and it’ll actually say you won’t get a depth of discharge curve for the Tesla Powerwall.  Citation?

If you’re using a solar inverter, you’re likely to be specking out an LG Chem battery. There are even higher-end lithium iron phosphate batteries that last a bit longer.  They will work with solar inverter manufacturers as Ill. About a year and a half ago, I selected a range of battery technologies and compared their cost and lifecycle cost. I selected a mid-range flooded lead-acid battery, a top-shelf flooded lead-acid, the Tesla Powerwall 2, and an ultra premium lithium iron phosphate that was made by Sony. I also selected a nickel iron battery technology as Ill.

 I have to get a little bit further into product selection to understand why this premium flooded lead-acid line was 105 kwh versus the industrial flooded blood acid at 122 kwh. You have to also understand how to estimate the size that needed, and you will soon. But, for now, what you’re really constrained by is the upper limit of batteries that will fit on no more than two batteries circuits. Within this particular premium flooded lead-acid product line, the largest size battery I could wire together to fit the 48 volt configuration was 105 kwh. 

There is a little bit more wiggle room within the industrial line and so I called the manufacturer and simply asked what was the most popular industrial flooded lead-acid battery, with the assumption that the most popular would likely be the best price point. That was the Tesla Powerwall. (1 or 2?)

So, I wondered if I spent ~$25,000 on a flooded lead-acid, could I instead use the Tesla Powerwall? How much capacity what I would I get? What would the user experience of a smaller battery bank look like?  The lithium iron phosphate I added for comparative analysis. This was to determine if the lithium iron phosphate for a 42 kwh battery was cost effective as well.

This is the top shelf, longest lifecycle maintenance-free battery, but the capital cost was beyond what the client was willing to spend. A competitor to flooded acid, called nickel ion, it’s also more expensive.  The difference was the that it is rated at a much deeper DoD for its regular cycle. So, what is the DoD?

 So I looked at the average DoD for the Tesla Powerwall for the nickel iron battery for the industrial lead acid and for the premium lead acid? How many cycles would I get out of it? I took the upfront cost of the battery, divided by the size of the battery. I looked at the size of the battery in the cost of the battery and multiplied those two together to get to the upfront cost.

 I have 105 kwh battery so I go into our energy analysis and I say our average DoD for this 105 kwh battery is going to be a 30% DoD.  I do a 122 kwh battery. My average DoD is only going to be 20%. The Tesla Powerwall 2 is actually rated for 100% DoD at 37800 kwhs, so their warranty is not based on the DoD.  It’s just based on the total energy output.

The reason is the lithium ions don’t have that same stress-strain curve where there’s a operating range in the mid range and top range of the battery is that the lithium-ion battery is more elastic than the lead-acid technology, and so you can fully discharge it and not worry about the operating range of the battery. 

However, once X amount of kwhs are used, the warranty expires.  Remember the Tesla provided inverter? They’re counting every single kwh that comes out of the battery and tracking it. Their warranty actually says you won’t get a Depth of Discharge curve for the Tesla Powerwall.  

Tesla says its warranty is for 100%, and this is how much energy you can draw out of it. ??  Technically speaking, you still can’t fully discharge a lithium-ion battery. It’ll give out and not charge back up again. Tesla plays a little game with their spec sheet where they actually will give you a little bit larger of a battery than what they say they’re delivering.   So they keep a little bit of reserve capacity in that battery that you can’t access to make sure you don’t kill it if you completely drain the battery.

Next, the nickel-iron battery!  The manufacturer says it’s rated for an 80% DoD as well, (and that they they’re more elastic than flooded lead-acid), although there’s much less market data available on nickel-iron.  With the flooded lead-acid technology types, there is a lot of comparative analysis between different flooded lead-acid manufacturers.

With nickel-iron, you’re really just taking the manufacturer’s word for it because there’s a small range of manufacturers. So, ask yourself what is the levelized cost of storage if battery is going to $X.  Then I’m only using 30% of 105 kwhs on a regular basis, or I’m only using 20% of 122 kwhs on a regular basis.

 I need to know how many cycles of 30% or how many cycles of 20% I’m giving and so that’s where I go back and I actually use these Depth of Discharge curves and so with a higher end industrial flooded lead-acid battery we’re getting this 20% range. See how you get a lot more cycles out of it then the mid shelf premium flooded lead-acid?

