My complete Off Grid system- Montague Township, Canada,

Karel
Karel Registered Users Posts: 4
Hi everybody,
I'm new here, in fact I'm a newcomer in Canada. I'm an electrical technician very much interested in starting my own business in renewable energy field. A few weeks ago I bought a piece of land south of Ottawa and thinking really seriously about making my new house totally off grid. After weeks and hundreds of hours of studying the matter this is the system that seem to have everything that me and my family will need:

1. 16x Astronergy CHSM 6612P-310 310w, Voc=45,42V, 4x4 strings (182Vdc), 2x8 pole mounted, total power 4,96kW (MN4PVHV combiner)
2. generator GENERAC CorePower 7kW (LPG) + 250gall underground tank (multipurpose)
3. Schneider 250A power panel with Xantrex XW-MPPT 80-600 high voltage charger, Schneider Electric XW6048-120/240-60 1 or 3-Phase Hybrid Inverter Charger, Schneider Electric XW System Control Panel (SCP), Schneider Electric XW Automatic Generator Start (AGS), Conext SW/XW Combox for Remote Access and System Monitoring
4. Battery bank 4x4 UB4D AGM 12V batteries (smart gauge) for total of 48VDC 800Ah bank.

My plan is to build a 2 floor 2350sqft house (already have a drawing) plus 500sqft basement and 480sqft garage/workshop.

Heating: air tight wood furnace, propane furnace
A/C: ceiling fans, one window mounted AC unit
Water heater: solar preheated, propane heated

I would really appreciate any advice which would make my system well balanced.
Is my inverter too small (60A)? I won't be using my electric power for heating, even my laundry dryer will run on propane and will be used rarely.
Battery bank OK?
Thanks

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,

    Welcome to the forum, eh? :D

    What you can expect from 800 Amp hours @ 48 Volts: about 4800 Watt hours per day @ 25% DOD. So the question is: is that enough? An A/C unit (window shaker) will pull 500 Watts and run half the time: 6 kW hours. Oops.

    Sixteen 310 Watt panels is 4960 Watts. Based on that you'd have enough to recharge that much battery, but realistically for end-to-end efficiency you've got perhaps 10 kW hours AC per day.

    Your inverter size depends on how much you need to run at any one time.

    As always you should start by coming up with a realistic daily Watt hour need and plan from there. Without that magic number there's no point in picking any equipment at all. You can start with existing electric bills and measuring wit a Kill-A-Watt meter any and all things you have now that you expect to run there. It will give you something to go on.

    Of course it is going to be said: why off grid? It is definitely not economical. Grid-tie may be far more practical for you.
  • Karel
    Karel Registered Users Posts: 4
    Re: My complete Off Grid system- Montague Township, Canada,
    Welcome to the forum, eh? :D

    What you can expect from 800 Amp hours @ 48 Volts: about 4800 Watt hours per day @ 25% DOD. So the question is: is that enough? An A/C unit (window shaker) will pull 500 Watts and run half the time: 6 kW hours. Oops.

    Sixteen 310 Watt panels is 4960 Watts. Based on that you'd have enough to recharge that much battery, but realistically for end-to-end efficiency you've got perhaps 10 kW hours AC per day.

    Your inverter size depends on how much you need to run at any one time.

    As always you should start by coming up with a realistic daily Watt hour need and plan from there. Without that magic number there's no point in picking any equipment at all. You can start with existing electric bills and measuring wit a Kill-A-Watt meter any and all things you have now that you expect to run there. It will give you something to go on.

    Of course it is going to be said: why off grid? It is definitely not economical. Grid-tie may be far more practical for you.

    Thank You for Your reply, and welcoming :)
    Maybe I did some miscalculations but 48V*800Ah should give total of 38,4kWh, right? And with DOD 25% that gives me what, 7.7kWh?

    I know that Grid-tie solution would be more practical but the rate that local hydro gave me for that solution is the main reason why we're even debating right now :)

    Anyway, A/C would be used only on hot summer days with lots of sun... and on days like that with my 4.96kW array I will be generating probably around 20kwh/day? Am I right so far? Even when my consumption goes sky high I still have a backup generator controlled by Schneider system which, based on my calculations would have to run max. 15hours/month to protect my batteries and back me up when I need extra juice.

