Homebrew Powerwall Sitting at 20kWh

Every now and then a hacker gets started on a project and forgets to stop. That’s the impression we get from [HBPowerwall]’s channel anyway. He’s working on adding a huge number of 18650 Lithium cells to his home’s power grid and posting about his adventures along the way. This week he gave us a look at the balancing process he uses to get all of these cells to work well together. Last month he gave a great overview of the installed system.

His channel starts off innocently enough. It’s all riding small motor bikes around and having a regular good time.  Then he experiments a bit with the light stuff, like a few solar panels on the roof.  However, it seems like one day he was watching a news brief about the Powerwall (Tesla’s whole-home battery storage system) and was like, “hey, I can do that.”

After some initial work with the new substance it wasn’t long before he was begging, borrowing, and haggling for every used 18650 lithium battery cell the local universe in Brisbane, Australia could sell him. There are a ton of videos documenting his madness, but he’s all the way up to a partly off-grid house with a 20kWh battery bank, for which he has expansion plans.

There’s a lot of marketing flim flam and general technical pitfalls in the process of generating your own non-grid electricity. But for hackers in sunny areas who want to dump those rays into local storage this is an interesting blueprint to start with.

117 thoughts on “Homebrew Powerwall Sitting at 20kWh

      1. You need a very deep hole with some nice ducting and vents away from where you want to breath, Fire retardant covers etc and some drainage.

        It’s just a Nightmare.

        I worked in a place with lots of lithium cells if it went up not much could stop it this was accepted just run.

    1. Actually, the type of lithium ion cells that come out of laptop batteries are almost impossible to ignite – they’re high efficiency cells that only do 1C discharge. They’re not like the high-discharge flight batteries that dump hundreds of amps when shorted out.

      (yes, we’ve tried.)

      1. Also, in this sort of installation every cell is individually fused (by using a small length of thin wire to attach to the bus bar) – so in a freak event where one cell in a parallel bank suddenly goes short circuit, the current flowing from the rest of the bank pretty much immediately blows the fuse and takes the cell out of circuit. It helps that each cell only needs to supply a very low current, so the fuses can be low value.

    1. I wouldn’t be so much worried about the fire (that is what fire supression systems are for — water / halon, etc)

      I would be more worried about an explosion that blows the house apart…

        1. Lithium reacts with water, strips away the oxygen and releases hydrogen. I can picture many modes of failure that would result in a bang. Especially with double glazing and good seals around the windows and doors.

          1. @Bill I do not know about the particular make and model used here, but older lithium batteries did indeed contain Lithium oxide and very pure Lithium metal.Which when opened changes from metallic (or a white encrusted metal) to black as it oxidises in Oxygen.

          1. Point being that lithium cells themselves don’t explode like M-80s but just pop open and vent fire. Of course if you put them in a sealed cabinet, the pressure builds up until something gives.

          2. @fennec, The pack as a whole didn’t pass regulation, but it was built using off the shelf cells such as these that had passed regulatory testing individually.

            @Dax, the cabinet wasn’t an air tight seal or anything, it’s more akin to a blast shield. It’s not sealed specifically because we don’t want to make a bomb and it still managed to blow it open.

          3. ” that had passed regulatory testing individually.”

            They aren’t tested individually, because the only way to test if one explodes is to stress it till it blows up. They’re tested statistically.

            “! and it still managed to blow it open”

            That was probably a gas explosion in the cabinet when the fumes from the batteries ignited. They usually smoke a lot before getting hot enough to catch fire, and that smoke is basically fuel vapor. The electrolyte is an organic solvent – think acetone for a simple example. The real stuff is some 40-letter word that nobody can pronounce.

            The battery is full of liquid fuel that boils and bursts out when the container breaks from overpressure. You got vaporized fuel, air, and an ignition source in a closed cabinet – little vent holes won’t help, it’s still going to throw the door out when it goes.

          4. I didn’t mean to imply that each individual cell had undergone regulatory testing, just that that particular make and model cell had already been qualified and was listed.

            “That was probably a gas explosion in the cabinet when the fumes from the batteries ignited. They usually smoke a lot before getting hot enough to catch fire, and that smoke is basically fuel vapor.”

