The Quick-Build PowerWall

Elon Musk isn’t just the greatest human being — he’s also a great inventor. He’s invented the reusable rocket, the electric car, and so much more. While those are fantastic achievements, Elon’s greatest invention is probably the PowerWall. The idea of a PowerWall is simple and has been around for years: just get a bunch of batteries and build a giant UPS for your house. Elon brought it to the forefront, though, and DIYers around the world are building their own. Thanks, Elon.

Of course, while the idea of building your own PowerWall is simple, the devil is in the details. How are you going to buy all those batteries? How are you going to connect them together? How do you connect it to your fuse box? It’s a systems integration nightmare, made even more difficult by the fact that lithium cells can catch fire if you do something wrong. [jehugarcia] is building his own PowerWall, and he might have hit upon an interesting solution. He’s built a modular system to store and charge hundreds of 18650 cells. It looks great, and this might be the answer to anyone wanting to build their own PowerWall.

Aside from acquiring hundreds of 18650 cells, the biggest problem in building a PowerWall is simply connecting all the cells together. This can be done with 3D printed battery holders, solder, and bus bars, with a few people experimenting with spot welding wires directly onto the cells. This project might be a better solution: it uses standard plastic battery holders easily acquired from your favorite Chinese retailer and a PCB to turn cells into a battery.

The design of this battery module consists of a PCB with sufficiently wide traces, an XT60 power connector, and a few headers for the balance connector of a charger. This is a seven cell setup, and in contrast to the hundreds of hours that go into making a PowerWall the old fashioned way, these modules can be assembled pretty quickly.

Testing of these modules revealed no explosions, and everything worked as intended. There was a problem, though: when drawing a high load, the terminals of these cheap battery connectors got up to 150°. That makes these modules unsuitable for high load applications like an e-bike, but it should be okay if you’re putting hundreds of these modules together to power your house. It might be a good idea to invest in some cooling, though.

106 thoughts on “The Quick-Build PowerWall

      1. I got to work with NeXT machines too. They were pretty good at the university running the even more innovative Wolfram Mathematica. But they treated it like a sacred artifact and there was NO FLOPPY DISK!!. USB was nonexistent and networking might have worked but back in the early 90’s everyone was doing their own protocol. Overall the experience was horrible, you couldn’t keep any of your work long term. You could do cool things on the expensive “special” machine but forget about it working with anything else. PCs running DOS were much friendlier.

        Overall. the NeXT computer parallels Elon Musk’s business model very well. Ever try to charge your electric Nissan LEAF at a Tesla powerstation. It simply cannot be done and Tesla doesn’t want it to be possible even though US funds have been spent to get them built. I can’t afford a Tesla of any flavor, even the model 3. I CAN afford spending less than $10K on a sturdy 3-year-old Nissan LEAF but the Tesla network is no good to me and is effectively suppressing the establishment of an alternative. Tesla is pulling EVs towards eletism but the masses are the ones who really need them. That said, the NeXT computer and the Tesla Model 3 have a lot in common. They were both very cool, but elete products. They both lacked major components that normal users would demand, like an instrument cluster for the Model 3, and a floppy disk drive for the NeXT. They both had “special”, even visionary feature s that no-one needed. They were just strong on cool-factor. Eventually the market will dictate what gets built but until then, expect Tesla to throw its weight around trying to pull in as much money as it can until their competitive advantage wanes. (Actually I think Musk is more adaptable than Jobs and is more likely to set ego aside when he is going down a bad path and follow a different trajectory, but we’ll see. Either way, I could be a friend to him, but never a minion. I wouldn’t want to get used up like Edison used up Nicola Tesla. )

        Deliberate incompatability is the trade mark of this kind of “inventor” (actually they invent almost nothing but DO effectively bring inventions to the market)
        Steve Jobs: A whole miriod of “stupid” “special” connectors when industry standard interfaces already exist, the Macintosh, and the smart phone
        Edison: The viable/marketable light-bulb and its E-connector. Audio and Movie recording tech and lots of other stuff.
        Musk: The whole Tesla car, power walls, solar panels trifecta (which is damn good idea), and then there’s Space-X and eventually satellite phones.
        As annoying as this kind of personality is, they do fill a vital function in the invention ecosystem. Steve Wozniac, Nicola Tesla and a whole constellation of brilliant minds were never wired to effectively put a product to market. The world needs technically skilled “snake oil salesmen” to close the gap between inventor and needed product in many cases. (Personally, think Elon is a lot less snake-oiley than Jobs.)

