Charging characteristics of my (house) LiFePO4 battery.

[pcatterall]

This is connected with my 6KW battery in my house rather than on the boat but I reckon the answers will have relevance for marine use.
Our new solar system is working well and has supplied most of our electricity in October.
Increasingly, though, we will have to import from the grid and are on Bulb's EV tariff (0200-0600) which gives us electricity at 10p/kwh rather than the standard 35p .
Getting the timing set for the battery to be charged does not seem to be easy for the user (though I am investigating that)
The installer has set the '3.3kw inverter' to charge the battery for just 1 hour in the off peak period which brings the battery from 50% to 100% charge.
As winter draws on we will need to get a full 6kw charge into the battery.
My question is... if we change the charging period to say 2 hours now will the inverter be working hard for longer that is required trying to shove more power into a full battery.
Sorry if my question is not 100% boaty !!

 

[Poey50]

This is connected with my 6KW battery in my house rather than on the boat but I reckon the answers will have relevance for marine use.
Our new solar system is working well and has supplied most of our electricity in October.
Increasingly, though, we will have to import from the grid and are on Bulb's EV tariff (0200-0600) which gives us electricity at 10p/kwh rather than the standard 35p .
Getting the timing set for the battery to be charged does not seem to be easy for the user (though I am investigating that)
The installer has set the '3.3kw inverter' to charge the battery for just 1 hour in the off peak period which brings the battery from 50% to 100% charge.
As winter draws on we will need to get a full 6kw charge into the battery.
My question is... if we change the charging period to say 2 hours now will the inverter be working hard for longer that is required trying to shove more power into a full battery.
Sorry if my question is not 100% boaty !!

There are quite a few posts on LiFePO4 (LFP) usage (including in the linked thread, below) which emphasise the principle of not leaving the LFP pack at full charge and, for a longer life, to avoid routine full charge and full discharge - with, say, 15% - 90% stage of charge being preferred.

 

[kwb78]

There isn't too much issue with using the majority of capacity of an LFP battery. Fully charging it is fine if you are frequently using it and the charger accurately cuts off bulk charging at the set level. You can increase cell longevity a bit by reducing the charging voltage from 3.65V per cell, which is a complete 100% charge, to 3.6v or even 3.55V without losing much in the way of usable capacity. Setting the inverter low voltage cut off conservatively will help with longevity as well - using something like 3V per cell will stop the battery going below about 10% SOC. Leaving LFP batteries fully charged is more of an issue if they are to be stored for a long period without being used, of the order months at a time rather than a few days or so. This does assume that the battery is LFP rather than something like LiPo or some other lithium chemistry which will have different cell voltages and behaviours and should be treated accordingly.

My question is... if we change the charging period to say 2 hours now will the inverter be working hard for longer that is required trying to shove more power into a full battery.

What type of charger do you have, and how customisable is it? If you are charging at 3kW (50%-100% in an hour for a 6kWh battery) then that's a charge rate of 0.5C which is no problem for the battery at all. If you want the battery to be full at the start of the day, then I would just set the charger to come on for the duration of the cheap rate - once the battery is full, it should stop anyway. Anything other than the most basic charger should behave properly when the battery is full and not try and overcharge it.

If you want a bit of capacity available for solar storage from the morning, then you would need to limit the mains charging - some chargers can monitor the battery state of charge and be stopped at a selected level, but others you might have to switch off externally. The optimum way to charge will depend on what power you use and when, and how much solar output you are likely to get. For example if most of your consumption is in the evening, you might want to have a bit of battery capacity available for the solar panels to fill rather than paying for it through overnight mains charging. If you use power throughout the day, you may find the solar panels do not keep up with your usage anyway and the battery supplies the extra. There's lot's of variables there, so you might need to do some experimenting to see what works best.

 

[pcatterall]

There isn't too much issue with using the majority of capacity of an LFP battery. Fully charging it is fine if you are frequently using it and the charger accurately cuts off bulk charging at the set level. You can increase cell longevity a bit by reducing the charging voltage from 3.65V per cell, which is a complete 100% charge, to 3.6v or even 3.55V without losing much in the way of usable capacity. Setting the inverter low voltage cut off conservatively will help with longevity as well - using something like 3V per cell will stop the battery going below about 10% SOC. Leaving LFP batteries fully charged is more of an issue if they are to be stored for a long period without being used, of the order months at a time rather than a few days or so. This does assume that the battery is LFP rather than something like LiPo or some other lithium chemistry which will have different cell voltages and behaviours and should be treated accordingly.



What type of charger do you have, and how customisable is it? If you are charging at 3kW (50%-100% in an hour for a 6kWh battery) then that's a charge rate of 0.5C which is no problem for the battery at all. If you want the battery to be full at the start of the day, then I would just set the charger to come on for the duration of the cheap rate - once the battery is full, it should stop anyway. Anything other than the most basic charger should behave properly when the battery is full and not try and overcharge it.

