I appreciate the engineering and time people are putting into these projects. I absolutely hate the way this article is portraying these people's work.
These are not "consumers who can't afford commercially produced powerwalls." These are people who could probably work professionally in the field if they're not already. These projects are not "a little research, invested time, and a little ingenuity," they are the culmination of years of experience and passion, along with extreme respect for the technology.
Don't get me wrong, I want people to do this. However, acting like people can safely dive in to high-energy electronics/electrical projects head-first is stupid.
I really agree with you. The idea of a homebuilt powerwall sounds really cool. But it's not something I would come close to working with.
Awhile ago I did a some volunteer work at a radio station. One of my tasks was to check for arcing. That involved going into the transformer room, disabling the safety systems, unracking some of the electrical components, powering the system on, turning off the lights and then looking for flashes from electrical arcs. It also involved going out at night with night vision goggles to check for arcing on the antenna. If we saw anything we knew what needed to be fixed. Safety was important there, but I developed a healthy respect for electricity. After working there I'm a little freaked out about the idea of dealing with high voltage systems.
High voltage, sure, but outside of an inverter that you can/should buy prepackaged none of this needs to be high voltage. The system in the embedded video has 14 sets of cells in series, meaning 40-60 volts.
Danger of electrocution is not particularly high, but can easy start a serious fire with such batteries. The energy capacity with create a seriously hot arc if shorted, and there is a potential for thermal runaway of the battery cells.
I once saw two marine lead-acid batteries explode and catch fire in the boat next to mine when the owner had dropped something big enough to short circuit and not disintegrate itself. Instead a fat copper cable exploded and then probably both batteries boiled, exploded and then everything caught fire. It took about 3 seconds.
Luckily he avoided the acid spray, but he had to empty 4 large fire extinguishers from the nearby boats to put out the fire and cool down the remains of the batteries enough so they did not immediately catch fire again.
I have a huge respect for what a huge solar array and a big bank of batteries can do after seeing what a relatively puny battery could actually do.
A standard AC system can't deliver even a percentage of that current.
I had something similar happen to me in a Datsun 180B once.
Driving home in the rain, and the tail light wires were submerged in water that was pooling in the trunk and shorting out. As I put my foot on the brakes - on the freeway in the pouring rain - the fusible link in the engine bay decided to just catch on fire instead of being a fusible link. The battery quickly boiled and I had a sudden and catastrophic engine bay fire on a crowded road.
A car battery has enough amps to literally use as a welder (two jumper leads, an arc welding rod and a 550CCA car battery is enough to stick most things together in an emergency :P ), and is more than enough to create a massive fire.
Yesterday I was putzing around with these exact batteries (swapping some old tooling that had great build quality but were NiCad junk). Those 18650s pack a punch as current sources. I had an ammeter hooked up and we're talking easily in the 10s of amps for a brushed DC motor of your standard cordless Milwaukee from the 1990s.
I did the same to fix one of those $40 dollar Swiffers (i.e., we're not really talking about powering industrial steel chop-saws powering through I-beams). It drew ~2.5 amps at no-load, no problem.
OSHA[1]:
"17-90mA" : Death is possible.
"90mA+" : Death is likely.
And remember breakers blow only when the sub-circuit's current exceeds the threshold (generally ~15amps in the US) so you can spec a standard gauge of wiring sufficient to consistently draw that current, not overheat and burn your house down. It's (generally) not checking to see if the current sunk = current sourced. You need GFCIs("RCD"s in other locales) to actually get that functionality.
The rule of thumb I've heard is 50mA passing through your heart is where the v-fib dangers begin. The real risk begins at as low as 20mA, because thats when your hands lose their muscular control. You can't let go of the source, and your heart enters v-fib (~100 mA) and you get oxygen starvation and brain death[3]. You have tales of people using one hand to support themselves on a grounded breaker box, using the other hand to just have a look-see, and not even realizing they're in v-fib, feeling a little 'off', sitting down for a bit to catch their breath and found dead 20 minutes later next to the box.
Your body basically acts as a resistor in parallel with the circuit. Here's the ASME's resistance model of an adult males' body[2]. As you can see, you're body is definitely not a 10meg resistor. If you're working in damp, humid, or hot (i.e., you're perspiring through your hands, and that Na+ is just looking for a donor electron!), the risk increases. Whether you're working on a 460v3ph 15 horsepower Hardinge lathe or a Swiffer, the best piece of advice I ever got was "have a healthy respect (and fear) for your tools". Use isolation transformers, current limiters, and CAT-rated gear and read a sufficient amount of information to inform yourself in advance of the potential risks[4].
[3] Brain death I've always heard is ~5 minutes for an average adult male with a decent pulmonary system and respiration capacity-- less if you're a couch potato with less lung VO2 capacity, more if you're Lance Armstrong. Obviously not a physician.
[4] I was googling to pick find a diagram to depict the resistance model of a human and stumbled across Allaboutcircuit's safety text (chapter 3, for those who are wondering). It's pretty good for "general" electronics. Obviously if you're working with things like tubes ('valves'), transformers, and other subsets, you're going to want to read about the safety precautions you want to take for those specific areas as well.
How many watts are we dealing with here? What's the actual power output of a 1 second long direct short? I'm pretty sure if you shorted that 40-60 volts across a hotdog it would explode. Even if we disregard if that voltage can stop a heart (which it can, 50v is minimum), the amount of power you are dealing with is the amount of power that causes things to explode.
I wouldn't go anywhere near a diy power bank without proper knowledge and training.
With batteries (and even small battery packs) there are two main safety issues: (1) chemical combustion (due to mechanical damage or stress or electrical stress i.e. over-charging) (2) Burns and fires caused by arcing.
For (2) the exact configuration basically does not matter, because generally speaking all battery packs capable of storing a large amount of energy are also able to release a large amount of energy in short time. So if you cause a short you'll always get a nice arc. Note that most kinds of protections are way too slow to suppress arcing, so you can always burn yourself. Arc-fault detection is more difficult in DC systems as well.