    Schneider/Xantrex is an absolute high-end, no doubt about it. Inverter is capable of giving me true sine true 6kW with 200% surge overload.
    My first conservative load estimate says my daily consumption will be between 7kwh and 9kwh (load correction factor taken into account). The same I should be able to generate with only 2 hours of sun per day (summer) or 3 hours a day (winter).

    I will be going 100% LED and energy star rated appliances.
    I just want to know did I make some wrong steps so far, especially with the sizing of my charger, battery bank and inverter?
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,

    Well if we take 200 Amp hours * nominal 48 Volts we get 9.6 kW hours so I guess I slipped a digit on that first calc. After AC conversion and not including inverter consumption you should have about 8kW hours to work with. So this is looking better.

    A/C could still be murder on this, depending on how much has to be allotted for everything else. Usually refrigeration takes up a substantial amount.

    Don't work from maximum sun hours on long days because you need to know the minimum amount of energy you can count on. There's some curious factors that will come in to play: hot temps will reduce panel Voltage and thus power, and even though the days get long the amount of equivalent direct sun on the panels doesn't increase linearly (especially not with fixed arrays).

    Then there is the fact the sun does not shine every day. That one can be killer, especially in our Northern climes. But that is why you allocated for a generator.

    Problem with battery-based system is that the efficiency isn't very good due to the losses incurred from having to put power in to and take it out of batteries. When you can make use of panel power during the day after the batteries are charged you gain efficiency, but it is not always convenient to turn things on midday.

    As such my original end-to-end estimate of 10 kW hours per day should be thought of as a minimum for sunny days. This will handle your <9kW hours daily load estimate. My main point is to be very sure of the accuracy of that number because it will make or break the system. If you have only 2 Hours of good sun on the panels that 'harvest' number will be cut in half. How much do you want to have to run the generator? The auto gen start may be a good feature here, and btw you can also use the XW with gen support to handle really large loads if needed (parallel output of inverter & generator).

    Xantrex equipment is not quite the wonderful stuff some may have you believe. If you search the forum for threads about the XW6048 you will see it has numerous bugs and a few operating quirks. Fortunately you will also find a lot of the solutions and workarounds too.

    (Okay: corrected over 30 typos in this post; it must be getting late.)
  • SolInvictus
    SolInvictus Solar Expert Posts: 138
    Re: My complete Off Grid system- Montague Township, Canada,

    With the Conext XW-MPPT80-600 charge controller's maximum of 600 V, you could put 8 PV panels in series to make two series strings. That way you buy only 2 parallel MC-4 connectors, do not need series fuses and can probably use smaller gauge wire. Because its MPPT range is 195 VDC to 510 VDC, 4 PV panels in series only gets the Voc to 181.7 VDC while the Vmp will be even lower which are both outside of the specifications.

    With sixteen 310 W PV panels, you will lose some power because the charge controller's output is limited to 80 A.

    16 * 310 W * .85 (efficiency) / 48 V = 87.8 A

    16 * 310 W * .85 / 57.6 V = 73.2 A

    The 80 A output occurs when the battery voltage is 52.7 V. It is a judgement call.

    It is best to have only one or two series strings in a battery array because it minimizes wiring and makes it easier to balance the voltages. You could use fewer batteries with a higher capacity to achieve that.

    Is there some reason why you want to mount them on a 2x8 (wooden?) pole instead of on the roof of the garage/workshop? Unless you need the pole for dual axes tracking, cantilever mounts always require strong components. I doubt that a 2x8 would be strong enough in cantilever (7.25"2 * 1.5" * fb / (8' * 72) = .137 fb at best). It would probably wobble and snap in the 2" nominal direction with only moderate wind. A ground mount rack would probably be less expensive and more durable. Concrete, concrete blocks, anchor bolts and EMT would be reasonably cheap.

