Four Years Later, Off-Grid Office Shed Still Rocks

About four years ago, [Russell Graves] created what was, to him, the ultimate work-from-home environment: an off-grid office shed. The shed might look a bit small, but it’s a considerably larger workspace than most people in an office are granted. Four years later, in the middle of a global pandemic, working from home has become much more common and [Russel] shares some thoughts on working from home and specifically reflects on how his off-grid, solar powered shed office (or “shoffice” as he likes to call it) has worked out. In short, after four years, it rocks hard and is everything he wanted and more.

Its well-insulated plywood walls let him mount monitor arms and just about anything else anywhere he wants, and the solar power system allows him to work all day (and into the night if he wants, which he doesn’t) except for a few spells in the winter where sunlight is just too scarce and a generator picks up the slack. Most importantly, it provides a solid work-life separation — something [Russell] is convinced is critical to basic wellness as a human being.

That’s not to say an off-grid solar shed is the perfect solution for everyone. Not everyone can work from home, but for those who can and who identify with at least some of the motivations [Russell] expressed when we covered how he originally created his office shed, he encourages giving it some serious thought.

The only thing he doesn’t categorically recommend is the off-grid, solar powered part. To be clear, [Russell] is perfectly happy with his setup and even delights in being off-grid, but admits that unless one has a particular interest in solar power, it makes more sense to simply plug a shed office into the grid like any other structure. Solar power might seem like a magic bullet, but four years of experience has taught him that it really does require a lot of work and maintenance. Determined to go solar? Maybe give the solar intensity sensor a look, and find out just how well your location is suited to solar before taking the plunge.

25 thoughts on “Four Years Later, Off-Grid Office Shed Still Rocks

  1. I think there’s a VERY big, almost understated, message right here in plain sight, which needs some serious consideration, and even more serious thought–

    The only thing he doesn’t categorically recommend is the off-grid, solar powered part. To be clear, [Russell]…admits that unless one has a particular interest [my highlighting] in solar power, it makes more sense to simply plug a shed office into the grid like any other structure. Solar power might seem like a magic bullet, but four years of experience has taught him that it really does require a lot of work and maintenance…”.

    This from a person who has four years’ worth of experience with what will seem to a lot of people–for a lot of different reasons, but, for the most part, reasons with no foundation–a hacker’s delight when it comes to “the perfect solution” to an ill-perceived and not-well-thought-out problem.

    Keeping batteries–which have a limited lifetime AND are very expensive–maintained and healthy IS very hard work. This from someone who does keep batteries maintained and healthy.

    There ain’t no free lunch.

    1. I’d argue “off-grid” can be more than “a particular interest.”

      This flip side comes up when local grid reliability is going down the tubes, as appears (anecdotally) to be the case in many locations on the east and west coasts. Where my folks live is not out in the boondocks, but in the last decade they’ve had multiple outages (5-ish) that lasted for a week or more. During those times any alternate workspace was packed to the gills and the local places that had power (coffee shops, libraries, etc.) were strictly enforcing occupation time limits.

      Imagine trying to explain to your employer: “I’m sorry I haven’t been working for the last week. Power is still out and there’s nothing I can do to fix it.” Their response is likely to be: “We’re sorry we’re not paying you any more. You haven’t been delivering enough work and there’s nothing we can do about it.”

      Having a generator is fine up to a point, but that needs maintenance, too. What you’re really facing then is that your grid is insufficiently reliable, and you have to balance the externalities of system maintenance against those of lost opportunity/profitability. So the maintenance requirements of an off-grid setup may be compensated.

      And if you’re in the boondocks, outages become longer and longer because the providers (rightly or wrongly; you can argue either) prioritize those repairs that bring the most people back on the system fastest. If you’re in a less populated area, you’re going to be living “the life primitive” for the duration.

      Of course this is also a more difficult problem; having limited power in your Shoffice but a house that is enjoying a back-to-nature holiday isn’t viable — though it is heaps better than nothing. I envy those whose grid supplies are completely reliable and have no more than one hour of outage per four years.

      1. ” I envy those whose grid supplies are completely reliable and have no more than one hour of outage per four years.”
        I think this is something the U.S. should tackle once everything has kinda settled down. I know the U.S. is vast and it would be a major task to improve the reliability of the grid significantly, especially for remote areas, but it is IMO crucial for economic stability since consumption (Netflix, Amazon, …) and production (remote offices, shops,…) just necessitate a stable grid.