Where 20% DoD is only giving us 4,000 cycles and not 5,000 cycles, the industrial flooded lead-acid battery that is $210/per kwh is rated for 20% more cycles than the premium flooded lead-acid battery that is 100 and $60 a kwh. Let’s pull up our calculator here and I put in $160 and I want to add 20% to that.

Still under $210 and isn’t using the premium flooded lead-acid isn’t that still going to be less cost of more cost-effective than 210 because 192 is less than 210?  So if you missed that, let’s go back.  What we’re trying to do is compare the industrial flooded lead-acid battery versus the premium flooded lead-acid battery.

The premium flooded lead-acid battery at a 20% DoD is giving us 4,000 cycles. The industrial flooded lead-acid which costs 25% more is giving us 20% more cycles. If you do an incremental cost analysis, it would appear at first glance that the premium battery is more cost effective than the industrial.

But what needs analysis is this: the premium lead acid battery is only giving a 35% range until battery degradation starts.   The industrial battery is giving us more of a 60% range before usage starts degrading the battery. The question is: how often will the client dip into this lower end. In order to do a very complete analysis, look at this. 

At 105 kwhs getting into this 30% DoD range that’s where I get our 2,750 cycles and then I also compared it to doing a higher end industrial lead acid and this is saying 5,000 cycles but be conservative. Looking at dividing the upfront cost by how many kilowatt hours 30% of 105, 20%  of 120, to  or assuming I can fully use the Tesla Powerwall at 30% to 7,800, I divide our upfront cost by how many KWHs we’re getting out of the battery.

 See the higher the upfront cost gets per kwh, the lower total operating costs gets, with the exception of nickel-iron. It  looks like nickel-iron stands out in the pack in terms of cost, delivering the lowest cost per kilowatt in total life. So when you present this to your client, then right off the bat the lithium ion battery is a non-starter because of client cost parameters.

When I went through this a few years ago, the Tesla Powerwall was not widely available, the client would have had to go on a waitlist, and so it really wasn’t a feasible solution. These things happen. And so we’re going to get into generator run time: the problem with the Tesla Powerwall and its smaller capacity meant I would have to run the generator more frequently.

This is how you tailor make a system to fit the client’s needs, and a great reason (if you needed one) to keep the client as involved as possible in the design phase.  One of the things that was important to this client, for sake of example, was that he did not want to listen to the generator.  It’s his house and his life so, that’s how we made that decision.

So, the storage capacity of the lithium ion technology for the price was another limiting factor. In the end, he preferred to spend more money up front on a lead acid battery.  He also was not afraid of doing maintenance work on his property. Every time, you have to match the needs of the client with the battery and its probably never going to be a perfect fit. 

Another valid solution would have been to spend less money today on a lower end flooded lead-acid, even though it would have cost more per cycle, use it until it dies and then in 15 years get a new battery that has all the technology and lower price that the future always brings.

A side note: the power company was going to charge him $20,000 to bring the power out to where he wanted to build his new house.  Since he had to spend $20,000 either way, he moved to solar, which he was going to do at some point anyway. This is going to be more clear as I go through our sizing exercise on how to determine which battery to select. 

Living off-grid is more reliable than having grid-connected electricity in Mississippi, where this client lived. Mississippi has above-ground power lines and in a rural area he would be at the end of the distribution circuit 

An important policy decision to consider when voting is the ability to form a power company for off-grid customers. Public Utility Commissions say they only grant electric cooperative monopolies the right to form a company because of reliability, which is paramount.  Obviously, any competitor would have to be able to provide reliable electricity to a region and to be able to decertify the monopoly in order to sell electricity to an off-grid customer.

And not just sell them the equipment, but actually install the equipment and sell them the electricity like any other power company. One of the successful arguments with a public utility commission is to demonstrate that your electric supply is more reliable than what the current power company can provide. 

For solar detractors, the argument is as follows. Actually, generating power on-site, not transmitting it over the electric grid, and storing it on-site is preferable.  Storage was the missing key, and once you generate your own electricity and store it, that will be very reliable. 