    We can not check the size of your battery array nor PV panels without having a detailed accounting of your electrical loads and how long you plan to run them. Does the propane furnace have an electric fan to circulate the air through a central ducting system? Since you have not built your house yet, you could design it to use less electricity. There are wall mounted, ductless propane heaters available. Two or four of them located strategically around the house could be adequate. Perhaps a wood stove?
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,

    To grid or not grid, that is the question. Some factors to consider:

    - connect charge for new grid connects
    - FITs, net meter etc, and their likely future
    - smart meters, future time of day metering.
    - grid reliability
    - current unit rate of grid power
    - labour rate for your PV install

    Our quote for a new connection was NZD23,000. That was a new pole, transformer, bore under road, and 100m of cable and trench.
    That put the payback for grid into the decades or generations. However it is a long term investment, likely recouped in house resale value, and failing that over a half dozen future generations, the life of the infrastructure. My cost benefit analysis, i decided to limit to 20 years for a few reasons. Im not getting any younger, and at the current rate of environmental destruction, im not sure we have many generations left ;)

    At 20 years off grid stacked up. Our off grid power costs about 40c/kWh cf 30c for grid. Up front capital cost was less than half the grid connect. For us , given we saved on installation labor, it made sense. Its partly a hobby, that wouldnt happen on grid.

    But looking longer term, its hard to say what the future holds. Grid for most people still makes sense, despite high connect charges. If carbon tax type scenerios kick in then the price of grid power is likely to build that in. Grid prices have increased at 4.99%/pa averge over the last 5 years. Thats less then many people, including me were projecting, and likely less than the real inflation rate, although higher thant he official inflation CPI rate of around 2%.

    So in our case at the end of 20 years im left with scrap value and a need for a new plan. Or another exiciting oppurtunity to design and build another new sytem?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • Karel
    Karel Registered Users Posts: 4
    Re: My complete Off Grid system- Montague Township, Canada,
    With the Conext XW-MPPT80-600 charge controller's maximum of 600 V, you could put 8 PV panels in series to make two series strings. That way you buy only 2 parallel MC-4 connectors, do not need series fuses and can probably use smaller gauge wire. Because its MPPT range is 195 VDC to 510 VDC, 4 PV panels in series only gets the Voc to 181.7 VDC while the Vmp will be even lower which are both outside of the specifications.

    With sixteen 310 W PV panels, you will lose some power because the charge controller's output is limited to 80 A.

    16 * 310 W * .85 (efficiency) / 48 V = 87.8 A

    16 * 310 W * .85 / 57.6 V = 73.2 A

    The 80 A output occurs when the battery voltage is 52.7 V. It is a judgement call.

    It is best to have only one or two series strings in a battery array because it minimizes wiring and makes it easier to balance the voltages. You could use fewer batteries with a higher capacity to achieve that.

    Is there some reason why you want to mount them on a 2x8 (wooden?) pole instead of on the roof of the garage/workshop? Unless you need the pole for dual axes tracking, cantilever mounts always require strong components. I doubt that a 2x8 would be strong enough in cantilever (7.25"2 * 1.5" * fb / (8' * 72) = .137 fb at best). It would probably wobble and snap in the 2" nominal direction with only moderate wind. A ground mount rack would probably be less expensive and more durable. Concrete, concrete blocks, anchor bolts and EMT would be reasonably cheap.

    We can not check the size of your battery array nor PV panels without having a detailed accounting of your electrical loads and how long you plan to run them. Does the propane furnace have an electric fan to circulate the air through a central ducting system? Since you have not built your house yet, you could design it to use less electricity. There are wall mounted, ductless propane heaters available. Two or four of them located strategically around the house could be adequate. Perhaps a wood stove?
    Well if we take 200 Amp hours * nominal 48 Volts we get 9.6 kW hours so I guess I slipped a digit on that first calc. After AC conversion and not including inverter consumption you should have about 8kW hours to work with. So this is looking better.

    A/C could still be murder on this, depending on how much has to be allotted for everything else. Usually refrigeration takes up a substantial amount.

    Don't work from maximum sun hours on long days because you need to know the minimum amount of energy you can count on. There's some curious factors that will come in to play: hot temps will reduce panel Voltage and thus power, and even though the days get long the amount of equivalent direct sun on the panels doesn't increase linearly (especially not with fixed arrays).

    Then there is the fact the sun does not shine every day. That one can be killer, especially in our Northern climes. But that is why you allocated for a generator.

