Solar-powered office shed
I have a client in southern Wisconsin that wants to build a 10×10 “Office Shed”. The plan is to build an above code insulated building with Zip system sheathing to achieve a high level of airtightness. The building will have a 15×15 slant roof with a 4-12 pitch oriented to the south.
Now to the actual question;
What would Solar potential be for PV panels on the roof? The client is thinking it would be nice to make the building “Off the Grid”. He is a computer programmer so there wouldn’t be a very large load on the system expect for heating and cooling.
My first thought would be to cover the roof in PV and use those charge a bank of deep cycle batteries.
This is my first “off the grid” build so any advice would be greatly appreciated.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I very strongly recommend a grid tied system and not a full off grid setup. Batteries are expansive, have a finite lifespan, and require maintenance (depending on the type). With a grid tie system, you have no need of batteries, no issues in extended cloudy periods, and no periodic maintenance items. It’s a much simpler way to go. You don’t need net metering to make this work, either, if you’re ok with the solar system only powering the shed during daylight hours and using grid power at night (not worried about net zero).
You certainly can go off-grid if you want to, in which case you need to house the batteries somewhere and I’d recommend they not be in the main living space unless they’re sealed cells. The downside to sealed cells is that they have a shorter life and are more expensive to purchase.
I see no reason you can’t install rooftop solar here, as long as you meet the obvious requirements for doing that (roof facing the right direction, not in the shade, etc).
Bill
It would nice to ride a unicorn.
You have picked a big and a hard goal that is not to say it can’t be done. It will require hard choices. The word nice makes me think it is unlikely they are will to give up much for this goal.
1 Are they willing to cut down any tree that would shade the roof between 10AM and 4 PM?
2 Are they willing to turn the building so its roof faces south?
3 Are they willing to change the shape of the building for this goal? The south wall is about 4 feet tall and 15 tall north wall and flat roof between them.
4 Are they willing to limit windows wall 5% of the walls? A good window is about R4 and this goal will not allow much window.
5 Are they willing to minimize power usage? No laser printer! No big screen monitors. No server rack, no tower computers, limit yourself to a laptop.
6 Do they understand the batteries will need to be replaces often? Much like the battery in their car 3-7 is normal.
Are they will to spend thousands more dollars to walk farther to a strange little building in the middle of a field to work in dark room on a slow computer with a small screen?
OK that is over the top my point is you will need to make compromises to get to this goal and it will cost more.
Life is much easier when you are on the grid.
Walta
FiveQuarterCon,
I built a small 540W off-grid, battery-based system much like what you are describing back in 2007. First, I should say that this is not really cost-effective if you already have access to grid power. I knew that at the time, but I did it anyway, mostly as an educational 'science project', to get some hands-on experience for when I 'moved to the country' (hasn't happened yet). I calculated payback for my system (vs. utility at $0.13/KW-H) at about 85 years, not counting battery replacement or any other costs. Realistically, payback is never. Off-grid PV only makes sense when you have no other option short of living without electricity.
I was never actually off the grid; I just had a stand-alone off-grid power plant that replaced one breaker in the panel. The system powered the electronics in a home office, basically 2 desktop PC's and a few low-power peripherals. The PV system output (120VAC) was routed to one side of a single-pol manual transfer switch. I could power the load either off PV or grid; the PV system itself was always isolated from the grid. I did this so I wouldn't have to just close up shop whenever the batteries ran down.
Capacity is largely a function of the size of the PV array and the battery bank. I used flooded lead-acid, which was and still is the lowest cost and most common battery type for off-grid. Lead acid batteries take a good bit of TLC and maintenance. You get longest service life by not discharging them too deeply. I tried to keep mine above 65% state of charge. During cloudy periods, when the batteries ran down, I just switched to utility. The sun always reappears before long in my area (Colorado). If the batteries discharged to 65%, I let them get back to full charge before putting PV back on line. This could take up to half a day. Load management is more automated with modern grid-tied + battery systems. Mine was manual everything. I had 540 W of PV (at 48V) and a 24V / 370 A-H battery bank. PV was still pretty expensive in 2007. If I built this today, I would double the PV to battery ratio so I wouldn't have to watch the batteries so closely.
You need to check the water level fairly often and keep them topped up with distilled water. Flooded lead acid also need to be equalized periodically, a type of deliberate over-charge which mixes up the electyrolyte. Many off-gridders use a generator for this, but I just switched load over to utility on a full sunny day, and once the batteries were fully charged (float condition) I equalized them with the PV panels. Equalizing rate was probably a bit low; I could have added a grid-powered battery charger, but I never did. It's good to check specific gravity for each cell from time to time. This requires a good quality hydrometer. You need to periodically clean the battery connections and at least wipe the battery tops with a damp cloth. Wear old clothes.