This has led, in part, to the decision by the new homeowner to oversize the system. In grid-connected solar, it’s very important to right-size your equipment because you will have to stay within budget.  (Tragic, I know.) Let’s say you put a 10 kilowatt solar array on a 10 kilowatt inverter you might possibly have been able to save ~$1000 by putting it on an 8 kilowatt inverter.

 Because of heat and inverter loss, and all the losses associated with solar generally you will undersized the solar inverter as compared to the solar array because you never get up to your full array capacity.  The 10 kilowatt array capacity is based on a standard test condition based on a 60° F day under full sun. Even then, you only get full sun at high noon, so it’s very rare to ever get up to your fully rated capacity.

 So, I undersized solar inverters because it’s cost-effective to do so. With a battery, if I oversize the battery and I aim for a 35% DoD instead I get up 25% DoD.  At the 35% depth of discharge, I get 2500 cycles and at the 25% DoD I get around 3300 cycles. The battery is going to last longer if I oversize it.  

The worst-case scenario of oversizing a battery is that your battery is going to last longer.  If you undersized your battery, you’re going to be operating in this degradation range and so it’s much worse. I still need to know what the battery bank size is going to be. I have is the monthly consumption of the client’s existing home. 

I know how much power he’s consuming on a month-by-month basis, and so then I can also know how much power he’s consuming and so I take the months of the year and his existing energy usage and I divide by days in the month. I can see that in January he’s using 48 kwhs a day and February he’s using 56 kwhs a day in August he’s using 74 kwhs a day, and so that comes back to where you look at the Depth of Discharge of the battery.

I model this and then updated the slides for the model you know what I can see here with the Tesla Powerwall is there’s going to be days where we’re using 100% of that 67 kwh battery.  It’s not the complete picture because our solar array is also going to be producing during this time. 

So perhaps in in August and Sepember, 27 kwhs maybe the solar array is on for one third of that time and so maybe one third of these kwhs come out of the solar array directly, rather than coming out of the battery bank. I know how many hours there are in a day and so if I know how many by 24, I get about two kilowatts of average power draw in January divide by 24  and I get about 2.2 kilowatts of power draw in February I go in into August and divide by 24, it’s 3 kilowatts of average power draw in the summertime.  

Our average power draw from winter to summer is ranging between 2 kW and 3 kW of course that assumes our two kilowatt power draw is completely flat from midnight to midnight. I have a perfectly even power draw from in here in the summertime from midnight to midnight, I have a perfectly even power draw.  That is not reality. 

In winter, if you have electric heat, it is going to go up in the morning and scale down as your home heats, and then turn the heater up later and coast through the night, cyclical. In the summertime it’s the opposite.  You’re going to be running your air conditioner during the hottest times of the day, and of course, this is where solar thermal, passive solar and energy efficiency merge.

 Do you put solar mass into your house so that you can run the air conditioner more consistently and maybe chill thermal mass at night so that your home remains cool during the day and so that you have a more even air conditioning load?  Can you get a variable-speed air conditioner so that when your home power becomes higher than what it should be you can ramp down the air conditioner?  

Do you have a control system with a smart thermostat so that when your power level exceeds a certain amount you dial back the thermostat?  All this is possible.  The home automation hardware isn’t advanced yet enough to fine-tune this experience, but there’s a lot of automated stuff out there for day trading your electricity.  In this example, we’re on a time-of-use rate, and so the thermostat is going to dial back.

 Then the time of use rate is gone and so the smart thermostat will dial up. These commercial settings do peak demand management, where the solar array reads the electric consumption of the building. Then, when above a certain power draw, it turns on the battery to match the power draw.  So, I shave off the peak, and that’s a setting on the battery inverter itself, but what I also assume is that it’s not just going to be an average power draw throughout the day.

It’s going to vary, and so, how do I know what our exact customer load profile is going to look like?  It is definitely location based, at least in part. This analysis is for Pennsylvania, where there are 4 distinct seasons.  Historically, homes in San Francisco, CA don’t have air conditioners because it just doesn’t get that hot in San Francisco.

Sometimes you already have access to data by logging in to your utility electric account.  If there is a digital meter installed on your home, the utility should be able to provide this data.  However, expect that sometimes they won’t give it to you.   I learned the hard way to tell the client what their power and outflow levels are going to be at X amount, and have them do their own economic analysis.