        I’m indeed lucky as I experienced my last power outage when I was little. It was rather exciting and I remember my parents lighting candles and my siblings and I having a fun time. I was honestly kinda disappointed that it only lasted for half an hour or so…

        1. I agree a stable grid is becoming a requirement for a modern civilisation. Though if that should be so in the way we currently expect it be is something we all need to consider – should be be expecting the grid to magic up the power for us all at any time or should be be managing the draw we put on it to match the supply. Working more staggered shifts, turning off AirCon and the like for a while during high load (many things like Aircon can be turned down or even off for hours without particularly inconveniencing the user) and local power generation which has lower transmission losses and can vastly reduce strain on the grid overall (not to mention provide some functionality should grid connection be interrupted).

          I’m luckier than you have been having never had a power outage in my memory (unless you count failed fuses and the like which I don’t – as even if it was caused by a spike/dip in the grid the power is still on just your own fusebox to sort out. And now I have solar with a relatively small backup battery anyway so as long as that system isn’t knocked out should at least have lights – wondering about building a CAES system as that avoids all the pitfalls of batteries with simple maintenance requirements – though I don’t really need it. But in terms of cost and reliability nearest thing to a free lunch of energy storage)

      2. You are absolutely correct, of course, and the linchpin is the “…particular interest…” part.
        I, for example, live in a fairly-populous area, served by one of the better electric utilities in the US. I am completely, thoroughly envious of anyone whose grid-reliability is anywhere near one hour of outage every four years. Even with heavy complaints to the state’s public service commission, outages occur almost every thunderstorm (but getting better), lasting anywhere from two to twelve hours, and the last (major) outage lasted a week (just like your folks’). But enough of that…

        It seems as though the electric-utility industry is falling prey to that same insidious problem which plagues a lot of our previously taken-for-granted benefits of living in a ‘modern society’: increasing unreliability due to lack of investment in basic infrastructure maintenance. Not infrastructure improvement, mind you, but the completely obvious need for the simple maintenance of the EXISTING infrastructure.

        There is an unfortunate reason for all the prime-time TV commercials for “whole-house” emergency back-up generating systems, nowadays. And then–which I haven’t found stated in any of those commercials–depending on the type of fuel chosen for your generator, you’d better be prepared to invest in a 120- to 250-gallon propane tank, or a large diesel-fuel tank, large enough to keep your generator running for a week. You did NOT think you’re going to get by with a few cans of gasoline or diesel fuel, did you?
        Just thought you’d like to know…


        @Joel, further down, mentioned Edison-cell batteries. This is an example of an extremely long-lived battery, but which is, even compared to lead-acid, very expensive and very heavy (NICKEL-IRON, after all). They are so reliable that some have been in service in railway signalling applications, where limited grid power is available–or was–for over half a century. If one can afford all aspects of its ownership, this battery is an excellent choice for a battery-backed-up system.

        Maintenance of any secondary-cell battery IS very hard work, but this type makes it much easier.


        There ain’t no free lunch.

    2. If you have the project capital, than a nickel–iron battery can last a long time. This company offers a 10 year warranty, and automated electrolyte management systems.

      Some Edison cells in early model 1930’s EV were documented as lasting over 50 years though multiple collectors’ care. Ultimately, the cells needed replaced as the seals finally began to rot out from age… rather than a capacity loss from regular use. If you have the physical space, than these may be worth reading about.

      I suspect one would need an air-conditioner in such a small space out in the sun.

      1. (Russel here, owner of the shed)I looked pretty hard at nickel iron before I went with lead acid, and I just couldn’t convince myself they were a great deal. They’re “lifetime” batteries, but have a ton of downsides to them in the process.

        Round trip charging efficiency is dreadful – on the order of 60%. So if you put 1kWh in, you end up with 0.6kWh out. Lead acid is far better, and lithium is better still. The gap between charge and discharge voltage on NiFe is quite substantial, and most of the surplus goes into gassing, so they’re both very loud while charging and require an insane amount of water compared to lead acid. I think you can put recombiners on them, but they like to fry recombiners from what I understand from the sheer amount of recombining they have to do (recombiners get hot in operation).

        There’s also problems with the electrolyte absorbing carbon dioxide and needing to be refreshed every 10-15 years, and you end up with an awful lot of hazmat on your hands as a result.

        They have their uses, but I couldn’t convince myself that my shed was such a use.

        1. @HACKADAY–

          I think that

          should be given a chance to respond before this discussion gets taken down.


          The simple–compound–statement,
          “…There’s also problems with the electrolyte absorbing carbon dioxide and needing to be refreshed every 10-15 years, and you end up with an awful lot of hazmat on your hands as a result.”…


          “…but have a ton of downsides…”

          are reasons enough to call into question the complete even-handedness of the comment.