        “marketer/inventors” Gotta love ’em. Can’t help but loathe them sometimes. I’m glad they exist and I wish Elon the best.

        1. Just a small side note… I charge my Nissan Leaf with a Tesla station. You see, while everyone was dragging their feet with a final connector design for the j1772 protocol, Tesla needed to move forward to actually make their cars. So they came up with a connector that was good and said everyone can use it. That was declined in favor of a slightly bulkier connector but they both work the same. meaning all you need is a very simple adapter and you can plug everything in.
          I don’t think of most of their stuff as being intentionally incompatible like Apple… Instead it’s a lot like Cutting Edge server Tech versus regular desktop PCs.. what they are doing right now may be incompatible with your machine, but in the future your new machine will be able to do it too.

          1. So you actually have an adapter for your j1772 port to work with Tesla chargers? Isn’t there some kind of Tesla account involved? I’d love to hear more details about your success story. :-) I’m using a 2015 Nissan Leaf. Actually there are three of them in my life, my dad has one, my brother-in-law, and mine. Sometimes my garage/driveway looks pretty interesting with red, white, and grey Nissan LEAFs all right there together on weekends. :-) We all paid <$10K for them and love them as run-about machines. Tesla's are out of reach and I'm not sure I could handle transition to having no instrument cluster. It smells too much like how so many appliances have gone to having just one button. When time is of the essense, like while driving, I want my info right there in front me. I don't want to go looking for it and I want to hit a readily available real switch or button to turn features on or off. :-)

          2. You just need to get a jdapter or a Tesla tap adapter… There is no Tesla account… I should note though that this works with Tesla Chargers everywhere except for the superchargers. The superchargers are the only ones that have an account associated with them.

        2. >” this works with Tesla Chargers everywhere except for the superchargers”

          But what’s the point then? Isn’t the regular charger just a 240 Volt outlet with a different socket?

        1. Yea, it took a few iterations to get something viable, kinda like redstone rockets or a certain fictitious castle I can think of. :-)

          “When I first came here, this was all swamp. Everyone said I was daft to build a castle on a swamp, but I built in all the same, just to show them. It sank into the swamp. So I built a second one. That sank into the swamp. So I built a third. That burned down, fell over, then sank into the swamp. But the fourth one stayed up. And that’s what you’re going to get, Lad, the strongest castle in all of England.”

          The future belongs to the relentless. (unless you let your sons grow up to be panzies)

    1. Baker Electric cars were being used in the 1890’s. Their batteries are still good too. They never go bad. You can leave them dead for decades, over charge them, keep them way too cold, nothing stops thorse batteries from working.

  1. luckily he fused the output of each module with a solid wire.. a smd fuse costs a couple of cents more but more reliable. Also don’t think the standoffs are a good conductor too, try replacing them with a copper one or use a proper thick cable.

    Why is he putting all the batteries in parallel using the .1″ header? If it is for voltage measure put a large (1-10k) resistor in series to the header, otherwise if a cell fail the equalizecurrent will go through this header.

    1. No it looks like the XT60s are forming a bus. Not transferring power on the standoffs. I’ve abused XT60 and XT90 connectors in lot of ways. End to end is pretty inventive. But also, only rated for 50amps in short duration. And only if you do a good soldering job.

    2. >Why is he putting all the batteries in parallel using the .1″ header? If it is for voltage measure put a large (1-10k) resistor in series to the header, otherwise if a cell fail the equalize current will go through this header.

