So, lets spec out and set up a system of LiFePo4 batteries and inverter

[inMichigan]

small LFP bank

I agree with PNJunction that unless you have some electronic engineering experience you should start off with a small system like his. There are two reasons for this, the first being that if things do go wrong you won't loose too much money, the second and more important reason is that the larger the battery the more energy that can be released quickly if things go wrong and the more damage that can be done.

Keeping an LFP battery safe and prolonging its lifetime is in my opinion more simple that keeping an FLA battery healthy. All you have to do is keep the voltage to any cell within the battery within the range of 3.6 volts to 2.5 volts. If you are charging from solar at a low charge rate the maximum voltage can be dropped to 3.45 volts or even lower if you want a longer life for the battery.

As PNJunction suggests this can be achieved by keeping the SOC of the battery between 20 and 80 % which equates to a voltage range of 3.4 to 3.2 volts per cell (13.6-12.8 volts for a 12 volt system), or by utilising a larger SOC range of around 95 to 10% and using some sort of Battery Monitoring/Management system to make sure that any particular cell does not go outside the safe operating voltage range.
Simon

This is the strategy I've adopted in my grid-tie hybrid using a mere 100 AH of CALB LFP at ~48V since July 2015. Looking forward to someday increasing it, but, so far so good.
inMichigan

[lkruper]

Thing is, do *we* need to always take into account the mainstream guy next door, or can we continue on for those willing to do so in this thread?

I am way behind you guys learning batteries, and will certainly start with lead acid. If I did anything with LIFEPO it will be tiny, and it is possible that threads like this will get me started, possibly even concurrently with FLA. What I was trying to determine is what advantage there would be for me to become involved with LI and if LI is known to be the battery of the future.

[karrak]

I agree with PNJunction that unless you have some electronic engineering experience you should start off with a small system like his. There are two reasons for this, the first being that if things do go wrong you won't loose too much money, the second and more important reason is that the larger the battery the more energy that can be released quickly if things go wrong and the more damage that can be done.

Keeping an LFP battery safe and prolonging its lifetime is in my opinion more simple that keeping an FLA battery healthy. All you have to do is keep the voltage to any cell within the battery within the range of 3.6 volts to 2.5 volts. If you are charging from solar at a low charge rate the maximum voltage can be dropped to 3.45 volts or even lower if you want a longer life for the battery.

As PNJunction suggests this can be achieved by keeping the SOC of the battery between 20 and 80 % which equates to a voltage range of 3.4 to 3.2 volts per cell (13.6-12.8 volts for a 12 volt system), or by utilising a larger SOC range of around 95 to 10% and using some sort of Battery Monitoring/Management system to make sure that any particular cell does not go outside the safe operating voltage range. I personally wouldn't operate an LFP battery without some sort of individual cell monitoring and alarm system.

With both these strategies it is most important to make sure that the all the cells have an equal charge in them at the start so they will all stay at roughly the same voltage across the range of SOC you want the battery to operate in. In my opinion it is easier to set them up at the same full level, i.e. top balancing. To do this you strap all the cells together in parallel and charge them all up to 3.6 volts. Leave them strapped together at 3.6 volts over say a twelve hour period to make sure they are all at the same state of charge.

It is important to make sure that the cells within the battery stay balanced over the longer term by monitoring the individual cell voltages and making sure that they do not diverge from each other when the battery is fully charged or discharged. If the cells do go out of balance they can be bought back into balance either manually or this can be done automatically by the BMS. I have found that if done manually it only needs to be done around once per year.

Depending on how bullet proof/autonomous you want the system to be will dictate how complicated you want the Battery Management System to be. It can range from you with a multimeter checking it at regular intervals, to a full blown automatic BMS with automatic cell balancing and automatic battery cutoff in the event of something going wrong.

If you are going to charge the battery from solar panels any Solar Charge Controller where you can adjust the bulk charge voltage down to 13.6 volts for a 12 volt system and the float charge voltage to 13.1 volts will do. It is also preferable to have some control over the Absorb time or end Absorb current but not essential. You will also need to disable any automatic Equalise cycles and temperature compensation.

Simon

[PNjunction]

Although a BMS may include some amount of digital logging and alarms, like the ill-fated Boeing Li battery pack, the actual voltage sensing and shut off or current bypass are just pretty simple analog circuits.

