Charging my battery

When I set my battery to charge during off-peak does going for 100% shorten the battery life?

Mjobe From everything I've read the battery will manage that quite nicely, however how it gets there so that you keep the heat in the battery and the inverter down is more important. I try to aim at a charge rate of about 2000w, so in your case a 4 hr charge window right at the end of your off-peak time is probably a good idea. That way you maximise the off-peak period use and keep the maximum for the peak period

Interesting and helpful, thank you.

The ideal for lithium-ion batteries (of any sort) in terms of prolonging cell lifespan is to keep the charge level between 20% and 80%. Never charge at full rate when cell temperature is below 5C either & don't charge at all if its below 0C.

In theory the gen2 batteries are overprovisioned/setup so when the SoC reads zero then there's actually 15% capacity remaining and when at 100% its actually 95%. I say in theory because firmware version 3013 clearly takes the battery cells WAY below 15% capacity so I am unconvinced this is actually still true (Givenergy said that's the way it worked in 2020)....

The GivEnergy batteries are LFP and as far as I'm aware those recommendations don't apply to LFP.

TX200

TX200 The GivEnergy batteries are LFP and as far as I'm aware those recommendations don't apply to LFP

Correct. In fact LFP need frequent 100% charges.

TX200 They're still lithium-ion batteries so yes they do apply.

LiPO4 batteries aren't magically exempt from physics any more than any other lithium-ion chemistry 🙂

For example on a LiPO4 cell, 80% DoD will result in around 50% more charge cycles than a 100% DoD before the cell hits 70% capacity. 60% DoD is around 250% more cycles and so on.

The fact that LiPO4 cells can do the same number of 100% DoD cycles as a NMC cell can do for 60% DoD cycles and still have more capacity left doesn't mean you should treat it differently in terms of preserving lifespan.

Zakalwe No they don't.

They might require that with the absymal givenergy SoC calculations in order to be anywhere close to reality but these aren't nicads and have no requirement to be regularly discharged.

For anyone wanting further reading then https://www.researchgate.net/publication/303890624_Modeling_of_Lithium-Ion_Battery_Degradation_for_Cell_Life_Assessment seems to be available for free. Its not what you'd call "easy reading" though 🙂

Edit - one of the tl;dr's in it is that the shortest lifespan is found in cells which were fully charged then discharged at maximum rate to 25%, then recharged to 100% at max rate, rinse/repeat. I suspect more than a few people who are doing max rate charge/discharge for export reasons are going to be well under 80% capacity on their GE batteries before 10 years have gone by. The thought occurred that the compulsory "health checks" on the new GE 12 year warranty T&Cs may have something to do with the way this market is developing in the UK. We'll see.

Oh and the above should obviously have said LiFePO4 - LiPO being the usual cell found in phones/laptops etc. I'd blame autocomplete except I didn't have it on 😉

all GE batteries are however limited to 0.5C which does help longevity.

hoggy Yeah I do understand there's lots of mitigation/design choices and GE will have done the maths in terms of maximising battery throughput during the warranty period while retaining capacity etc. I have to say I wonder if whoever did the maths all those years ago thought people would import energy to export sometimes minutes later as well as use the battery to supply the house. Probably not so much 🙂

The point is that LFP batteries are just a variation on the batteries we're all used to in phones etc. They're not a magic bullet & all they really do is trade energy density for increased cycles (and not start unmanageable fires as there's no cobalt). Same rules apply in terms of best temperature/SoC/charge rate for maximising lifespan....

If it helps my pack has done 13.99MWh (so close!) Throughput and still does the same 6.5kWh in & out it always has since installed 3 years ago.

hoggy Which shows it likely has a buffer of some sort, either firmware/software or hardware (over-provisioning of cells). If its the 8.2kW battery then doesn't that reserve the lower 15% and upper 5% in firmware?

For example you'll get SEI formed on the electrode for the first couple of dozen charges - that will permanently reduce the cell capacity by around 3-5%. Pretty easy to make that disappear in firmware/software, not so easy in hardware if someone is counting in/out.

Given the recent decision to allow a class action against Apple in the UK for "historically" manipulating battery data/device performance you can see it is/has been a widespread issue across lots of devices.

Vestas Yeah it's an old 8.2 with 80% DOD before they changed cells.
Buffers, yes they mostly do 15%-95% across the range of sizes. None are truly 100% DOD but they deleted the explainer on that for some reason?
I'm not sure if it's a reporting thing or the 9.5 packs run slightly different DOD but I've never really seen those report much more than about 200AH actual capacity (vs 186AH/9.5kWh usable) so those are the only ones that are a bit odd as that would make the buffer much smaller.

Edit - until the registers are released for GivTCP the AIO buffers are just based of the sticker on the packs (15.972kWh, 13.6kWh usable so about 85% DOD)

hoggy I suspect they deleted the "explainer" as its no longer true -assuming its the 24/11/20 one?

If they are actually doing 100% DoD* on the Gen2 9.5kW batteries then that's not ending well.

*3013 is definitely doing 100% DoD - I've had to setup automations to stop the battery going down to below 45V (less than 2.8V/cell) as the firmware is simply a complete disaster. No other description fits. Hell it even did a SoC crash when a full calibration was run - that lasted 4 days then another SoC crash, same SoC figure the calibration crashed. Total garbage.

NB - the figures in/out for the battery seem reasonable on a day to day basis but the SoC is as reliable as a chocolate teapot.

Vestas They might require that with the absymal givenergy SoC calculations in order to be anywhere close to reality but these aren't nicads and have no requirement to be regularly discharged.

Wrong.
Lithium iron phosphate batteries work differently from NMC batteries. Even Tesla recommend 100% charging on their LiFePo powered cars.

Show me the technical data which says you have to regularly fully discharge/charge LiFePO4 batteries.

You can't & you know it.

"“If your vehicle is equipped with an LFP Battery, Tesla recommends that you keep your charge limit set to 100%, even for daily use, and that you also fully charge to 100% at least once per week.”"

https://insideevs.com/news/557527/tesla-model3-lfp-charging-recommendations/

"According to Tesla, following the above guidance maximizes available range and improves the vehicle’s ability to accurately determine the state of charge and estimated range."
So it's not necessarily about best care for the battery, more about the best experience with your car.
Specifically, Tesla probably don't want regular SoC crashes when the BMS lost track of how much power is in the battery ! Because of the flat voltage curve for this chemistry, it's harder to infer SoC from voltage.