      Yeah, I didn’t understand why he’s doing that either, seems like he doesn’t really understand the purpose of those things , which is to monitor individual cells, not a bunch of them in parallel and as you say, if one has a voltage below the other, the current will go through that puny terminal.

        1. The PCB guy doesn’t have a clue either as long as de DRC comes out ok it’s ok for him. jehugarcia is responsible for the choice of components (connectors) and the schematic. He should have ask an EE for reviewing the design choices he made, especially when high currents are possible.These battery packs are dangerous stuff and in case of a thermal runaway his house prolly burns down. I hope he’s got a good insurance and a freestanding house with no nearby houses.

          The sad thing is this video gets on the internet and other people get the idea it is ok to use a resistor leads or some fiddly tracks as a fuse. Or use iron standoffs as high current traces.

  2. Greatest human? Greatest inventor? Guys, do some fact-checking, OK?

    Have you seen the way he attacks people who dare to question him? Like the way he labelled Vernon Unsworth as a “pedo” for shunning the submarine he planned to rescue the Thai boys? Or the way he sets his fans on people on Twitter?

    Re-useable rockets have been around for decades – not just the Shuttle, but the DC-X, Roton, Hyperion, HOTOL, ROMBUS…

    Electric cars have been around for well over a century.

    Powerwall is not really an invention, it’s just a big Li-ion battery, a scaled-up UPS.

    Thanks for the article, but no thanks for swallowing the Musk hype.

    1. I get where you are coming from, there were certainly DESIGNS for reusable rockets before the Falcon 9, and in the case of the DC-X at least some early test flights, but the F9 is certainly the first operational one.

      And no, the Shuttle is far from a reusable rocket. More like rebuildable rocket.

      1. Does Buran-Energia count as potentially reusable (well, if economics of former soviet union haven’t killed the project)? For Energia busters – there were plans for reuse, though not implemented at the time. Buran – reusable.

        1. Tom Mueller had more to do with F9’s design than Elon Musk and without him there wouldn’t even be a Spacex.

          The first reusable rocket to reach space was the X-15.
          Sure it’s sub orbital but so is the F9 first stage if they threw a couple of billion dollars at it and made one the size of a jumbo jet they could have evolved it into the first stage of a launch system.
          There also were plans to try and make the Saturn rockets at least the S-IB and S-IC partly reusable.
          The fly back S-IC almost became the first stage of the shuttle and the program probably would have earned a much different reputation as the vehicle stack interactions wouldn’t be as bad and there would be more margin to go to things like safety systems.
          Most ambitious of all was the SASSTO which would have been a true SSTO RLV or used with a first stage for larger payloads.

          1. Musk is an “inventor” in the same way Thomas Edison was. I don’t know what the case is today, but in my day we were taught that Edison was the greatest American inventor to have lived. But, if you dig into it, he was the greatest American manager and recruiter of inventors and fundraiser for inventors who also enjoyed getting his hands dirty with them. That is Musk. And that is what we have needed to restart the engine of invention in the heavy mechanical realms we used to excel in.

            All of those past attempts at reusable space vehicles failed because they lacked a Musk – someone to say we can and will do it and stir up the funding to make it happen and also to push the design hard toward simplicity and elegance.

        2. Elon got into Tesla when they had a prototype car, the T-Zero , a two seat roadster with an insane power to weight ratio. They also had a vision of a connected EV charger standard for load levelling the grid. The current Tesla models under Elon are not particularly well built: Drivetrain losses are high around 45% from the socket at low speeds and it is only getting 200WH/kM average which equates to 159g CO2 / kM, worse than most gasoline cars!!

          1. No Foxpop, that’s not the calculation you wanna do: US electricity production (wind coal nuclear…), including distribution losses and plants running part load accounts for 620g CO2/kWh (2017 figure) at the outlet. A Model S with a 85 KWh battery pack usually takes 70 kWh from empty to full and that is 43400g of CO2. On that it drives 420 kM if you are careful and that is 103g CO2/km. However charging is not optimal and winter time charge efficiency drops and the battery heaters also uses wall power, so on average 159g CO2/kM. This is not different from an ordinary gasoline car, but you are spared for NoX pollution in the city air, but it doesn’t have the slightest impact on global warming!