            Yes, this is exactly what happened. I supposed the more clear language shouldn’t be “the cells exploded” but “the cells caused an explosion.” We’re actually not allowed to use the word explosion at all, but this is anonymous so I feel ok using it. We repeated the test with a ton of ventilation to try to prevent another explosion and managed to make a fire tornado. I just want to make sure people who are putting homebrew projects with huge banks of lithium cells in their homes know all the possible consequences if something goes wrong and are taking safety seriously and using every precaution. Professionals don’t even get it right on the first try every time, but when that happens usually someone just has to clean up a mess in test chamber, in someones home something like this with that many cells, even if it has to be a perfect storm of conditions to happen, could absolutely destroy something if they managed to smoke and cause an explosion.

      1. The problem with lithium battery fires is that they’re self-oxidizing. When these batteries burn, they will burn until they burn themselves out. Everything they need for a fire is entirely self-contained.

        Even less dangerous battery chemistries will happily ruin your day. I saw a 40kg LiFePo battery pack go into thermal runaway at an F1 track a couple years ago. The F1-rated fire suppression system barely made a dent.

        1. Different battery chemistry, drastically different configuration. The battery you saw burn would have a massive discharge rating. This powerwall uses individually fused cells that only do 1C; the fuses pop before enough current flows to make them get hot, and you pretty much have to *throw them in a fire* before they catch fire.

          Remember, the cells are designed for *laptop* use – they’re extraordinarily safe. The incidents you see with batteries catching fire in personal devices are always of the LiPo pouch type (e.g. the recent Samsung Note 3 debacle), *not* 18650 LiCo cells.

  1. Guys there are so many ways to address potential fire/combustion/continuity issues, rather than disparage
    with facile one-liners offer a few meaningful thoughts to go somewhere ie its a “no-brainer” as positive
    criticism instead please, you have several to start from such as & in no particular order:-
    1. Extinguish – water is best, high specific heat, cheap
    2. Array temp sensing
    3. Smoke/volatiles sensing ie Latter as batteries can release volatiles but below initial stoich ie prior combustion
    4. Switch out banks re 2 & 3 at least & offer mission critical if continuity needed re handling 1
    5. Other – the ubiquitous arena not yet considered – over to others…


    1. Only one comment/modification – Water is NOT best for electrical fires. There are enough extinguishers out there that are safe for electrical fires and easy to add to either an automated or “panic switch” type system.

      1. Water is not the best in lithium battery fires, because the metals in the battery are very keen on reacting with water exothermically, creating more fuel and heat for the fire.

        The energy remaining in the batteries basically splits the water into hydrogen. It’s a similiar problem with CO2 extinguishers, which react to produce carbon monoxide which is toxic and burns.

        Lithium battery fires are problematic anyhow because lithum reacts with moisture in the air at room temperature, which re-ignites the fire after you’ve put it out. You have to remove oxygen, water, and heat at the same time and then cover it up in some sort of inert foam or oil which prevents further reaction. Then you have to dismantle the wreck without exposing it to humidity in air or it will again heat up and ignite.

          1. That’d be a good idea if he didn’t have to manually balance them. As it is he has to have access to both ends to disconnect and isolate them. Unless he makes an automated solution pulling those out would be super messy.

        1. These are Lithium ION batteries. They do not contain metallic lithium. But as somebody else pointed out, they contain flammable organics. Probably very fine water mist would be a good choice of extinguishing system, it has extreme cooling capacity. But you don’t want too much conductive water in the electrical DC system. CO2 sure would work.

    2. I have one comment/modification for this. Water is NOT the best for electrical fires. There are a number of extinguishers that are safe for this type of combustion that could very easily be added to either an automated system, or a “panic button” type deployment.

      1. I know right, meaningful thought in this instance, meaning not actually thinking or knowing even the first thing that you’re talking about while trying to maintain an authoritative tone. I mean it’s not even being slightly dull about chemistry, but that fact that it’s electrical as well.

          1. No, they don’t. That’s why they are called lithium ION.
            The “Po” – polymer – refers to the type of construction and the nature of the electrolyte. So they are basically glued together with the polymer-bound electrolyte and do NOT need the steel container. But they are of a lithium ion chemistry. The common round cells like 18650 are cells of lithium ion chemistry using a steel envelope.

    3. Oh I’m not ragging on him, I think this is a great project and I’m sure he has the safety aspects under contro, but the fact is that he has a huge array of Li batteries, used, and we know the dangers of this stuff. I hope this system serves him well for his/it’s lifetime, but building one would scare the crap out of a lesser man, like me for instance!