 Remember, the utility will always prefer to talk the client out of doing solar.  At the same time, they don’t provide the customer or the solar company with access to the same data, even though they charge the summers for the meter fees. Use sizing software to estimate what the load profile is going to be. Start with the region of the country of the project. Then, do they have a cold winter or a hot summer?  The other aspect involves the electrical devices used in the home. 

With net metering, you never had to consider this. Solar design was reduced to considering just annual electricity usage and the size of solar array. But not every state has net metering, and even in net-metered states it is becoming increasingly difficult to offset most of your electric bill with solar array.

An optimal amount of bill reduction, using a solar array and a battery is becoming more economic. In some circumstances, it might be more in line with the customer to take the whole house off grid.  It’s like the industry’s coming back full circle. A few years ago, the combination of solar without storage and simplified electric policy had seemingly reduced the need for detailed energy analysis. 

But that also creates system inefficiencies – for example some solar arrays could be more cost effective if the customer has a better control over when and how they are using their energy.  Now we consider load shifting to better match up solar production with site consumption. 

In the summertime, the solar production is going to closely match the load profile, so the batteries are rarely used.  It’s really in winter, when solar production is highest the middle of the day and energy uses are higher in the evening and at night.

Imagine a client currently uses electric heat, but when they build their off-grid home, they will be doing wood chip heat.   The client wants to how much electricity is attributed to her heater.  How different will her electric bill be?

Well, start with what you know.  I knew her monthly electric use.  Then you ask the client if is there a swimming pool, electric heater, or air conditioner? What kind of water heater, light bulbs, etc   Then go to a software called Aurora. 

 What they’ve keyed into is within a given region certain residential load profiles look the same. While everyone’s electric usage is different, there are some generalizations. The shift of the work they do, whether they work from home or not, all this will create generalized load profiles. I may have to make some assumptions.

Here’s my monthly consumption data and then here is my PVWatts data and what I can see is that in the month of January I’m using 1500. Here I’m only producing 1100 and then I go down into summer and here in July I’m using 2200 and then here in my PVWatts data, in the month of July, I’m only producing 16800. In PVWatts, I’m not designing on an annual basis to produce as much energy as on consuming. I’m designing on a on a month-to-month basis and on a daily basis as well, but to a lesser extent.  How much energy am I taking out of my battery bank?  

I want to fill back up and so I plugged in an 18 kilowatt solar array and looking at the PVWatts data in January, I see that I’m producing 1650 kwhs.  That means I’m overproducing in January and that’s good. In fact, I am overproducing into May. That’s just something you have to deal with off-grid; in the spring and fall, you’re going to over produce.

 It’s really in the summer and winter that you have to examine, and so I see in in July now I’m producing 2,400 and here in July ~2,300.  It actually looks like the 18 kilowatt solar array is a little bit oversized. That would be assuming my daily average consumption is equal to my daily average production and that’s not true.

 You know the old adage, “If you don’t like the weather , wait 5 minutes”.  We know that weather can vary immensely, especially lately with an increase of all kinds of weather issues and global heating. You have to take into that into consideration, and that is where PVWatts becomes very useful. 

If you scroll all the way down to the bottom of PVWatts, you get the hourly production data. This data will take weather reported from low weather monitoring stations. Here I have beam irradiance and diffuse irradiance. The beam irradiance is a sunny day and the diffuser radiance is a cloudy day.

 On cloudy days, you’re going to have thick clouds and or thin clouds and sunlight still gets through thin clouds, etc  You still get production out of your array on overcast days, but a cloudy day is by no means a standard type of day. 

You’re still going to get a little bit of power out of your array, I mean this is our 18 kilowatt solar array and in the middle of the day I’m still getting 1.5-3 kilowatts of power. So in the middle of the day, for maybe five to seven hours of the day in the winter on that overcast day I’m still running off solar power.   For the remaining third of the day my batteries are draining, and then on the next day it’s the same thing. 

 You might be running this is on an overcast day, but it’s thick clouds thinning out, and so I’m actually running off solar for six hours a day.  I’m charging it for another two except  that’s still only one third of the day. Look at my PVWatts

Data, and get an hour-by-hour output of my system.