Another thing - lead acid (other types too) have a preferred temperature range. For lead acid it's similar to humans, not too far from 25C (77F). My batteries were located indoors--barely--in a mostly unfinished out building. It seldom goes below freezing for long, but the batteries were basically too cold for about 8 months of the year. A too-cold environment caused the electrolyte to stratify, which reduces capacity and shortens life. You don't ever want to let the batteries freeze (can destroy them). Best preventative is to keep them charged, which lowers the freezing temperature of the electrolyte.
My batteries were in a sealed box, vented up through the gable wall. This is a safety issue, particularly in a tight building, because the batteries expel hydrogen gas. No ignition sources allowed. Lead acid batteries, especially flooded, take up a lot of space, so you want to plan for that. You don't want them in a living area.
My batteries were mid-spec L16 class. If I actually lived off-grid, I would want industrial-class batteries (one source is Rolls-Surette). They have a greater number of charge/discharge cycles for a longer service life. My batteries lasted 11 years, reasonable for L16's. It was a good run, but I haven't replaced them. I'm satisfied that I know how to do it, I don't need to keep doing it. If I installed solar today, I would put in a grid-tied system. If you live in an area with different time-of-use rates, a grid-tied+battery system (charge up during the day and run off battery in the evening) might make sense. They do cost more. A programmable controller manages power sources and loads based on what works best for you.
The primo authority for all things off-grid, on this site anyway, is Martin Holladay. He has lived off-grid for a long time, and knows all the stories. Check out his articles. Most of my PV 'schooling' came from Home Power magazine. Sadly, they stopped publication a couple years ago. Last time I checked, the articles were still viewable on their web site, however.
I hope I haven't discouraged you too much, but your proposal probably isn't very practical. Good luck.
Don
Lets do a bit of math. Assuming 2x6 construction with batts, vented roof with R28 batts, R10 sub slab and say 10% R5 glazing.
In my are with a 2F design temperature, a 15x15 building with 8' average wall you'll have 480sqft of walls, 250 sqft of roof and 50sqft of glass. You loose 1700btu through the walls, 800btu ceiling, 500 floor and 700 windows. Total around 3700BTU, add in some air leaks and ventillation, you are probably looking at 4500BTU.
With a decent cold climate heat pump, at design temp you are probably a bit above COP of 2, say 2.5 to be optimistic.
To heat the office for 8h a day, your heat pump would consume around 4.5kWh. Your lighting and misc loads would probably add at least 1kWh to that. With our low winter sun, you expect to see 2.5 sunhours per day, so a 2.5kW array would probably carry it. That easaly fits on a 15x15 roof and not too expensive, probably even worth while to increase the array a bit.
To get a couple of days of autonomy, you need at least a 15kWh pack. You can probably get away with a single power wall but two would be saver. Lithium ion has come a long way, maintenance and life should not be an issue in this type of application.
Overall, you are probably looking at at least a $35k price adder to the project. You can save a bit of solar/battery cost by bumping up your insulation, this is definitely the case where doubling the R value of the assembly and better windows are worth it.
Definitely doable and not out there. Solar and battery costs are pretty low, provided you get decent winter sun, it should work.
It would definitely be an interesting project.
At 10' x 10' he could probably heat it with his body heat and a couple candles.
A couple of Goal Zero 3kw kits with their 4 breaker transfer switch of 120v might get it done with LED lights, computer and printer. Bring your own wood stove.
Maybe doable with a couple of large Czech men doing the beer polka for body heat.
Would also be pretty hard to concentrate though.
And if they wanted to stay in there full time. Left unoccupied over night I'd bet it would take a crew of dancers to get it warm.
I noticed my 27" iMac was super hot to the touch today (by design). It'll put out 485-682 BTU/hr at 142W-200W based on CPU load. That'll cover a good chunk of the season's load and it doubles as a computer. Win-win.
I realize this thread has been dormant for a while, but Akos’ proposed solution of using PV + Powerwalls sounded promising, so I decided to research it some more.
Tesla’s web site is vague about using the Powerwall 2 off-grid. Under Supported Applications, they describe off grid capabilities as ‘coming soon’. The owner’s manual likewise says nothing about off-grid. The current model Powerwall 2 is an AC coupled device with a built-in inverter/charger. The only user access is through the AC port. You need a second inverter, connected to the PV array to charge the Powerwall from solar, via the power bus in the AC panel. Tesla refers to this second inverter as the ‘solar inverter’. This is the same type of unit used in basic grid-tied systems. It can be a string inverter or a set of micro-inverters.
Several people who have PV + Powerwall 2 have run off-grid simulations by shutting off their main utility disconnect. These tests have run for days, maybe weeks, but not years or ‘indefinitely’. The simulations were halted when Powerwall discharge appeared imminent, which occurred for a variety of reasons. These tests reveal the difficulty with using the Powerwall 2 off grid: By regulation, grid-tied inverters need an active grid, or some simulation of it, to operate. In this case, the Powerwall functions as the grid. If the Powerwall fully discharges, which it can safely do, the solar inverter loses its reference and shuts off as well, right when you need it most - to recharge the battery. You’re locked out, even if you have sun on the array. There is probably some initialization procedure to cold-start the system, but I doubt it’s automatic. I’m curious what Tesla has in mind for off-grid support.
The above situation is avoided by connecting to the utility grid (or standby generator). In that case, you could only discharge the Powerwall completely by exceeding PV capacity during an extended utility outage. I believe the Powerwall app warns you when the battery is getting low and can command the Powerwall to disconnect from the AC bus when some programmed discharge level is reached, saving some capacity to ‘wake up’ the solar inverter when the sun returns.
AC coupling works well if you already have a basic grid tied system and want to add battery storage, or if you’re installing grid-tied with batteries from the get-go. You are better positioned to deal with the curtailment of net metering and the imposition of time-of-use rates than you are without the battery, and if you already have a battery-less system, you don’t have to replace anything. The Powerwall 2, with its integrated inverter/charger, fits perfectly here, and I believe this is where Tesla sees their primary market.
If you’re building off-grid and expect to stay that way, I think DC coupling, with charge controllers and an off-grid inverter/charger, still makes the most sense. Power conversions are minimized, and you can charge the batteries from solar any time it’s available. If you don’t want the involvement of lead-acid batteries, there are a variety of 48V and 24V lithium ‘drop-in’ replacements (higher voltages too if you have a compatible inverter/charger), which are smaller, lighter, can cycle deeper, and which require almost no maintenance. In return, they are more expensive.
One good thing about lithium batteries is that prices are dropping steadily. A couple years ago, I looked at the SimpliPhi 24V/3.5 KW-H battery to replace my ailing lead-acid battery bank. It cost $4117 at the time, but now, their slightly larger 3.8 KW-H model sells for $2690 (same vendor). A 3.8 KW-H lithium, which can cycle down to 80% depth of discharge (DOD), is equivalent to an 8.88 KW-H (370 A-H @ 24V) lead acid bank that cycles down to 35% DOD. Cost per usable KW-H is still higher for the lithium ($885/KW-H @80%DOD vs. $414/KW-H @35%DOD), but the payoff comes with a much longer cycle life: 10,000 cycles for lithium (27.4 years) vs. 2600 cycles for L16 class lead acid (7.1 years). If you can hang in there for 15.2 years, the lithium battery becomes a better value. I passed on the Lithium replacement 2 years ago because I had more pressing needs for the $4K, but if you’re in for the long haul, Lithium has some advantages.
If the OP truly wishes to operate a pure off-grid power plant, there are a few other things to keep in mind: During an extended cloudy period, you can easily blow through your designed days of battery autonomy. Average sun-hours for any location are just that, an average. Also, with panels installed on a 4:12 roof (~18 degrees tilt), snow will pile up, during which time you won’t generate anything. Keep your snow rake handy. Finally, you have to decide – How critical is system availability? Can you work somewhere else or take the day off (woohoo) when the batteries are low and you have no sun? If availability is critical, I still think grid-tied with batteries, whether DC or AC coupled, makes more sense.
Most serious off-grid inverter/chargers include an AC input to connect a standby generator, although you could connect the grid here if your inverter/charger is a multi-mode or ‘hybrid’ model that’s listed for grid connection. These units have algorithms to minimize grid use, if that’s your intent. With correct sizing, you will still be essentially off-grid almost all the time and still keep going during extended cloudy weather. Many solar distributors have sizing worksheets on their web sites. These worksheets are easy to use and work well.
One last thing – you need to decide for yourself how you feel about the environmental aspects of lithium extraction and processing. I wouldn’t say lead is especially ‘green’ either, although almost all of a lead-acid battery is recyclable into new batteries. Don’t know about lithium.
Well what is the plan for heating and cooling? A small propane system for heat will use no electric. The smallest AC might suffice. As far as the electronics, first do an analysis of what kind of power draw you will be needing. Add up all the appliances wattage that you will be using and estimate how long they will be used for. Once you have the total daily power usage then you can begin to size the system.
Go with AKOS above. Good air sealing and super insulated like my home. Easily heated with heat pump at low watts. Go grid tie as it's very inexpensive and payback is fast. Doesn't matter what direction the building lies if you go 4:12 roof pitch as you can face PV East and west. In the near future you'll get paid more for your electricity as that's when other solar facing south can't produce. Good incentives yet with FOCUS in WI. PK- solar installer, net zero builder in Appleton, WI.
Good LEDs are way better than more windows. Also, Solar thermal air will bring you way more warmth and no heat loss as with windows. We have lots of sun on the coldest days here in Sconnee!
Right paul, was gonna mention LED's and yes a single SHW panel with glycol loop attached to a baseboard or solar hot air as you suggest may work out also.