A big reminder that the Boeing battery was NOT LFP, but a more unstable chemistry where only .1v of overcharge can send them into thermal runaway. For LFP, that take about 30v overcharge, although yes, you will be damaging the battery and likely to see swelling, but the initial reaction is not so dramatic.

That's the question for another thread I guess - why would any SANE person every allow anything BUT the LFP chemistry on board an aircraft, where the greedy little phosphate holds on to oxygen atoms even when abused? That's not to say that improper infrastructure could be set alight, but the industry that thrives on news-bytes devoid of facts loves to spin.

[PNjunction]

If that is the case, then skills obtained early on will be academic once the batteries become mainstream. That is, unless the advances are in the chemistry to make the batteries bullet proof as opposed to computerized systems.

Thing is, do *we* need to always take into account the mainstream guy next door, or can we continue on for those willing to do so in this thread?

[PNjunction]

I suppose that if a bank were to accidently be fully discharged it would be considered abuse. With FLA, the batteries can recover, but not LI.

Not true. Lifepo4 recovers better than any other li-ion chemistry, but like any battery, try not to do that.

1) Get to the discharged cell(s) asap.
2) Recharge with no more than C/50 until a cell voltage of 3.2v is reached, whereupon you can apply a normal charge.
3) Do a sanity-check for balance.

If a BMS failed and one was there to monitor it, one could prevent a catastrophe. I would imagine that a BMS has quite a bit of software programming in it, and wonder how rigorously it has been tested for anomalies.

For our uses, the house-power-bms from CleanPowerAuto llc is popular. There are others.

Or, if you are experimenting, you may find that with monitoring, be it simple or complex, (your choice), and *conservative* charge values and limited DOD's no more than 80% and preferably less, you can go a long time without having to balance.

But we've been through all this before.

[PNjunction]

LFP are pussies and will die with just one little careless mistake. You can completely discharge a FLA to nothing, and recharge it just fine. Do that with LFP and you have boat anchors.

The first mistake is not using an LVD. If over-discharge DOES happen the following applies:

1) Get to them ASAP. It takes time for them to deteriorate, but the longer you let it go, the worse your recovery will be.
2) Recharge at NO MORE than C/50 until the cell reaches 3.2v, whereupon you can apply a normal charge rate.

Even commercial lifepo4 chargers, like the Tecmate-Optimate 291 LFP charger do this for you automatically, and also watching for abnormalities in internal resistance and charge slopes. I've brought some small powersports (Braille, Shorai etc) back from the dead, but I did so quickly, and not let them sit around for a month like that.

That gave me the courage to do the same with my 20ah cells. Whaddya' know - if gotten to in time, and with a small deep-discharge recovery current, they survived, and no apparent degradation was detected when I did a capacity test with my West Mountain Radio CBA-IV.

So the knowledgeable can bring them back. The followup is to find out why your LVD is either missing or failing in the first place, and also why you have designed your load to go beyond 80% DOD and hit the LVD anyway.

Just ask Nissan Leaf how much it cost them to learn that lesson replacing 2 model years worth of batteries. Yeah 2 years is all they lasted. My Chi-Coms lasted 7 months.

BOTH of which are being used in an EV application, and not in our tame low-current application. The Nissan is also NOT USING LFP, but a more unstable li-ion chemistry, and wrapped tightly in a cramped mobile steel box in the desert. Let's not do that to ours. Heck, bring 'em inside if you prefer.

Chi-Com LFP are default choice for DIY EV, and we love/hate them with a passion.

Anyway, it's just a bummer that we keep going over this again and again and again about fitting the battery to the application, which in this case is not EV, nor is it the guy'-next-door's battery.

Those truly interested in LFP experimentation just walk or keep silent as the signal to noise just never gets low enough to actually get to the information they need.

[inetdog]

I suppose that if a bank were to accidently be fully discharged it would be considered abuse. With FLA, the batteries can recover, but not LI. This reminds me of the exercise I have seen on sitcoms to show a teenager how much work it is to care for an infant where they have to carry around an egg for a couple weeks. How does one take a vacation when one has an expensive battery baby to care for?

If a BMS failed and one was there to monitor it, one could prevent a catastrophe. I would imagine that a BMS has quite a bit of software programming in it, and wonder how rigorously it has been tested for anomalies.