(1) is mostly a matter of doing things right. This means active temperature monitoring and measures (cooling, venting, shut-off) for larger packs. Independent over-voltage and over-current monitoring. And of course the little things, like wiring things up the right way and placing sense wires at the proper locations. For example, this is a simple thing of doing things wrong:
Designing these kinds of circuits correctly requires a lot more experience and detail knowledge than one may think at first. (Besides miniaturization this is another reason why one-chip battery management chips are so popular in the industry).
——————
I did a few things with a few types of batteries (lithium ion and lead) and these little cells can cause an impressive amount of uh "disturbance" in short order. I admit I am almost as careful (in a different way, obviously) with them as when handling high-voltage things like the HV oscillator in an old 'scope.
What is being sensed here and why is it wrong? If voltage is being sensed and the busses are fat chunks of copper, this seems entirely proper. What am I missing?
EDIT: Wait, I think I know. You also need independent current measurement at each cell to detect internal shorts. And probably a way to automatically isolate such cells.
"Charger" would of course also be for discharge. "Sense" can either be charge sense, but that's not very common (due to CCCV charging), or monitoring. In either case, in this way of wiring things up the uppermost cell has a lower resistance connection to the charge/discharge port compared to the lowermost cell. So that uppermost cell will handle higher currents and overall will degrade more quickly. (This frequently applies when connecting "identical" parts in parallel in power electronics, e.g. capacitor banks).
If the voltage feedback (assuming there is any) is taken from the lower side, then a reduced voltage (by U=Rbus×Ibus) will be measured, which is lower than the voltage present at the cell terminals. Siblings mentioned that you want protection against internal cell shorts, using fuse wires or similar.
If each cell is connected to the bus bars with fuse wire (like in the Tesla battery pack for instance) this connection diagram seems valid. Same way it is connected in the car/powerwall packs that tesla makes at least.
Yeah i have read enough scary stories about people dealing with 18650 cells (effectively AA sized lithium batteries used in high powered flashlights etc), that trying to build a powerbank from used laptop batteries with enough capacity to handle domestic appliances is not something i would undertake.
I have a sneaky suspicion that their insurance isn't going to cover the fire this may cause. These are lithium batteries, with who knows what for quality control, being used by people with who knows what for education. This could be a tragedy in the making.
The idea that insurance won't cover fires from unpermitted, DIY stuff is mostly urban legend. Most homeowners' policies are what are called "all hazards" policies. They cover any kind of damage for any reason, with explicit exclusions like war, earthquake, flood. Outright arson / fraud by the policyholder would also obviously be excluded.
Banks would object to policies that excluded unpermitted work, since it's relatively common and this type of exclusion would leave them with lots of exposure.
So, they may decide to drop you like a hot potato afterward, but a fire from your homebrew powerwall would likely be covered. Read your policy fine print to be sure.
They're lithium-ion batteries, which contain very little elemental lithium and do not result in lithium fires (which are very difficult to put out). True lithium batteries (the kind that can cause lithium fires) are not rechargeable.
The batteries being talked about are 18650s, which is mentioned pretty clearly in the article and apparent from every single picture of batteries in it. This makes sense, because the vast majority of laptops sold across the world use a set of 18650s as their batteries. These are also the form factor of batteries being used in current Tesla powerwalls (though I believe they will also switch to the 21700 form factor as production of that ramps up).
18650s are just as safe as any other battery chemistry in everyday use, but they're a high energy lithium cell. In particular, if you do bad things to them, they may light on fire. It's not that hard to take appropriate precautions to avoid that, and reasonably sized lithium-ion battery fires can generally be put out with a standard househould fire extinguisher.
It also appears the community is collaborating to spread good information about powercell design to avoid/manage fires inside the units, in the unlikely event a cell starts a fire. If the container for the powercell is designed correctly, it can contain a complete unit burndown: https://electrek.co/2016/12/19/tesla-fire-powerpack-test-saf...
In short, you're overestimating the risks and underestimating the level of safety attainable with these products.
> In short, you're overestimating the risks and underestimating the level of safety attainable with these products.
No. When it comes to things that might produce fire inside your house, there is no such thing as overestimation. You cannot be over protective when it comes to fire safety.
"You cannot be over protective when it comes to fire safety."
Ridiculous. There are obvious examples of overestimation of risk, why not prohibit electricity in homes entirely for example? Should I get rid of my kitchen stove?
This kind of absolute rhetoric is banal, obviously absurd, and is not helping anyone.
Effective safety measures involve accurately quantifying and balancing risk.
> there is no such thing as overestimation. You cannot be over protective when it comes to fire safety.
Have you ensured your house is built exclusively of fire resistant materials, and avoided buying modern furniture with inflammable fabric covering and interior padding? Have you installed a Halon system in your home and all physically adjoining structures?
I'm not too familiar with those, but it looks like they just shut down the circuit. We were looking for arcing to pinpoint where the problem was. It was a short wave radio station, so it had a large antenna. Arcing in the antenna was probably caused by a broken wire. That could be fixed by climbing up and replacing it.
DIY projects always get glorified like this. As if people figured out this secret trick to bypass consumer capitalism. When the whole idea of capitalism is about not having to invest a ton of effort into learning a new craft every time we want to use a new product/service. We pay companies to master that craft for us with money we earn from focusing on our own individual skills.
It'd be more accurate to sell it as a rewarding experience to do as a hobby rather than a clever cost-saving commercial solution.
If money and ROI is the primary focus then most specialized knowledge could be better utilized in other ways than simply building a product once for yourself in your spare time.
That's the early on naive version of distribution of effort and mastery. The market evolves into a very average product quality and all kinds of tricks (focus on shallow quality, brand name versus longevity, repairability...) to improve benefits, which is the core signal in capitalistic feedback loops.
It's also good to do stuff on your own. I have to admit, I am too dogmatic about not buying new things. It's only good if you have time to invest it's true. But after a threshold it's of great value. I remember the fear of electronics and household appliances, it's now gone. Capitalism turned optimization into blindness.