  Then I can go into PVWatts monthly consumption which I’ve converted into a daily average which then I divide by 24 and get into an hourly figure. Go into PVWatts and if this is my AC system output in watts, I can make a column for AC output in kilowatts by dividing by 1000 (that’s my solar production) so I might say production and kilowatts.

  Then I’d go and I’d have a different column for consumption and kilowatts. Put in again the in January I’m averaging two kilowatts a day and then I have a column that would be my difference where I subtract the two. Not just copying it but doing a cumulative.

I’m at plus my production- my consumption?  Now I’m getting my difference, and each one of these lines is one hour. This is my kwh column and I can do this for the entire year. I’d have to go in and customize it for each individual month. This example assumes a 2 kilowatt average power draw, and then go in and look at the minimum level.

So I’m still falling and falling and see now I’m at 14 kwhs and then I’m adding two and subtracting two, so my battery is maintaining its state of charge. Then I’m adding 4 and so now I’m charging my battery back up. Then I get up to 70 kwhs the next day.  I’m shrinking back down.  I’m at a hundred. 

 Then, it’s charging back up. Let’s go back into that cloudy weather day. Here we’re at in a sunny day and then I start getting into the January 6 and 7 and 8th.  On January 9th; we’re going through our cloudy weather days.

As you go back to the battery bank, ask yourself how far up did the state of charge get? 150. Well I’m not buying 150 of 150 kwh battery bank, so what you do is modify the formula and make sure that you’re not exceeding.  

In our case I bought a 122 kwh battery bank and so I need to make sure that the way that you would do that is say use your Excel formulas. If 122 then make it 122, otherwise keep going and change your column like that. I’ll often just go up to the top and make a cell for my battery capacity.   Make it 122 kwhs and then use Excel formulas to hard code.

 I’m not going to do the complete model here but, by using Excel you can hard code this data so that your battery capacity never exceeds the maximum. I have to make sure that it still never gets above 122 regardless, and so now I start to see our battery bank drop back down. Then it gets fully charged again and starts dropping back down.  

This exercise is modeling the battery capacity for every single hour of the year based off the actual performance of the stored array. It is also based on the actual size of our battery bank. We’re getting into the cloudy weather system and what was a 100 20 kwh battery bag only has 3 kwhs left, only has 2 kilowatt hours left.  Now we’re at – 6.  

Let me just extend this on down a ways see how deep I go into our deficit and we end up -30 and so you could do is one of two things. I could increase the size of our battery bank. I need at least 150 kwhs, otherwise we’re going to run completely out of power. I don’t want to discharge our battery bank more than 80% and so 150 divided by 80% you know I really want 180 kwh battery bank or I can say okay Ill as soon as I get down to 80% or 20% of 120 is 24.  

As soon as our battery bank dips below 24 kilowatt hours, we’re going to turn the generator back on and  fully charge it again. I’ve turned the generator on and recharged our batteries bank.  And so, by doing this exercise, I can go from monthly consumption to what our daily average consumption will be.

 Combine that with PVWatts to determine the appropriate array size, so that I fill our batteries back up every month. and I might even want to oversize the solar array. I said that an 18 kilowatt array would produce our monthly production for what I needed, but I might actually want to do even larger array.   

Overcast days are what I really want to design around, and here’s the secret to making this more cost-effective. Buy by the pallet!! I don’t do custom designs for a rooftop. This improves your shipping logistics and your budget.  That’s how to get below $0.60 a watt. So in our case, the next pallet size up was 22 kilowatts and so I adjusted the PVWatts figure for a 22 kilowatt system.

So, that gives us an average cycle throughout the year of ~30%, and I can go back to the DoD curve and for my off-grid house even though sometimes on getting down into the 50 or 60% range, there’s a lot of times when I’m just up here in the  top range and my average is in this 30% range. 

I’m averaging a 30% DoD my cycles are going to be 4000 cycles and so finally I get back to here we’re going with our 122 kwh battery, and it turns out to be $200 a kilowatt hour. Now I know our upfront cost, our average depth of our cycles, our average DoD, and then our total levelized cost. Use PVWatts and your monthly electric bill to determine array size. Export it into Microsoft Excel and finish your model. Then you cheat by assuming that once the batteries deplete to a certain level, you’re going to have a generator click on and charge the batteries completely. 