Although a BMS may include some amount of digital logging and alarms, like the ill-fated Boeing Li battery pack, the actual voltage sensing and shut off or current bypass are just pretty simple analog circuits.

[lkruper]

I believe the battery of the future will be some type of Li chemistry version. It has a lot of benefits for deeper discharge and potential more cycles then FLA. But that battery isn't here yet.

The biggest hurdle is the charging requirements. Will the standard user understand them enough to follow the proper procedure to recharge one of these batteries.

Remember the majority of the populace that have been exposed to Li type batteries use them in computers and mobile devices. There isn't much they need to know about charging them since everything is a simple "plug and play".

The future home energy storage system will need to have simple instructions on how to use and maintain it so they can get as much life out of it as possible without the user screwing it up.

When systems get to that point, there won't be any access for an end user to tweak the internals because they will be like modern automobiles that are not suited for the average person to work on. A finely balanced expensive system would likely have a warranty that would prohibit tinkering.

If that is the case, then skills obtained early on will be academic once the batteries become mainstream. That is, unless the advances are in the chemistry to make the batteries bullet proof as opposed to computerized systems.

[lkruper]

This subject has been brought up in the past here, in passing. As someone who was considering LFP, my memory is my thought was the cells are thought to be reliable if not abused, and show signs of failure gradually if not abused. I could be flat out wrong, but that is my memory. I was anticipating ordering a new bank at a certain loss of capacity, and have at least a month for delivery and replacement.

I suppose that if a bank were to accidently be fully discharged it would be considered abuse. With FLA, the batteries can recover, but not LI. This reminds me of the exercise I have seen on sitcoms to show a teenager how much work it is to care for an infant where they have to carry around an egg for a couple weeks. How does one take a vacation when one has an expensive battery baby to care for?

If a BMS failed and one was there to monitor it, one could prevent a catastrophe. I would imagine that a BMS has quite a bit of software programming in it, and wonder how rigorously it has been tested for anomalies.

[SunEagle]

It sounds like the OP wants to get a jump start on learning how to use LI. My question is this: Is LI definitely the wave of the future, and/or will the skills acquired with LI better prepare someone (compared to FLA) so that they have a competitive advantage when the next generation of batteries become mainstream?

I believe the battery of the future will be some type of Li chemistry version. It has a lot of benefits for deeper discharge and potential more cycles then FLA. But that battery isn't here yet.

The biggest hurdle is the charging requirements. Will the standard user understand them enough to follow the proper procedure to recharge one of these batteries.

Remember the majority of the populace that have been exposed to Li type batteries use them in computers and mobile devices. There isn't much they need to know about charging them since everything is a simple "plug and play".

The future home energy storage system will need to have simple instructions on how to use and maintain it so they can get as much life out of it as possible without the user screwing it up.

[Living Large]

Following with interest. Question for you. Your OP says this is a test system to prepare for the possibility of making an off-grid system in the future. One thing I like about FLA batteries is the fact that they are easy to acquire and set up. If you are off-grid and your system fails, how long would it take you to replace your LI battery bank compared to FLA and what would happen in the mean time?

This subject has been brought up in the past here, in passing. As someone who was considering LFP, my memory is my thought was the cells are thought to be reliable if not abused, and show signs of failure gradually if not abused. I could be flat out wrong, but that is my memory. I was anticipating ordering a new bank at a certain loss of capacity, and have at least a month for delivery and replacement.

[lkruper]

Sammy it can be done as I've pointed out.

It isn't all that drastic as long as one follow basic battery safety. And sounds like you've done a bit of homework.

It sounds like the OP wants to get a jump start on learning how to use LI. My question is this: Is LI definitely the wave of the future, and/or will the skills acquired with LI better prepare someone (compared to FLA) so that they have a competitive advantage when the next generation of batteries become mainstream?

[lkruper]

SK,

So what happens with your big Pb bank that gets down to 50% SOC and the genset is "offline" for the next week or two? So much for Pb is cheaper then LFP now - right?

Following with interest. Question for you. Your OP says this is a test system to prepare for the possibility of making an off-grid system in the future. One thing I like about FLA batteries is the fact that they are easy to acquire and set up. If you are off-grid and your system fails, how long would it take you to replace your LI battery bank compared to FLA and what would happen in the mean time?

