LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[ebaugh]

RC,

You have the best data, and perhaps the best cell performance we've seen to date. I for one appreciate the effort! The boss just bought one of those testers, but I haven't had a chance to tinker yet.

I sort of have a baseline now from the pics I just posted. Will do a few full cycles back to back and see what happens. The discharge rate is our dockside loads of 2-3A, and recharge rate I will set to about 80A. Will probably go to 3.65 one time, I want to quantify what lies between 3.55 and 3.65. But forget the lost cell, the baseline today is around 800Ah on 1200 when new (maybe?).

I wonder if there is value in going below 3V to achieve full cycles. This won't be easy unless I take a cell out of the other 3 strings.

Going to take several weeks to collect meaningful data.

The biggest problem with my bank is probably heat, it's > 80F 6 months a year down here in the engine room.

[Maine Sail]

Originally Posted by ebaugh

RC,

You have the best data, and perhaps the best cell performance we've seen to date. I for one appreciate the effort! The boss just bought one of those testers, but I haven't had a chance to tinker yet.

I sort of have a baseline now from the pics I just posted. Will do a few full cycles back to back and see what happens. The discharge rate is our dockside loads of 2-3A, and recharge rate I will set to about 80A. Will probably go to 3.65 one time, I want to quantify what lies between 3.55 and 3.65. But forget the lost cell, the baseline today is around 800Ah on 1200 when new (maybe?).

I wonder if there is value in going below 3V to achieve full cycles. This won't be easy unless I take a cell out of the other 3 strings.

Going to take several weeks to collect meaningful data.

The biggest problem with my bank is probably heat, it's > 80F 6 months a year down here in the engine room.

In all honesty I suspect heat is a much bigger factor in LFP cell degradation than has been previously alluded to in the industry. I also suspect floating or holding voltages that keep them near top of charge with a 3.35 +/- holding voltage is also bad news for them.

Combine this and heat and I think it could makes for an early demise. Combine heat with higher charging voltages (3.55VPC +) and floating/holding voltages and I know they can be murdered in less than 80 cycles...

I am only taking my cells to 2.9V for a full capacity test but this is at just 30A on a 400Ah bank so 2.9VPC works fine and by the time the bank hits 2.9VPC at 30A most of the capacity is gone. Below 2.9VPC at 30A it drops off so fast I can barely extract another Ah or two so it's just not worth it. I am holding 13.8V / 3.45VPC however until tail current drops below 10A. I have not noted much more capacity by taking the cells to 3.5VPC - 3.55 VPC+....

[nixsee]

Originally Posted by Maine Sail

I am only taking my cells to 2.9V for a full capacity test but this is at just 30A on a 400Ah bank so 2.9VPC works fine and by the time the bank hits 2.9VPC at 30A most of the capacity is gone. Below 2.9VPC at 30A it drops off so fast I can barely extract another Ah or two so it's just not worth it

Originally Posted by Maine Sail

I suspect I did not see as much gain with the back to back 100% DOD cycles as Terry had because I am normally discharging this bank to 80% DOD & then recharging at .23 - .3C to less than a 10A tail current.

Maine Sail,

Would you mind explaining to me how you determine 80% DOD? This would help answer the question I posed a couple days ago - I am regularly discharging to 2.9VPC right now as that is where my HPBMS (warning) LVC is set (protection is at 2.6VPC). I was under the impression that this was a safe level, but have noticed (as you have) that the voltage drops very quickly after this level meaning it is likely close to 100% DOD.

Do you use a certain voltage level (e.g. 3.1VPC?) or do you count the current to determine 80% DOD?

Thanks so much.

[Maine Sail]

Originally Posted by nixsee

Maine Sail,

Would you mind explaining to me how you determine 80% DOD? This would help answer the question I posed a couple days ago - I am regularly discharging to 2.9VPC right now as that is where my HPBMS (warning) LVC is set (protection is at 2.6VPC). I was under the impression that this was a safe level, but have noticed (as you have) that the voltage drops very quickly after this level meaning it is likely close to 100% DOD.

Do you use a certain voltage level (e.g. 3.1VPC?) or do you count the current to determine 80% DOD?

Thanks so much.

I use a carefully calibrated Ah counter (Link-Pro/E-Expert-Pro) that is physically and manually reset to full each time the bank hits 13.8V & less than 10A of charge current. Voltage only is very difficult to use with LFP and especially at low current draw...