    1. Sure does, if you watch enough of his videos you will see how much effort it took to put it all together. Doing it worked for him, for a lot of different reasons, but I can’t see it being ideal for most people, even if they have the required skills.

      He does a costing (sans man hours) and his solution comes in at 10% of an off the shelf FeNi battery system, but unless it last 5 years without requiring any maintenance or part replacement it is a false economy in the long run.

      I’d rather pay a lot more up front for a system that will last almost as long as the house will, with zero fire risk and almost unkillable chemistry. The only problem with the FeNi is discharge rate on peek loads, but that is where a relatively small number of super capacitors can make all the difference. The super caps can also use tech that has an extremely long life that matches the batteries.

  2. From my (admittedly limited) knowledge, charging lithium batteries are like chains of LEDs. If you stick some in parallel, one will take more current – and get warmer.
    If you stick them in series, and one goes short circuit, you get a higher voltage across the remainder, and THEY get warmer.
    Ambient temperature in Brisbane come summertime means they will ALL get warmer.

    I would suggest at least one temp sensor per bank, plus a few ambient sensors. Any difference beyond a point should trigger alarms, or at least isolate the warm bank.

    1. Well, best practice really is to monitor at least bank temperature. At multiple points, if possible. Hence why you’ll find multiple temp sensors in battery packs for notebooks, just to name an example.

      When connecting/using/charging LiPos – no matter if P or S config…or both – you want to have their voltages quite close to each other. And the internal resistance should match aswell, if possible. The cells of a pack in P configuration will balance each other out, the issue there is just the current draw. If you build a 4P pack with three full and one empty cell, the three full ones will rapidly charge the empty one until all four are at the same voltage. Hence the reason why LiPos should always be charged with a CC-CV PSU.

      If connecting the cells in series, a balancing system is required. A proper one which not only measures, but also charges/discharges the individual cells of a pack. As you said, if one goes dead – which mostly results in a short – and the charging circuit doesn’t notice, they all go dead.

    2. Lithium cells don’t have the negative temperature coefficient that LEDs do, so it is safe (and standard practice) to put cells in parallel. They self-balance when in parallel. Series is another story, so yes, you have to take steps to balance the voltage across any series string of lithium cells.

      1. During charging and discharging, some cells have higher/lower internal resistance and the current isn’t shared evenly among parallel connected cells. At low currents the differences are minimal and basically just causes uneven wear on the battery pack because the low resistance cells get discharged deeper and then re-balanced when the load goes away, adding to their cycle count.

        At high currents you can get a cascade effect going on where one cell pops before the others.

    1. Can you quote some common homeowner’s language that makes this invalidate your policy?

      I don’t have direct knowledge for or against this. I specifically asked my agent about a few scenarios and she told me that “insurance covers stupidity”. That said, if this is done without proper building license and inspection I would think you would be in a pickle.

      Would love to hear from someone with relevant knowledge and experience.

      1. Here in the U.S., homeowner’s insurance usually doesn’t cover damage resulting from major work by unlicensed/uncertified contractors, so if you decide to rewire your house yourself and it results in a fire, you’re on your own. Similarly with plumbing and water/sewage damage. The workaround for DIY types is to have the work inspected/verified by a licensed contractor or an inspector before it’s put into service.

        This varies from policy to policy and by location, and of course there have disputes where a homeowner has replaced minimal wiring (as little as a light switch or a plug), there’s been some disaster and the insurance company tries to avoid payment but that seems to be more about money grubbing than a contractual problem.

        1. In my experience, properly permitted and inspected work (including major electrical, structural, plumbing, and hvac) is covered in all cases regardless of the worker. In 3 of my local 14 jurisdictions; there are licensing requirements for some work, but these are exceptions and not the rule.

          1. The key word is “inspected”. Yes I can do my own work but I still have to get it approved by a licensing authority to be considered “safe”. My insurance won’t cover anything that doesn’t have the electrical safety authority approval.

            We recently got inspected by our insurance and the first things they looked at were the electrical panels and wood burning appliances. The inspector basically said if there is no sticker there is no insurance.

            Two years ago we had a neighbour get told to replace their oil furnace within 90 days or lose their coverage because heating oil is now being considered to risky to cover with very high environmental cleanup costs.

            Insurance companies also aren’t willing to insure former grow op homes due to electrical wiring and mould issues. Tell that to the mortgage company when your loan gets denied because insurance won’t cover your premises.

