The people that are saying, "a battery's a battery, I'll just put it in a shed" are dabblers. They don't understand the technology well enough to do it safely, so they just build it with the assumption that it's going to go FOOF someday.
The people that are saying, "a battery's a battery, I'll just put it in a shed" are dabblers. They don't understand the technology well enough to do it safely, so they just build it with the assumption that it's going to go FOOF someday.
Accommodating for failure modes is not lack of understanding... most chemical batteries have some inherent combustive or explosive danger, lead acid can produce hydrogen explosions, LiPo can slow burn when the layers are punctured through expansion etc.
Failure in a car is one thing (make sure it doesn't happen violently so that it can be controlled and escape is possible), but in a house you just can't afford for it to happen at all, sticking it in a separate building is a very reasonable and simple precaution for such a large quantity of chemical batteries on one place.
It's bad enough when small laptop batteries burn up, and those have to pass safety test... imagine a few crates... in your house... under high load from having to power all of your appliances.
You know, hacking with a depth of understanding. I think you are responding to a person like this.
Talk to any telecoms vet and they will have stories about accidents with DC power hopefully funny ones not tragic ones
But the implication that you need to use high voltages isn't true. You can easily feed 24/36/48 volts into an inverter. The high current potential only acts weirdly when you short it out, which is where a couple fuses can prevent issues.
One of they ones I heard was about a painter who when painting the walls of a small exchange put the paint can across the bus bars which shorted and exploded the can of paint.
There are certain things that will make sense to someone with a pretty extensive knowledge of the subject at hand but are still in the category where the factors at play should make one pause and consult a professional. One example would be people who hook up generators to their 220V dryer connection in order to power their house in an emergency (it's called backfeeding). Technically, that should work but if you don't think to throw the main you'll wind up injuring or killing the lineman expecting a deenergized line when they come to restore power to your neighborhood.
http://blogs.sciencemag.org/pipeline/archives/2010/02/23/thi...
The was one company making a good product called Battlepacks (for lightweight - super heavyweight classes). If you were doing anything in the smaller weight classes (antweights through the 30lbers), you were typically hijacking a lot of remote aviation resources.
The options available now are much better than they used to be.
One would argue even if you know exactly what you're doing, unless you're replicating those same safety features as well, you should put it in a shed anyway.
That said, totally agree with what you're saying.
The logical conclusion of this is someone mentioning ICE vehicles and gas tanks, so allow me to be that person. If you park your gar in the garage and it's an ICE vehicle, you're parking quite a lot of latent energy in a fairly volatile form in your house.
If you wouldn't feel comfortable rigging a bunch of old gas cans in tandem to a little engine your scrounged as a DIY generator and putting it in your garage, maybe you shouldn't rig a bunch of batteries up in a similar arrangement.
For that matter, with that perspective, how far away for your cinderblock outhouse is far enough?
Any energy storage medium that holds a good amount of energy and has a methods to release it in a fairly short time frame (either through a failure mode or through intended use) deserves some real thought. Which is basically what you were saying...
- Natural Gas is brought in through VERY small pipes, and is never kept anywhere, it's just used immediately.
- Jerry cans are found regularly but are also sealed up as much as possible while still remaining usable.
- Vehicles have sealed fuel systems, and are designed to not allow the fuel to combust all at once if at all possible.
In fact, now that I'm thinking about it, pretty much the only time you hear about homes exploding/bursting into flame/etc. is when one or more of these systems is compromised.
Medical oxygen plus ignition sources (especially cigarettes) seems to be another (for “bursting into flames” more than “exploding”.)
That said, a firewall is not a fire proof wall, it's a wall designed to burn slower to prevent the fire spreading between those areas too quickly. Its purpose is to retard the fire u til it can be put out or so you have enough time to escape with your life. I suspect with failed batteries the latter is more likely your best chance.
The safety features you're so worried about are just fuses. And maybe some diodes if you're that worried. I can say that installing a bank on a piece of plywood on your garage wall is obviously stupid, but a simple opening of the NEC book will show the code for a battery bank, shall have clearance from a wall and rest on non-flammable/non-conductive material, usually in a nuclear power plant or a substation we would build a rack out of metal then install ABS plastic shields.
Source: I'm an industrial electrician and went through an IBEW apprenticeship,
A LiFePO4 will happily catch on fire if you charge it with the same voltage/current curve as a LiPo(3.6v vs 4.2v).
[edit]
To expand a bit, this[1] is a fantastic reference on exactly how complicated and involved the different chemistries can be. This is talking about Lead -> LiFePo4 but a lot of stuff applies to LiPo and other variants as well.
> Please note that in all 68 of these RUINED LFP banks, that I have to date, there was not one incident of fire, explosion or dangerous off-gassing.
So you can destroy them easily by overcharging or other mistake, but LiFePO4 is probably the safest of the lithium chemistries as far as fire goes.
Most lithium chemistries have a nominal voltage near 3.3 - 3.7V per cell.
When you stack lithium cells in series for any sort of high-performance application, you need additional balancing circuitry to bleed off energy from the cells individually (tiny variations between cells results in them having different capacities, which means some will charge up at different rates, resulting in them DISCHARGING at higher rates, which results in pretty serious negative effects for the life of the battery pack). The only exception here is extremely well matched cells, which is how Tesla is able to get away with not using much balancing circuitry (they are very picky about the cells they use, and have enough cells in stock to pick ones with very similar capacities).
And the safety features include a lot more than fuses - there are over-voltage protection circuits (extremely important!), over-discharge protection circuits, temperature protection circuits (you can overheat a cell even without drawing too much power out of it).
In short, large stacks of lithium cells are a difficult and dangerous beast to handle, and a lot of work goes into keeping them safe. These cells tend to have very low internal impedance, and as a result a chain of them can produce staggering amounts of power.
Source: I work on the embedded protection systems for a company which makes specialty lithium-ion battery systems for industrial, aerospace, and defense customers.
Why didn't they use them in the first cars? When they either over or undervolted, they tended to explode. We didn't have the electrical circuitry to handle lithiums.
Now we do, mostly. Bad "spray fire and death" style accidents still happen.
Incorrect. Lithium Ion batteries are 3.7V (nominal) with different chemistries varying from 3.2V (LiFePO4) to 3.8V (LiHV).
>The safety features you're so worried about are just fuses. And maybe some diodes if you're that worried.
Also incorrect. It won't just have fuses, but also low voltage (over-discharge) protection, high voltage (over-charge) protection, and transient current limiting with auto reset, typically done with an IC + FETs per small pack of cells or even per cell.
So before you can build your pack you need to make sure that the cells all have the same chemistry and charge level, and that the ICs are programmed to cut out at the same voltage and current, otherwise you get nasty surprises when connecting in parallel (the voltage mismatch) or in series (current mismatch) or any hybrid of the two.
Source: specced and sourced cells, safety ICs and chargers for portable payment terminals. And 15 years experience in electric RC aircraft.