Justifying Solar Panels
I am planning on adding a workshop/garage/multipurpose building in the near future. I’ve been mulling over the idea of including PV solar as part of this project, but I’m having a hard time coming up with a good justification, either economically or environmentally.
The main limitation is that even though I have grid electricity at my house, I cannot tie into it. No selling excess power, not even net metering. So I would have to have a system that runs off batteries, and can also smartly switch from PV to grid feed when the PV isn’t producing enough. That’s an initial investment that will almost certainly never be made back on electricity bills.
The environmental picture isn’t any better. The grid here is already 96% carbon free generated, so the carbon cost of a bunch of solar panels and batteries also seems like it could never be recovered either.
Am I missing something, or does solar PV just not make any sense for me? I’ve had years of “green” propaganda convincing me that PV solar is “good”, so my gut is having an argument with my head on this one.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Trevor,
Your instincts are correct. Grid-tied PV systems make sense. Battery-based off-grid systems do not.
I'm in agreement with you and Martin. And I would normally favor utility scale solar versus residential. However, your utility situation could change in say, five years, so it may still make sense to plan for a solar installation. This would mostly be orienting the roof of the building correctly and perhaps getting the trusses, etc made with engineering stamps that certify that the roof is capable of handling increased wind loads. Just an idea.
I would be tempted to go for a small hybrid inverter/charger setup, something like an Outback VFXR3524 that can do grid zero mode with a small battery bank. Just get enough panels to cover your baseload but not enough that you use the storage much if at all. In grid zero mode, it will not export power, it will just supply enough to feed the downstream AC loads.
If you sharpen your pencil, you can probably put a kit together for under $5k. Still won't make sense on ROI, but now you have a backup that can run your lights and internet for a day or two.
I don't see the need for batteries. It doesn't make economic sense to try and save your excess production on-site if you can't sell it.
What you want to try is called "peak-shaving" where you try to produce enough of your electricity at times when you need it to make it worthwhile. The break-even point is a much smaller system than if you have net metering and are trying to go net-zero. But it's still possible for you to have a system where everything you produce is usable.
As a ridiculous example, I doubt your usage ever falls below 5w. So if you had a 5w solar array 100% of its production would be usable. You need to figure out what your baseline usage is and size to that.
From an environmental perspective I don't feel it matters that your local utility is 95% renewable. Electricity generation is a national and even international marketplace. The electricity you consume could be sent somewhere else that doesn't have such a favorable composition.
DC beat me to mentioning peak shave systems, which I frequently advocate here on GBA as a way to use solar without need of net metering rules. Peak shave just means you offset your normal grid load whenever your system is producting, and at night you use grid power normally. This is a great way to put in a system without a lot of expense too -- just size the system for your average daytime load and use a grid tie inverter and you're good to go. If you have "time of day" utility rates in your area, peak shave systems pair well with that since peak solar production is usually close to the peak pricing on those time-of-day rates, which means you maximize your savings (off peak rates are typically much lower).
To comment on DC's last paragraph though, electricity is NOT a national/international marketplace. It's regional. There are several distinct regions in the US that are not interconnected at any significant level (and they're not synchronous with each other, either, so it's not easy to interconnect them). Most of the West coast is part of what is known as the "Pacific Intertie", Texas is it's own system for the most part, and most of the Eastern US, including the midwest, is part of the Eastern Interconnection. Quebec runs their own system up in Canada for their region of Eastern Canada. These systems can't really move power between each other. Note that these systems due interconnect Canadian and US grids in their respective regions though.
You can't really move discrete units of power long distances either. What the grid operators do is called "wheeling", and they load shift. Say a utility three systems over needs an extra 500MW. Utility #1 supplies 500MW to utility #2, ,then utility #2 supplies 500MW to utility #3. It's utility #2's generation supplying utility #3, with utility #1 primarily supplying utility #2, not all the way to #3 where they're selling "their" power. What this means is that while it may look like you're offsetting generation hundreds of miles away, you're probably not really doing that. If you're worried about emissions, this matters -- your offset in a region of primarily local hydro power doesn't necassarily mean you're offsetting distant coal-sourced electricity.