      2. “quote some common homeowner’s language”

        Thats a laugh – obviously you never read your insurance policy and legalese. Nothing common or home owner oriented at all in there.

        You will like this clause

        “There insured shall notify forthwith the insurer of any vacancy or non-occupancy extending beyond 30 consecutive days, or any increased hazard that shall come to his knowledge and that every increase in hazard (not permitted by this policy) shall be paid for by the insured – on reasonable demand – from the date such hazard existed, according to the established scale of rates for the acceptance of such increased hazard, during the continuance of this insurance.”

        I think an uncertified/un-inspected home-brew 20kW power wall and a trip around the world lasting more than 30 days would affect your insurance coverage if something happened and you didn’t tell them about it.

        Check out this article

      3. I live about 1.5 hours north of Brisbane, off-grid with 2.5kW* of PV, 1320ah* of lead-acid batteries, controllers, a sine-wave inverter, and a house dual-wired with 24VDC and 240VAC. Plus a backup generator. It’s been upgraded twice over the last 20 years, and to get the subsidies, it had to be installed by a licenced solar/low voltage specialist, and the 240VAC system inspected and certified by an electrician. The insurance company only wanted to know that it had been installed and certified, and the estimated replacement cost.

        *Some of the panels are over 20 years old and I doubt they’re supplying their original wattage. The current (ha ha) bank of batteries are about half-way through their expected life, so I expect 1320 amp hours is a bit optimistic.

        As a matter of interest, I was cold-called by an installer recently to give me a quote. I told him I might be interested in replacing some old panels, and maybe a new battery bank. He ignored me completely, and quoted a turn-key system with new panels, controllers, inverter, and LG Chem lithium batteries – and throw out the entire existing system. I chuckled and told my regular installer about it and he said that the CEO of LG Chem had visited Australia and had explicitly stated that their lithium batteries WERE NOT WARRANTED for off-grid use, as few people were willing to install auto-start generators to avoid over-discharge of the batteries.

        1. Why would you need an auto start generator? You just need a proper deep discharge protection which switches off the batteries in case of deep discharge if you accept this power outage. You could even switch off non essential circuits at 10% or 20% residual charge to keep the freezer and a small light in each room running longer. And then you go outside and pull the generator to start, if you need/want it.
          I have only a very small system in a garden house, 50W panel, two old car batteries and some LED lights for weekend use. If I need more power then I start the small 2-stroke gasoline generator.

  3. He knows what he’s doing, he’s watching his charging carefully.You would be more likely of cell shorting than thermal runaway. We should be applauding this man for use of recycled goods.

    1. There’s not much recycling left in lithium cells because the breakdown mechanism is exponential.

      Once past the knee point of the curve (67% capacity), there’s very few full cycles left. If the original battery was specified at 700 cycles to the breakdown point, it’s got maybe 70-100 left beyond that


      1. Sure, if laptop cells were individually replaced when they died. But it doesn’t happen that way. You have to replace an entire pack when it dies, and they *almost always* die because only one or two cells kill themselves early, whilst the rest of the cells are still in prime condition. Eventually self-discharge kills the remaining cells in the pack, but if you recover them before that happens, you get a lot of very good cells for almost nothing.

    2. Odds are at some stage it will get taken out by a gecko, the only question is the scale of the damage. The gecko will not survive. Speaking of geckos and electrical damage, good luck with warranties and perhaps even insurance if that happens because I know people with external heat-pump units (air con) have had their claims refused on those grounds. It is the owners responsibility to do the impossible and keep animals out of the gear. Just a little and very specific detail to ask about when you are looking for coverage.

      1. That reminds me of the time my inverter started to shut down under moderate load. I noticed the cooling fan wasn’t kicking in so I took it to the supplier. The fan was replaced and I was handed a mummified frog in a plastic bag. Said frog had sacrificed itself across the power terminals of the fan.

  4. If my understanding of the chemistry of lithium ion and polymer batteries is correct, isn’t the electrolyte a lithium based salt and the electrodes are commonly aluminum and copper or the like? Thus the many comments about lithium reacting with water are not really relevant since it isn’t lithium metal but an ionic salt solution. I was under the assumption that the “exploding” batteries was due to a shorted battery heating up, venting the boiling electrolyte and that electrolyte being flammable and not that there is metallic lithium reacting with the moisture in the air. Please correct me if I am wrong.