When you design battery solar arrays, keep in mind: they a lot more flexible than you think!  If a certain manufacturer states their system is not compatible with others, really what they want you to stay within their product line. But ,when you you talk with the battery inverter manufacturers, they often say you can just AC couple your battery inverters right next to an existing solar array.

Generally, the solar companies want you to use their battery inverter alongside their existing system. Not all solar companies have off-grid products, though. SolarEdge does not support an off-grid array.  The battery inverter is really only for protecting critical loads. 

They don’t warranty their system for an off-grid purpose and they don’t provide an inverter for off-grid, which means that technically speaking, you will void your SolarEdge inverter warranty by using it in an off-grid setting. That doesn’t mean it won’t workoff-grid, though.  SolarEdge is an average, non-battery inverter.  It’s looking for a grid signal. As long as the battery inverter provides a pure sine-wave grid signal, the SolarEdge inverter doesn’t know the difference between the battery order and the grid signal. 

It’s more SolarEdge making sure that your battery inverter is, in fact, feeding its inverter system. Or, it could be that its inverter system might be back-feeding your battery bank. They don’t want to be responsible if you use a battery inverter that doesn’t have a frequency shifting, where if the battery banks are full, it has to be able to shift the frequency to turn off the solar array. 

One reason why that’s important is for generator selection. There are generators that are specifically built to be compatible with renewable systems. These generators are basically variable-speed generators that will throttle up and down based on the load, rather than just supplying a constant stable power output. So, they’re energy efficient. 

It’s difficult to say how important it is to use a renewable generator on an off-grid system. A renewable generator is more important if you are trying to use your solar array in an off-grid setting, so that if your solar array is providing when your battery isn’t. 

Let’s use the example of a solar array on a home and a hurricane knocks out the power grid. You might think of the breaker on your service panel?? and crank the generator and provide a grid signal to the solar array. Turn the solar array on and run the home off solar, and let the generator idle at that level, where you have your load jumping up and down. This is a solar array that’s providing some power, but clouds might drop the power production of the array.

Then, a variable-speed generator that can throttle up and down based on the load is advisable. But really, it’s the battery capacity and range of the battery capacity. It is also capable of taking any surplus power from a generator or surplus pin or from a solar array and using it to store in the battery.  It’s when you don’t have a large battery is when you want that variability in a generator. 

It’s easy to have a problem with a renewable generator on an off-grid site. There are so many different settings in a battery inverter, and now you have a generator that has a hundred different settings.  When I was just getting started, I was auto starting a generator and trying to charge the batteries.  Because the array had enough power to supply the home, the generator didn’t uptick and throttle upwards to charge the batteries.

 It “thought” all the load is being supplied, so I don’t have to output any power. It was idling when the batteries were at 50%. You have a generator control system built in to these off-grid battery inverters, and so you don’t necessarily need a generator that provides its own control system.  If you’re trying to get away with a small solar array with a small battery, that might make your generator run more efficiently when it’s in use.


Honestly, not even the generator companies know exactly what the best configuration is for a generator + battery +solar array. Let’s not forget, we are still in the nascent stages of renewables, and there is much to learn, especially with battery configurations.  Always feel free to contact solar experts, including myself or online forums and ask for advice.  It’s what we all do. 

  Puerto Rico, where the grid has been obliterated, had some generators, but the problem is they require gas. It would it be nice to assist the generator with a solar array + battery.  When running your whole house or building on a generator, generator signal quality becomes important. However, if the generator is simply playing free safety, to rapidly charge the batteries to move the building back onto the solar inverter, then the generator quality needn’t be so important. Ironically this means an “off-grid” or “renewable” home generator isn’t absolutely needed for an off-grid solar battery, instead, it those labels are more important when running off the generator full time.

 There are a lot of components that are built into or added on to a battery inverter that you don’t get with just a solar inverter.  A solar inverter takes the array power and outputs it through the inverter into the load.  It doesn’t take an input to charge the batteries. A battery inverter is going to be more expensive than a solar inverter because it goes in both directions. 

A battery inverter is bi-directional and a solar inverter is not. You could have a battery inverter that is built for DC coupling and is not bi-directional where it gets charged from the solar array not from the grid. Some battery inverters are built for either the European or Australian market on a 230 volt or 50 Hertz signal. 

