Extending battery life by periodically limiting SOC

[berksrunner]

Hi Folks, I have a self-converted promaster van which has 900W of PV on the roof and 6.2 kwh (24v) of diy lifepo4 storage. The van goes through periods of non-use, and during those times, it makes no sense to rev the SOC to 95%+, particularly in hot weather; it does nothing but significantly shorten the overall cycle life. What I'd like to do is effectively limit the max SOC to somewhere around 70% when the van is not in use, and I'm trying to figure out the best way to do it (and it'd be nice if it was a way that could easily be disengaged when it comes time to use the van and then re-engaged once the van goes back into storage).

Here's a description of what I've got available to work with:
Smartsolar 150/70
BMV 702
Overkill Solar BMS (3x)

I've considered the relays built into both the BMV and the mppt, but I'm not sure exactly how to use them. I could also manually set a lower voltage (maybe around 26.3) on the mppt, or I could do something similar to the BMS's (but I'd need to do it 3 times since I have multiple).

Any thoughts? TIA!!

 

[snoobler]

During periods of non-use, it makes sense to keep the batteries stored at 50% and to not charge them at all.

Best way to do it would be to turn off the "charger enabled" in the battery settings in the Victron Connect app and turn it back on a day before planned use - one simple toggle in your phone app.

Simply reprogramming the absorption and float voltage to a lower level will ensure the LFP battery is at a lower SoC. You might consider even making the change permanent as a lower peak SoC will improve cycle life as well. You will have to experiment with what peak voltage produces the best results for you.

 

[berksrunner]

During periods of non-use, it makes sense to keep the batteries stored at 50% and to not charge them at all.

Best way to do it would be to turn off the "charger enabled" in the battery settings in the Victron Connect app and turn it back on a day before planned use - one simple toggle in your phone app.

Simply reprogramming the absorption and float voltage to a lower level will ensure the LFP battery is at a lower SoC. You might consider even making the change permanent as a lower peak SoC will improve cycle life as well. You will have to experiment with what peak voltage produces the best results for you.

I guess I should have included some additional details. When my van is not being used, I nonetheless like to keep security related items going (some wifi cameras that notify me of any motion around the van), as well as lights (I just keep the LED strips on 24/7), fan for temp control, the fan for the composting toilet, etc. I will even sometimes keep the electric water heater going. Therefore, I do need to continue cycling to a degree -- I just don't want to cycle it all the way up to near 100%.

 

[snoobler]

Then simply lower your absorption/float voltages to something that gets you to a lower peak state of charge. Maybe start with something around 13.2V and see where that gets you.

 

[softdown]

If cycle life is what you seek a 40 to 80% SOC range will get you there.

 

[berksrunner]

Then simply lower your absorption/float voltages to something that gets you to a lower peak state of charge. Maybe start with something around 13.2V and see where that gets you.

Thanks! That's kind of what I have done in the past, but just wanted to see if my thinking was in line with the group.

 

[berksrunner]

If cycle life is what you seek a 40 to 80% SOC range will get you there.

Yup that's where I want to be although I don't want to get so low that I'll hit low voltage shut off in the event we have consecutive rainy days.

 

[Forbisher]

If cycle life is what you seek a 40 to 80% SOC range will get you there.

Why buy 100% capable LFP if you are only going to use 40% capacity?
Yeah, don't discharge lower than 3V per cell and maybe don't always charge to 100% but limiting to 40% is kinda wacky.

After 5 to 10 years use there will probably be better and cheaper batteries.
It's not like berksrunner is using his batteries every day so they will last a long time anyway.

 

[nebster]

Yeah, float them at lower voltages until you see the SOC you’d like to achieve.

Don’t listen to the folks advocating for holding packs at 100% all the time. It makes sense to take simple steps to (maybe) get even longer life out of the cells.

 

[softdown]

softdown said:
If cycle life is what you seek a 40 to 80% SOC range will get you there.

Why buy 100% capable LFP if you are only going to use 40% capacity?

The whole point of the sentence is contained in the first few words regarding cycle life. I got the "40 to 80%" idea from one of Will's podcasts - where he suggested that such cycling of lithium batteries could result in decades of use. Will kind of works with batteries for a living - last I checked. And you?

 

[Gazoo]

Yeah, float them at lower voltages until you see the SOC you’d like to achieve.

Don’t listen to the folks advocating for holding packs at 100% all the time. It makes sense to take simple steps to (maybe) get even longer life out of the cells.

I think this is the best advice and what I am planning. Everyone knows most manufacturers rate the max cycle life of a LifePo4 cell at 2000 cycles with 80% capacity left. This is with full charge/discharge cycles. We also know keeping the cells at a full state of charge will reduce the capacity of the cell over time. Everyone's usage will be different.