Contrary to popular misconception Ah counters are not very accurate beyond a few cycles on any battery, let alone LFP, thus frequent manual resets to 100%, when the bank is known to have met full criteria, are really a requirement. 2.9VPC at house loads is basically 0% SOC.

[ColdEH]

I very rarely if ever take my cells below 3.2 v per cell or 12.8 volts overall.

There stills seems to be lots of capacity down there but I like to keep things out of the knees , would like these batteries to last a long time .

Regards

[nixsee]

Originally Posted by Maine Sail

I use a carefully calibrated Ah counter (Link-Pro/E-Expert-Pro) that is physically and manually reset to full each time the bank hits 13.8V & less than 10A of charge current. Voltage only is very difficult to use with LFP and especially at low current draw...

Contrary to popular misconception Ah counters are not very accurate beyond a few cycles on any battery, let alone LFP, thus frequent manual resets to 100%, when the bank is known to have met full criteria, are really a requirement. 2.9VPC at house loads is basically 0% SOC.

Yes, I have only just realised that 2.9 is probably 0%. I have a Ah counter (EV Display from CleanPowerAuto) but have zero confidence in it being precise. Is there a voltage level that is associated with 20% SOC or is that too far into the flat section of the voltage curve? If not, I suppose I can use my Ah counter to conservatively approximate 80% DOD and stop there.

[nixsee]

Originally Posted by typhoon

I very rarely if ever take my cells below 3.2 v per cell or 12.8 volts overall.

There stills seems to be lots of capacity down there but I like to keep things out of the knees , would like these batteries to last a long time .

Regards

Yes, keeping above 3.2 seems like a reasonable thing to do. Perhaps I'll change my monitoring to cut off loads around there as well. I haven't been using these long so doubt I've done major damage yet (never been below about 2.87VPC).

[toddedger]

One other factor that concerns me is vibration.

They drop, shoot holes in the case, over and under charge for saftey, but I wonder about vibration. It's been stated that the larger cells are not recommended for boats because of internal structure.

Ebaugh's cells are in his engine room and he has had a complete failure of a 100 amphour cell. I'm very curious about that failure. That's the first individual failure of a small cell that I've heard about.

[Maine Sail]

Originally Posted by nixsee

Yes, I have only just realised that 2.9 is probably 0%. I have a Ah counter (EV Display from CleanPowerAuto) but have zero confidence in it being precise. Is there a voltage level that is associated with 20% SOC or is that too far into the flat section of the voltage curve? If not, I suppose I can use my Ah counter to conservatively approximate 80% DOD and stop there.

If you watch my video in post 5235 you can see that:

400Ah Rated Bank: 4X Winston 400Ah Cells (2009)

30A Constant Current Load (80% DOD Based on Cell Rating)
-319.8Ah = 12.775V / 3.19VPC Under Load

30A Constant Current Draw (80% DOD Based on Tested Capacity)
-344.7Ah = 12.661V / 3.17VPC Under Load

Bottom line... Keep the cells above 3.2Vor 12.8V for a 12V pack. ...

[OceanSeaSpray]

Originally Posted by ebaugh

Yes, thanks! The old post interesting, but don't know yet if I can "retrieve" capacity by cycling. I think there is a "formation" charge at the end of cell manufacture. This some number >3.6V. I wonder if after a period of time this should be repeated? While rebalancing this bank, I did observe far more charging going on in the 3.5V range than on install. When new, at 3.45-3.46V it was done, maybe another 20-30Ah to 3.6. When rebalancing it was easily 100Ah, possibly as much as 200Ah, I didn't watch that close.

The formation charge is something different, it is the process during which the SEI interface layer is produced in the new cell.
From memory, what Terry had done to recoup the out-of-reach capacity was absorbing the cells for longer at a higher voltage (3.8V/cell specifically) a few times. It is the higher voltage that makes all the difference. As you observed already, repeated low charging cycles make them harder to recharge at the top end, so it has a compounding effect.
The first thing I noticed with my bank is that, short of charging properly at the top, it leaves quite a lot of capacity on the table and the memory effect is going to grow from there.

