LiFePO4 - The future for off-grid battery banks?

[SunEagle]

Just found this site by mistake and found it interesting so joined. We've been using lifepo4 energy packs for almost 6 years and was surprised to see the claims of them only having lives of 2-3 years. That's not our experience and we have a number of installations with life times of 4-5 years and our first pack, is nearly 6 years old and is used every day on job sites. None have suffered any loss of capacity and have changed the lives of those we have installed for. Now our customers cook using them, heat hot water and use almost whatever they want.

We only install lifepo4 now, gave up agm and gel almost 2 years ago when we'd proved the concept of lifepo4 in off grid installs. We also use lifepo4 on job sites instead of taking a genny with us because they work so well, providing much more energy and extremely rapid charging.

We carry 120ah packs and with one 250w solar panel, have power all day. We designed our own charge systems and are in the process of designing and building a real dedicated liffepo4 charge system that has no need for BMS or active cell balancing. Presently the solar charge controllers we are using, remove the need for a BMS and if our customers were intelligent enough, we wouldn't use BMS at all. However many customers have no idea how to use energy system properly, so we are making a charger which will control the system without any intervention. Currently we use active cell balancers and a BMS for customer installs, our own work pack have no BMS or balancing and they work fine on our solar charge controllers.

You could look at it this way, our first pack, has been used for at least 300 days a year for over 5 years working and on non work day, we use it for leisure and camping. In fact I use the dam thing for just about everything I need for power and carry it in the tractors on our farm. So it could be classed as cycling at least 1500 times. We don't see it that way, because of the charge/discharge parameters we use.

We also have lifepo4 as starter batteries in all our vehicle, work and personal. The two (24v) in the tip truck have been in use for over 4 years and they are amazing for staring engines, so much power they for instantly, except in very cold weather.

One thing we learnt early, is not to use big capacity cells. Our first commercial packs were 100ah cells, big mistake. When one cell collapsed, the entire system went down and we lost a lot of money using big brand cells and almost made us give up on lifepo4. So we went for 40-50ah military/industrial prismatic cells, which is what our first two packs consisted of and have not had a problem since.

Sounds interesting. I would like to know a little more about your energy packs. Although they seem a little small at 120Ah & 250watts to be be able to generate enough power for most off grid users or what most electric hand tools need to run which would be closer to 1000 watts.

[PNjunction]

Hi Dax and welcome!

Your experience sounds typical, although I'd like to know what those big-brand 100ah cells were, and if they were new, or perhaps new-old-stock that had previous owners? Also interested in your current cell manufacturer. Here, we are mostly familiar with Winston, GBS, and CALB, and on the smaller scale, A123.

If you read through all the threads, you'll see we've covered the whole bottom-balance vs top-balance scene, individual cell-bleeder boards, which I and others eschew, and even shoehorning solar CC's designed primarily for lead into LFP use by careful selection of voltages, disabling temp-comp and so forth. Or just bite the bullet and grab a Mastervolt, or Genasun LFP controller for the smaller stuff.

It isn't hard, but it isn't ready for the guy next door. Much like a Rolls-Surrette isn't for the guy next door who doesn't know what a hydrometer is...

Again, welcome, but like with all LFP threads across the globe, keep your flamesuit on.

[dax]

Just found this site by mistake and found it interesting so joined. We've been using lifepo4 energy packs for almost 6 years and was surprised to see the claims of them only having lives of 2-3 years. That's not our experience and we have a number of installations with life times of 4-5 years and our first pack, is nearly 6 years old and is used every day on job sites. None have suffered any loss of capacity and have changed the lives of those we have installed for. Now our customers cook using them, heat hot water and use almost whatever they want.

We only install lifepo4 now, gave up agm and gel almost 2 years ago when we'd proved the concept of lifepo4 in off grid installs. We also use lifepo4 on job sites instead of taking a genny with us because they work so well, providing much more energy and extremely rapid charging.

We carry 120ah packs and with one 250w solar panel, have power all day. We designed our own charge systems and are in the process of designing and building a real dedicated liffepo4 charge system that has no need for BMS or active cell balancing. Presently the solar charge controllers we are using, remove the need for a BMS and if our customers were intelligent enough, we wouldn't use BMS at all. However many customers have no idea how to use energy system properly, so we are making a charger which will control the system without any intervention. Currently we use active cell balancers and a BMS for customer installs, our own work pack have no BMS or balancing and they work fine on our solar charge controllers.

You could look at it this way, our first pack, has been used for at least 300 days a year for over 5 years working and on non work day, we use it for leisure and camping. In fact I use the dam thing for just about everything I need for power and carry it in the tractors on our farm. So it could be classed as cycling at least 1500 times. We don't see it that way, because of the charge/discharge parameters we use.

