Try Alibaba

Posted By: Ed DaviesPerhaps there's a business to be made breaking up car packs, weeding out the weak cells and reassembling for domestic use.

There was a report about this a couple of years back.

As Ed hinted, LiPo are quite fussy about being charged correctly. It is easy to kill the battery, fairly easy to explode it even, if overcharged.

Since a single cell is typically 3.4-3.7V any battery useful for more than a phone will be quite a lot of cells in series... with complications for charging in series once things go out of balance. Designing around the failure modes is complex and usually needs to be application specific.

It's bad enough if the typical laptop 12-20V 2AH setup goes wrong but larger capacities and voltages could be very disastrous. No-one with a decent reputation to lose would make cheap batteries without appropriately carefully designed chargers to go alongside - and the liabilities are pretty huge.

It is just about feasible for cars where the charging is carefully controlled and of course you have to pay appropriately. Even then the battery packs are carefully controlled/restricted and in many cases you don't even get to officially own them.

So yes, they are great for raw cost vs capacity, but don't expect cheap robust general purpose battery solutions any time soon.

There are lots of different types of lithium ion batteries with very confusing names and abbreviations for the names:

Lithium ion just means using lithium atoms with extra or fewer than normal electrons as the charge carrier. Lithium iron cells are ones which use the metal iron as one of the electrodes (I think). There are also lithium phosphate cells like LiFePOâ‚„ ones which should not be confused with lithium polymer LiPo cells.

AIUI, the LiFePOâ‚„ cells which are often proposed for domestic use are a lot less spectacular when things go wrong than some of the other technologies. They don't have the same thermal runaway effects that some do. If overcharged they swell and eventually burst letting out unpleasant but not too disastrous materials. They're much less likely to burst into flames than some.

They can also be damaged by discharging too far. That just stops them storing charge but I don't think it has any more overt effects.

All cell types (lithium ion, lead acid, nickle cadmium, etc), when formed into a series battery, need some mechanism to keep them in balance. With most cell types you do this by overcharging them a bit. The leading cells in a lead acid battery get rid of their excess energy by bubbling off a bit of water by electrolysis. NiCds do it by getting a bit warmer.

The downside to lithium ion batteries is that you can't just do this overcharge to balance the cells. Instead you have to measure individual cell voltages then charge the weaker cells or discharge the stronger cells periodically to keep them balanced.

Some battery management systems do this balancing continuously. However, this has the disadvantage of wasting a bit of energy and also of holding the cells at the maximum charge voltage for a while when it's better to get them up to charge then stop charging and let them relax back to their natural voltage once charging has completed (i.e., not to do a float charge as is done with lead acid cells). Also, it's more complication in a high-energy system which is probably safer kept as simple as possible. I've read of cases of homebuilt EVs catching fire on charge where it looks like the original source of the problem was a fault in the battery balancing system.

Practical experience has shown that actually cells track pretty well most of the time and that they only need to be balanced pretty rarely. E.g., cells being charged and discharged daily for a year or so without balancing.

My plan, therefore, is to have a system which monitors the cells continuously and makes sure that no individual cell goes outside the range 3.0 V to 3.6 V (most of the energy goes in and out in about the range 3.2 to 3.4 V). Once in a while I'll do a manual balance if that proves to be necessary but will not have a balancing system connected in normal operation.

Some of the newer ARM processors which are available pretty cheaply have multichannel ADCs with enough resolution to sensibly track the voltages of individual cells in a 24 V, maybe 48 V, battery so monitoring should be pretty easy. Arduino ADCs don't quite have the resolution to do this with sufficient margin to show the progress of cell charge or discharge.

Posted By: SprocketYou pretty much have to physically remove the failed cells to recover a useful battery.

Why would that be a problem? Just disconnect the failed cell and connect another one - it'd only take a couple of minutes if you had a spare to hand.

You'd end up tinkering with it all the time

That's not the experience of the two people I've read who've set up domestic-scale batteries using LiFePOâ‚„ cells. It seems to me that they're less hassle than wet lead-acid cells - not needing topping up, of course.

Found some guidance here..

http://www.homepower.com/lfp-precautions

The original question was something like "where are the battery manufacturers?". I am just providing some context and explanation as to why there are not many and no big names and reputations.

Companies like Tesla have tried to make this business cost effective but it is marginal and still only worthwhile because there is an expensive car attached and they get to sell you the whole system including charger and even then for a very specific and controlled use. Renault will sell you a car but will not allow you to own the battery, you have to lease that.

Dickster asked for 24kWH.
Even if we do that with prismatics like you suggest we have what... 50 cells at 3.2V and 160Ah in simple series. At a cost of what... at least $60 each for volume or $200 each for lots of 50 (anonymous Chinese product).

Then there is the BMS. To monitor each of 50 cells you need a fair bit of circuitry, for reliability preferably some at each battery. Cost of that is anything you like but in power electronics you get what you pay for.

And you have 150V DC so still need an inverter for normal use. Cost should be comparable to a PV inverter.
Then development cost? And testing? Just how safe is this? Can you even get product liability insurance?
And where will you put them? Can you get household insurance if you have these in your house?

Yes it is easily feasible. No it is not really commercially viable (yet)... hence no complete system products with useful warranties from the usual reputable big-name companies.

I am still using the LiFePO4 battery. It has a built-in BMS.

http://www.earth.org.uk/off-grid-stats.html

Companies such as Nedap offer turn-key domestic Li-based battery solutions right now, as it happens.

Rgds

Damon

Small; enough to run my server for a couple of days at ~200Wh.

But I have a background project to integrate ~20kWh (3 days' worth) of a similar chemistry battery with our 5.2kWp PV system to keep us entirely off the grid for most of the year.

Rgds

Damon

Ta for all the info.

It's all a bit worrying when you read about Winston/Thundersky batteries. They went bust. Mr Winston seems to be in trouble with the authorities, but I can supposedly still order the batteries on the internet.

The problem is parting with large amounts of money with no real security or guarantee that you're going to get what you pay for. it would be great if someone stocked them in UK, I'd be willing to pay a plussage for peace of mind.

If you pay by credit card, the goods aren't as described/faulty, you return the goods to the retailer & they don't give you a full refund including shipping costs then you are entitled to a refund of the total cost from your credit card company. Its up to the credit card company to manage getting the money back from the retailer & in the meantime they should refund the full amount to you. This applies equally to goods purchased from foreign retailers:

http://www.oft.gov.uk/news-and-updates/press/2007/149-07

David