    Problem with battery-based system is that the efficiency isn't very good due to the losses incurred from having to put power in to and take it out of batteries. When you can make use of panel power during the day after the batteries are charged you gain efficiency, but it is not always convenient to turn things on midday.

    As such my original end-to-end estimate of 10 kW hours per day should be thought of as a minimum for sunny days. This will handle your <9kW hours daily load estimate. My main point is to be very sure of the accuracy of that number because it will make or break the system. If you have only 2 Hours of good sun on the panels that 'harvest' number will be cut in half. How much do you want to have to run the generator? The auto gen start may be a good feature here, and btw you can also use the XW with gen support to handle really large loads if needed (parallel output of inverter & generator).

    Xantrex equipment is not quite the wonderful stuff some may have you believe. If you search the forum for threads about the XW6048 you will see it has numerous bugs and a few operating quirks. Fortunately you will also find a lot of the solutions and workarounds too.

    (Okay: corrected over 30 typos in this post; it must be getting late.)

    Thank You both for your replies, I'm really glad that you found my post worthy of having a debate on the proposed system config.
    I will start by quoting of the best and most important posts published on this forum:
    I would bet that if you did a random survey and asked people what the most expensive thing is about living off-grid that you would not get the right answer from most folks. It is batteries. Not solar panels, or inverters, or controllers, or generators, or fuel in the generators. Batteries.

    Sandia Labs did a study a few years back on off-grid deep-cycle batteries related to DoD and battery life using Rolls-Surrette batteries. They determined that the amount of usable energy gotten out of a battery over its life is greater if you deep cycle it vs shallow cycle it. I'm sure the study could be googled up for those who are interested. So people who buy extra battery capacity to prevent cycling them below 70-80% SOC are actually spending more money over the long term on batteries than those who deep cycle them to 50% and replace them more often.

    Where generator support in an off-grid inverter comes in, is allowing you to buy the bare minimum in battery capacity to run your normal daily loads and let the generator help with the big stuff instead of just throwing more batteries and solar panels at the problem. Many folks consider it the "pinnacle" of off-grid to not have to run their generator. In reality, that generator fuel is a lot cheaper than the extra battery (and RE generating) capacity it takes to never run the generator.

    In most off-grid homes the normal loads are pretty light, and the heavy draw stuff is very intermittent. Instead of spending the extra money on batteries and multiple inverters to meet the peak demand, generator support is a lot cheaper. So it not only saves you money on batteries, it also saves you money on extra (or bigger) inverter equipment.

    And the last thing it allows you to do is save money on generator fuel. Too many off-grid people have a 10, 12 or 14 (and some even over 20) kW generator on a XW6048 inverter or similar. This is ridiculous. The generator should not be any bigger than required to get either maximum charging amps from the inverter, or C/10 charge rate for your battery bank, whichever is smaller. Running a 8 or 10 kW generator to charge batteries with a 6 kW inverter is like using a semi to haul an armload of wood. Inverters that only have generator pass-thru support are limited to either what the inverter can run, or what the generator can run. And that means if you want to power a 6 kW load with a 4 kW inverter you need a 6 kW generator, and you have to run that 6 kW generator for charging when you only need 3,500 watts to max out the charger in the inverter. With gen support you can save the gen fuel and run your 3,500 watt generator for battery charging, and still power that 6 kW load with ease.

    These are all the reasons why, to my way of thinking, an inverter that does not have generator support in it is not a real off-grid inverter. I know way too many off-grid people who have over-sized their generator because they got Outback FX-series inverters and need to power an occasional heavy load that only the generator can run.
    --
    Chris
    http://forum.solar-electric.com/showthread.php?18459-Demonstration-of-Generator-Support&p=141978#post141978

    SolInvictus, you could be right about using 2x8 instead of 4x4 strings. But regardless of max voltage rated on XW-MPPT 80-600 I still think that inverters don't really like voltages too high (or currents for that matter). On a really cold day instead of 182Voc I'll probably have 200VDC and below 25A which my inverter will chew with pleasure and spit out 50 and something volts needed for my battery bank. I am aware that my current calculation is on margin but as Cariboocoot stated moments of full output power from an array are rare so I think I could live with a few hundred W hours wasted.
    EDIT: Oooops, you say its MPPT range is 195 VDC to 510 VDC? 2x8 then, you're right.

