Understanding Lithium Batteries

[Clunegapyears]

Please don't turn this into a for and against Lithium batteries; that has been covered elsewhere. We have them and we are very happy with them. I just want a better understanding of our system, and how to manage it.

We have had LifePros since August 2018, but not actually tested them for real until this last week. We had winter 2018 under a tiled roof and didn't set off in our motorhome until March 2019. Since then, it has been pretty sunny and / or we've driven some distance.
............................................................

Three days ago ... they were close to dropping off ... I had just switched on the inverter and it didn't like it and then it sounded like the batteries were bleeping at us. We ran the engine for 20 minutes, but then the bleeping started again at midnight ... so ran the engine again. We have had 5 days of rain and very overcast skies, so I KNOW we will not have been harvesting much solar. Below is a print of the solar coming in from the MPPT 100/50 ... and I can see that it has only been bulk charging for the last 12 days, so I guess, the batteries have not been 100% full in this time ... I've highlighted the cell when the power almost went.

Questions:
If bulk charging is putting as much power into the batteries as possible, what are absorption and float charging?
Should I be looking at the voltage in the evening and if it below say 12v, should we then know to run the engine for 20 mins or so?
............................................................

A retro fit we had just before we set off on this trip was a Viktron Smart BMV 712 bluetooth battery % monitor. I've been monitoring this monitor .... and most mornings by about 11.00, it has shown 100%. On the evening when the power just about went, it was showing 70%. Even the morning after the power nearly went off, it was showing 74%. This is clearly not the case.

Question:
So basically should I just ignore the % monitor as it is inaccurate? Waste of money? To gauge how full the batteries are, should I look to see if, we have moved from bulk to absorption, float charging? Look at the voltage?

I am sorry to be back on as the technical numpty ... don't know what we'd do without you lot!
Many thanks
Katherine

 

[QFour]

Running the engine for 20 minutes is a waste of time. All you will do is upset your neighbours and wreck your engine. Lithium batteries require a totally different way of charging to standard wet cell batteries. The manufacturer of the batteries is the best person to ask. Overcharging lithium batteries is not recommended and can cause serious damage. Tracking battery voltage with Lithium cells is a waste of time because the cells will show pretty much full voltage all the way through the discharge cycle whereas a wet cell will drop voltage gradually and can be monitored.

 

[Nabsim]

My BMV works perfectly with my Lifepo4 bank so it may be you have a parameter setting not quite right? I know initially mine was showing battery’s back to 100% when I was only harvesting one or two amps. I knew it couldn’t be right as I needed to replace at least 50amps for it to be 100%. For me it was one of the parameters in the meter for solar.

Dave or Phil May be the ones to advise you though

 

[Nabsim]

Running the engine for 20 minutes is a waste of time. All you will do is upset your neighbours and wreck your engine. Lithium batteries require a totally different way of charging to standard wet cell batteries. The manufacturer of the batteries is the best person to ask. Overcharging lithium batteries is not recommended and can cause serious damage. Tracking battery voltage with Lithium cells is a waste of time because the cells will show pretty much full voltage all the way through the discharge cycle whereas a wet cell will drop voltage gradually and can be monitored.

I agree that tracking voltage on lithium is pointless but I didn’t bother with a BMV until I changed to Lifepo4. For me the amps used and SOC is invaluable

 

[Clunegapyears]

My BMV works perfectly with my Lifepo4 bank so it may be you have a parameter setting not quite right? I know initially mine was showing battery’s back to 100% when I was only harvesting one or two amps. I knew it couldn’t be right as I needed to replace at least 50amps for it to be 100%. For me it was one of the parameters in the meter for solar.

Dave or Phil May be the ones to advise you though

I hope you’re right about the settings ... would you be able to screen shot your settings and send them over .... if you’ve the same as us ...

I know in theory Lithium’s put out a steady ampage, then switch off. But in my table you can see a slight drop ...

 

[Admin]

If bulk charging is putting as much power into the batteries as possible, what are absorption and float charging?

LifePo4 batteries only use bulk charging, once it has finished the charger switches to absorption which is for lead-acid batteries, this should only be for a short time and is only used for cell balancing by the batteries internal BMS). Finally once fully charged the charger switches to float and just keeps the battery topped up.

Should I be looking at the voltage in the evening and if it below say 12v, should we then know to run the engine for 20 mins or so?

If the battery voltage is down to 12v then they are at about 5% capacity, running the engine for 20 minutes will add about 10% extra. The voltage of Lithium batteries drops near the end of the usable capacity, in the test results I have seen for RB100's the batteries don't drop below 12v until they have delivered 100Ah.