It's a lot easier to justify solar systems when you think about your own load and not the grid scale stuff. If you go with peak shave, and don't try for net zero (which I'm not a fan of) or selling excess power, you can usually make a smaller solar system make financial sense with today's prices, especially if you can also get a time of day rate from your local utility. I think using net zero as a goal puts solar installations out of reach of many people who would benefit from smaller, cheaper, systems that could be used as peak shave.
Bill
So in this peak shaving scheme it's possible to be drawing power from the solar array and the grid simultaneously? How does that work?
Not only do we not sell excess electricity here, we pay to dump the excess.
As Bill said, you use what's called a "grid-tie inverter." An inverter takes the direct current that PV cells produce and converts it to the alternating current that is used in household electricity. "Grid-tie" means it's connected to your electric grid. There's a couple of reasons for that. First the alternating current provided by the grid runs at a precise frequency, you want the electricity you produce to be at the same frequency. Second, if the grid goes out you want to stop producing power, so that someone trying to repair the outage doesn't get electrocuted. With a grid-tie system when your PV panels are producing your draw from the grid is just reduced accordingly, it automatically switches between them and maximizes your use of solar.
The question is what happens when the PV produces more than you're using. With net metering you put that electricity back onto the grid and your electric meter literally runs backwards. Without net metering the simplest thing is to put the electricity back into the grid anyway for free. If your power company won't allow that then you have to figure out a way to limit your production when that happens.
Yeah, they won't. I'm pretty sure they don't even allow the tieing in the first place. Which kind of makes sense, if they don't want you feeding back into the grid, it's a lot easier to just prevent you from physically connecting anything than expect a home owner to devise a system that will somehow limit the production.
I just ducked "peak shaving", and it sure seems like it relies on batteries. Also seemed like it was all about grid level or industrial usage rather than individual residences.
It's common in the commercial world to do peak shave with onsite generation, often diesel generators. This is because on-peak rates are sometimes VERY much higher than regular rates. Many of the datacenter facilities I work with at work get on interruptible industrial rates because they are cheaper per kwh, and electric costs are usually the biggest operating costs for datacenters. Since the datacenters have backup generators anyway, if the utility calls for interruption, the datacenters just run on generator power for the duration of the peak/emergency the utility is experiencing. Overall, this saves the datacenter operator money.
I'm not aware of anyone who uses batteries for peak shave purposes aside from the new utility scale battery systems. The reason is that batteries just don't store enough energy to be particularly useful here, and they are both very expensive and another maintenance item. The reason generators are so good for peak shave in my example above is because they are ALREADY part of the system design, for backup purposes. As a result, all of the maintenance is already taken care of, and the equipment is already purchased. Using the equipment for a few peaking events in a year pretty much just adds fuel cost, and that fuel cost is offset by the ongoing power savings on the reduced rate you get with interruptible electric service.
In your case, you would just be using the solar power as much as possible while your panels are producing. As the solar output tapers off as you get into the evening, you start using more grid power. There is no "switching" between solar and grid power, it's all handled by the grid tie inverter which is in parallel with the grid (that's the "tied" part). If you use more power than the solar system is producing at the time, the extra is pulled in from the grid. If the grid drops out, so does the grid-tie inverter (be design, for safety). Some of these inverters can be configured to ensure they NEVER export any power to the grid, so if you use LESS than your solar system is producing, the inverter will just back off what it's supplying to make sure your draw from the grid never gets below zero (assuming "negative" here means you're sending power TO the grid rather than drawing power FROM the grid).
Bill
The device you want is called a "zero export grid tie inverter."
Your public utilities commission is probably the best source to find out whether they are legal where you live.
I worked on a commercial install out near Woodstock that had similar limits. The solar could be grid tied but could never export any power. The onsite generation went to feeding local loads, once the battery was full the solar was curtailed. Generally on most day with decent sun, there was little grid draw during the day.
In terms of installed solar capacity, it is not the best use of panels since the PV does spend a fair bit of time being curtailed. It really makes no sense not to be able to export especially if the grid actually needs the peak power such as during hot summer days but those are the rules.