One of the really disappointing things about Tesla is how incredibly proprietary all of their products are. For a company that has such lofty goals as "saving the world", it feels wrong to me for them to then sell DRM cars and unrepairable black boxes. Maybe I'm just too idealistic, but surely if we have an opportunity to create a new market for solar devices why would we decide to recreate the proprietary world of the 1980s? Creating your own powerwall from recycled batteries is a pretty neat hack, and I'd hope that people are publishing free (as in freedom) hardware designs for said battery banks.
My annoyance with Tesla is somewhat related. I don't like that I can't simply buy a Powerwall or solar roof tiles. I would have to hire one of their approved installers, which increases the cost remarkably. I don't know the markup, because you can't get flat materials pricing data from Tesla, or couldn't last time I checked, but I know labor for solar installation and electrical work is quite high.
I understand why they do it, but it means that for the segment of the population that enjoys that kind of thing, and knows how (I've done my own solar installations, and helped others, on multiple RVs) I have to use much less awesome components. I wanna play with the tech and save a few thousand dollars, but that's not an option for Tesla solar and Powerwall gear. I also want to use it in ways Tesla won't allow (Powerwall can't be installed in RVs or manufactured housing like a tiny house on wheels, for example, even if one were to design the house around safely doing so).
I can go to Home Depot or Lowe's, right now, and walk out with a breaker box, conduit, a bag of breakers, a couple hundred pounds of copper, and install my own household 220V/150A electric service. Local codes or the electric company may require an electrician to sign off on my work before I grid tie, but the companies who make the stuff aren't going to refuse to sell to me. Nobody seems to think me killing myself by making mistakes with that would be the fault of Home Depot or the manufacturers, and I'm reasonably confident that is more dangerous than installing a battery box (unless Tesla has built an inherently dangerous device, in which case it won't matter who installs it).
I can also go to the same store and walk out with a roof-sized pile of roofing tiles. No one will stop me from putting on my own roof.
Hell, Tesla sells cars. The single most deadly consumer product, by far, in the world. Admittedly, people have tried to sue them about their autonomous features, but still, cars are really deadly and no one blames manufacturers for those deaths (though, we do, as a nation, impose some safety standards...but, it doesn't stop tens of thousands of deaths every year). Tesla is arguably much better about not killing their customers than some other manufacturers, but they already operate in the most deadly industry (aside from, say, coal or fossil fuels, but the deaths caused by those industries are generally an amorphous statistical blob of health problems rather than distinct deadly events).
My point is: Unless Tesla has built a dangerous product, there's no safety-related reason it should be treated differently than other home improvement items that require the same types of skills to safely install them. Electrical and solar power are well-understood by plenty of non-professionals. I do see the appeal of making it turn-key; they're aiming for average home owners, particularly at the time of construction or roof replacement, when there are already contractors involved and a mortgage to pay for everything. A large-scale solar power installation is expensive, and it makes sense for most people to work it into their mortgage.
But, it's frustrating for me; when I build a house (and I'm kinda shopping for land now in a casual way), I plan to self-fund it and do a lot of the work myself because I enjoy that sort of thing. There won't be a mortgage, and I won't be paying for contractors to do things I can do myself. Electrical work is well within my comfort zone. So, it's frustrating that when I install solar, it'll have to be the somewhat less attractive panels rather than nice-looking tiles.
It's certainly Tesla's right to sell their product however they like. But, I reckon I can complain about it, too.
Humans are bad at separating novel risks from familiar likely ones. News is what's novel - not what's likely, so we have this horrible situation of putting our attention and efforts on the wrong thing.
Everything you said is 100% correct, but just look at all the extra news and scrutiny that's given to every Tesla crash. Or the people warning of the dangers of fidget spinners:
Unless you're an EE and an expert in power supplies, this is way more dangerous than splicing some copper wires and following some building codes.
Which ppl manage to f. up all the time anyway (missing ground, neutral and live swapped, splices used in lieu of a longer length of cable, improper gauge, poor cabling technique)
This community takes it as a given that "email is too hard" for an amateur to do right. Then how is wiring a house considered easy when the electrical code books are thick as bricks?
"This community takes it as a given that "email is too hard" for an amateur to do right."
I don't.
I work on software (http://www.virtualmin.com) that helps over 100,000 people manage mail servers (and web servers, and databases, etc.) for themselves. So, I disagree with the premise that email is too hard.
"Unless you're an EE and an expert in power supplies, this is way more dangerous than splicing some copper wires and following some building codes."
Is it? I don't understand how. It is a battery bank. You hook up your charge controller (two wires) and your inverter (again, two wires). Same as you would for any battery bank...it's just a better battery bank, and battery installation aint rocket surgery. (Here's a diagram: http://i.imgur.com/jXuag5l.jpg )
If one can understand a solar panel installation in the general case, it seems pretty clear that one can understand Powerwall installation. In fact, building your own battery bank is more complicated than the Powerwall, since you have to wire up all of the batteries, and figure out how to protect them from over-charging and over-discharging. Tesla provides the smarts and bundles up all of the batteries for you.
Some people like to be self-reliant and understand how things work.
Again, I can't stop Tesla from selling their products this way. It's certainly their right, and I'm still pleased with Tesla's existence (and I have a standing buy order for TSLA if it ever drops low enough, as I regret selling the TSLA I was holding a few years ago). I believe in Tesla. I just wish I could buy Tesla products and use them the way I want to.
If you know what you're doing go ahead. I certainly won't stop a hacker from hacking. Don't expect your insurance company to share my general liberalism towards DIY.
If you don't know what you're doing, go right ahead anyways. It's not my house and my family in there.
But this is hard.
Did you know that DC can't use AC switches rated for the same voltage?
How about cooling the packs? whats the appropriate distance between the packs as the geometry of the heat dissipation changes from 1D to 2D to 3D?
Are you going to cool with forced air or passively? If forced, what if there's a power outage during battery bank failure (highly correlated know that you wired the bank to the mains). If passive, how do you ensure that the environment is always providing suitable cooling?
That's a bit of domain knowledge that most ppl who know how to wire a socket don't know. That's also a house fire type mistake when your dealing with a battery bank.