So basically should I just ignore the % monitor as it is inaccurate? Waste of money? To gauge how full the batteries are, should I look to see if, we have moved from bulk to absorption, float charging? Look at the voltage?

It sounds like the BMV has not been set up for Lifepo4 batteries. This is a common error when the fully charged voltage is not adjusted from 13.2v (float voltage for lead-acid), as a lifepo4 batteries voltage normally sits at 13.2v (unlike the 12.8v for lead-acid) the BMV thinks the batteries are fully charged and resets the state of charge to 100%. So they could be at 10% charge but the BMV would show 100%. The Peukert exponent and charge efficiency also need adjusting otherwise the unit will calculate incorrectly.

Victron, Battleborn, Relion and other battery companies all recommend the BMV-712. You will commonly see BMV-712 units fitted in professional LifePo4 installations. Battleborn test every one of their batteries using BMV-712 test rigs

My Advice:

1) Get the BMV set up correctly so that you can see the state of charge of the batteries.
2) Make sure your solar panels are clean and no shadows are being cast on them.

 

[Clunegapyears]

Thanks Phil. I didn’t have to read that twice to understand it either. Until the peukert exponent!
We are regularly cleaning panels and I do check for shadows after your video about this.

Is setting up the BMV something I could do? You know me so have an idea of my limited abilities!!!! Is it settings on the phone? Away till August!

 

[Sharpie]

My understanding is that the Victron BMVs, when used with Lithium and correctly set up for it, have no way of telling what their state of charge is unless starting with a fully charged battery, then counting current (actually Coulombs) out and in from there, and applying proprietary algorithms to deduce the efficiency losses during the process. They also need to know the true capacity of the battery, AFAIK unlike lead ones they cannot deduce this and learn it from straightforward voltage measurements.

Naturally they cannot be 100% perfect, particularly under low loads such as the small but significant parasitic drains from connected equipment, even so there will also be some drift expected over a series of incomplete charging cycles.

Furthermore I understand (but have no personal experience of Victrons, nor Lithium), that if the power trips e.g. the Lithium's internal BMS trips out to protect the cells, then the BMV assumes that it is 100% full as soon as it goes back on charge, in the absence of any other information

By most accounts the Victron is one of the best.

A totally depleted LiFePO4 12V battery is at 10V (no load). But you saw only 9.53V at one point, that might account for the beeping, warning you that this was far too low. My inverter has a beeper to warn, then cut off, set for typical lead batteries, but will take those far too low IMO. Maybe this is what you heard, I'm not aware of the batteries themselves having these.

To reset the BMV I think it is necessary to fully charge the batteries, e.g. by connecting to hookup, then the BMV knows that the battery is full and can carry on from there. This will happen automatically if the battery is periodically fully charged, but if constantly operated in a partial state of charge I would expect it to gradually drift out, in either direction.

From your data I can see that you have been getting very little input from solar at times. E.g. divide the Wh figure by 12 to get a feel for the Ah (this very approximate).

As little as 11Ah on day 3. Day 6 was the best, 94Ah. But that did still not complete the full charge, it only got up to 13.5V

Only on the days that the batteries reached 14.5 V would they had have had a chance to approach 100% capacity.

You can see that in your data. Only on days 13, 14 and 15 did the bulk phase reach this level then switch over to absorption, even then only for five minutes. Typically at least 30-60 minutes are required in absorption for LiFePO4 to complete the charge. Nevertheless they did switch over to float after that. You might want to look at your MPPT 100/50 settings to see if just 5 minutes is correct for your battery. It seems a very short time.

Battleborn for example recommend 30 minutes per 100AH of the individual battery's capacity.

See https://battlebornbatteries.com/programming-a-victron-smartsolar-charge-controller/

Some equalise the cells automatically as they charge, others require a longer absorption phase to bring up any individual cells to identical levels. Good makes will select and match the cells at manufacture to have near identical capacity, poorer ones will just assemble them, with wider tolerances, if doing any testing at-all.

Remember, this is for a perfect continuous charging regime such as on hookup, whereas a typical solar installation could be fluctuating as light levels change, so it is not quite as simple as bulk followed by absorption followed by trickle. And the controller may be fooled if say five minutes into absorption a cloud comes over, the current drops, then the sun comes out again, the controller sees the battery is once again at 14.5V assumes it is full, so drops straight into trickle. Perhaps clever ones have a memory timer and would go back into absorption until complete, perhaps not.

Technically a nominal 12V four-cell LiFePO4 is only 100% charged by 14.6V, slightly more than a typical vehicle alternator produces, but longer life is usually achieved by limiting then to a lower state of charge, say 14 V at 95%. The internal BMS should look after this. Each manufacturer has their own preferred voltage, and the BMS itself drops a little.