            The energy content in gasoline is irrelevant in your calculation if you compare them to electrical kWh when you don’t factor in the pollution from fossil fuels, with the associated losses, used to generate the electricity.

          2. I think my equation still stands since a powerwall owner might (should) also have adequate photovoltaics installed to keep it charged and then there is no CO2 involved at all/ (ignoring initial setup) Those batteries in both the powerwalls and the cars are huge and can store plenty of energy to get both the house and the car through the nights with reasonable planning. You need the whole system to get the best of it. Plus, not all grid electricy is produced via methods that release CO2 so that needs to be taken into account. <100% of an EV's power comes from CO2 releasing. Exactly 100% of a gas/deisel operation involves CO2 releasing. The electric vehicle and powerwalls are only part of the boat, but it is a vital part, and once it is in place the other energy production methods become more practical. People also argue about the environmental impacts of making all those batteries, but anyone who knows batteries well, knows that they are very recyclable, especially when they are large in both quantity and size.

      1. In contrast to spoken text, the problem with written text is that is is very hard to understand the intention it was written with. So I think we have to give Brian the benefit of the doubt here. I don’t think it is a question of bad history knowledge, lack of research or just trying to troll. Nope, I firmly believe that this is just plain old sarcasm.

        Unfortunately sarcasm is very subtle even in spoken text, so trying sarcasm in a written text is a challenge. Perhaps some more practice from the writer, but most certainly some more understanding from the readers is required.
        Because sarcasm is easily misinterpreted if it is not expected.

        But people.. everyone who believes that Brian was serious when he said Elon invented the car or the rocket, well… come on… from a person writing technical related articles that CAN ONLY BE sarcasm.
        I’m pretty sure Brian is aware of the actual facts and shame on all of us if we fail to recognize this well placed sarcasm.

        Then again, sarcasm or not. It doens’t matter if you invented something or not if you succeed in promoting the technology to a point where it is widely used and accepted, well then you are (in my book) equally important as the inventor. Sure credit must go to the one to who the credit of the invention belongs. But for the real facts (that’s partially sarcasm) we always have wikipedia and the internet.

          1. Yeah, and Brian could also have added “nighttime caped crimefighter” or something even more obvious. But really, a writer has to assume a certain degree of intelligence on the part of their readers. For those that fall below the bar, I guess wind ’em up and watch them go…

        1. I got halfway through the first paragraph, saw the setup unfolding, and I was already laughing because I knew exactly what I’d find in the comments section.

          The actual comments section did not disappoint. I’m dyin’ over here.

          Brian, you’re a puppeteer par excellence, and your marionettes don’t even see their strings. Masterful.

    1. anyone that pushes lithium as a house battery system is a dipshit.

      it is the worst possible battery chemistry they could be used for such an application. lithium only niche is Mobile devices because of its weight to power ratio mean you phone doesn’t way 6 pounds. out side of that they fall short of all other battery tech on safety, charge cycles, recyclability and availability of materials.

      in less then ten years your going to be replace these batteries for a fresh batch for a boatload of money.

      but if instead you used a nickel iron battery the physical size may be larger but the batteries will lost a 100 plus year, need a little cleaning, then last another 100 years off continuous charge and discharge cycles. it is made of some of the most common and recyclable materials on the planet and you can pound on it with a sledgehammer and it will never explode in a fiery ball of death that take the whole house with it.

      and that’s just this if you want to talk about how stupid the mars plans are or how the hyper loop scam falls apart if you start running the numbers we’ll be here all day

    1. He bought a company that had an electric power and drive design. While attempting to shoehorn it into a modified Lotus roadster, the engineers Musk hired had to pretty much redesign the whole thing, along with much of the Lotus chassis.