Difference between messing with mains wiring and messing with high capacity batteries and/or solar panels is that mains always has some kind of fuse somewhere on the power company side and thus you will not see circuit that is completely unfused and will supply high currents until something blows up spectacularly and dangerously. This holds for significantly smaller things than DIY powerwall, eg. car batteries and even batteries for desktop UPSes.
Edit: for instance accidently shorting net 30 to ground when working on car electrics might well end in the car being completely destroyed by ensuing fire (I have friend who totalled his car in this way, he is licensed electrician). On new cars the fuse between 30 and 30a is usually placed directly on the battery to prevent exactly this issue, but even 30a will supply enough current to wreak havoc.
I can't imagine Tesla would ship that battery bank without fuse protection inside (and I imagine it is a "no user serviceable parts inside" kind of design...it looks like a black box device). Every solar deployment design has fuses on either side of the batteries.
One neat engineering trick in tesla battery packs is that the last resort fusing is part of the battery pack construction itself (the spot-welded connection to the actual cell is intentionaly surprisingly weak to act as a last resort fuse)
On the other hand, the difference is in that for the mains there is always going to be some kind of fusing, that is not part of what the tinkerer owns and probably does not even know where it is located. For solar and huge battery packs you own the whole thing and can mess with anything in the system, even more so when you build the thing from individual cells instead of buying COTS solution with bunch of "no user servicable parts inside" labels.
The other car manufacturers seem to cope, it's not particularly obvious why Tesla should be unique in that respect and why they need to lock everything down to such an extent.
I've heard way too many Volt-on-fire related jokes to agree, unfortunately. Tesla had a very aggressive response to the Tesla accident a few years back, and still suffered as a result of it.
Big retailers like Lowes cultivate a reputation of catering to construction workers while also being approachable to weekend warriors, which insulates them from the downside of product safety issues, somewhat.
Tesla supported Kevin Mahaffey and Marc Rogers when they were hacking at their friends model S and trying to get root access. The model 3 repair support guide just came out and has diagrams of the battery pack down to the cells. Yes, they dont put the design of the car on github. They are hardly "incredibly proprietary" though.
Yeah but if you hack your Tesla even to change the color theme of the display, you'll probably loose warranty, loose certification status or whatever it is called (so you can't have any new parts installed unless it is expensively recertified), you'll get a call from Tesla asking what the hell you are doing, and so on.
On the other hand, I can replace almost every part in my conventional car without ill effects from the manufacturer.
Not for long. You can still do this with older cars, but newer ones really restrict you. I just bought a new car and am not even allowed to jump start someone else's battery without risk of voiding my warranty.
I don't know what car you've got but I have a brand new Mercedes-AMG and I can do a surprising number of things with it without voiding the warranty. It's certainly as easy to work on it as on older cars, I don't need to go to the dealership to replace the brakes, the battery, the bulbs, filters, engine oil, transmission and diff oil. Manual explains how to jump start the car but certainly doesn't prohibit it. I've heard some horror stories with newer cars (especially French and American) but it's not like all new cars are horrible to work on.
In large part thanks to the CAN bus and how jump starting is likely to cause enough spiking to fry the boards.
Was trying to help a relative with some issue with their car the other day, and the wiring diagram for the part and related controls almost seemed to run in reverse depending on how the switch was set.
doing work on your car always voided your warranty, with good reason. That's not a realistic thing to ask. It's realistic to ask for the UI and apps to be open-source (say, something like android), but even that would be an incredibly open thing for a non-software company to do. I still think it's very unfair to call them extremely closed.
However, I really do wish they'd adopt a more android-like ecosystem.
Unrepairable black boxes? This isn't really true of Tesla.
Sure, it would be nice if they could make service manuals and service parts more freely available, but the same could be said of all car manufacturers.
There are guys on YouTube who have re-built entire Teslas from near-scratch using salvaged parts from multiple vehicles. Guys with no formal training and no previous experience with EVs. Tesla does not do anything to make their cars hard to repair.
Going the salvage titled Model S route is, from the accounts I've seen (likely featuring the same videos from YouTube), is thoroughly unsupported by Tesla, to the point that Tesla will refuse to sell parts to people if Tesla knows they are trying to rebuild vehicles that have a salvage title.
There's making parts "more freely available" and then there's "you can't have this from us, for any price". Fortunately the ludicrous amount of power in the Model S and people's inability to control the car (or themselves) means there's a good sized underground market of salvaged Model S' bodies and parts, even if there's no office service manual available to amateurs.
Meanwhile, Chevrolet will just sell you the engine from a corvette in a crate. (Okay, it's ~$12k, but they're willing to do it.)
I say this as someone with a Model 3 reservation, mind you. Tesla wants to build the iPhones of the electric car industry - no user-serviceable components inside, and the battery isn't removable by us mere mortals who don't work at Tesla. (Abandoned battery replacement tech demos from a few years ago aside.)
From what I've heard the Model 3 will be a bit different.
But certainly on the Model S, the battery is not that difficult to remove/replace - again, there's plenty of YouTubers who have done this.
You can even, in many cases, swap a higher capacity pack into older/lower-spec models if you're so inclined. There is no DRM or software checks that prevent this from working!
You're probably thinking of activating for supercharger access. Tesla may not let vehicles connect to their superchargers if they've been reported as written off. It also won't be able to connect to Tesla's servers for OTA software updates, etc.
If you can prove that the vehicle has been appropriately repaired (by a qualified body shop, since it's a safety issue in the case of structural damage), it can be re-activated.
But Tesla does not remotely disable vehicles or anything like that. Even if it's been written off, theres nothing to prevent it being driven just like any normal car once repaired.
This isn't unique to Telsa, all vehicles are becoming more difficult to repair as computers and software replace what used to be handled by mechanical systems. I would actually much prefer to drive next to a Tesla knowing that the guy behind the wheel (potentially staring into space as it drives itself) hasn't implemented any "hacks".