Originally Posted by ebaugh

Without a different charger (or a new board in the Magnum), I can't change the 12.7 to 12.8V where the charger switches from "FULL" charge back to Float. I will raise the Float setting to get the high end SOC higher. But how high? This cycle is non-configurable logic by the charger to not overcharge FLA.

A full cell sits at about 3.35V OCV (top of the discharge curve) and it will only charge extremely slowly at 3.4V if the goal is a proper charge. The strategy I have implemented is charging and absorbing properly to 3.5V/cell and then stop. Following this I try to maintain 3.35V/cell using solar. The result is extending the duration of the discharge cycle very considerably.
Using shore power is a bit different in the sense that it won't allow discharge at night, so I would choose a voltage that allows enough discharge to get into the natural voltage range of a battery in storage, say 3.3V/cell.
This would effectively place the bank in storage while on shore power. Why cycle when there is no need?
Note that the charger DC has to be very clean, no ripple. Sometimes the only sensible option when equipment is inadequate is discarding it. I threw out two solar MPPT controllers on recent installations, a Morningstar and a Tracer. They were horrible for that application.

Originally Posted by ebaugh

I have to think more about PLC/BMS control of the charger. I always liked the idea it didn't really "control" anything, but was simply the fail safe monitor. Putting the BMS in control creates a single point of failure since to actually control the process, it would need the ability to charge to a pretty high voltage. You could add a CC/CV charger limited to 14.4V which normally wouldn't be an issue, but could be if control was lost.

The BMS only really needs to signal whether the battery is allowed to be charged and the rest of the gear must be made to comply. It can see the end of the absorption (if it measures battery current) when the other gizmos typically don't even care and it allows implementing strategies. The charging sources should be voltage-limited as to provide a first line of protection against dangerous overcharging.
Incidentally, in the bigger picture beyond the DIY group, a LFP system with a failed BMS is not an operational system, it is just a liability.

Originally Posted by ebaugh

But let's say I can do exactly what I want....what is that? Is it different at anchor than at dock?

I think it is. With shore power and LFP, the goal posts are elsewhere:

• When you have a LA battery and shore power, you charge it 100% full and try to hold it there without too much gassing etc. There is no meaningful cycling and the battery is not really being used. This kills LFPs.
• A LFP battery needs to be left at around 50% SOC if not being used.

In the first case, it requires a charger; in the second, it requires a constant voltage power source at a voltage just high to maintain a balance around a half-charged bank and the BMS can help with switching it for that matter.

[kmacdonald]

I charge my 20AH LiFePO4 battery with a 60W solar panel to 14.2V. When it reaches 14.2V I turn off the solar. It powers LED lights, depth finder, VHF, cell phones, and tablet PC. I monitor the battery with a Victron BMV700. The battery is fully charged by the solar panel even on cloudy or rainy days. I am going to change to a smaller and lighter 30W panel sson. This has worked fine for several years of weekend use. I never check the capacity and will just replace the battery when it no longer has the capacity to meet my limited electrical needs. For those that are thinking of adding refrigeration, consider the cost of additional electrical and charging capacity into the equation-----it's very significant.

[greg_kulosa]

I finished reading the whole thread! (There should be a merit badge or something). Thanks for all the excellent discussion. I really learned a lot, and feel like I have a good handle on LiFEPO4 now.

Long ago on this thread - there was talk of a BPS - a Battery Protection System. That's what I'm looking to install in my RV. There was also talk from various people that they were going to develop one. Has there been any progress on this?

Features that I would like:

- NO BALANCING RESISTORS!
- Adjustable LVC/HVC per-cell voltages
- Adjustable cutoffs for pack voltages
- Disable charging under 0 celsius

Those are the most important things that I see to protect the LiFEPO4 battery from damage due to other things going wrong. Other nice to haves would be alarms before we get to the cutoff voltages, some kind of display, etc.

My own system is going to be much simpler than most of you. My RV is a trailer, so no alternator to worry about. Just solar charging. I'm going to use the Bogart SC-2030 charger with the TM-2030 display/meter. It looks like that can be adjusted to "absorb" at 13.8 volts until it reaches an adjustable amp level, and then stop charging. When it stops charging, it can be programmed to reset the SOC meter to 100%.

The 2 options that I see for a BPS are:

- Use the popular Housepower BMS, and modify the cell boards by removing the resistors to make sure there can be no accidental unbalancing.