If you want a bit of capacity available for solar storage from the morning, then you would need to limit the mains charging - some chargers can monitor the battery state of charge and be stopped at a selected level, but others you might have to switch off externally. The optimum way to charge will depend on what power you use and when, and how much solar output you are likely to get. For example if most of your consumption is in the evening, you might want to have a bit of battery capacity available for the solar panels to fill rather than paying for it through overnight mains charging. If you use power throughout the day, you may find the solar panels do not keep up with your usage anyway and the battery supplies the extra. There's lot's of variables there, so you might need to do some experimenting to see what works best.

Thanks for that! The charging is via a growatt3600 hybrid inverter. The dashboard indicates up to 100% charge at times but that may just be manufacturers 'speak' to mean it is charged as fully as their settings require.
There is a massive disclaimer warning me that I enter 'settings' at my own risk and i rely ( at this time) on the installer to alter settings. My instinct is that the manufacturer Will have put settings in which maximise battery longevity. I guess that November onwards will be the time when a full off peak charge will be needed most days, anything extra from solar can go back into the grid for a bit of dosh.
Thanks again

 

[kwb78]

I had a quick look at the manual for that inverter and it is pretty customisable - enough that a cursory glance is nowhere near enough to understand the implications of adjusting anything in it, so I won't tell you what you should or shouldn't do with it as there's a lot in there with the potential to mess things up. It does look like there are options for limiting the SOC of the battery however, so it should be possible to do that without any external control.

It may not be worth worrying about if you can buy at 10p a unit to charge though - if you are getting at least that amount per unit for exporting then you might as well export what you can. If the amount you get back is less than it costs to charge, then it benefits you to use as much of the power you generate yourself. You'd need to get some idea of how much you can expect to generate and your likely usage during the day to know what capacity to keep available. As the battery is relatively small, the numbers aren't going to be huge either way so it may not be worth the complication. Might make an interesting exercise if you like playing with data though.

 

[William_H]

A bit of fred drift (appologies) about domestic PV .
Raises question is OP house 3 phase or single phase power input? If 3 phase does he have a 3 phase inverter?
ol'will

 

[sailaboutvic]

After doing quite of work with LIFEPO4 and solar on our boat , I be looking into turning my house once we move back in to partly run on batteries and solar .
Good luck with your new system Paul.

 

[GHA]

There isn't too much issue with using the majority of capacity of an LFP battery. Fully charging it is fine if you are frequently using it and the charger accurately cuts off bulk charging at the set level. You can increase cell longevity a bit by reducing the charging voltage from 3.65V per cell, which is a complete 100% charge, to 3.6v or even 3.55V without losing much in the way of usable capacity.

Any links to the research where these numbers came from please?

 

[Poey50]

Any links to the research where these numbers came from please?

While much in LFP technology is still relatively new, there seems to be a big consensus around the above figures fed by multiple sources.

 

[pcatterall]

While much in LFP technology is still relatively new, there seems to be a big consensus around the above figures fed by multiple sources.

I have followed (or at least tried to follow) your very interesting thread on LFP builds. There may be a parallel to be drawn in that, on the one hand there is a group of very knowledgeable DIY experts who get down to and understand the nitty gritty of emerging systems (and often spear head them!!) and those like me who prefer to buy off the shelf. There are similar groupings when it comes to private house solar. I bought off the shelf, partly
because of my lack of knowledge but also because I had to secure the MCS certification. When I go on line, though, to improve my knowledge I can only find DIY experts and find it hard to understand it all.
I found that many installers who I approached for quotes may be great at fitting a system but have little understanding of how the ability to change settings as the seasons change is important (or how to do it !!)
Not sure if there is a similar issue on 'oven ready' boat systems.

 

[Poey50]

I have followed (or at least tried to follow) your very interesting thread on LFP builds. There may be a parallel to be drawn in that, on the one hand there is a group of very knowledgeable DIY experts who get down to and understand the nitty gritty of emerging systems (and often spear head them!!) and those like me who prefer to buy off the shelf. There are similar groupings when it comes to private house solar. I bought off the shelf, partly
because of my lack of knowledge but also because I had to secure the MCS certification. When I go on line, though, to improve my knowledge I can only find DIY experts and find it hard to understand it all.
I found that many installers who I approached for quotes may be great at fitting a system but have little understanding of how the ability to change settings as the seasons change is important (or how to do it !!)
Not sure if there is a similar issue on 'oven ready' boat systems.

I suspect many professional installers are just following a template that someone else has devised as relatively foolproof, so the installer doesn't have to hold much specialist knowledge. That's perfectly fine until something goes wrong.