A similar argument exists for why DIY battery banks are not ideal. It's not inconceivable that some dedicated hobbyist engineer pulls it off in a safe way. But how would you feel if you were over in your neighbors garage and noticed he had haphazardly wired 600 18650 cells together based on a design he found on the internet?
Security precludes freedom. The trend from just about anyone being able to build their own home a half century ago to not being able to even make your own battery pack just shows the trend away from the freedom to experiment.
The right to repair died once magic computers came into the world. Before that trying to tell anyone their mechanical / electrical systems were undocumented proprietary and warranty breaking if you tried to repair them would have caused a riot. But since computers were sufficiently magical that a critical enough mass of people wrote off trying to understand them, the industry and all resulting industries undermined the right to actually own anything related to them.
This stuff isn't black magic, it's science and engineering. As more hackers participate in this space, the open source components and techniques will get better and better (I'm not familiar with open source battery management systems but I would bet there is already some good stuff out there). I think there are a lot bigger dangers to hand-wring over than nerds needing out over this stuff.
I think there are as many or more car hackers than ever - CAN bus opened car hacking from the world of the gear head to the world of the computer geek.
> As more hackers participate in this space, the open source components and techniques will get better and better
I so strongly disagree with this assumption. We got a nascent not even close to open source scene on x86 and a few hobbyist tweak boards. They still use proprietary hardware, the firmware is still almost always proprietary, the drivers are often proprietary, and for general users 99% of their OSes are proprietary.
And the consequences of that world that doesn't care about right to repair or software freedoms or having control of computers impacts me. I can't get an x86 pc now without a hardware backdoor, I can't use 802.11an wifi without proprietary code, I can't display visuals to a screen without proprietary code. The screen itself is running a ton of proprietary code. My hard drive has a computer in it and thats wholly proprietary.
And that was in an ecosystem where moddability was handed to us on a silver platter with ACPI which only existed for IBM and Microsofts sake, not for anyone elses. It was not a charity. That is why we still have no mobile platforms that use a standard hardware abstraction layer that you can run a generic OS on.
The Internet of Things never developed an open source ecosystem. SmartTVs never developed an open source ecosystem. Set top boxes and consoles never devleoped an open source ecosystem. Cars almost certainly will never develop an even remotely functional open source ecosystem - companies will use open source code, because someone else did the work for them and they can save money. They won't contribute back, they won't respect their users, and they won't respect the developers that put in thousands of hours of free labor by sponsoring them. That isn't open source winning, that is corporate profits winning.
Look at John Deere. That is where we are going. This is not going to be a simple matter of find the ethernet jack, telnet into a shell, and start running code. This is signed payloads, read only rom, and no way to access the firmware.
If there is an open source car ecosystem, it will be like the open source phone OS world. A joke, that cannot practically run on anything, that at best is ripping half of the Android equivalent out of itself to even run. And I guarantee Ford et al will never be as philanthropic as Google was in open sourcing Android to release their car OSes middleware like that.
About the same as I feel about 95% of software running on the internet being copied from StackOverflow. Rather Meh. I would worry if the engineers at Boeing did it, but a bunch of crappy javascript apps running on the web? No.
As for the neighbor, I don't live on a zero lot line, so let him burn his house down all he wants.
>About the same as I feel about 95% of software running on the internet being copied from StackOverflow. Rather Meh. I would worry if the engineers at Boeing did it, but a bunch of crappy javascript apps running on the web? No
This is exactly the type of attitude that leads to the proliferation of vulnerabilities on the web.
"It's not like we're using it to fly planes, so who cares if we're not checking every SQL statement for injection vulnerability"
Then the code gets used on some ecommerce site, the site inevitably gets hacked, and customers' PII gets leaked.
This opens Tesla up to an immense amount of liability with no benefit.
If you want to build your own open source battery storage system, battery management system firmware, and distribute it online (think Backblaze), I'd encourage you to do so. But Tesla has enough to worry about as it is.
Tesla‘s are literally killer machines. Accessing the high voltage cables is not difficult (they are only covered by a plastic cover which is easily removable). Those power lines can easily kill you. Even if you are very careful, even a small mistake with the wiring can have fatal consequences. There is a reason why you are not allowed to do wiring and a lot of electronics in your house by yourself.
I certainly hope that Tesla will provide manuals to highly trained electricians at some point, but if Jone Doe tries to fix his electric vehicle by himself and makes a fatal mistake (or maybe even starts a fire), that will only make national news and scare more people into thinking electric vehicles are inherently more dangerous then traditional cars (which they are not).
I can understand why Tesla does not want to take that risk.
So are all cars. Under-skilled people routinely (and legally) repair and maintain the systems that make these vehicles drivable (brakes, steering, suspension, etc).
Some folks are killed in doing so, much like any activity, but not any number that is out of proportion and typical with any activity where small mistakes can cost lives.
>There is a reason why you are not allowed to do wiring and a lot of electronics in your house by yourself.
That's not true in many places, much of the United States included. Only inspection is mandatory in most cases, barring specific regional laws. Again : Those that do their own work aren't just dropping like flies. A few accidents occur, but nothing of any proportion to require legislating more strict laws regarding such things.
>I can understand why Tesla does not want to take that risk.
My understanding of why Tesla acts the way they do has nothing to do with the complexity in their systems and danger in working around such complexity, but rather their interest in more protection for their IP, their desire for keeping customers within their own 'walled garden' of service options, and for controlling public perception of their product, all to the disservice of the customer (much like John Deere's behavior these past decade or two).
We aren't strangers to 'killer machines'. This isn't any new phenomenon. It's being framed that way by some to rationalize over-protective and controlling corporate behavior towards the consumer, and Tesla isn't alone in exercising such power.
> There is a reason why you are not allowed to do wiring and a lot of electronics in your house by yourself.
I've found that a few people here really don't understand some industries; yet are adamant about their opinions on said industries.
A few weeks back I was just in an argument with someone (may have been another account, though) about high-voltage lines in the US used for dryers, and how "magically converting a 120V line" (paraphrasing) to 240V was just super dangerous and a death sentence. Point is, arguer had no knowledge of two phased 120V lines in US electricity and combining the two was sure to burn a house down. But they sure did have an argument.