We also have lifepo4 as starter batteries in all our vehicle, work and personal. The two (24v) in the tip truck have been in use for over 4 years and they are amazing for staring engines, so much power they for instantly, except in very cold weather.

One thing we learnt early, is not to use big capacity cells. Our first commercial packs were 100ah cells, big mistake. When one cell collapsed, the entire system went down and we lost a lot of money using big brand cells and almost made us give up on lifepo4. So we went for 40-50ah military/industrial prismatic cells, which is what our first two packs consisted of and have not had a problem since.

[Sunking]

That's the thing - I don't care what Tesla is doing for the following reasons:

Calm down al I am pointing out wheen it comes to charge rates there is no real difference between an EV and Solar Application. If anything with solar you will charge at a faster rate than an EV.

On the discharge side my comment was to point out even discharge rates are similar in Fractional C-Rate. With a Tesla cruising you are talking about C/6 discharge rates, unless you expect to go 300 mph for an hour.

What I think we agree upon overall Solar is less forgiving, not in charge or discharge rates, but rather environmental conditions. We do not need the level of Automation an EV needs. You do not need any active BMS or really even a BMS if you know what you are doing. All you really need to do is run PSOC and set your LVD correctly so it is impossible to ever have an over discharge.

One thing we can do is use some of the same ideas, the biggie is not allowing yourself to use full capacity. You never fully charge, and leave plenty in the tank. No EV will alow a customer full access. Tesla did open most of the capacity in Roadsters. Unknown to the public Tesla upgraded everyone capacity by 15 Kwh on their older Roadsters. It was done while their customers were sleeping. All Telsa did was upgrade the Firmware while you were sleeping. It is by limiting access that allows then such long battery life. Chevy and Nissan also do the same, they just do not have as much to play with.

[PNjunction]

That, and the whole Fiskar / A123 thing. Old history, and times and manufacturing procedures has moved onwards towards the better. A123 is still around, whereas the average consumer thinks of bankruptcy as out for the count. Not true, but makes great fodder for the WSJ.

Tell ya, what - If I were in charge of Volkswagen, I'd have no hesitation going LFP with A123 and doing it right and inexpensively. Who knows ...

[inetdog]

If they used LFP, I doubt they would have been able to deliver the 1500 amp currents required for their Insane and Ludicrous Mode acceleration figures.
Maybe not a requirement for a useful EV, but definitely a marketing point for their high end buyers.

[PNjunction]

I agree to some extent, but there is over lap, especially when you compare a pure EV like a long range Tesla.

That's the thing - I don't care what Tesla is doing for the following reasons:

1) They aren't using LiFepo4, but another chemistry that is higher up the scale in unnecessary energy density and less stability. LFP at the bottom end of the rung, is already far more than sufficient for our needs and basically the safest. Longest life is when run at no more than .5C, and while they can take 1C or even more in bursts, most of us will never achieve that, nor do we NEED that, unless you are running a very high powered genny.

2) They are using non-lfp non-prismatic cells which adds to the complexity of the overall system - frequently requiring the use of proprietary or walled-garden support electronics. With large prismatics like Winston, CALB, GBS, etc, one can diy it like you would with lead, as long as you follow basic safety and operational rules. These non-lfp cylindrical cells have even higher performance characteristics than the large prismatics, so all that performance is wasted in our application - unless one is trying to make a ridiculously small battery bank.

3) While the price of these non-lfp, cylindrical cells is coming down, the safety and complexity issues remain. Even if they came down to 10-cents a cell, for us, we should only by paying about a penny due to the waste of performance. We are also back to the safety, stability, and complexity even at a penny a piece. No thanks.

This is what I mean about EV use when applied to solar-housebanks. Unfortunately, all roads seem to lead back to Tesla, and everyone forgets the basics about why LFP in large prismatic form was chosen for our application in the first place.

[reed cundiff]

??

"... comparing Apples to Apples charge and discharge rate..." Is this not the way to properly compare? Or am I missing some artifact in reasoning?

[Sunking]

Due to the much smaller internal impedance(resistance) of LFP batteries you can use a much smaller battery than you would need to with an FLA battery if you need to run large loads. In my case I can draw over 6kWs from a 9.5kWh LFP battery (~0.6C) with less that a 10% voltage drop. Try doing that with an FLA battery, especially if it is a few years old.

Simon

Hey one-trick pony go get some more help, your all wrong again. . Not even talking about FLA. It was in reference to EV and Solar capacity size.

[SunEagle]

Are you saying that AGM and LFP batteries only last 2.5 years (913/365)? My LFP battery is two and a half years old and shows no signs of any variation from its initial specifications, still has same storage capacity and same internal impedance(resistance). Unless you can give me proof to the contrary I can't see any reason why an LFP battery will not last around ten years or more in off-grid applications. If you use the ten year figure you end up with the same price as good quality FLA batteries. This was the conclusion I came to before I purchased the LFP battery and the reason I purchased the LFP battery.