As for batteries, you don't need something expensive or large. A pair of decent lead acid batteries are good enough (group 34C or 4DM). The idea is to only cycle them as needed to cover PV drops due to clouds, but don't discharge them to try and load shift. As long as the lead acid is kept full most of the time and not discharged excessively it will last.
Depending on the size of your PV, a diversion controller to a hot water tank could also be useful. This could cover most of your water use and better ROI than a heat pump water heater.
As much as some utilities are pushing back on PV, residential grid tied solar will be the norm in the future.
Take the money that you would spend on solar and donate it to some cause that is working on getting politicians to properly regulate utilities - so that utilities shut down coal and install the far more cost effective utility scale solar (and wind).
On a macro scale, residential scale grid tied solar is a waste of money. Hopefully it will be replaced by utility scale solar.
There is no need for batteries in a grid tied solar system. It's much simpler to just "waste" some of the solar output if it's not needed at the time it's generated rather than trying to bank it in a battery. Batteries are expensive, they're a maintenance item, and they have other safety issues. For the relatively small amount of power they store, why bother in a grid tied system? The power was free to make anyway, so wasting a bit isn't a big deal, and you save the cost of the batteries and don't have to worry about replacing them down the road, either.
Residential peak shave solar systems make financial sense today. Payoff is usually 3-5 years or so, depending on a combination of how sunny your locale is, what you're electric rate is, and how much electricity you use during the day when your solar system is producing. Peak shave systems act to REDUCE daytime loads during peak times, which is a good thing. There is no need for fancy storage systems or net metering rules for residential solar to make sense. People who think that are either overthinking things, or trying to make money off the net metering rules (like the solar system rental companies).
The utilities actually have pretty good engineering reasons why they don't like net metering, and there are some valid cost reasons too. Utility scale solar and wind have their own problems, and aren't a magic solution to energy supply issues.
Bill
Bill ---
I've seen it posited on the Internet that if you take a zero-export microinverter, put a cord on it and plug it into a wall outlet, what results doesn't need to be permitted nor need utility approval. What is your considered opinion? Setting aside for a moment the question of how anyone would ever know if you did.
Its a code violation to put a plug on a hard wire device. So it is a non starter from there.
If I wanted to build a guaranteed no export grid tied setup, I would go for a DC couple setup.
Go for a larger 24V or 48V high efficiency power supply feeding a DC bus. Wire your MPPTs (and/or batteries) to this DC bus from there run a quality inverter to supply your local AC power. This way if there is sun the MPPTs will push up the bus voltage above the setpoint of the power supply and all the power will come from the PV, if not it will be supplied by the grid. Since the entire setup is DC coupled, there is no chance of backfeeding.
Does add an extra power conversion and associated losses. Power supplies can be had with 95% efficiency, so a loss but not that much.
Proper zero-export is much simpler though.
You tend to get in trouble with testing lab certifications (UL, CSA, etc.) by "modifying" a hardwired device to have a cord on it, but code says everything must be "listed for the purpose", by which they mean listed by one of those testing labs (NRTLs, as they are known -- Nationally Recognized Testing Laboratories). That means you're not suppose to do it :-)
I'm not so sure the utilities can really tell you you can't hardwire such a device though. They can't tell you you can't put in an ATS and generator, for example, and they can't tell you what kind of light or ceiling fan you can use. There was a case a long time ago with ATT and the hush-a-phone that might give you some legal background here (not sure how applicable it is though).
Essentially I'm not so sure the utilities can bar you from putting in a grid tie system IF it meats all applicable codes. The code regulates things inside your home, not the utility. As far as I know, all grid tie inverters will automatically disconnect if line sync is lost (such as if the power fails, so that you won't backfeed), so you shouldn't be able to backfeed the grid and cause problems. What the utilies CAN regulate is net metering stuff, saying they won't pay you for anything. If you still have a mechanical meter, they won't even be able to tell if you did feed them some power as long as you didn't go negative in a billing period -- any backfed power would just deduct from the count on the meter.