          Potassium Hydroxide is no more a ‘hazmat’ threat than is the sulfuric acid used in lead-acid batteries. And it must be changed every 10-15 years? IF true, this is much, MUCH longer than any lead-acid battery’s lifetime.

          Your comments are deeply appreciated [Russell], and highly informative, but I don’t think you’ve taken all aspects of Ni-Fe benefits into account.

          ”Comparison of commercial battery types”

          1. I don’t see why the conversation would be taken down, it’s certainly relevant. I looked, very hard, at nickel iron before going with lead acid. I simply couldn’t convince myself that the benefits of nickel iron outweighed the rather significant costs in overall system design.

            If you don’t believe any lead acid battery will last 10-15 years in daily cycled solar service, that would indicate you’re not familiar with the state of the art of lead acid – look at the Trojan industrial solar line, and you’ll find L16 format cells designed for 15+ year daily cycling use. They’re rated at 3600 cycles at 50% DoD, and in some rather abusive accelerated lifecycle testing, accomplish 15-17 years of simulated use before starting to show capacity loss.

            If I build a 48V pack around the Trojan SIND 06 1225 (400lb batteries – have a forklift or a few friends handy), I need 8 of them, which should run me around $10k. That’s 48V/1225Ah. The comparable Iron Edison pack (48V, 1000Ah) is $42,250. Under the cycling of a well designed off grid system, those Trojan batteries should last 15-20 years (shallow cycling during the longer summer days, deeper cycling during the winter) – at least. For $40k, I can have an exceedingly nice lithium bank (LiFePO4 even!) that will be long lived and well behaved, with radically better charge/discharge efficiency than NiFe.

            I know where to recycle lead acid (including the acid): Anywhere that sells them. The recycling chain for lead acid is wonderfully well established. I have no idea where to dump 50 gallons of potassium hydroxide, but it’s generally not regarded as something you just put down the drain (I certainly wouldn’t wash it into my septic field).

            Modern lead acid cells are also using various carbon additives to the plates that help prevent hard sulfation from building up during sustained partial state of charge cycling, with the current state of the art being the Firefly Energy batteries that behave a lot like lithium in terms of PSOC cycling behavior and recovery.

            I won’t argue that nickel iron looks pretty solid based on the spec sheets, but go find people who actually live with them – there are a few over on some solar forums you can find and read through their reality. NiFe is staggeringly expensive (partly because they’re quite niche compared to lead acid), they’re thirsty, they’re very loud while charging, and the round trip energy efficiency is horrible. If you compare them to lead acid from 20-30 years ago, yeah, they look appealing – but compared to modern lead acid? Or to lithium? They’re just not solving a problem well enough to justify their downsides.

    3. What is this “very hard work” it takes to maintain batteries? Of all of the rechargeable batteries I’ve ever had, most were maintenance free, the exception being lead-acid, requiring occasional topping-off with water. Didn’t seem very hard.

  2. Totally concur about deep thinking and productivity theme. Author struck a nerve on that one. A lot of meetings these days are to solve someone else’s problem using a group hug.

    Eliminate the battery and go grid-tie inverter. That uses the grid as the virtual battery, so no battery maintenance or replacement. Something as small as this shed system, you can put power back into the house in a net-zero way. However, in the low end of this solution, you have no battery to power through a blackout.

    1. If I went grid tie, I wouldn’t bother with solar out on my office. The net metering requirements are a pain in the rear, as I’m learning as I go through the process for the house (see current posts on my blog for the process there).

      As soon as I touch the grid, I’m required to meet the local interpretation of the national electric code (and building code), and that starts to require things that make very little sense in this particular situation. I abide by NEC in terms of wire size and such out here, but it would be an expensive pain in the rear to run grid power out here. I’d either need another meter drop (couple grand), or to trench an awfully long way in terrain that is utterly unsuited to trenching (you can safely model the property as a bunch of rock with a thin layer of dirt on top).

      Being standalone and under 200 sq ft, nobody cares what I do out here, so I can do things like have a separate DC power bus if I want without pulling permits and getting inspections for my battery bank.

  3. i was expecting the warning about solar from the headline, so i was glad to see it made it in.

    as a contrast, a friend of mine has a solar setup which i think is eminently practical. he has a house in BFE with unreliable electricity, and DSL. he has a small number of solar panels (removed from paying customers’ installation due to viible damage), and a regulator to allow them to charge a pair of old 12V lead acids (liberated from dead trucks, i think), and a DC-DC converter to generate 5V. he doesn’t power his whole house, just his DSL modem, his router, his wifi, and i think maybe some sort of NAS device. they all run on 5V or 12V, so no inverter. in a pinch, he can even charge his laptop.

    the total draw is so low that the batteries will last for multiple days without being topped off, and even a few solar panels on a cloudy daya deliver about 10x what is really needed, so there is a huge margin in everything. it’s just enough that he doesn’t have to resort to (spotty) cellular data when the power is out.