    1. No you are partially right, everyone else is an idiot. Anode is usually graphite on copper, and cathode is usually a lithium intercalation compound on nickel or aluminum. No metallic lithium is present.

      1. “lithium intercalation compound”
        “No metallic lithium is present”

        Intercalation is basically layering of lithium atoms between layers of another metal. It’s not quite a separate as a metal, and not quite an alloy but something in between – and the point is that the lithium metal comes easily out of the intercalated positions, and as such will react with water.

        But most of the stuff that burns in a lithium battery is the flammable organic solvent used in the electrolyte. The exothermic reaction with water and lithium provides the ignition source, and the electrolyte provides the fuel.

  5. I’ve been very interested in playing with Nickel Iron batteries for this purpose. I know they have some big disadvantages, higher self discharge rate, maintenance required and the necessity to vent due to production of hydrogen gas. But…

    More modern types of batteries are such throw away items. I don’t like to buy expensive things like house sized battery banks if I am going to have to keep re-buying them every few years. I don’t see how anyone can save any money that way! I’ve read that nickel iron batteries have been known to last 50 or even 100 years. They just need topped up with water and every decade or so they might need new electrolyte. That sounds like something worth experimenting with to me. Perhaps the higher self discharge rate can be compensated for with a greater number of solar panels, wind generators, cow farts or whatever one prefers to power their house with.

    1. Yah, these… you almost can’t kill them. They are large, but that shouldn’t really matter too much for a house. Have ’em sitting along your back wall in deck boxes or something.

      I’ve been trying to brain up a way to halfass them from junk, like dump a bunch of metal in a rubbermaid tote and pour in the lye kind of deally. Seriously if you could get the capacity of a car battery in a tote, for say quarter the price of a new lead acid, but you could hide it somewhere and ignore and abuse it, it would be very usable.

      Self discharge doesn’t worry me too too much, realistically you’re not likely to be planning to store power weeks or even days ahead. Sure, goes flat calm and cloudy you put off your power hungry activities and try to last it out. But you’re never going to have a ski cabin where you think you’re gonna store sunpower all summer and use it up all winter or something.

      1. Jay Leno’s 1908 Baker Electric still has it’s original 100+ year old batteries. They just need a wash out and new electrolyte periodically.

        Another nigh indestructible type of battery was the wet cell Nickle Cadmium. They were used a lot in 1940’s airplanes, especially during WW2. Their stainless steel cases made them physically very durable. The largest battery manufacturer in the world in the 1950’s (IIRC Union Carbide) bought manufacturing rights to them (after the war) for the North American market, then proceeded to do absolutely nothing with those rights – while the world pretty much forgot about that type of battery. They even fell out of use for military equipment.

        Just wouldn’t do to have automobile batteries available that would only need fresh electrolyte once in a while.

        Another interesting thing is the old book on battery rebuilding, from the early 20th century when car batteries were made of hard rubber cells in a wood box, sealed up with tar. They were very similar to the absorbed glass mat design. The lead grids with lead oxide packed into the squares were pressed together with thin sheets of wood veneer. Hot steam was used to melt the tar so the cells could be uncapped and the rubber jars pulled out of the box. After disassembling the battery completely, bad plates, veneer sheets, cracked jars and caps were replaced and the box repaired or replaced as needed. Then the stacks of plates and veneer were pressed and lead bars soldered across the top of the plates. With the jars reinserted in the box, the pressed plate stacks would be quickly set into the jars. Put on the caps, pour on tar and let cool. Solder external bus bars to connect the cells, fill with electrolyte and screw on the filler caps. Rebuilt good as new, ready to charge up.

        What would be an interesting hack is to build a 1910’s style battery with modern* lead grids and fiberglass matting in place of the veneer.

        *There hasn’t been a huge amount of improvement in the past century, but there has been some.

    2. The bigger problem is the overpotential required to charge the nickel iron battery, which wastes about half the power you put in and makes the battery unable to accept very large input power without boiling over, so with the cheap and durable battery you just basically doubled the cost of energy, or doubled the payback time of your solar panels.

      Henry Ford planned an electric car right before WW1 and he tried to use the Edison Ni-Fe battery for it, but ran into the same issue: poor power handling, poor efficiency, too expensive overall – and that was when cars weren’t required to go faster than 30 mph.