The issue is you don't want to be sued and be some utility's legal test case here. They will have much deeper pockets than you do. From a technical side, which is what I am most qualified to speak to, I don't see a problem IF applicable codes are met, which includes the auto--disconnect function in case grid power is lost. That's the safety feature that is there to protect utility crews.
BTW, Akos' idea about a DC link (what we call "online double conversion" in the telecom world, would work too, at the expense of additional system losses under ALL conditions (even when running directly on grid power). You don't need batteries for this, you need a power supply, and inverter, and a DC power supply that can keep a contant output voltage from your solar system (not sure if a charge controller will do this). You need two beefy blocking diodes, one for the line power supply and one for the solar power supply. Set the solar supply voltage higher than the grid power supply's output voltage. This will reverse bias the grid power supply's blocking diode so that the inverter draws power from the solar setup. You won't get very good load sharing though.
I actually designed a system for this some time back to share power between two DC sources (a DC generator and a battery bank). I did it with a switching power supply with a special PWM controller I designed that did every other pulse from the opposite power source. By controlling the width of alternating pulses of the PWM waveform, I could I very precisely control power sharing from the two sources. It's not terribly difficult to implement this these days with some FETs and a PWM capable microcontroller (I used a PIC from Microchip Technology), along with associated drive electronics. The tricky part is the magnetic component design. If you design the DC link to have dual positive and negative riding rails of ~170v DC or so, you can build a very simple inverter stage using some IGBTs in half bridges. Two of them running 180 degrees out of phase will give you 120/240 split phase residential power service. Note that this isn't a DIY project for someone without some pretty high level electronic design experience so keep that in mind if anyone wants to try it.
Bill
This is somewhat of a personal desire on your wants vs. needs. You want batteries if your desire is to be independant of the utility grid system after sunset.
It's peice of mind but it is expensive. I started off with a manual transfer switch and it became a pain after a year. I have never seen a financial payback from my batteries but once when the power went out and the neighborhood went dark we popped some popcorn and watched Netflex (the local cable repeater was in a different grid system down the road than what affected us) during a complete outage.
Here is a video I made... jeeze, 10 years ago already?! As I was experimenting...
Prices have changed, technology has changed (Moore's law) and power prices increased. Would I do it again? Yep. https://youtu.be/CinOm23_-FQ
Can anyone point to some particular products for the inverter? I tried searching using the keywords mentioned here, and came up empty handed. Mostly just blogs talking about the devices in theory, and even when there's a specific product mentioned they're not listed for sale.
Almost all manufacturers offer a setup that will do grid zero but usually requires some extra hardware for the controls.
You can look at Enphase with their Envoy box. The box can be programmed to a number of grid profiles and one of them is power export limit. This setup requires no batteries but they do have a battery option that can be added on later as well. Make sure to check the Envoy documentations as not all of their micro inverters support zero export.
SMA also can do export control if you get their cluster controller and energy meter. Again, no batteries needed.
Most (if not all) of Outback's grid tie inverters can do grid zero with no extra hardware as long as all the loads are downstream of the inverter so the inverter can measure load current directly. Most models need a battery and separate MPPT for the solar panels.
Schneider Conext XW is similar to the Outback units but they also support a remote energy meter if you want to zero out your whole house load not just the loads downstream of the inverter.
Fronius also supports zero feed in mode on most of their inverters but need their external power meter (Fronius Smart Meter).
Make sure to talk to the manufacturer you choose (usually go for one your installer if familiar with) to make sure you get all the right equipment and documentation on how to set it up. Generally solar setup are pretty simple but for some reason setting up any of these "special modes" always end up with some hair pulling and cursing.
+1
Example: I have a Fronius + smart meter positioned in front of the utility meter. By law here in Germany, you can only export 70% of the installed peak power back into the grid (residential installation) - after the house load is consumed now you rarely see that cap. For you - use 0% instead of 70% and you are done. Standard installation workflow - the installers are used to it and normally do not mess that up. If you ask for more (datalogging etc.)then your hair suffers again..