Here are some figures from the real world. My Winston battery made up of thirty two 90Ah cells which equates to a nominal storage capacity of around 9.5kWh cost AU$4213.44 or AU$443.52 per kWh in April 2013.

From power usage information logged via my BMS, for the year to date I have drawn 18374.54 Ah from the battery which equates to around 1.8kWh per day. This equates to around 6500 kWh over a ten year period. If we divide by the cost of the battery we get a cost of around $0.65 cents per kWh. For the past couple of years we have not been living at home full time. If we were I would expect our use from the battery to be around 3kWh per day so would bring the cost down to around $0.40 cents per kWh.

There are also other savings from using an LFP battery. I can have a smaller solar array and solar charge controller due to the high efficiency of the battery and not having to keep the battery charged as close to 100% as possible to stop sulphation. From data logged by my BMS, the overall efficiency of my battery over the past few months is around 94%-95%. I have no need of a generator as I can comfortably keep the battery at a PSOC (partial state of charge) i.e. not full.

Simon

Depending on the manufacturer and actually duty cycle of the battery an AGM or LFP lifespan can exceed 2.5 years. But based on the history the average usage of those type batteries rarely have lasted beyond 3 years.

To say that a battery has been deemed to last 10 years because of accelerated testing is misleading because based on how they are used in a real life situation the lifespan could be much shorter.

If you have gotten 2.5 years out of your batteries they may go for another 2.5 years or just rapidly fall off the curve of their rated output. So to be safe you really can't say that all LFP batteries will perform that same as yours or if "any" will last 10 years without better "real life" testing.

I hope yours last long although there isn't any solid history to say all LFP will last long. I am just being realistic. No one ever stakes their reputation on a single data point.

[karrak]

Charge Rates from solar

I agree to some extent, but there is over lap, especially when you compare a pure EV like a long range Tesla. Do that, you now are comparing Apples to Apples charge and discharge rates, and loose any debate about Sub C-Rates when they are similar. Another point in the Sub C-Rate debate in most cases Solar is charging at a higher C-Rate than an EV. Solar is in excess of C/6 and can go as high as C/2 if you live in areas with low Insolation. So on the Charge side there is no Sub C-Rate debate, Solar is charged faster. On the discharge side, Sub C-Rate comes into play. You are much higher risk of over discharging a cell in an EV with short burst of acceleration using high C-Rate than with Solar. That just means you need more automation on an EV with Respect to LVD.

I find the charge rates you say puzzling, in my case my maximum charge rate is a little over C/10, most other people I know with off-grid systems with LFP batteries charge at less than C/5. Your design criteria that the battery should have three days storage implies you only need an average charge of C/3 per day. Even if you spread this over only a few hours you end up with a maximum charge rate of C/6. If high charge rates do cut down the life of an LFP battery and I had an oversize solar array I would look at limiting the charge rate into the battery to say less than C/5.

Last is capacity. No meat on this bone. A Solar LFP battery can be smaller, equal, or larger in size. No C-Rate issue here.

Due to the much smaller internal impedance(resistance) of LFP batteries you can use a much smaller battery than you would need to with an FLA battery if you need to run large loads. In my case I can draw over 6kWs from a 9.5kWh LFP battery (~0.6C) with less that a 10% voltage drop. Try doing that with an FLA battery, especially if it is a few years old.

Simon

[karrak]

Cost of power from LFP battery

You can buy and operate a FLA battery for 5 and a few up to 10 years. There are a couple out there. A really good FLA cost around $200 to $250 per Kwh. If it last say 5 years means just in battery cost alone you paid $1250 / 1825 days = 69-cents per Kwh just for the batteries. You want AGM or LFP? They cost twice as much and last half as long. That means $2500 / 913 days = $2.74 cents per Kwh or 4 times more than FLA.

Are you saying that AGM and LFP batteries only last 2.5 years (913/365)? My LFP battery is two and a half years old and shows no signs of any variation from its initial specifications, still has same storage capacity and same internal impedance(resistance). Unless you can give me proof to the contrary I can't see any reason why an LFP battery will not last around ten years or more in off-grid applications. If you use the ten year figure you end up with the same price as good quality FLA batteries. This was the conclusion I came to before I purchased the LFP battery and the reason I purchased the LFP battery.

Here are some figures from the real world. My Winston battery made up of thirty two 90Ah cells which equates to a nominal storage capacity of around 9.5kWh cost AU$4213.44 or AU$443.52 per kWh in April 2013.