The same may be true in boat installations with the additional problem that there is, as yet, no clear template. I watched a YouTube video about three years ago of someone highlighting their professionally installed system on a catamaran before setting off to go round the world. It certainly seemed a smart set-up. The next I heard of him was on the Lithium Batteries on a Boat Facebook Group. He was stuck in the Philippine's with very little power, only enough to power up his mobile phone to message the group. He assumed that he was repeatedly getting a low voltage cut-off and that his pack couldn't hold any charge. The professional installer had no idea what was wrong and basically washed his hands of the whole situation. In fact most of the cells were charged but an imbalance had development between cells because of some poor installation practice. His BMS was programmed to trigger a cut-off when the voltage difference between cells (delta reading) rose above a certain point. He had no manual and the installer had not included an add-on that allowed user-adjustments to be made. In the end the problem was temporarily solved by him getting a car headlight bulb, soldering on some wires and manually bleeding off power from the higher cells. Since then I have always carried a bundle of resistors to do the same and a 3.65 volt, 5 amp charger to boost a lagging cell.

It's not a popular viewpoint but, a) the best DIY boat installations of LFP starting from raw cells are better than anything yet available off-the-peg (other than high-end integrated systems) including most drop-ins. And b), DIY is complex and best carried out by owner-nerds!

 

[GHA]

While much in LFP technology is still relatively new, there seems to be a big consensus around the above figures fed by multiple sources.

Agree, but little or no data confirming, reliable data is nice ? All the data I can find so far is lab results. Maybe just too early as they last so well.
Rod Collins did a report somewhere of his long term bank which was holding up very well, can't remember his charge profile though. in Google somewhere.

 

[Poey50]

Agree, but little or no data confirming, reliable data is nice ? All the data I can find so far is lab results. Maybe just too early as they last so well.
Rod Collins did a report somewhere of his long term bank which was holding up very well, can't remember his charge profile though. in Google somewhere.

I can't recall it either but, in general, the consensus on target voltage for charging has come down considerably over the years based on experience. 14.6 volts was common at one time, now many - including me - use 13.8 volts. Anything above 13.6 volts will eventually fully charge LFP. A lower figure gives a longer time for the BMS balancer to do its job.

 

[kwb78]

Any links to the research where these numbers came from please?

As mentioned above there are many sources, but as an example here is the data sheet for some LiFePo4 cells manufactured by EVE, as they are good quality cells that are quite typical: Datasheet.

The standard charging current is given as 0.5C at 3.65V, with cut off at 0.05C. Standard discharge cut off is 2.5V.
The tested capacity using repeated charge and discharge cycles at 0.5C is 6000 cycles at 25°C - that's over a 16 year life before they drop to 80% of their original capacity if you do a complete charge and discharge cycle every day. For most use cases they won't get used anywhere near as hard as that, so are more likely to die of old age than capacity loss.

Here is a graph of recorded state of charge against cell voltage (not mine, but it's basically the same as data I've gathered myself):



You can see that the cell voltage changes very little between about 12% and 96% state of charge - only about 0.15V for the majority of the capacity of the cell. If you reduce the charging voltage from 3.65V to 3.4V, you are still charging to about 97% capacity while avoiding the higher stress portion of the charging curve - in practice as long as you don't go over 3.65V and use a charger that cuts off and that doesn't try to hold the voltage there for long periods after the cell is full (eg an equalisation cycle on a lead acid charger), there should be minimal effect on capacity. The BMS should take care of protecting from higher voltages.

The datasheet recommends a nominal SOC range of between 10% and 90% for maximum life, which would mean charging to around 3.3V with a low voltage cut off of around 3.1V. I don't think there's any need to be that conservative though.

I have followed (or at least tried to follow) your very interesting thread on LFP builds. There may be a parallel to be drawn in that, on the one hand there is a group of very knowledgeable DIY experts who get down to and understand the nitty gritty of emerging systems (and often spear head them!!) and those like me who prefer to buy off the shelf. There are similar groupings when it comes to private house solar. I bought off the shelf, partly because of my lack of knowledge but also because I had to secure the MCS certification.

That's a perfectly valid approach - not everyone has the time or inclination to learn how to do it themselves. There's a lot of waffle posted in various forums about what you should or shouldn't do with LiFePo4 cells and much of it is people overthinking things. As long as you recognise that they are not the same as lead acid batteries and treat them accordingly, stick to the manufacturer's design spec, and use good quality chargers and BMS they will last a very long time.

 

[sailaboutvic]

I can't recall it either but, in general, the consensus on target voltage for charging has come down considerably over the years based on experience. 14.6 volts was common at one time, now many - including me - use 13.8 volts. Anything above 13.6 volts will eventually fully charge LFP. A lower figure gives a longer time for the BMS balancer to do its job.

I always had mine set at 13.8v then dropping to 13.5v and once a month I would rise that to 14.1 v so they had a good balance.
where always within 0.02v at 14.1.
if as I did to test which I did a couple of times then set it to 14.4v once they reached 14.2v I found cell 4 would start to stray .
The same happen to a friend set up which I put to gather for him ,
personally I think that was to do with the Smart 123 bms as since then others who used the same BMS he found the same problem.
@13.8v I consider that about 97% but even so they would not be there for long as they where being use all the time.