    And about those poles... I was actually thinking about simple racks. My English language skills could still use some improvement :)
    I can't use my roof because it is oriented W-E. I will place one of my racks directly towards south and the other will be just slightly tilted (7deg) S-SW. That way I will lose some of the max power available at noon and early afternoon but gain some later when it's really needed.

    Now I want to say something about that quoted post from Chris. I have every intention of programing my charge controller to about 30% DOD and thus (according to Sandia Labs research fact/assumption with which I agree 100%) have more available kWhrs per day and prolong the life of my battery bank which is made of 16 AGM batteries 12V/200Ah each.

    I will take your advice on wall mounted propane heaters along with my already taken-into-account air tight wood stove.

    It is a complete misconception that Ontario being colder and further north has substantially less sunny hours than most of the USA. A quick look at my solar map gives me more than 2000 sunshine hours/year and total production of 6000 kWh (2000hours*3kW average output) which is more than 16kwh per day. I know I won't be able to utilize 100% but still... With my 30% DOD (11.5kWh/day) and just about right sized generator backup I should be able to have and enjoy all the benefits of going off grid.

    Thank you for pointing to some bugs noticed in XW6048. From what I hear most of those should be resolved with the latest firmware and minor modifications, all based on customers feedback to Schneider.
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,
    Karel wrote: »
    I'm really glad that you found my post worthy of having a debate on the proposed system config.
    Welcome to the forum,
    Obviously you're new, or you would realize that we'll debate just about anything :-)
    Karel wrote: »
    I will start by quoting of the best and most important posts published on this forum:

    Generator support is good. It is what we call "AC coupling". It is very tricky to get it working right. It can, potentially, be more efficient than DC coupling.

    In favor of DC coupling is its robustness, and the ability to mix and match components of various types and manufacturers (multiple wind, hydro, solar, and generator charging sources). Some folks use a DC generator in their systems to increase overall efficiency of their systems.
    Karel wrote: »
    my battery bank which is made of 16 AGM batteries 12V/200Ah each.

    Big mistake. Four parallel strings is too many, especially with AGMs. The lower internal resistance of AGM batteries makes them more susceptible to becoming unbalanced by slight differences in the resistance of all the battery cables and fuses in each string of batteries. Short discussion here:
    http://forum.solar-electric.com/showthread.php?14674

    By the way, you quoted Chris Olson. He has admitted that his six parallel batteries are a mistake and he intends to replace them with a forklift battery. He periodically dismantles his battery bank and load tests individual batteries, and will shuffle their location in the bank as needed to try and keep them balanced. If he keeps up with the routine, he will catch batteries as they age and fail and remove them from the bank. The bank will get smaller and smaller... 6 strings to 5 strings to 4 strings... This is probably a better outcome than the usual, where the whole bank fails quickly.

    The reason for rapid failure of parallel battery banks is that good strings may mask the bad strings. You get to a point where, after being charged, some battery strings are discharging into other parallel strings. By the time you realize the generator is turning on every hour or so, there isn't much battery to salvage.

    I'm not saying that parallel batteries can't work, or don't work. I am saying that they are not optimal.

    one more thing... Chris Olson, in trying to rationalize parallel batteries, referenced and quoted a research paper that showed hat you can mix and match batteries of different types and capacities in a parallel battery bank. It's true.... BUT this paper was about large standby battery banks that spend most of their lives in float and seldom get cycled. So to clarify what I am saying: Parallel batteries are not the optimal solution for batteries that are frequently cycled.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Karel
    Karel Registered Users Posts: 4
    Re: My complete Off Grid system- Montague Township, Canada,

    OK... to put it simple, neither this Attachment not found. or the fact that I'm using AGM batteries will help me overcome some disadvantages of 4 parallel strings? AGM technology is all about 'maintenance free' and nothing else?
    My plan B was to get 16x Surette 6V Ah S530 and have the bank (2x8) of the same voltage and capacity as the one we're discussing about. SmartGauge, MPPT and AGM made 4 string option look a bit better... Until now.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,
    Karel wrote: »
    OK... to put it simple, neither this Attachment not found. or the fact that I'm using AGM batteries will help me overcome some disadvantages of 4 parallel strings? AGM technology is all about 'maintenance free' and nothing else?
    My plan B was to get 16x Surette 6V Ah S530 and have the bank (2x8 ) of the same voltage and capacity as the one we're discussing about. SmartGauge, MPPT and AGM made 4 string option look a bit better... Until now.