Some battery inverters are coming from small cabins that might only be on 120 volts.??? RVs are on 24 volt batteries with 120 volt outputs. SMA makes a 120 volt battery inverter. Sometimes you have to add a 240 volt transformer on top of that.  There are not a lot of 240 volt off-grid only inverters. 

An inverter that is not listed for a grid connection might be cheaper because it hasn’t gone through the UL listing process. Another reason for reduced price would be that it isn’t using safety provisions built into its hardware and software to disconnect from the grid during an outage and transfer onto a critical load panel.

You could avoid some of that if it’s off grid.  Likewise, local AHJ jurisdictions that require external disconnects and external transfer switches and you might get an off-grid inverter that has all of that built into the inverter cabinet itself. 

If you tried to use those components, you wouldn’t get through a grid connected inspection. Not that they are in violation of National Electric Code, but they might be in violation of standard interconnection requirements from the utility.

Adding a transformer on top of that is that an additional failure point in the system, but they also produce extra noise. Battery inverters can produce more noise than solar inverters, which are generally quiet. Battery inverters have multiple modes of operation in them. Schneider inverters can be used for off-grid.  They can also be used for grid connection.  (this sounds like a child said it)

Grid connected battery inverters are used in different ways.  It might just switch to a critical load panel during a power outage. In Hawaii, in order to interconnect a solar array to the grid, you have to put it into a zero export mode so that it can still receive power from the grid but it won’t push power out.

For commercial design, it would be in a demand management mode, where it’s monitoring how much power comes in from the grid.  Then, run the battery to make sure your amperage draw stays within a predefined threshold. There’s another mode of operation, when you want to prioritize using the battery rather than using the grid.  

In the long run, it might seem more expensive to use the battery, but the customer is paying for it, at least in part, to reduce their electric bill.  Remember that each client has their own needs and desires, and it is our job, as much as possible, to give them what they pay for. 

Some clients want to know exactly what their bill is going to be, rather than select some other, more cost-effective configuration, that can seem more abstract. Also, the monitoring systems of these battery inverters are more complicated. For those reasons, you may not just be monitoring the production of the solar array, you may also be monitoring the current that flows through your main service panel.

In the case of load management, you have to be able to monitor the electrons coming out of your service panel.  You’re also monitoring the electrons out of the battery itself. The Tesla inverter (remember, it’s counting the kwhs) that come out of it. You’re also monitoring the temperature of the battery, which helps the inverter not pull too much power out of the battery. In fact, you’ll get into additional specs on the battery where if you need to shed some circuits because you’re drawing too much power, you can turn off auxiliary circuits.

Some additional features included are if your batteries are at a full state of charge, you’ll turn on certain circuits that might not otherwise come into play. This prevents overcharging your battery or so that you can use the power that you might not otherwise use. This could allow for a water feature on your property, like a water fountain in your yard that turns on only when your batteries are at a full state of charge.  Another example would be a heated swimming pool.

 When you’re producing surplus power that you’re not using there is a visual indicator for the customer. Obviously when the client has a healthy state of charge, why not use it for something fun?  They are paying for it. 

To assist your customer with energy management here’s something that I ran into that you might not expect.  Off grid design with battery-less solar is designed around a National Electric Code upper limit of 600 volts. You’re wiring your modules, positive negative, positive negative positive up until a 600 volts cutoff.  That would mean I would have ~8 circuits for the 22 kilowatt solar array. 

With lot of off-grid products, the charge controllers are not first 600 volts. Solar modules were more expensive when they first came out, obviously, and instead of using power optimizers for shade mitigation, they just used less solar modules and more circuits. Clever at the time, the shade only impacted one circuit and not the others.

A lot of your solar charge controllers that are available for off-grid are only 150 volts charge controllers. If you use a grid-tied, high-voltage solar module that’s common for the current solar industry, you can only get about three modules a circuit. If you’re doing a 22 kilowatts solar array with 80-plus solar modules and have three modules a circuit, you will end up with 30 different circuits for the off-grid solar array.  If you put that on a roof, the wiring is just going to become a giant mess.