FWIW, this form is so much more argumentative than others online about other topics not understood. I think that's part of the IT/techy/hacker culture.
I can tell by your post that you never did any sort of meaningful repair on a car.
Yet, for some reason, you seem super concerned about what other people might do.
Is there a term for this kind of bizzare nanny behaviour?
Why pretend that you know what you are talking about and then take a step further and not only suggest ridiculous 'solutions' but push for regulations.
If they're from laptops the cells should be safe (as opposed to unknown new cells from AliExpress which may or may not be to spec); albeit not necessarily at maximum capacity. As long as appropriate protection circuitry is used, I don't see how using then hard would make any difference (laptops use batteries hard too).
opposed to unknown new cells from AliExpress which may or may not be to spec
Many of those are actually recycled old cells, likely from laptop batteries, and could still have a lot of capacity left --- often, what happens is that one of the cells in the pack fails/goes out of balance and the protection circuit disables the whole thing, meaning the rest of them are still very much usable.
Most of the videos on the subject I've seen have a simple idea: depending on the battery, one cell could turn a whole pack off for safety. Pulling them apart, and rewiring them to a new (and probably better) Battery Management Chip, could make the N-1 pack working. They also often charge them a few times individually to assess their stability.
[1] one bad cell could force too much current elsewhere or overheat IIUC
Kind of dangerous seems like quite the understatement, considering the nature of lithium batteries!
It'd be fun to play with something like this for the learning, but I wouldn't actually run one unless it was in its own shed, many dozens of yards from the nearest anything else. Lithium fires are no joke.
I've read claims (theoretical and practical) from both sides, so apparently the safety of lithium cells can vary widely based on manufacture quality and maintenance.
But also, on the video, they shows each pack of cells he uses has individual over and underchange and temperature sensors, so the system is able to tell if any individual section is going bad. It's a pretty neat, and external installation, so I don't think it is very harmful on a "massive heat incident".
BTW I've seen lithium cell safety bags, that can hold the flame enough. I wonder how a case could be done in a similar way to at least contain the heat and gas.
Also, could there be a thermoelectric powered fan to slow down the heat rise when cell(s) starts to fail or burn ?
You absolutely can design a firewall system able to fully contain a runaway pack. They have small systems that look like briefcases to nab runaway laptops on planes, for example. But when a cells goes, it goes. No fan is going to remove anywhere near enough heat. And you wouldn't want it to spread the vapors, as HF is a genuinely dangerous byproduct. The only real solution is full containment. And being 100% confident in your containment solution would require a lot of destructive testing- it isn't a trivial problem.
Ok a fan was ridiculous, but I didn't mean to diffuse gaseous products; just tap into the heat to activate some endothermic reaction around the system to, well, delay or dampen the process.
Can this thing be put in a fire-resistant compartment outside house? Even a couple layers of drywall would probably be enough to isolate the fire. A design with inner fireproof walls would prevent one cell from lighting up most of the pack. All this is neither hard nor expensive, and would limit the consequences of a fire severely.
No way I'd do this inside my house, either, but if you watch the video, the Australian fellow had his in two little aluminium sheds out back, set on concrete. Looks like, most likely, a catastrophic event would trip the wires and (at worst) burn down the sheds and couple computers/Arduino boards.
Now, insurance folk will always fight a claim, but you might be able to get past... maybe =/
The hyperbolic response isn't warranted, as these are nearly identical lithium cells of known origin. Were they all different random cells, or if disassembly were required, then one might catch fire or explode like a small firecracker.
This isn't much more dangerous than putting batteries in a toy car, or putting gasoline into a hybrid.
I'd rather use old cells than shady new cells. New cells are the dangerous ones- manufacturing defects can make it through QA, but everything gets found after a few years of use.
I'm not sure how relevant it is as an appeal to authority measure, but in the US there is a Professional Engineer license. It's mostly relevant when you have to sign off on designs for regulatory bodies.
>the only person (in most states) who can legally call himself an Engineer
Someone must have a PE to sign off on certain things when dealing with regulators and so forth. Which makes getting a PE in certain situations important. But not having a PE is hardly equivalent to calling yourself a lawyer if you haven't passed the bar in a state.
That’s not true for any of the United States. You can’t call yourself a Professional Engineer without being licensed, but anyone can call themselves an engineer. Train engineer, growth engineer, happiness engineer, and software engineer are all perfectly valid titles without a P.E. license.
The funny thing about that is, at least when I worked in the oil business, I'm not sure I've seen another industry that threw around the "engineer" with more abandon for technicians of all stripes.
I'd note that's a case where a business is representing themselves as an engineering firm and, and in some situations, it would be reasonable to expect that meant they employed licensed engineers. (The specific case is pretty silly though so is calling themselves engineers.) But individuals certainly call themselves and are called engineers all the time who aren't licensed in Texas and everyplace else.
Professional Engineer. There's an Engineer in Training certification first, typically gained by passing a Fundamentals of Engineering exam in college. Then you work for a few years under a PE and take an exam to become a PE yourself.
Luckily, home owners' insurance covers stupidity as well as acts of God. So as long as there is no evidence of intentional arson, then insurance would have to pay.
In the case of electrical improvements (like hooking your hacked battery up to mains power) your insurance company just needs to do a little digging. They just need to find out if there were improvements without permits or if those improvements weren't made to code.
If it's not in the policy "we don't cover homemade lithium projects", your covered. Very few things a home policy won't cover like nuclear war, terrorist acts for example. If your operating as a home business building these then that could be an issue, otherwise you can be an idiot and build whatever you want and your covered.
That sounds really unusual I feel, the insurance company should be within their right to deny a claim if the home was destroyed due to gross negligence, like someone deciding to make a pool on their roof without asking an architect, or throwing a raging party with 700 people and the floor collapsing.