    AGM's advantages over flooded cells: lower weight, higher current capacity, no maintenance.
    Disadvantages: greater cost, easier to destroy, no way to check SG with hydrometer.

    When it comes to having multiple parallel battery strings the best thing to do is use a bus bar to attach all equal length wiring from each string to. This is to be sure the resistance of each string stays fairly close, keeping the current flow in and out even. Each string should have its own fuse/breaker on it too.

    Be advised that the potential problem with uneven current in multiple battery strings shows up less in 48 Volt systems than in 12 Volt systems. And it is not a matter of "you WILL have trouble" either; it just increases the likelihood. Heck, you can have trouble with one battery string if a connection gets loose or corroded.

    FYI that 600 Volt MPPT charge controller costs as much as two 'standard' MPPT controllers and there is no improvement in the conversion efficiency factor.

    A couple of basic design info threads:
    Batteries: http://forum.solar-electric.com/showthread.php?15989-Battery-System-Voltages-and-equivalent-power
    PV arrays: http://forum.solar-electric.com/showthread.php?16241-Different-Panel-Configurations-on-an-MPPT-Controller
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,
    Karel wrote: »
    My plan B was to get 16x Surette 6V Ah S530

    Why not just get 2 volt cells. Many fewer points of failure (cables, lugs, crimps, fuses, buss bars, etc). Easier maintenance (fewer pools of sulfuric acid to deal with). You have better electrical access to your cells... you can read the voltage across each cell. If you have a bad cell you can remove it and set your inverters and chargers to work at 2 volts lower.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,

    FYI 2 Volt cells tend to cost more per Watt hour capacity than most other battery types, including forklift units (which have their own problems).

    With some careful consideration you might be able to run with 395 Amp hour 6 Volt L16's in two parallel strings of eight.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: My complete Off Grid system- Montague Township, Canada,

    Yeah i was wondering about the 600V controller too. Midnite recently come out and said that they will never build a 600V controller. I havent looked at the efficiency curves at all, so i dont know how efficient they are. On the usual 150v controllers, the higher the pv voltage the less efficient they are, and this is reasonably serious considering the amount of heat you are potentially asking the controller to dissipate. 90% efficiency means 500W heat on a 5000W array. According to the FM80 datasheets when the PV voltage is as close as possible to the PV voltage the efficiency can be as high as 99%.

    The XW has two AC inputs i think, grid and genset. You could use a grid tie inverter on the grid in, and run your 600V from the array to the grid tie inverter(s). You dont say why you want a high PV voltage, so we can only assume its distance. That distance would need to be pretty great before you'd contemplate a HV controller IMO.

    AC coupling does sound quite useful for bigger systems. It allows you to spread your arrays out, and everything taps into an ac mini grid. But if you go that route SMA gear would be worth a closer look, as its designed to run AC coupled off grid systems. Money doesnt seem like an issue for you, and SMA gear is reputable. The europeans know a thing or two about solar. For an documented system using that approach see 'Gulf Island Off grid system':
    1. http://forum.solar-electric.com/showthread.php?10780-Gulf-Island-Off-Grid-System
    2. http://forum.solar-electric.com/showthread.php?13277-Gulf-Island-Off-Grid-System-%97-Part-2
    3. http://forum.solar-electric.com/showthread.php?18715-Gulf-Island-Off-Grid-System-Part-3&highlight=gulf+island
    4. http://forum.solar-electric.com/showthread.php?19610-Gulf-Island-Off-Grid-System-Part-4
    5. http://forum.solar-electric.com/showthread.php?20959-Gulf-Island-Off-Grid-System-Part-5-(Surge-Suppression)
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar