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

[john61ct]

Originally Posted by OceanSeaSpray

]The data in fact indicates that the lower the SOC the better for storage.

I realize I'm being nit-picky, and **I** realize you don't believe what your words actually mean, but it is possible some noob may believe it.

A fully depleted 100% discharge, 0% SoC is NOT better for storage than 5%.

In fact discharging that completely will permanently damage the bank. I believe there would be some longevity loss anywhere south of 10%.

Perhaps 10% SoC is better than 15%, but personally I don't plan to allow mine to go even briefly below 20%, maybe 15% if needed in an emergency.

So what low SoC do you believe is actually "the best" for long-term storage at average 70°F?

Say for five years, needing to account for some level of self-discharge.

[OceanSeaSpray]

Originally Posted by john61ct

A fully depleted 100% discharge, 0% SoC is NOT better for storage than 5%.

In fact discharging that completely will permanently damage the bank. I believe there would be some longevity loss anywhere south of 10%.

What are you basing this belief upon?? Completely discharged means they stopped at 2.5V/cell at whatever current they were discharging at and then the voltage would have bounced back up a little. The chemistry is stable at that voltage, there is no acid eating the plates, nothing happening. The reason for not storing at 0% SOC would be allowing for any minimal self-discharge that would take them down below 2.0V, that's it.
Yet the guys who wrote the paper I was referring to stored them for 9 months at 0% and then cycled them: they were the best cells of the lot.

Originally Posted by john61ct

Perhaps 10% SoC is better than 15%, but personally I don't plan to allow mine to go even briefly below 20%, maybe 15% if needed in an emergency.

Why? If the bank is managed at cell level, you can discharge it all the way down to 0% at zero risk. We use 90+% of the capacity at times and never even think about it. In winter, there is nothing uncommon for them to get close to 3.0V/cell.

Originally Posted by john61ct

So what low SoC do you believe is actually "the best" for long-term storage at average 70°F?

Say for five years, needing to account for some level of self-discharge.

You are asking me to guess what the self-discharge rate is. I don't know. 3.25V/cell is 25% SOC and I suspect they would easily last 5 years on that. A university had some A123s in storage for 10 years at 20% SOC and then recharged them, but was that 20% at the start or when they took them out again? It wasn't a controlled storage experiment, so it is a bit fuzzy.

If I was to store my bank in a disconnected state today, I would discharge it down to 13.0V and then try to have a look at it once every 1-2 years maybe. If I was able to keep an eye on it, I would take it down to 12.0V.

[OceanPlanet]

FWIW...if any on this Li battery Forum are in or near Hawaii... We have a client who is replacing his 5yr old Genasun battery system with a new OPE-Li3 system for his next cruise. The Genasun kit was originally two 24v banks of 360Ah, each bank being two 12V/360Ah packs in series. One of the 24V banks is testing out great (will have a capacity test done soon), however one bank has a cell pair that appears to giving up the ghost.


So the one higher-performing bank could remain a 24v bank, or could become one or two 12V banks (twice the Ah's, half the voltage). The other bank could be one 12V bank, half the size of the other. The weaker pack could be kept for spare cells.


These aren't new, and are long out of warranty, however could be a heck of a bargain. If you're cruising near HI, drop me a PM if interested. Packing & shipping these batteries is not really practical. But moving them to another boat in HI is something that could be done.

[john61ct]

Originally Posted by OceanSeaSpray

What are you basing this belief upon?? Completely discharged means they stopped at 2.5V/cell at whatever current they were discharging at and then the voltage would have bounced back up a little.

Aha, now I understand.

I am working with bare cells, DIY infrastructure.

So to me 0% SoC isn't an arbitrary "lowest safe" level defined by a BMS but dead flat 0 volts could put your tongue on a dead short & feel nary a tickle.

Thanks for clarifying the misunderstanding.

[OceanSeaSpray]

Originally Posted by john61ct

Aha, now I understand.

I am working with bare cells, DIY infrastructure.

So to me 0% SoC isn't an arbitrary "lowest safe" level defined by a BMS but dead flat 0 volts could put your tongue on a dead short & feel nary a tickle.

Thanks for clarifying the misunderstanding.

SOC curves never end at 0V and by the time you get to 2.5V, there is no energy left. Anything less than 2.0V is a throw-away cell.

There is nothing arbitrary, it is the way the battery chemistry works. If you flatten a LFP cell, there is a point where everything suddenly changes and its polarity even flips around.

If by "bare cells" you mean that the bottom is 0V... this is how people have very successfully incinerated EVs, by recharging destroyed cells.

[john61ct]

Yes that is exactly why I was so befuddled by your "the lower the better".

So as a general statement, sounds like say 15-25% SoC would be a good long-term storage level.

[Rebar_Caruso]

OK... going to have to follow this thread even though the jargon has lost me a bit. Had my boat since Nov 2015 and still wrapping my head around alot of different boat systems. I've got 8 cells / 800ah total. They came with the vessel and I'm glad they did. Love the lithiums, now I've just got to learn more about them. Prior owner (PO) had the charging profiles set wrong and the 12v alternater on the engine had the regulater burn out about a year before we bought it. oops. so... I'm tryin to make up by doing a crash course & will be following this thread. Thanks for posting it. Here in Australia, they're very leery of Lithiums and go out of their was to slam them, but in my view, it's the way things are going.

carry on.

[thomasow]

Originally Posted by john61ct

Problem is most chargers float forever. We were talking about charge sources that don't allow "just stop".

To me float means "charging is finished, bank is full". All my intended charge sources DO have full programmability, sufficient to ensure Absorption doesn't terminate too early

But I'm still looking for an Alt VR and solar controller that has "just stop".