      1. Sure; plenty.

        Simply type “12v-20v up converter” in A***n’s search window, and you’ll find 20 pages of the things.
        Here are two:

        Gowoops 2PCS 150W DC-DC 10-32V to 12-35V Step Up Boost Converter Module Adjustable Power Voltage
        ASIN: B00J1X4XXM


        Yeeco DC Boost Converter, DC-DC Step Up Converter Board 5-32V 12V 24V to 5-50V 36V 48V Voltage Regulator Booster Module 15W Adjustable Voltage Step Up Transformer Stabilizer Power Supply Module
        $ 8.59
        ASIN B074J9D278


        You certain you don’t want a 20 v-to-12v DOWN converter?

        They’ve got those, too.


        “I keep my eeepc 900 laptop, which can be powered by 12v,,,”

        Neat machine; I keep mine, too. Matter of fact, i Just upgraded the 4 GB SSD to a 64 GB mSATA SSD. 128 GB was only $5 more, but I said, “How in the HELL will ever need, or use, 128 GB on this machine?
        Probably need it next week.

        1. The part he left out, I think, is that the REASON he keeps his eeepc 900 is that he can run it directly off of 12V. The overwhelming majority of laptops need 18-20V to operate.

  4. I moved to raw land in the boonies in 1979 and have been off grid since then.
    There is no free lunch. However, if you order the right stuff, lunch can be tasty – if not too cheap.
    Some things have changed. Back then, Solarex PV panels (owned by BP!) were ~ $6/watt!
    Now decent panels (Chinese) are to be had around 80 cents/watt. That’s kind of a big deal.
    Room to put them out of the shade is now more the limit. I have enough such that any day that isn’t dead dark gets me by, and a good solar day lets me charge my Chevy Volt, and then run welders, lathe, mill and so on. It’s nice, you just
    have to roll with the weather.
    Batteries…sigh, yes, they su*k. And going for initially cheap costs far more in operating costs. Many people don’t realize that a pack that might go 1000 full cycles or so before it’s too degraded, will often easily go 10,000 10% cycles. Go big! The big stuff has things like catalytic caps that reform water from the evolved H and O2, and require little maintenance especially if a temperature sensing charge controller is used.
    Electronics tend to be very good. Maybe I’ve been lucky (I’m a retired EE and careful, which might help). I started with Trace, and have followed them through a couple changes of name and ownership. I think one unit of something has failed over those decades (and they fixed it).

    I do run a few buildings off my system, but I only live in one of them – the others are storage and my workshop/lab.
    My base load varies, but it tends to be around a couple hundred watts, between fridge, network (mine is big…), some fans to ban mold in crawl spaces, a vacuum system in the lab and a few always on “Lan of things” computers (mostly pis or odroids) that control and log solar stuff, water collection and purification, HVAC(!) and you name it.
    I’ve worn out a few generators (to say the least) – even a Lister diesel, before I got wise. It’s good to need them less, and the standard tradeoffs people use are not the best for this. Back in the day, I computed cost of generator electricity at around $1.64 a kwh. Ouch, that’s not inflation adjusted!

    But! Like with solar panels, everything has gotten vastly better over the time I’ve been doing this. And it helps hugely to “go big or go home”. Things that are loafing last longer and wind up costing less over time – a lot less. Batteries are better, even lead acid ones, there’s more emphasis on making good ones for solar these days than there was when it was more a niche.
    Generators, for crying out loud, now have nice things like oil filters, fuel injection and so on – on anything a bit nicer than the harry homeowner hardware store versions – and they last a metric crap-ton longer while getting better mileage on fuel and oil.

    When I was running a couple small 12v batteries and 600w of panels to run a boom box, brake light, and a soldering iron – it was super expensive and hasseliferous. Now that I have 6+kw of panels…and 2500 amp hours at 24v of serious pro 2v main batteries…with spares on the spares – even a couple nice inverter generators – it’s nice. I still have a job if things go wrong, but they just don’t go wrong often anymore.