[Sunking]

So what happens with your big Pb bank that gets down to 50% SOC and the genset is "offline" for the next week or two? So much for Pb is cheaper then LFP now - right? So we can throw away our "cheaper" Pb bank when its ruined or its life shortened because it sat under charged for weeks. LFP will happily sit at low SOC for extended periods then charge right back up.

You can take a good FLA down to 20% just like LFP, so not so sure what you are talking about. You are in the same boat either way. FLA are tuff suckers, much more so than lithium. LFP are pussies and will die with just one little careless mistake. You can completely discharge a FLA to nothing, and recharge it just fine. Do that with LFP and you have boat anchors.

One thing you have overlooked is capacity comparisons. A FLA usable capacity is 100 to 50% under normal operation, and can go down to 20% when needed without much harm. Just do not make it a habit. But here is what you have not looked at. It takes a 70 AH LFP to equal a 100 AH FLA. Sounds good right? However, they both have the exact same amount of usable power. Both can only give you 50 AH under normal operating conditions. FLA still has the advantage here, and it is significant. I can use 80 AH from the FLA from time to time no problem. You can also use the full 100 AH in a pinch and still recover OK. With your 70 AH LFP you can only use about 65 AH in a pinch. Anymore than that and you will likely destroy the battery. LFP has ZERO tolerance for over discharge. FLA can handle it. If you live where it is warm or cold, and FLA has a huge advantage. LFP batteries require Temperature management and thermal control. They do not like being warm or cold. Just ask Nissan Leaf how much it cost them to learn that lesson replacing 2 model years worth of batteries. Yeah 2 years is all they lasted. My Chi-Coms lasted 7 months.

The other "biggie" for me is that LFP can take a huge charge. I can fill em up fast while i have the opportunity...which in my case, that charge opportunity may be gone in a few hours and not return till tomorrow...... or a few days later.....or sometimes even a few weeks later (worst we had so far was 3 weeks with no grid and a fuel shortage to go with it - a double wammy !!!) Stuff happens out here on the perimeter.

You are being short sided or wearing blinders. You can charge lead acid just as fast. You just use either AGM or hybrid. Still last as long as LFP and cost less. You are confusing cost with quality. Yes FLA is 1/3 the price of cheap Chi-Com batteries. You have not heard any bad things about LFP batteries because you are looking in the wrong places. You are looking in Solar Forums where they are not used except rarely. Look where large format prismatic cells are used. Can you think of any thing? How about DIY EV's? Ever looked around those Forums? Those are where you find the real users with experience. You wil also find a lot of professionals on those forums. Here is the most popular of the EV forums.

Chi-Com LFP are default choice for DIY EV, and we love/hate them with a passion. Why do they use them. Same reason you have to use them. Because you cannot afford or even buy the good Prismatics. Only OEMS can purchase them. Factories are over booked trying to keep up with orders. A123 Systems backed up solid until 2018. Companies like Sanyo, Panasonic, LG Chem, and others who make the equality cells do not bother with consumer markets. Commercial has them over loaded. What th eDIY guys really prefer is a salvaged Leaf, Volt, or any commercial EV and rob the batteries. I just replaced my 16 GBS 100 AH LFP. Had them in service for 7 months. They were starting to fade. Capacity loss was acceptable at 4% in 7 months. But Internal Resistance had doubled which suks acceleration down. Me and on eof my neighbors got a crack at a set of Nissan Leaf batteries and split them. The GBS batteries were pampered babies. The warm climate here eats batteries alive. The GBS had maybe another year. Will see another guy bought my GBS and using in his NEV.

Look I am not trying to tell you not to use LFP, but you have false assumptions about them and making a purchase decision based on false assumptions is a plan for disaster. A 48 volt 100 AH system is not a learner system. Right now just the batteries will cost you $142/cell x 16 = $2247. That is not a toy or experiment. Neither is a 24 volt 100 AH set. I am just sugggesting you tap the brakes because you still have a lot to learn about LFP, and LFP is not ready for Solar. At least not the public in any DIY application. Chi-Coms will cost you roughly 50-cents per watt hour for a 2 to 3 year battery. Quality LFP like those from A123 and LG if you could purchase them are 3 to 4 times that much money when they become available and is rare when they are available. All I am suggesting is buy you some 18650 or 26650 cells to learn with because you wil in all likelihood destroy them. Much easier to accept loosing $8 cells than $150 cells.