I attempted to PM but not sure it was successful: Can’t help with solar (yet), however my alternator regulator will accomplish what you want: Arduino based Alternator Regulator

Configure the ‘float time’ to 1mS and the regulator will directly pass into post-float mode (charging disabled). Float can actually be configure to hold a voltage, regulator battery current to 0A (or other value), or pass to post-float mode. Your choice.

[john61ct]

Hi Thomas, no PM from you, but we have been in touch in the past. Still haven't gotten into Arduino nor your Smart AR project yet, but plan to once I get my HO alt mounted and running.

[exMaggieDrum]

Originally Posted by thomasow

I attempted to PM but not sure it was successful: Can’t help with solar (yet), however my alternator regulator will accomplish what you want: Arduino based Alternator Regulator

Configure the ‘float time’ to 1mS and the regulator will directly pass into post-float mode (charging disabled). Float can actually be configure to hold a voltage, regulator battery current to 0A (or other value), or pass to post-float mode. Your choice.

I was wondering when there would be Arduino based devices for regulators. Thanks for passing this on. I'm not sure I am a candidate for installing one but I will take a close look at it.

One thing I really would like is a device that could recognize a connection to ground from a lithium BMS to turn off a battery via a positive momentary pulse on/off battery switch rather than a battery contactor that is controlled by a relay that needs the constant drop to ground to open.

A forum contributor was kind enough to design a device but I would have to build it myself from scratch, which I will do if not other options are available. I am sure someone could, or has, developed an Arduino circuit that would do that.

[senormechanico]

How about replacing that "relay that needs the constant drop to ground to open" with a bistable relay, i.e. pulse the winding to open, pulse to close?

My boat has a relay like that for a 3 way light in the saloon. On at one end while passing thru, and off at the other end while exiting. Momentary contact switches on the bulkheads kick the relay.

You could have an automatic "stop charging" via the BMS signal and a "start charge" button when initiating charge.

[thomasow]

exMaggieDrum, and others. Ya, this regulator has been available for about 4 years now. I started it because I was unable properly charge my house batteries - and given we depended exclusively on them most of the year that was a problem.

Don't be mislead by the 'Arduino' label: it is only the dev environment I decided to use (simple for folks to install if they are so inclined). Though up until recently is was needed as I only had blank PCBs available. But this winter I went ahead and had a batch professionally assembled. They are available fully assembled, flashed with latest firmware, conformal coated and ready to install. Just hook up the wires, put in some type of case, and set the DIP switches. For many installs that is all that is needed. More advanced configs is done via a USB cable and a serial terminal program on your computer.

Of course if folks wish to do the whole hand-soldering, bootloader flash and firmware compile/install - I do have some blank PCBs hanging around

To your other point: (If I understood it correctly) --> One capability is a feature-in port which is begin used by some LiFeP04 installs to allow an external BMS to signal to the regulator to go into 'float' mode (which can be different things, depending on how the regulator is configed - ref above post). Others are letting the regulator make the transition decisions and using feature_in as a backup in case something goes wrong.

My goal for this design was a reliable, accurate, and flexible alternator regulator. There is more, but not sure how appropriate it is to get into details here... Am happy to answer questions.

[john61ct]

Originally Posted by thomasow

One capability is a feature-in port which is begin used by some LiFeP04 installs to allow an external BMS to signal to the regulator to go into 'float' mode (which can be different things, depending on how the regulator is configed - ref above post).

The ideal would be for that signal to trigger ending all charge input to the LFP bank's charge buss.

If no loads, output is no longer needed, then zeroing out field current to the alt.

If loads are active would be nice for the alt to carry them, without putting any more current to the bank.

Can you see a way to set up your SAR to do that, perhaps in conjunction with something else available off the shelf?

Ideally, the same logic would be applied for any charge sources present, but I realize that is a different project?

[exMaggieDrum]

Originally Posted by senormechanico

How about replacing that "relay that needs the constant drop to ground to open" with a bistable relay, i.e. pulse the winding to open, pulse to close?

My boat has a relay like that for a 3 way light in the saloon. On at one end while passing thru, and off at the other end while exiting. Momentary contact switches on the bulkheads kick the relay.

You could have an automatic "stop charging" via the BMS signal and a "start charge" button when initiating charge.

That would be great and solve the problem. I haven't seen an off the shelf device that will do that yet. I got the House Power BMS and it won't do that. I haven't seen a relay that will do that. It has to trip on a voltage level event.

[exMaggieDrum]

Originally Posted by john61ct

The ideal would be for that signal to trigger ending all charge input to the LFP bank's charge buss.

If no loads, output is no longer needed, then zeroing out field current to the alt.

If loads are active would be nice for the alt to carry them, without putting any more current to the bank.

Can you see a way to set up your SAR to do that, perhaps in conjunction with something else available off the shelf?

Ideally, the same logic would be applied for any charge sources present, but I realize that is a different project?

Not sure what a SAR is so can't answer that part. I have a House Power BMS that is basically a fail-safe if my alternator or solar regulators fail and don't stop charging. I will program them as best they can be and will shut them down manually with an alarm circuit to tell me if voltage gets too high. I won't be running the alternators except when I am there and do not intend to just run the solar all the time. Most people just put a contactor that can open on a connection to ground via a BMS.

There is a separate solid state AC relay to stop AC input from my inverter/charger on a HVC.

I wanted to have a main battery remote switch that I can open/close with a separate switch rather than a contactor. Much less power consumption just sitting there as opposed to an always on contactor. It would be the main battery switch so would shut off everything in and out of the house battery. I could reconnect with a pulse switch separate from the BMS after I shut down the faulty charging circuit. I have a separate contactor to shut off loads and it works on a closed ground if LVC. I realize it will have some draw to stay closed but haven't figured out a way to avoid that. It's a fail-safe too and should remain closed normally to keep loads powered.