    I would mention as an aside that with this approach I’ve saved hugely in property taxes as none of this assess as high value and it seems the county has let me slide as part of the “keep floyd weird” program. With a big spread and 4 large buildings that paid for the setup fairly quickly – and having the spare change that results from “no utilitiy bills” is a really big deal in lifestyle freedom – if you work, you keep the money, if not, you don’t starve either. For me, that was the motivation – Being cool with nature came along for the ride, and I appreciate that better now than I did when I was treating this all as an engineering challenge…

    And I’ve been working from home, when I worked, this whole time. Y’all are beginners at this telecommuting thing.
    It’s nice to be an engineer…I create information that others use to make things…info goes over DSL just fine. And that’s my only bill.

  5. “Great report; thanks for all the benefit of all your experiences.
    You’ve given us all some terrific ideas to think about, and to put into practice. Here’s one of my favorites, and one which tends to not even be thought about by most people–

    “… Things that are loafing last longer and wind up costing less over time…”.
    Spoken like a true engineer, with a ton of experience.

    AbsoLUTELY true, and one of the major reasons for spending the extra money to get a generator which runs at 1800 rpm, (as well as paying the extra bucks for much-larger batteries) rather than almost all of the ‘toys’ which run at 3600 rpm. They’re 3600 rpm because that’s cheap to produce, and that’s the ONLY reason (I own a 3600 rpm unit, but only because it was drop-dead cheap, brand-new (along with pretty good quality) and it makes for great insurance to run my well pump in an emergency)

    Once again–thanks for all the tremendous insight. All the best to you.

  6. Well i been running my house on solar for over 12 years, 7500sq it works great, I have 5kw panels and edison cells for the storage, considering i am 59 they will be lifetime batteries. if you set things up right its smooth sailing the whole way.

    1. What about it?

      It wasn’t available (in terms of “I could get my hands on one”) when I built the office 4 years ago, and still isn’t rated for off grid use, as of last time I looked.

      My battery bank (~10kWh) was around $1500, my inverter/charge controller were around $1000 between them. So $2500-$3000 for the parts of the system that would be replaced by a PowerWall. How much do they cost?

      Practically, I only use 15-20% of my battery capacity on a daily basis, except for a month or so in the winter when it gets hammered on pretty hard. A Powerwall doesn’t solve any problems I have, but costs an awful lot more in the process.

    2. I looked long and hard at various Li type batteries and wound up categorically rejecting them.
      I think the reasons were good, and more of them will stay true for a long time.
      They cost much more. That may change but it matters now.
      They don’t last any longer than state of the art flooded lead acid batteries.

      So, they’re smaller per kwh, and lighter. But this is a _home_. I’m not going to be driving around in it, or flying to the next county. I have room!

      Li can make more peak amps going either way. At the size of a home battery, this makes no difference whatever.
      L-A can do a few hundred amps at C/10 where it’s happy. More than that is probably some kind of dangerous fault.

      L-A doesn’t need quite the management of Li, and the extra complexity of a per cell battery management setup is a liability. You really do have a more stringent requirement than some lab kludge, or a car you’ll replace in a few years here.

      Total cost of ownership – higher –
      Advantages – none that matter. Decision – easy.

      I did try nicads – I got some surplussed from locomotoive use back in the day. Like other alt tech, they have serious issues – even surplus they are also expensive, but that’s not most of it.
      Inverters and charging systems can’t handle the wider spread of voltages these alt technologies have – the difference between the lowest under load when nearly depleted and the highest needed to finish off a charge is bigger. Many inverters and chargers either can’t handle this, or do so poorly and lose efficiency one way and another. That kind of adds to the cost. And that thing about the electrolyte. Yep…it does go bad, and it’s a nasty thing to deal with.

      Yes, as an EE I can build things like that – and have, I did one of the first if not THE first MPPT trackers right here in my lab, and kludged it up using a weller soldering station sans the 24v transformer, bootstrapping by running the handpiece right off my 24v pack. And I’m decent at high power electronics.

      The cost of one failure of kw level things – all those fets and PCB go to magic smoke – means the expense of development is high – (no there is no fuse fast enough that won’t nusicance blow too – and the attempts at that aren’t free either).

      At some point you have to do other things around the homestead and discover that going with the high volume products is the best way by far for total ownership cost, not to mention having a life. High volume is of course relative, a vendor estimated that there were around 20k off grid homesteads when I started out….that’s not a lot of volume!

      I got a pack of Crown brand 2v cells almost 10 years ago. As far as I can tell, they’re good as new, still. The system is monitored by my “lan of things” using arduinos and pies, and I have data in an SQL db going back years minute by minute, amps, amp hours, voltages, temperatures and so on – it’s a big investment so I keep an eye on it.

      Things have really changed from the old Trojan L16 days – even then, with care, I got nearly a decade out of a batch, where most people got 4 years – paying attention and not abusing them does make a difference.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.