      Then just as Tesla got rolling with production on the roadster, Lotus calls up and says “Y’know all those suspension and other parts we’ve been making, along with the custom chassis etc? Yeahhh, we’re discontinuing our version of the car so we won’t be needing to make any of the stock pieces your roadster has in common.”

  3. “Elon Musk isn’t just the greatest human being — he’s also a great inventor. He’s invented the reusable rocket, the electric car“

    I really hope this is sarcasm

  4. Comparing Musk and Jobs is like comparing an apple and a pear. Jobs was a visionary who inspired people to build to his vision you didn’t actually build those things. He inspired people and gave them a direction and the definition and expected nothing but the 1%. Musk on the other hand is a true engineer, and brilliant just as brilliant as Jobs was it being a visionary. I’m sure he doesn’t build everything himself either but I’m sure he has to technical knowledge to deconstruct and reconstruct and make it better.

    I agree the author has too high of praise and is using the wrong words. But it doesn’t take away from the fact that Musk is the man of the century at this point. The hard push for EVS wouldn’t even be on the table still if it wasn’t for this man pushing the industry. The man that pushed Titans of Industry to keep up. You really can’t deny that.

  5. “Elon Musk isn’t just the greatest human being — he’s also a great inventor. He’s invented the reusable rocket, the electric car, and so much more. While those are fantastic achievements, ”

    I think BB is being tongue in cheek.
    One way or another.

  6. Ol’ Musky… I can only hope when we are all cosmic dust he will be Steve Jobs’s power bottom with Dyson just watching from the corner, “underway sock” in hand ;)
    Snake oil is the lubrication for most of these things it seems.

  7. Can we just pause the circle meat rub by appreciating that jehugarcia came up with a solution so brilliant in it’s simplicitiy that it will make powering electric stuff much easier?

      1. For a powerwall application you will never see the load he was subjecting it to. That was more of a slam it against the wall and see what happens test and not representative of the intended application.

        In a powerwall setup each module would see 1/20th or less of the load he subjected it to.

    1. Down on power? Battery pack overheating? Head for the back country to find a washboarded gravel road and/or a few cattle guards. That’ll jiggle all the cells to refresh their contact.

  8. If there was a low melt temp solder paste that could be put on the contacts…

    Assemble each module with fully charged cells and a skim of solder paste on the contacts. Put enough draw on to heat the contacts to the solder melt temp. The solder melts and makes a lower resistance contact. The contact temps then cool below the solder melt temp. The lowered resistance prevents the contacts from getting hot enough again to melt the solder.

    How about holders that have contacts with several sharp barbs punched in? Those could dig into the ends of the cells a bit to make better contact. Apply silicone grease to contacts and cell ends first so that air is sealed out.

    Copper standoffs at the power end should be a better choice. The monitoring bus connectors should be available with longer pins or taller connectors to bridge the gaps.

  9. Power walls using lithium tech are silly, in the long term, because they should last longer and don’t need to have the energy to weight ratios that an electric car battery has. You can build them bigger and safe and have them last as long as the house in some cases, but you really do need a wall of them, not a problem if it is designed into the overall structure and doesn’t pose the fire hazard that lithium based cells do.

      1. Actually the old Edison style Nickel-iron battery is even better, if you incorporate the latest nanotech and materials research into the construction of the electrodes to boost their surface area and conductivity, while reducing the mass of nickel require per W/hr of storage. No need for a very low leak cell when you cycle them every day, what you need more is 50 plus years of cycles. They are very hard to break and rejuvenating them is very easy. In a big fixed install you could even avoid the use of flammable plastics for the cases. I am talking about something that would be part of a house design and construction, serviceable but designed to not need it. Even their hydrogen production risk can be removed with a small converter that uses platinum to cycle it back into water, so you can have them sealed. pair them with super capacitors and the hydrogen issue is even less likely as they have the charge/load peaks smoothed off and unlike most other rechargeables they will come back from zero charge with no extra work and be fine again after a few cycles.

          1. Yes, but crudely. I imagined putting money aside as an investment to fund new lead acid batteries on a regular basis and still was money ahead compared to the purchasable models I found. I even looked at bulk nickel sheets to see if I might try to make my own (academically, not so much serious) Nickel is disapointingly expensive. Iron, not so much. :-) Prove me wrong. I would love to to be wrong here. Make me eat crow. That kind of crow would be yummy. :-)

        1. Lead acid batteries have extremely high charge efficiency. You get 95% out of what you put into them. energy density is not relevant when the battery is stationary. However Li-ion costs fewer $$/Wh installed, but their life span is shorter in contrast to properly managed wet cell lead acid storage batteries

    1. Amen. Home storage is silly. Storage is needed, and even distributed storage, but it doesn’t make sense to provide storage at residential premises. Most places have a reliable power grid, and the storage should be located at substations, distribution transformers etc where it is maintained properly and can be economically size.

      Lithium battery technology is an environmental disaster and an unnecessary use of rare earth minerals for a long life but low energy density application.

      1. From an engineering perspective your statement is spot on. In the context where I live where the utility company is owned by its customers, your recommendation would be wise. When the utility company is for profit and consequently not the servant of the customers, having your own powerwall of any practical flavor and personal energy production would likely be superior to having to “pay the man” for every joule. As usual, the human equation makes solutions more complicated. (Adam’s curse at work – we’re not good enough for such a perfect world.)

          1. Line the right pockets and regulatory bodies can be neutered. (the normal state of things) It’s best to keep control of vitals like energy as close as possible. Otherwise you will have more than you want when it’s abundant and none at all when it’s scarce because there is always someone else out there with deeper pockets and/or greater influence than you to redirect the scarce resources their way. Pooled resources combined with unscrupulous leadership is a recipe for disaster. I’d say it’s best to own your own powerwall than to share a large one with the public. People are too selfish for group ownership to work well. Take a look at the great prosperity of present/former communist/socialist countries. That says enough right there. People just aren’t good enough to make collective ownership ideas work without some incorruptible compulsory force keeping everyone in line.

      2. Well that would be ideal, but you will find that even harder to get implemented than to do it yourself, and there are a lot of more remote properties in my country Australia that do need there own storage too.

        1. Yup, regional diversity leads to diversity in optimal solution. Local central energy storage works well in places with higher population densities and where the service can be counted on to serve its customers with high value, like when it is owned by the customers as a CO-OP, municiple utility, or regional public utility like mine. (For profit companies tend to cut corners to pay dividends rather than build up spare parts and other options in the event of failure) It’s a winner Idea in my neighborhood for reasons just stated, but I think that is more of an exception than the rule. Usually the population density is too low or the electric company is too greedy. I’d like to be wrong here but I don’t think I am. :-)

  10. That board wasn’t designed by Jehu, That was made by a guy called Justin and did it in his own time. He sends me his update of the board layout each time he works on it and then sent it to Jehu.

  11. Elon is a mouthy putz with money. His products never live up to his hype and his customers always seem to be taking him to court. His mouth gets him into all kind of trouble, I am not a great Jobs fan either but at least Steve made some really cool products that did meet customers expectations.

  12. If the editor left out the totally unnecessary verbiage abut elon, the designer of this battery bank might even get a few comments. From what I have seen so far he did a nice job. I have looked a bunch of builds and I really like this one a lot. Gives me some ideas, though it looks nice enough to just co opt as it is.

  13. Two things. One a company called Baker was working on their Townmobile rig, and needed a battery for it, they asked Edison. And of course two, one of the companies best known for making back generators, read fuel driven monstrosities for home use, also makes exactly what he has in mind there. Oh and Baker still exists today in their corner of NJ, they make forklift trucks that are considerably better then the gas powered things most businesses that need them use.

    I see this idea as a good one only because it is something most of us can wrap our minds around. However, BB has another problem, he’s surrounded by about two million hungry tribbles.

  14. Although I feel Brian and these articles are the best examples of what’s wrong with HaD, I’d probably be harvesting comments in the same way if I was in his position. So I can’t really fault him. Money is money man.

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.