      1. Those Edison batteries are awesome for home storage. You cannot kill them by overcharging/over discharging. They require very little maintenance and will probably last 50+ years if they are built to original specs. They can be over-discharged quite a bit and still charge up to original specs, refurbishing them only requires a cleaning and new electrolyte. The down side is that they are very heavy, and large. So not space efficient. But building a 12v cell with those is a 10cell job, they are only 1.2v per cell. They also can be charged rapidly, not as fast as lithium but you can abuse them quite a bit.

        1. Except for the bit where they waste half your power.

          If the payback for just the solar panels and inverter are on the order of 20-30 years, adding the battery will make the whole thing pointless.

          1. I don’t remember from my previous research that they waste half the power all the time, just if you expect them to suck it up as quick as a lead acid of same capacity. I think there’s some work function type thing of the nickel and if you exceed that you’re making H2. Anyway, impression I got left with was, install plenty of capacity, and if you wanna soak up surges to take advantage, like really brisk winds on your genny all day, then use a smaller amount of a different tech as a buffer.

    3. Discharge rate is low enough to not be relevant in a domestic load scenario with top-up every day. Super caps smooth the load and reduce gas production plus a catalytic converter will return it to water anyway. They are not high maintenance (where did you get that from!) and venting is not an issue as any type of power bank will need venting due to cooling requirements. If over the long term you have a bit of water loss so what you can top it up easily, or have a system to do it for you that means you don’t need to look at them very frequently at all. Power density could be an issue, but how far were you planning to drive your house?

      Every 40 year or so you may need to drain them and add some fresh water and potassium hydroxide. Not a big deal really.

      1. I think they’re ripe for reinvention/reimplementation too that could reduce weight and bulk while leaving them just as sturdy. Something like a conductive carbon nanomaterial that layers of nickel and iron are plated onto. Possibly involve some material like polonium that’s a hydrogen sponge and keeps free hydrogen from gassing out, and lets it recombine with the O2 electrically.

        1. Are you sure about using Polonium? If I had enough of this I would build a RTG. :-) But there are enough non-radioactive hydrogen catalysts. At least Platinum works, although expensive – but still less then Polonium. Perhaps Nickel is also possible.

  6. Good video. I don’t really ever leave comments. But just wanted to say I like what you are doing, and like the community supporting you to do it. The time you spend building, filming and editing is a lot of work, and the donations you are receiving are well deseved. In return we are all getting great videos from someone we can relate to and with a great ititude. Win win!

  7. Pretty amazing project. All power too him.
    But why he isn’t protecting his batteries with a BMS System is kind of beyond me. Balancing with a straight busbar is pretty how you doing. I guess OK if the difference is only a few millivots.

  8. “I don’t like to buy expensive things like house sized battery banks if I am going to have to keep re-buying them every few years.”
    Tesla has been AMAZINGLY quiet regarding this aspect of of his batteries, both house and car, hasn’t he?

        1. A guarantee means money back if they don’t last, so there’s an incentive to make sure they do.

          They also sell pre-emptive battery exchanges for $12k that you pay now, and get a new battery 8 years later.

          So put one and one together and you get…

          1. What you get is precisely the situation that [Me] is unalterably opposed to–
            “…I don’t like to buy expensive things like house sized battery banks if I am going to have to keep re-buying them every few years…”

            You’re supposed to forget, sheeple, that Elon Musk’s Grand Vision In The Sky, while making money for him, costs YOU money. Big time.

          2. I don’t really care about your conspiracy theories and personal grudges.

            The point is that the warranty is for 8 years or 160,000 miles whichever comes first, and they’re making damn sure they won’t be needing to refund too many people or get a class-action lawsuit over it, so that sets a practical example of the expected lifespan of lithium batteries: 8 years to a decade.

          3. $12K less in your pocket and no absolute guarantee that 8 years down the road Tesla will still be manufacturing batteries that will fit your old car. But they may offer you a discount on a new car, with no interest accumulated, at least not for you.

          4. @Galane:
            If they give you a guarantee they will have to stick to it, except the company had ceased to exist (gone bankrupt). So they will have to manufacture battery modules compatible with your car.

      1. The man is a self-serving asshole who’s been running multiple companies in the red for over a decade without delivering anything truly groundbreaking, on a continuous stream of investor capital and government subsidies. Even SpaceX is just him playing Thunderbirds with little hope of actual commercial success in recycled rockets, but it has a specific purpose: to make it look like something’s happening.