- Use the Junsi 8M cell logger with a relay on the alarm port to disconnect both load/charge buses for either high or low voltage events.

Frankly, even though the Junsi seems kind of fragile and "hobby/amateur", that's the direction I am leaning. In my case, I can get away without separate load & charge buses, but I can still use two relays/contactors with a delay on the load relay to avoid any high voltage spike from getting onto the load bus when I disconnect.

Are there any other SIMPLE options for a Battery Protection System?

Thanks.

[NahanniV]

Originally Posted by greg_kulosa

I finished reading the whole thread! (There should be a merit badge or something). Thanks for all the excellent discussion. I really learned a lot, and feel like I have a good handle on LiFEPO4 now.

Long ago on this thread - there was talk of a BPS - a Battery Protection System. That's what I'm looking to install in my RV. There was also talk from various people that they were going to develop one. Has there been any progress on this?

Features that I would like:

- NO BALANCING RESISTORS!
- Adjustable LVC/HVC per-cell voltages
- Adjustable cutoffs for pack voltages
- Disable charging under 0 celsius

Those are the most important things that I see to protect the LiFEPO4 battery from damage due to other things going wrong. Other nice to haves would be alarms before we get to the cutoff voltages, some kind of display, etc.

My own system is going to be much simpler than most of you. My RV is a trailer, so no alternator to worry about. Just solar charging. I'm going to use the Bogart SC-2030 charger with the TM-2030 display/meter. It looks like that can be adjusted to "absorb" at 13.8 volts until it reaches an adjustable amp level, and then stop charging. When it stops charging, it can be programmed to reset the SOC meter to 100%.

The 2 options that I see for a BPS are:

- Use the popular Housepower BMS, and modify the cell boards by removing the resistors to make sure there can be no accidental unbalancing.

- Use the Junsi 8M cell logger with a relay on the alarm port to disconnect both load/charge buses for either high or low voltage events.

Frankly, even though the Junsi seems kind of fragile and "hobby/amateur", that's the direction I am leaning. In my case, I can get away without separate load & charge buses, but I can still use two relays/contactors with a delay on the load relay to avoid any high voltage spike from getting onto the load bus when I disconnect.

Are there any other SIMPLE options for a Battery Protection System?

Thanks.

Junsi also makes the powerlog 6S that will do everyting the 8M can plus: monitor and trigger an alarm for temperature, monitor current with a built-in shunt rated 40A continuous/130A Peak.

Cheers,
JM.

[ebaugh]

Maine and Ocean,

Both of you provide lots of things to consider. Thanks for posting.

Sorry for the delayed response, I am following this thread again. But not every day due to non boating stuff....it stinks when life gets in the way of cruising.

I will add another observation from another install using an Orion Jr BMS. In that case, the charging device, a Victron Inverter Charger shows huge (+/- 100 mV) readings of string voltage under charge. The designer/developer claims these readings are real. As explained, there is little to no averaging going on in the readings. But a Fluke multimeter, or anything else I attached does not show these deviations, perhaps explained by automatic averaging in the firmware/software of other devices. Hmmm. Not being able to tell 3.45 from 3.55 is a issue. Further, there is a feeling by the Orion BMS company that the peak voltages are limiting. Not to beat up Victron, shop testing on a Magnum device shows the same.

Not directly related, but a consideration when pushing to 3.6V or more per cell.

With the new data I have discovered from my install, the Magnum charge logic will only allow "Float" charge for 4 hours when cycling from "Full" to "Float" to "Full", regardless of the voltage targets programmed. So best case...I add 400 Ah from 12.7 - 12.8V to +400 Ah, then it waits to 12.7-12.8 to repeat. That's about 10% to 50% SOC...if you use the current capacity. Otherwise more like 10% to 40%.

Will keep tinkering as time allows and report results...





Sent from my iPad using Cruisers Sailing Forum

[transmitterdan]

Most battery chargers do not output clean pure DC current. They output DC that varies considerably depending on the internal switching frequency. The design details of the particular A/D converter can result in variations or errors due to inadequate filtering which results in aliasing. When testing the A/D with a lab DC power source the problem does not show up.

An A/D used for BMS may require fast response times in case something goes wrong. So it's a tradeoff. Lots of digital designers don't know how to "fix" the digital display to show the true average of the A/D reading while preserving the fast A/D sampling rate. Fluke engineers know this very well.