[brian_]

In the mean time you can have one thing connected to show you use the power and then disconnect everything else. You end up paying a minimum bill to cover paper costs but it keeps you legal.

If you are going to have all of the overhead costs of having a connection, why not use it? Sell excess energy to the grid, and keep the ability to buy as much as you need for anything you have in case you need it. This is a hybrid system... and to be fair it is the most expensive (due to complexity) of the choices (which are off-grid with battery, grid-tied without battery, grid-tied with backup, and hybrid).

Also, since the only reason for a requirement for power grid connection is to ensure habitability of the residence, a grid connection to some token device may fool power company administrators, but wouldn't pass any inspection.

 

[brian_]

200 amp at 240v is about 48 KW - I would be very surprised if you could draw that much!

Decades ago, much lower capacity service was normal; my first house was built in 1954 and I think it had 60 amp @ 240 volt service. That was not enough for everything that a typical family would have, and by the 1970's 100 amp service was normal in Canada. Since then it has escalated, and 200 amp service is typical, at least for electrically heated houses. One problem might be that the total of the all of the circuit breaker capacities in the panel must not be more than some defined multiple of the panel and service capacity, to minimize the chance of overloading the service. Since there is much more branch circuit capacity than would ever be used at once - largely to provide outlets where ever they might be needed - the numbers are large.

I agree that the residence is unlikely to ever use 200 A @ 240 V even momentarily, but some components will need to be sized for that.

Actual peak draw is an interesting question. Typical North American homes have an electric kitchen range with multiple stovetop elements plus the oven, with 50 A @ 240 V service. The clothes dryer is typically electric: 30 A @ 240 V service. Kitchens have multiple 20 A or even more 15 A circuits, to handle appliances such as toasters, kettles, and other cooking appliances. Air conditioning is common - no idea offhand what that takes, but it's significant. While modern LED lighting is no big deal, the incandescents of a few years ago can easily be a kilowatt over the whole house, and peaks can be much higher with stuff such as outside floodlights. The piles of electronics are small change compared to heat-producing appliances. The really variable part is what is plugged into outlets - maybe next to nothing, maybe multiple 1500 watt portable heaters, large woodworking power tools, and a 5 horsepower air compressor in the garage. While most people here use natural gas for both space heating and water heating if gas is available, millions of people in Canada have electric heat and even more have electric water heaters; space heating alone is over 20 kW for a typical house, and whole-house water heaters run about 6 to 9 kW. With unfortunate timing, a 20 kW peak in a home even without electric heating wouldn't be implausible.

Of course a solar user going off-grid likely won't use plain resistance electric heat (more likely would use a heat pump or another energy source), and would likely be reasonable with other loads.

 

[brian_]

I'm using a simple system - the solar panels feed the inverter and it makes AC - which it delivers to my house
I'm also connected to the grid
Basically I use power as I want - if I'm using more than Solar then it comes from the grid
Less then the surplus goes to the grid (and they only pay a pittance)
This is all automatic

Classic grid-tied system. This is optimal if

• power during grid outages is not needed or important, and
• the utility buys excess energy at or near the price of energy purchased from the grid.

This covers a lot of solar installations, but of course not everyone, depending on local factors. Here in Alberta utilities pay enough for "microgenerators" output and in most areas power delivery is reliable enough that a basic grid-tied system is a good choice.

Even this system without storage is not trivial to build correctly. A safety risk in this sort of system is that if not controlled properly it can create an electrocution hazard if the grid connection fails near but outside of the property.

Of course if using on-site storage (for any reason), this basic grid-tied system is not applicable.

 

[brian_]

Per his O/P, grid tie inverters are expensive. How long would it take to repay the investment of $15,000 at grid buy-back prices for his excess electricity?
...

First, that's a very high price for grid-tie inverter: in a randomly found listing of grid-tie inverters the highest-power inverter (11 kW) cost US$2736 and the most expensive was US$5863.

Then there's battery storage: the OP has a specific battery pack, but in general adding storage costs money, and using the grid connection avoids the need for storage, except as back-up.

Finally, the proposed scheme to which I was responding was to isolate everything except a token device (for legal purposes) from the grid: that would mean a solar-powered installation capable of handling the household's peak demand, which would be relatively expensive, instead of whatever size of solar system is desired and the grid providing any peak power desired. In a non-grid-tied system the inverter must be sized to handle peak demand; in a grid-tied system it only needs to handle peak solar output (which in most cases will be substantially less). Is the large non-grid-tied inverter much cheaper than a smaller grid-tied inverter matched to the solar system output?

 

[brian_]

Also, when I last looked many grid-tie inverters WOULD NOT FUNCTION if the grid was down. Have they fixed that?

That's what I recall, too; however, what I found when this thread came up was that grid-tie with battery backup is now available, which means that it does operate when the grid is down - a substantial functional improvement.

As in all things, the market changes quickly. You would need to perform a full analysis of your requirements and price a variety of solutions to fully optimize your results.

I agree!

 

[brian_]

So, there is a new category of Inverter which works in either grid-tie or stand-alone mode. I can probably search myself, but if you already looked it up can you tell me if it is smart enough to flip a breaker to isolate your home from the grid?

I didn't keep links to specific hardware, but there's no way these things could be legally installed if they didn't isolate the power source from the grid connection in case of grid failure.

Now, of course, I need to do more research...

The ideal solution would be one where a computer somewhere in the system (could be in the inverter, could be stand-alone) acted as a programmable brain to determine when to pull from the grid (i.e. modify how much battery discharge to allow when, say, dawn is approaching so as to resist some "hard rule" that says "buy power" when in fact the sun is about to come up) and when to sell it back (batteries are almost full, sun is shining, still 6 hours to sunset so sell some portion during peak hours at a premium).