        He is simply running his mouth faster than the engineers can make stuff work to keep on the upslope of the hype curve, because as soon as he stops talking and starts cashing in his promises, he crashes into the valley of broken illusions and the empire he built out of a bouncy castle makes a big farting PPPFFFFT…

  9. I wonder if you could reduce the risk of thermal runaway by putting more separation between each cell. It appears as though he’s got about 5mm. Is there a distance at which you could isolate a fire in one cell from spreading to the next?

    1. yes. DougM gets best comment in my book. there will be a single-cell thermal runaway event, it is all but guaranteed. i don’t know how to isolate the cells from eachother to prevent the one cell from lighting the next, but i think a little spacing is an important part of it.

      1. These are high-efficiency, low discharge-rate cells. They also don’t go into thermal runaway – the resistance goes up as they get hotter. Also, the cells are individually fused, at a level well below the point you could make one ignite (which is trivially easy since each cell doesn’t supply more than an amp or so max in normal usage.)

        Also, have you actually *tried* to get one of these things to ignite? They’re not like the lipo pouch cells you get in phones or drones. We’ve tried shorting them out, ramming nails into them, and actively overcharging them. The latter got them too hot to touch, but that was about it (it didn’t even damage the 3d-printed holder I had them in.) You want one of these to burn? You’ll need to throw it in a fire.

  10. Balancing the batteries by making a solid parallel connection between them looks like a bad idea to me. For sure voltages will equalise (they have no choice) but this can’t be good for the batteries. A setup of this size could use an activeky controlled balancing system. Alternatively he could sacrifice some energy by having balancing resistors (much easier).

    1. You don’t randomly throw cells together and make them equalise, you get them all to the same voltage first before putting them in a parallel bank. Once that’s done they’re fine. You’d just blow the individual cell fuses if you didn’t, anyway.

      Balancing is *only* required between series cells.

    2. What’s your concern about this, specifically? Every laptop battery I’ve seen (and I’ve seen plenty) used muliple cells in parallel, with these then wired in series. It’s standard practice.

      Everybody is freaking out about this, seeing all that many lithium-ion batteries in one place and interpreting it as an omen of the apocalypse.

      Hey: I’ve got news for you: there are shipping containers and warehouses full of laptop computer batteries that literally dwarf this setup.

      1. I am not freaking out about the cells – only that I do not own at least 1/10 of them :-)
        Of course balancing is a no-issue for parallel cells – they just have to stay voltage balanced.

  11. I would not say don’t do it, but if you want to play like that, I hope you live in a place *far* away from other people and very tolerant of large fires. I don’t really care if you burn *your* house down but I care if you take mine with it. And I really care if you say start a wildfire. So I am not saying don’t do it, but if you have to play with a hazard like that, do it someplace safe. As hacking efforts start to get bigger and bigger, this is going to be more and more important. I think it is pretty safe to say we are all good with an adult taking personal chances, but make sure that you will be the only person effected by your experiments should something go wrong.

  12. No battery bank in parallel self balances. The batteries with a lower cell voltage will act as a load on the batteries with a lower cell voltage and dissipate the stored energy difference as heat.

    There are battery PMICs made for exactly this purpose, to balance cells in series and parallel, they are used in tool battery packs. These provide individual cell over-current, over-voltage limiting, over discharge, etc.

    There are supposed to be thermal sensors on the battery cells or pack to prevent charge or discharge of the pack outside of normal operating temperature ranges.

    This system does not seem to take safety into consideration, but hey that’s up to the individual’s tolerance for taking risks.

    Good for you for making a prototype. Work on the safety aspect of it next.

    1. I have seen tool packs 2p4s with no balancing at all and only one thermal sensor. The drawback: one cell was dead, had no capacity left, has gone high impedance. So the usable capacity of the whole pack was only 1Ah instead of the expected 2-3Ah. But that’s the way these cells normally die: loosing capacity and increasing internal resistance to the point of unusability.

    1. Ah religion, you’re kidding but, if even a little serious, bear in mind all religions share common & low attributes
      – All claimed deities very bad impotent communicators
      – Nothing claimed can ever be verified
      – Proponents achieve; Status, Authority & Power over the intellectually feeble & emotionally meek
      There are several others but, that’s it for me, you were kidding I guess Kenneth as no-one who has
      observed details of Nature could ever take a (personal) god seriously, a parenting example they are not ;-)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s