From power usage information logged via my BMS, for the year to date I have drawn 18374.54 Ah from the battery which equates to around 1.8kWh per day. This equates to around 6500 kWh over a ten year period. If we divide by the cost of the battery we get a cost of around $0.65 cents per kWh. For the past couple of years we have not been living at home full time. If we were I would expect our use from the battery to be around 3kWh per day so would bring the cost down to around $0.40 cents per kWh.

There are also other savings from using an LFP battery. I can have a smaller solar array and solar charge controller due to the high efficiency of the battery and not having to keep the battery charged as close to 100% as possible to stop sulphation. From data logged by my BMS, the overall efficiency of my battery over the past few months is around 94%-95%. I have no need of a generator as I can comfortably keep the battery at a PSOC (partial state of charge) i.e. not full.

Simon

[Sunking]

Notice how all discussions seems to revolve back to commercial EV's as the only source of experience?

All discussions rapidly get out of context of OUR application of relatively low voltage / low current, something these cells laugh at, unless you have purposely used an insanely small bank to begin with, drawn to the promise of 80% DOD leaving you with no autonomy. Mistake #1.

I agree to some extent, but there is over lap, especially when you compare a pure EV like a long range Tesla. Do that, you now are comparing Apples to Apples charge and discharge rates, and loose any debate about Sub C-Rates when they are similar. Another point in the Sub C-Rate debate in most cases Solar is charging at a higher C-Rate than an EV. Solar is in excess of C/6 and can go as high as C/2 if you live in areas with low Insolation. So on the Charge side there is no Sub C-Rate debate, Solar is charged faster. On the discharge side, Sub C-Rate comes into play. You are much higher risk of over discharging a cell in an EV with short burst of acceleration using high C-Rate than with Solar. That just means you need more automation on an EV with Respect to LVD.

Last is capacity. No meat on this bone. A Solar LFP battery can be smaller, equal, or larger in size. No C-Rate issue here.

Something not mentioned either is that in a fixed installation, these cells aren't going to be subject to physical stress / vibration etc. They are just going to sit there and laugh at the load your inverter places upon them - again of course assuming you were duly diligent in sizing your bank in the first place.

This is really the only major difference between Solar and EV. In Solar you do not need a very constrictive box, that is insulated thermally, and uses Thermal Management. You do not need much if any thermal management with Solar assuming the batteries have ventilation above 32 and below 100.

Technically the only difference between an EV and Solar System application is an EV must have Thermal Management, Active LVD, and Automated systems to watch over because the user cannot do that while driving. In Solar some think you must you the high level of Automation that is used in an EV which is False. EV's do not charge to 100% SOC normally, and do not allow the customer to access full capacity of the battery. You can do the same thing without a high degree of automation or special equipment if you know what you are doing. Otherwise, then you should use a high level of automation, but tune it so it limits the amount of SOC to less than 100% and never allow you to go below 10 to 20%. .

[lkruper]

I lurked here for education from the start. With a low insolation, cold environment and shading after 2:30PM, I desired fast charging for periods where the genny is needed. I also was looking for low toxicity and high energy density. I was willing to pay a premium to take advantage of multiple characteristics of LFP after considering other options.

However, what happened is I realized at some point late that I was not willing to rely on a solar system for a year round house in the boondocks in the mountains. If I can't get POCO because my HOA will not permit where I want to run it, I will be selling my cabin. It's partially a matter of principle, but there was definitely a shift away from the enthusiasm I had for the idea in the first place. The irony is that the conditions that made LFP attractive to me have convinced me I shouldn't go solar.

There certainly are a number of comments related to politics, trolling, etc, but there is wheat within the chaff. I suppose it is a fair question how many will get something they can use.

Sorry to hear you would sell your cabin. Perhaps if it comes to that you could supply a well-thought out plan for solar that could help sell the cabin.

[Living Large]

Thing is, there really is no point in discussing LFP, because we always get sidetracked with non-solar housebank applications, be they 48v or lowly 12v. And quite frankly, 99% of the comments come from those who will never use it, but just want to fan the flames for entertainment, and not education.

I lurked here for education from the start. With a low insolation, cold environment and shading after 2:30PM, I desired fast charging for periods where the genny is needed. I also was looking for low toxicity and high energy density. I was willing to pay a premium to take advantage of multiple characteristics of LFP after considering other options.

However, what happened is I realized at some point late that I was not willing to rely on a solar system for a year round house in the boondocks in the mountains. If I can't get POCO because my HOA will not permit where I want to run it, I will be selling my cabin. It's partially a matter of principle, but there was definitely a shift away from the enthusiasm I had for the idea in the first place. The irony is that the conditions that made LFP attractive to me have convinced me I shouldn't go solar.

There certainly are a number of comments related to politics, trolling, etc, but there is wheat within the chaff. I suppose it is a fair question how many will get something they can use.