Yes, this is extremely dangerous. I'm sure some of the people mentioned have the expertise to pull it off, but for most this is like building a massive bomb in your garage. There's a reason that 18650s aren't generally sold directly to consumers, in most applications they need dedicated circuitry to be used safely. And that's for brand new cells, it's another story entirely to be using random collections of "old" cells, at various points in their life cycles, different rated capacities, and subtly different chemistry. No thanks!
Yeah, I'm aware you can buy them. I have 4 cells sitting in front of me right now. But there is a big difference between someone specifically buying them for use in a high-discharge device like a flashlight or vaporizer versus picking up a pack at Walmart next to the AA/AAA. Because of the high discharge you can hurt yourself or start a fire much more easily. Just look at the stories of people with burns on their thighs from throwing them into a pocket with keys/change.
But there is a big difference between someone specifically buying them for use in a high-discharge device like a flashlight or vaporizer versus picking up a pack at Walmart next to the AA/AAA
You can find 18650s for sale on walmart.com . I'm not sure if the physical stores carry them, because there aren't any where I live, but I know that in Shenzhen/HK and other parts of Asia you can easily find 18650s for sale amongst other batteries in physical stores.
Because of the high discharge you can hurt yourself or start a fire much more easily. Just look at the stories of people with burns on their thighs from throwing them into a pocket with keys/change.
NiCd/NiMH can also source huge currents if shorted.
Flashlight geek here. Several people on flashlight forums have reported finding the Westinghouse branded 18650s in physical Walmart stores. They have a low capacity - 2000 mAh, compared to 3000-3500 for the latest cells from Sony/LG/Samsung/Sanyo/Panasonic. They also use a cell of unknown manufacture, so I can't be sure of the quality.
Specialty battery stores, higher-end outdoor gear stores and vape shops have 18650s, but it's best to get them online from dealers that know what they're doing. In the US, good options include Illumn, MTNelectronics and Li-ion Wholesale. These places have the gear to test batteries to make sure their suppliers didn't send them fakes, and usually have very competitive prices.
Can I ask a flashlight question? :) How do I tell how many lumens a flashlight really is? I go on Amazon and I see XyzFire flashlights that claim to give out many more lumens than would make sense, and many times they're inconsistent on the page itself. I don't know how to really tell. (If it matters, I'm trying to find a flashlight that fits in my pocket, is < $20, is ideally zoomable, and is as bright as possible. Best one I've found so far is this: https://www.amazon.com/gp/product/B011B1U2QS)
You need to find a review written by an enthusiast with measuring, or at least estimating equipment. Assume all the *fire brands are outright lying. Cheap stuff claiming more than about 1500 lumens per 18650 is usually lying. There is, however a pocketable, $40 flashlight that makes an honest 4000+ lumens on a single 18650: the Emisar D4. There's a catch though: it's dissipating about 80 watts of power to do that and thermal throttling kicks in after a few seconds. To give it more HN cred, it has an open source firmware.
As to your request, a couple general recommendations first: skip the zoom, and don't chase output numbers. Zoom is pretty optically inefficient, and small zoomable flashlights don't end up with much more throw distance than similar size fixed reflector flashlights. They don't do wide-angle illumination as well either. A good reflector with sufficient output provides both at once. As for brightness, human perception of brightness is roughly logarithmic (with a base between 2 and 3), so it takes four times the output to look twice as bright. Anything over about 500 in a pocketable flashlight is pretty decent, and most 18650 lights have more.
Specific recommendation: the Wowtac/Atactical (both names show up on Amazon sometimes) A2. This is one of the smallest 18650 lights, has shortcuts to low, high and last-used modes from off and has a right-angle configuration that works equally well handheld of as a headlamp. A pocket clip, headband and 18650 battery with built-in USB charging port are included. It costs $20, and I have no idea how they make a profit given all the stuff it comes with. Go with the neutral white tint if you can find it; this looks more like sunlight and won't wash out colors as much as the cool white.
A few of the budget brands like Convoy, Eagle Eye, and Maxtoch have official Aliexpress stores. I'm not sure about linking Aliexpress here, as the filter might think I sold my account to a spambot, but their store IDs are:
This is Hacker News. Open source is generally appreciated here even when there isn't a direct, concrete benefit to most end users.
But if you're here, there's a good chance you're the sort of person who could get an AVR programmer and tweak something if you don't like it. Different modes? Timed stepdowns to prevent overheating or not, or more sophisticated thermal regulation for the attiny25 based drivers that have a temperature sensor? A firmware with completely different behavior like near-continuous ramping instead of fixed modes?
The A2S is not worth a 50% price increase for most users, but if you really want more output (in the form of a bigger hotspot, probably not more intensity or throw distance), it's only $10 extra.
Theres no way to detonate a pack like this. 18650s have a pressure release, so the cstch fire but dont explode. In the very unlikely event that fails, a cell could burst, but it cant cause other cells to burst unless the entire pack failed to vent under extreme pressure.
> Sure, there's a fire risk if you have faulty or unprotected cells.
Or if an unrelated fire occurs in the garage/basement where it's located...
This is certainly enough stored energy to seriously ruin a firefighter's day if they didn't know it was there and how to properly deal with (or, more properly, to not bother dealing with it until it burns out).
That, to my untrained eye, looks like enough batteries to take down nearby structures if they went up, and packaged in such a way that a runaway in one cell that caused it to vent would propagate to other cells nearby.
Granted they're pretty reasonably protected by 2-3 layers of electronic monitoring and on-cell fuses, but lord, a small fire would burn the neighborhood.
I think all that it would take to render them marginally safer is a firewall between the different packs made of some good steel or refractory material.
Semantics. Is a battery undergoing thermal runaway/venting "exploding"? Or merely "ejecting scorching hot gas and catching fire?" Maybe it's not blowing up like an IED, but either way I don't want it happening in my garage.
The difference between "can be ignited" and "massive bomb" is not semantics. Even new pouch cells are way safer than they used to be. Unregulated 18650 cells are quite safe if the correct circuitry is used. The only dependable way to ignite them is to smash completely open them at full charge.
Cool, you caught me being hyperbolic. They are not literally a massive bomb.
> Even new pouch cells are way safer than they used to be.
