Lithium's Not Fully Charging

[TimH]

I've got a bit of an odd problem with the Lithiums in the Tvan ( 2 x 90Ah Fusion batteries installed by TT as part of the new build).
They seem to be only charging to 66% of their capacity.
Not every time. Sometimes they will charge to their full capacity.

I thought it may be the 240V Projecta charger since it doesn't have a Lithium charging profile, just AGM/lead acid.
However the Fusion batteries are supposed to be able to be charged by an AGM charger.

Last time I was able to get them to accept a full charge, it was a combination of Projecta 240V charger + solar.
However just tried repeating the experiment and it cut out at 66% charged again.

Lithium charging is a bit different to "normal" battery charging. They charge at a constant current with rising voltage till they are charged then the current drops away to zero.
I'll have a couple of runs towing the Tvan over the next couple of days so I'll see if being charged via the BCDC1240D makes any difference.

Just wandering if anyone has struck a similar issue with their Lithiums?
Tim

[leachy_9]

I've heard other similar reports of Fusion Lithium batteries not charging properly using standard type chargers, despite what Fusion advertise. Even if you did manage to get them to charge from a standard charger, it would be significantly slower than via a lithium specific charger, negating one of the major advantages of lithium being their fast recharge.

[Nomadic]

I have had this before. The reason is sometimes the coulomb counter (which measures amps in & amps out to give you state of charge) gets confused and has to be reset. It also happens if you have auxiliary solar that doesn't go through the BMS (coulomb counter). I'm only familiar with Victron and BEP gear which lets you do a reset back to 100%. You may find it in your manual.

Your LiFePo4 batteries may be fully charged and the charger has cut out but the counter hasn't measured all the amps going back into it - hence its showing less than 100%

It can also happen in reverse if you connect loads to the battery without going through the BMS. It won't be counting the amps out and therefore won't be showing the true state of charge of the battery.

[TimH]

Yep I've been trying to refine the Peukart's and efficiency settings on the Victron to get a reliable indication of state of charge.
I've also been trying to get workable parameters to have it automatically reset to 100% SOC when fully charged.
Easier to do under controlled conditions such as when charging with the 240V charger. Harder to do when charging off solar.
I've been thinking I will probably have to rely on doing a manual reset when I know the batteries are fully charged say after a long drive.

So when I have been testing the batteries I have tried not to rely on SOC on the Victron.
I have tried to use constant current loads and time to determine Ah consumed.
I have also tried to use voltage measurements compared to charge and discharge curves to validate what I think is happening.

So when I "fully" charged the battery and then ran a fixed load over time it confirmed I could only get around 110Ah from the batteries, before the voltage drops away, instead of the expected 180Ah.
Similarly when the batteries have actually been fully charged I have been able to confirm that I could get at least 155Ah and the voltage was still around 12.8V indicating I still had some capacity left.

Every chance I may have got myself confused with all of the testing though  .

The Victron is supposed to measure everything that is going in and out of both batteries but that assumes it has been wired correctly.
I'll have to do a few more cycles to try and nail down what's happening and maybe trace some of the wiring to check how its been put together.
Tim

[Nomadic]

Tim you are definately on the right track with your testing. What charge efficiency have they got the Victron set at? I presume the Puekert factor is at 1 (or close to it)?
If you are only getting 110ah before the voltage starts to drop away (under no/low load conditions) then your batteries definately weren't charged to their stated 200ah capacity. If this is the case it's either the Fusion internal BMS or the battery charger cutting the charge short.
Unfortunately the integrated battery /BMS makes it harder to fault find.

I have done several Marine LiFePo4 installations now using the Winston cells with external BMS so we can easily test the individual cells. Our first conversion is still going strong after 8 years. We lived aboard for two years and the coffee machine, two fringes and a freezer gave them a good workout.
I love them but it takes a while to get your head around their quirks.
As you are finding out they don't behave like the AGM and you can't easily determine  SOC with voltage. The very flat discharge curve means you can only determine when they are nearly flat. The high voltage also fools some smart chargers into thinking that the battery is full so puts it into 'float' mode.
I would continue testing and even try a different charger. I guess if your Tvan is new I would shift the problem back to TT and Fusion. You should definately be able to get 180ah out of a 200ah bank.
I'm happy to take a call if you want to discuss. PM for contact details.
Cheers
Allan

[TimH]

Thanks Allan

Just heading off for a few week's camping so I'll see it goes with being only being charged via the Redarc BCDC.
This will mean I'll be working off a different charger to the 240V one so it may narrow things down a bit.

Only problem is trying to actually work out how charged the batteries are without running them down till the voltage drops away. Not ideal for keeping the beer cold when you are in the field.
So I'll just have to be a bit cautious with the power and resume serious testing when I get back.

I'll be pointing this back towards TT to fix under warranty however I am trying to work out what the problem is first.
I think there is limited Lithium expertise around so if it goes to a random auto electrician it may not get resolved straight away.

With the Victron settings I had started at their suggested setting of 1.05 for the Puekert but that seemed to still produce SOC errors so was working my way back towards 1.
Definitely a different kettle of fish to AGMs.
Tim

[VK3GJM]

In reality, no single Lithium battery profile and onboard BMS is the same. I would advise not to parallel 2 Lithium batteries unless both BMS can talk to each other to share battery charge state and equalisation. The issue might be that one battery is being charged and the other starved a little each time therefore both battieres never see 100%SOC after a good discharge.

I suggest a good Lithium battery charger like the projecta IC5000L or similar and charge each one seperate and use that as a baseline moving forward.

The alternative is replace both 100a/h with a 200a/h, you will save extra kg and no inter BMS hunting issues.

I do a lot of high end 48vdc 100a/h LiFePo4 installs, I am talking about 2000a/h+ Per bank. Each sub rack within the massive bank has a data bus (RS-485 CAN bus) to ensure all batteries share SOC and internally they all balance according to each internal battery sub-assembly need and profile rather than one that may have a slight imbalance and lower internal resistance and needs more/less time to charge.

Best of luck.

Gerald

[TimH]

Yep I can see why having two Lithium batteries with separate internal BMS is not ideal.
However given the space constraints in the Tvan it would be difficult to fit a 200Ah battery so I will have to put up with any imbalance between the batteries.

After a recent holiday the battery's charging seems to have come good for the moment. They allow discharge of at least the full 180Ah and they fully recharge.
A bit annoying since I don't know what changed and therefore whether the problem will return.

It's not all good news though.
The Redarc BCDC is throttling the charge rate down to float charging (around 6A instead of 30A) when the batteries are only around 75% charged.
This is when they are charging from the car while driving so it means that I'm not getting a full charge unless I am driving for quite a few hours.

Redarc have asked me to check how the BCDC is wired up and look for poor connections etc in case the BCDC is not reading correct voltages and is getting confused.
So back to poking around in the electrics to see what's going on.

The Lithiums are certainly good at delivering a lot more power than I could reasonably expect out of AGMs so happy with the concept.
I just want to optimise the charging to get the best out of them.
Tim