Getting close to the float voltage sweet spot for SOC is what I am going to try to do. I don't know exactly what that will be yet, but it will be between 80% to 90% SOC. Since my usage will be for a UPS, in the event of an extended power outage, I have no problem letting the cells drain down to 3 volts per cell.

I can't put solar into the equation at this time because I am in an apartment.

 

[nebster]

Yeah, the exact voltage is really dependent on the things in the circuit between your battery and the thing measuring the voltage (e.g., wires, fuses, relays, BMSes), and then the thing measuring the voltage has some error in it, too. Since even a few millivolts can yield a pretty different float SOC, it’s really not possible to prescribe a single magic voltage for everyone to use when targeting something in the flat part of the voltage curve.

Instead, you just adjust it, iteratively, over time, gradually tweaking and observing your battery monitor’s count/SOC. (Remember to recharge back to 100% occasionally so the counter can reset/resync to a known good point.). You’ll quickly be able to converge on something you like. And it will probably end up being a little bit different voltage than the next guy’s, even for the same SOC.

 

[Gazoo]

(Remember to recharge back to 100% occasionally so the counter can reset/resync to a known good point.).

Yeah that's good advice too and kinda doing that with my Valence batteries. Charging to 100% also lets the BMS balance the cells. It will be the same with most of the passive BMS's. The coulomb part of my cheap Juntek meter is fairly accurate from what I can tell. The capacity indicator I am not so sure of. I need to do some more testing with it. Seems to be a problem with most of the cheap Chinese meters. But for everything else it's great. From what I have read here the Victron meter is accurate in all respects. And it's fine if one has the money to invest in it. But I don't. So I may get another Juntek for this build....lol

 

[toms]

The whole point of the sentence is contained in the first few words regarding cycle life. I got the "40 to 80%" idea from one of Will's podcasts - where he suggested that such cycling of lithium batteries could result in decades of use. Will kind of works with batteries for a living - last I checked. And you?

Last i checked, Will was an amateur DIY’er who has been playing around with lithium batteries for a couple of years.

I’ve worked closely with engineers in designing Lithium cells for over a decade- my own off-grid system has been in use for over 8 years cycling from 20-80%.

Always fun to get into a pissing contest, but better to stick to facts backed up by real experience. Basically unless you have been using LiFePO4 for over a decade it’s worthwhile looking at the people who have and noticing how they have been managing their cells.

 

[rbreding]

It sounds like you have remote connectivity to your van?

On our trailer that is in storage I leave the SCC off for around 10 days. The various devices that I leave on along with the exhaust fan on a timer consume around 24 watts. Once I get a notification from the Vrm portal that my charge level is low I connect and turn on the SCC and let them charge back up to full and repeat the cycle (full being 13.9v in my config).

 

[amaclach]

In my vehicle, I either disconnect solar or pull the alternator circuit fuse to manage SOC below 100%.

 

[Ampster]

Yeah, don't discharge lower than 3V per cell and maybe don't always charge to 100% but limiting to 40% is kinda wacky.

I would agree most of the time. However the example being discussed was when the van was in storage. It is exactly what I do with my Tesla. I generally keep it charged between 40% and 80% when I am just doing daily driving. When I am planning on going more that 80 miles the next day I charge it overnight if it is not already at 80%. If I am taking a longer trip I charge it all the way.

 

[amaclach]

Has anyone used the Victron BMV712 relay function to enable/disable charge circuits? I'm thinking aloud here, but set the relay to open at 80% SOC to disable the alternator charge and close at say 70%...

My Redarc charge controller isn't configurable apart from battery chemistry...

 

[nosys70]

possibly an idea would be to add a cell in serie (possible only if you have a multi-cell battery pack.)
so instead having 8 cell for 24 volts, you would get 9.
so the charger would work 1/8 lower (12%)
then instead working in 20-90% range with same settings you are working in 8-78% range.
for a 12V 4 cells, you can get a 25% margin, but you have to check that the low voltage is not too low.

for example an LFP 3.2V is full at 3.6 , so 8*3.6=28.8
with 9 cells at 28.8, each cell is really 28.8/9= 3.2
on the low side if your system is set to 3.0V (8*3.0V=24V)
then with 9 Cells, you get 24/9= 2.66V

 

[softdown]

Last i checked, Will was an amateur DIY’er who has been playing around with lithium batteries for a couple of years.

I’ve worked closely with engineers in designing Lithium cells for over a decade- my own off-grid system has been in use for over 8 years cycling from 20-80%.

Always fun to get into a pissing contest, but better to stick to facts backed up by real experience. Basically unless you have been using LiFePO4 for over a decade it’s worthwhile looking at the people who have and noticing how they have been managing their cells.

That is open to road interpretation. This is not:

softdown said:
If cycle life is what you seek a 40 to 80% SOC range will get you there.