Right, because these packs filled with recycled cells are going to have the latest protections.
> Unregulated 18650 cells are quite safe if the correct circuitry is used.
Which is my whole point, I'm not sure I trust the guy down the street to correctly wire and have adequate circuit protection for his DIY battery pack in his garage.
> The only dependable way to ignite them is to smash completely open them at full charge.
Lithium ion batteries are able to experience thermal runaway caused by internal shorts. This can be mitigated by quality control and proper handling (a lesson Samsung recently learned), and also other chemistries (like LiFePo, which is much more stable).
> There's a reason that 18650s aren't generally sold directly to consumers
They are, but IME the sourcing on some of those (especially through Amazon.com, which comingles items from different sources) is hinky. That 18650s can be dangerous if misused or if counterfeit isn't an argument for restricting availability, it's an argument for making them available directly to consumers from legitimate authentic sources.
I take old laptop batteries apart and use the cells for USB power banks. You can buy cases off of eBay for a couple bucks that hold two cells and have USB connectors and a charge controller built in. It's no Powerwall, but they're still very useful. I power my bike light with them.
DIY Powerwalls aside, the idea of procuring cells across various "dead" batteries is amazing for the environment. A lot of times, people just throw out the entire battery even though only 1 or 2 cell is dead which shuts off the entire thing. But this is literally the 2nd-R in "Reduce, Reuse, Recycle", which comes before recycle. I know there's a notion that recycle makes everything so much cleaner but there's a reason why its the 3rd-R in the motto.
Sadly "Reuse" has found no reflection in the laws and practices in most industrial countries. For example, giving replaced lab equipment to hobbyists is a perfect way of reuse, however, very few companies do that. Most rather send everything to the crusher or exchange it with the manufacturer for a discount who will do the same.
Question.. All my device lithium batteries seem to lose ~70% of their ability to hold charge after 500-1000 cycles, however the tesla car batteries/powerwall claim to give a decade+ of use despite being built of the same cells. How does that work?
If you never discharged below ~30% or charged above ~80% of their true capacity, and made sure they always stayed within a comparatively narrow thermal envelope, your batteries would achieve the same lifespan.
The last quarter-charge in either direction is where ~80% of the lifecycle wear comes in, and temperature excursions account for much of the rest.
From what I gathered, lack of internal voltage allow for different chemical reactions and structural changes that either create resistance, or actual shorts.
There is more battery in the box than what they advertise/offer up for you to use.. I.e. there are spares built in. Also they have spent a lot of time/energy/hardware making the environment the batteries live in, as close to ideal as possible (i.e. thermal cooling, etc).
Are they built of the same cells though? There's not just one single type of lithium battery, there are multiple different lithium battery chemistries, with different strengths and weaknesses.
I believe most of the physical damage done to a battery occurs either near 0 or 100% charge, and assume that automotive batteries are less frequently fully depleted.
I don't think the trouble of scavenging, testing and grouping the laptop batteries is worth it. The cheapest quality cells I've found costs around 210 euro for 1 kwh. To reach the same capacity you would need ~17 healthy laptop batteries @ 60wh.
With all the added work, buying just one type of cell seems way better. But really, why not buy deep cycle lead acids and an auto-watering system? No hype!
One thing that often strikes me is the use of 18650 cells in an unpractical, though technically and economically correct way. Even low power laptops today employ series of multiple 18650 cells in parallel, so why not make an effort to bring down the price of bigger cells such as the 26650 or 32650? today the 32650 costs more than the sum of the 18650 cells it replaces, which clearly explains the wide use of 18650s, but using that one would also reduce the number of cells by 2 or 3x, therefore saving some good money in both time and parts (wires, fuses/protections, holders etc). I'm talking about generic lithium cell production and marketing, not this project which is truly great even just for avoiding the burial of used cells in some 3rd world landfill.
As an ecig user / moderate enthusiast, this horrifies me.
18650s are great! It's awesome to know you can easily and cheaply fix an old laptop battery to better than it's ever been by replacing the cells with some nice new VTC5s or 25Rs.
Hooking up tons of them with varying capacities and levels of resistance is surely not wise.
Ecig user knows using even an un-married pair of 18650s in a dual 18650 mod is a recipe for blowing up in your face sooner or later.
Idk, maybe there's something with such a large number that makes it less likely to fail catastrophically?
In a use case like this you're sticking a hundred cells in parallel and [dis]charging them at a fraction of C. This makes the variances in capacity and resistance nearly meaningless. All the cells in a block will safely track the same voltage, and you can easily put a protection circuit on each block to keep that voltage in spec.
I'm an Amateur Inventor. I used to work in Insurance. You might be able to guess where this went: Ctrl+F "insurance" - no results.
Okay, fine, take some risks you innovators you! Just don't expect me to play a sad song if the experiment burns down your house, your neighbor's house, and results in criminal charges. Really, I'm not being glib. I will continue not being glib if what I foresee comes to pass, with potential serious damage to life and/or property. GL, HF.
There is an entire section in the article about how most people's first reaction is "don't burn the house down" and how the community worked together to create safety systems. The community is aware and working on issues like that
From what I've seen (and I haven't really been following this), they're rather using old electric car batteries to do that. Electric cars have been around long enough for there to be old/broken ones available now and most of the cells in their banks are still good.
I wonder, they don't seem to put protection / monitoring circuitry in on the per-cell level (which would be prohibitively expensive). How do you know when a cell or a block of cells go bad? Or does it not matter?
I think it could be bad if the resistance of one increases and you have a bunch of them in series...
Isn't it a pain in the ass to find/test all those cells. I think I saw either an article or a instructables page that had like 100 cells or something. Maybe it was a thousand. Looked tedious.
These are not "consumers who can't afford commercially produced powerwalls." These are people who could probably work professionally in the field if they're not already. These projects are not "a little research, invested time, and a little ingenuity," they are the culmination of years of experience and passion, along with extreme respect for the technology.
Don't get me wrong, I want people to do this. However, acting like people can safely dive in to high-energy electronics/electrical projects head-first is stupid.