I was seeing descriptions as" grid-tie with backup", which suggests that the logic will just maintain the backup battery, and sell or buy net power production or demand as long as the grid is available; that wouldn't cycle the battery at all except in case of grid failure. I agree that a more optimal strategy is possible, but I don't know if it has been implemented.

An analogous situation is a gasoline-electric hybrid car: the most simplistic approach would shut the engine down (like a grid power failure) for coasting/braking and until the battery is run down, and use the battery only while the engine is shut down (regenerative braking) and for initial acceleration (until a switch to engine power). It would work, and would deliver some benefit (like any stop-start system, plus regeneration) but it's not worthwhile or at least not an optimal use of the hardware. Real hybrids use a more sophisticated control strategy to make use of engine power.

A utility might manipulate the logic of a grid-tied system with storage for a couple of reasons:

1. to provide power the grid when needed to handle daytime peaks,
2. to buy power from the end user when it is cheap and sell it when it is expensive, if the area has time-of-use pricing.

These two factors seem contradictory.

I can guarantee you that if you buy such a solution from the power company then the algorithm is going to favor the power company, not you. I think I can also accurately predict that such solution s WILL be sold, and marketed aggressively by the power companies.

Maybe, but utilities here in Alberta are required by regulation to allow grid-tied systems, with minimum buy-from-end-user prices, and I haven't seen any of them pushing any kind of system with storage. Perhaps the pricing is the reason: even with logic favourable to the power utilities, "microgenerators" (as end-user installations are called here) are probably not an attractive source. Also, we have a bizarre semi-regulated utility system in which energy generation utilities, energy distribution utilities, and energy retailers all separately operate - it might be difficult for any of them to deliver and manage a grid-tied system to their benefit.

According to the Solar Energy Society of Alberta, there are 2,343 solar installations in the province, with installed capacity of about 40 MW. This organization's website describes only the simple grid-connected configuration (no on-site storage, for backup or otherwise), so anything more sophisticated is at least not common. A local energy retailer offers leased solar systems, for which the description is clearly grid-connected, but it is unclear if backup or more flexible storage system is available. An Alberta solar-only company has this to say about storage:

The simplest and most cost-effective way to harness solar energy is with a grid-tie system. A grid-tied system allows you to use solar power when it is available during the day, and grid power when it isn’t, sparing you a huge cost in batteries, or from meeting any power requirements. Batteries are bulky, expensive and require regular maintenance, and won’t last as long as panels or inverters. Therefore we don’t recommend battery back-up systems unless you live or work in an area that experiences frequent power outages, and even then there might be more affordable solutions (such as back-up generators) than storage batteries.

... which implies that storage in a backup mode is available here, but doesn't suggest use of on-site storage while the grid is up.

Since pricing here essentially pays the solar system owner the same for energy as it costs them to buy energy, there is no incentive to cycle a local battery, so a battery is only for backup. Different pricing in other areas will drive different system choices.

 

[brian_]

Grid-tie with battery: example #1

So, there is a new category of Inverter which works in either grid-tie or stand-alone mode. I can probably search myself, but if you already looked it up can you tell me if it is smart enough to flip a breaker to isolate your home from the grid?

This is an example:
Solar Plus Storage - Grid-Tied Solar Power Systems with Battery Backup
As the title suggests, the battery appears to be only for backup...

When the grid is available, power produced by your solar system is used by the home and excess energy is sent into the utility grid. In the event of a power outage the backup inverter will automatically turn on to power critical appliances such as the fridge/freezer, lights, well pump, computer, phone and more. The combination of solar energy and battery storage works seamlessly to provide power to essential appliances during an outage, working similar to an off-grid system.

The system design addresses the isolation issues by using the normal features of a grid-tie system (using multiple solar panel powered microinverters), with a separate inverter for operation from the battery:

Magnum Energy’s MicroGT 500 Microinverters are paired with a powerful Magnum Energy MS4448PAE battery backup inverter.

The wiring diagram is part of the purchased package, so details are not freely available, but it appears that the "power center" connects the loads to the backup inverter when the grid-tie inverter automatically disconnects. How the battery is charged is not clear to me.

 

[brian_]

Grid-tie with battery: example #2

Schneider offers two configurations using their inverters and control hardware for a grid-tied system with battery backup.
Solar Grid-Tie Inverters with Battery Backup
Diagrams with specific equipment models are provided, and which one is used depends on the metering method allowed by the grid connection.

• FIT (feed-in tariff): all solar output goes to grid through one meter, all power used comes from grid (when up) or from battery (when grid down)
• Net metered: one inverter provides AC power from solar, another inverter provides AC power from battery; power flow bi-directionally as required to absorb solar output and meet load

There is a pile of documentation on this Schneider site. One would need to plow through it to determine if the system can use battery storage in the desired way while also using a grid connection.

Schneider knows their stuff. Even without digging into the details, I expect their systems work well.

 

[brian_]

Since pricing here essentially pays the solar system owner the same for energy as it costs them to buy energy, there is no incentive to cycle a local battery, so a battery is only for backup. Different pricing in other areas will drive different system choices.

You are right it doen't make sense yet, but I think it will in the near future.

Cycling a battery doesn't currently make sense in a grid-tied system here in Alberta, but it likely does in other areas, where the price paid for energy sent to the grid is much lower than the cost of energy from the grid.

Time-of-day metering matters as well, but may favour grid interaction rather than battery use.