Lithium Ion Batteries

Most of that talk is complete BS. There is no such thing as a 7000 cycle 10 year battery. 1000 to 2000 and 5 to 7 years is more realistic. As for prices the only way to make it work is to use LiFeP04 Chi-Com made batteries as they sell for roughly for 45 to 50-cents per Watt Hour, Catch is none have made it past 1000 cycles or 5 years. To get to the 7 year 2000 cycle battery battery you are talking $1.50 to $2.50 per watt hours using say A123 or Enerdell LiFeP04 prismatic cells.

Charge rates are good up to 2C and 3C. Meaning you can charge a 100 AH cell at 200 to 300 amps. But a nice slow C/10 works just as well.

Next in line you had better now exactly what you are doing because just one mistake will turn a Lithium into a Brick of a boat anchor. A LiFeP04 (aka LFP) can take a little abuse from over charging, but does not tolerate over discharging. One discharge too deep and you have a brick.

So the answer to your question Lithium is not ready for Solar unless you know exactly what you are doing and can afford it. It is still more expensive, and no longer lived than Lead Acid in the end today. A LFP battery to provide you with 1 Kwh per day cost you about $1500 and last about 5 years using Chi-Coms. A top end USA made 1 Kwh per day FLA will cost you around $1000 and last about 5 years. Note both examples are made with real 3 day autonomy being a 3 Kwh LFP and 5 Kwh FLA.

The lithium-ion batteries we are talking about are called LFP, or LiFePo4 (lithium-IRon-phosphate). They are related, but differ significantly chemically in operations from the li-ion in your laptop or cellphone. Just so you know.

While they can take up to 2 to 3C charge, the highest suggested rate for general-purpose charge/discharge is no more than about 0.5C. Thing is, depending on your solar system, you may have a very hard time reaching that - so possibly no worries there unless you run a genny.

Sunking has exposed some of the realities of lifepo4 life. Cycle life is not the ONLY thing you should be concerned about. What extends its life greatly is as you've seen not going deeper than 80% DOD on a regular basis - BUT-BUT - don't fall into that 80% trap with solar and accidentally UNDERsize your bank to save money. One still has to do their solar homework to figure out if they can recover or have enough autonomy despite the advantages of lifepo4.

Keep in mind that when doing solar with lifepo4, we are typically in what is called a "fractional-C" low-current application (compared to battery capacity of course), and these cells which were mainly designed for EV use are generally going to be pampered unless one uses too small of a battery capacity and just hammers it.

There are other technical details discussed elsewhere such as balancing, monitoring, and under-over voltage concerns. It is do-able, but read up first - generally common-sense stuff that goes along with any battery system, but as Sunking mentions, going to extremes of abuse will cost you.

In other words, if all you are looking for is cycle life, and you have a system working well now, I'd just be tempted to replace the batteries more often. One chooses lifepo4 not *solely* due to cycle life claims, but operational paramaters that fit your needs, ie PSOC, no need to charge to 100% or float, no sulfation, 1/3rd the weight of lead, high voltage under load..

If these advantages other than mere cycle life are actually useful to you, then yes look before you leap.

+1 pn

Thanks PN.

I agree. Currently my system is running well with AGM's and it is well balanced. I have more money than time right now and my system needs to run itself hence the AGM selection. I was just curious to see if I could go with a smaller bank and use a higher charge rate. I am not about to redesign and change things up now.

With AGM you cannot really run a smaller bank, but they can be charged at a much higher rate then FLA. AGM has its applications like high charge/discharge rates, extreme cold, unusual installation orientation, and moble applications where spills are unacceptable. But there is a huge trade-off and pain that comes with it. AGM cost roughly twice as much as FLA and last half as long. So in the end will cost you 3 to 4 times more than FLA.

But consider this. LFP prices bottomed out last year and starting to go up slightly like solar panels, but today a LFP pack long term is less expensive than AGM, and to some extent initially less expensive than AGM if you ignore the equipment cost.

With an AGM or FLA you size them for 5 day autonomy. Not the case for LFP to have roughly equal 2.5 day reserve only requires a 3 day autonomy. So if you need a 500 AH AGM/FLA you are looking at a 300 AH LFP. Assuming 12 volt system a 500 AH AGM is going to bite your wallet with $2200 bill. A 300 AH LFP is the same $2200. But the kicker is the AGM will need replaced in 2 to 3 years and the LFP if not abused or destroyed from ignorance should last you 5 to 6 years.

So when it comes to most bang for you buck, long term investment cost, FLA is still the winner. A 12 volt 500 AH 5 year FLA will cost you $1000 to $1200. A 12 volt 300 AH 5 year LFP around $2200, and a 12 volt 500 AH 3 year AGM will cost you around $1200. Based on that if you need what AGM offers, LFP is the better value. But you had better know WTF you are doing because one mistake will turn a LFP into a Brick or nice Boat Anchor. Automation mentality sold today for LFP will not prevent a Brick.

How do the numbers look for purely emergency backup?

AGM or even GEL batteries are used in UPS systems all the time.
They still cost more than FLA, but the cycle count is not as important as the shelf life on Float and the low maintenance required.
If you are not using it every day, you may not be thinking about it either.
Not sure about the details of the numbers. SK may have some input there.

Depends on the application. Emergency lighting, small form factor UPS, alarm panels etc generally demand low maintenance set and forget it. AGM and Gel make good choices. Additionally very high discharge rates demand AGM like say a UPS which some are designed for 15 minutes or 4C application. So it really depends on your application.

One mistake we see a lot here is someone finds a great deal on used Telecom or UPS batteries. They are designed for Emergency Float service. They can supply the very high discharge rates, but there is a Trade-Off. To get high discharge rates on lead acid batteries requires sacrificing cycle life. LFP does not have that trade-off. Keep in Mind LFP or any Lithium comes with some heavy technical baggage.

I want my cake and eat it too. I suspect a small chance of being stranded at the cabin for a natural disaster but if that happens I want a good few weeks at least of solid performance to supplement my genny.

Good point Sunking. My system is 100 amp / hr agm. Runs mun Dometic fridge and cell phone booster like a champ 2 -3 weekend per month. I am small potatoes.

SK, can you tell me what data you base this statement on?

Simon

These batteries have not been used in Renewable energy applications for 10 years in order to demonstrate 7000 cycles in that environment, as far as I know. I just read an article from 2013 trying to make a case for using them like this. I have also read many theories as to how to accomplish this and about failures. So, the question is, who has data to support that this technology does already provide this type of performance?

There is always a question of the validity of tests under extreme conditions and their ability to be scaled to predict long term normal usage.
Theoretically such testing is only valid if you understand the details of all possible failure or degradation modes.

One example that was missed early on, AFAIK, was that Lithium chemistry batteries degrade faster when left idle at full charge than when left idle at 2/3 charge.

For my particular low sun hour in winter application, I was drawn heavily by the high charge rate I could use from the genny, when needed. I didn't want 4, 6, 8 hours of gen run time. Every app is different, but this was one parameter that was attractive to me.