 If you’re doing a ground mount solar array, doing a bunch of wiring is not as biga deal as if you’re doing a ground mount solar array and you have lots of wiring. When a mouse comes and chews some wires, losing one circuit out of 32 is better for risk mitigation. And, that will happen.

But when the off-grid system is going up on a rooftop, you don’t want to spend time and do all this extra cable management.  Then, you want a big long 600 volt circuit. (why? What about squirrels then?) Them more you know, the  more obvious design selection becomes.  In this case, the only companies that supply the off-grid market are Schneider and Morningstar. (is that important?) should we tell them we are mentioning them in the class?

 pìkô and pika is a start-up and Morningstar 

These companies make a quality product, but they don’t have the brand name recognition yet that Schneider does. Surprisingly, Outback, SMA, and even Magnum, all of which have stellar reputations do not have a 600 volt or even today have a 600-volt charge controller. Ultimately, I selected Schneider for this off grid project.  

I needed a charge controller to be made by the same company as the inverter because I wanted to charge controller and the inverter to work together flawlessly.  (why? Does someone want ot do a project where they don’t run flawlessly?) Running an Ethernet cable between the charge controller and the inverter so they can communicate with each other is a good idea.  So that’s why I chose Schneider in that particular example.  

I’ve started to look at even more obscure battery inverter companies particularly ones that are not rated for grid connection.  (why?)

Another question you want to ask at this stage is are the inverters expandable? Not all of them are.  Usually if you have a grid connected, capable battery inverter, it’s going to be expandable, but sometimes the inverters that are designed to be strictly off-grid. The company, Ames, is an example of one that’s not listed for grid connection.

Again, it’s substantially cheaper than any of these companies because it offers less. It’s a pure sine wave inverter with surge ratings and a lot of good features built into the cabinet, but it’s not stackable or expandable.  This means  their maximum inverters size is 12 kilowatts. If you’re doing a higher-end home, this would not be a good fit. They don’t make a product that would serve that higher demand. 

I talked about combining multiple battery chemistry’s out of all of these inverters only tyka said that it supports multiple battery chemistries.  However, when I called and asked, they stated they don’t support multiple battery chemistries if they have a lithium-ion o and a lead acid option

 In this same design, I went with two inverters due to  expandability. Using two inverters rather than one is something solar professionals prefer, because with off-grid it’s always nice to have backup. 

The Schneider inverter is quite loud, so as a transformer built into it and it does make a lot of noise so I put it in a dedicated control room inside the garage.  (Remember, this client was willing to pay more for less noise.) 

There are 600 volt charge controllers. I had eight circuits and originally I remember thinking with eight circuits I would put two in each charge controller. However, they could handle three circuits (why??) I went with a circuits of three, three and two. 

This was a little bit of a controversial decision because our solar array was 22 kilowatts and each charge controller had an output of just under five kilowatts and so if I had six three circuits three circuits and then two circuits, basically I had five kilowatts going into here, and twenty to eight times three and then nine kilowatts and nine kilowatts going into here and here. 

 I grossly oversized the DC capacity input on the charge controller versus the DC capacity output.  If I instead had gone with two circuits on each of these, they would have been better sized and I would have gotten more power output out of the system.  The reason why that was preferable is because what we’re doing that 22 kilowatt oversized array specifically to produce more power on overcast days. 

 and so when We’re putting eight kilowatts or you know nine two two nine kilowatts of DC capacity onto 5 kilowatts of DC output capacity, on those overcast days, we’re still only feeding it with four kilowatts of input. Five kilowatts is the output, and so  we’re over sizing the DC side of things specifically to produce more power on overcast days.

On those sunny days they now have surplus power.  If the client adds an electric vehicle, heated pool, or something like that, I may put in another charge controller and run those circuits over there. 

Back on that battery bank size: You don’t just have one battery plugged into your inverter.  Even this 12 volt battery is inside the battery. It’s comprised of multiple two volt batteries that that stack up to 12 volts. Within the battery itself, you might have different what are called “cells” in the battery, and if you have two batteries and they’re plugged into the inverter, you get resistance out of these cables.

Tags: ,
Not Permitted for Redistribution. Licensing available on request.

Posted August 28, 2019 by John Cromer in category "Classes

Leave a Reply

Your email address will not be published. Required fields are marked *