A big inverter for example might also easily exceed the allowable current for a single battery, if loaded to full rating. Say if the battery internally limits the current to 50A, that could only supply (or accept) approximately 600 Watts before tripping.

Depending on how your BMV is installed it may not even be able to measure all the input and output to the battery, i.e. from alternator, mains charger and solar. It depends how the shunt was connected. As a retrofit that might have required significant wiring changes to do properly, if so you should also be able to see the battery recharging when running the engine for example and judge how much say 20 minutes on tickover actually puts in. I suspect taking it for a 20 minute drive might put in more, and be much better for the engine.

 

[Okta]

My BMV works very well with my LiFePO4. As previously stated it is worth checking the BMV settings and the manual gives helpful suggestions for LiFePO4 values. It may also be worth checking the wiring of the BMV, I recently came across one where there were several connections to the negative terminal of the battery and the BMV was not registering all the loads. However, it was registering all the charging so it showed full far too quickly. Easily checked, all the connections to the negative terminal must go through the BMV.

Battery voltage on its own is no guide to depth of charge, particularly with LiFePO4 but it can also be very misleading with lead acid too. The only way to really know what is going on is to count all the amps in and out and make proper allowance for charging efficiency and the Peukert exponent. I chose a BMV because all the parameters are adjustable, unlike some others.

 

[Nabsim]

Your comments on the BMV sound about right to me Sharpie, you do have to understand a few basic ‘rules(?)’ on how they work. As I said I have found mine invaluable and as I use around 60amps a day and off grid in U.K. I needed something to help. There may be other meters do the same I have no idea, well I do know Relion market one that looks similar lol.

if you have the app Catherine you will be able to set it up okay, mine is a 700 so much more fiddly unless you add the Bluetooth module (which I soon did). I made sure absolutely everything connected to the Lifepo4 went through the shunt as the instructions say then it was quite easy to see what items actually used rather than what you think they do. Once you know your normal daily usage you soon realise if a SOC reading is telling porkies then it was just working through things. I did everything but had an error os a voltage setting I think it was on the solar setting on the BMV.

my battery’s aren’t Relion but may be same settings apart from bank size but no saying mine is yet 100% right although it seems has been perfect for a couple of months. Think it’s better if Phil or Dave advise on settings rather than me tell you something wrong. I dont mind screen shorting them and sending but I don’t have their knowledge

 

[Sharpie]

It may also be worth checking the wiring of the BMV, I recently came across one where there were several connections to the negative terminal of the battery and the BMV was not registering all the loads. However, it was registering all the charging so it showed full far too quickly. Easily checked, all the connections to the negative terminal must go through the BMV.

I had to sort out a similar poor installation on a friend's yacht. Not Victron, Lead batteries, but the shunt was wrongly wired up so that not all of the loads, or feeds, went through it. Afterwards it worked much better.

In my van it would also be quite complicated to arrange this, there are several current paths on the negative side that would not easily be re-routed and connected together at one point, to simply bolt onto the shunt, without stripping out the Electrobloc entirely.

The Electrobloc does have an internal shunt in roughly the right place, but I don't think it is compatible with say a Victron, quite different mV/A, unless that could be changed in the settings. Actually I'm not sure that the Victron shunt feeds a voltage signal to the rest, it looks as if it has some electronics built in and actually communicates digitally.

Further, here is some seemingly useful information about the characteristics of LiFePO4 cells.

How to charge Lithium Iron Phosphate lithium ion battery packs including packs with high current and High Capacity.

High capacity LiFePO4, Lithium Iron Phosphate batteries.

www.powerstream.com

Now there seems to be some confusion about LiIon, which is a generic term. For our applications, retrofit habitation batteries, Lithium Iron Phosphate is the preferable choice, currently, and seemingly the most durable.

Other Lithium Ion chemistries exist, with inferior characteristics for our application, and potentially worse safety characteristics. Properly managed in e.g electric vehicles, they can be superior, better energy to mass ratio and mass produced in Gigafactories or elsewhere.

In all of them ions flow in a solid electrolyte so they are all technically some sort of Lithium Ion.

If considering these, be quite certain that you are choosing a Lithium Iron (Phosphate) battery, not something quite other.

 

[Admin]

I have contacted Kathrine and she has sent me photos of the install and it is wired correctly.
I have also gone through the set up of the BMV for LifePo4 batteries based on Victron's and Relion's recommendations. Katherine has now configured it correctly.

 

[Clunegapyears]

A big thank you to all who took the trouble to explain ... my knowledge is ever increasing, albeit slowly!

And also to Phil for the contact today. Mr patience on a white charger!

Really appreciated. Kx

 

[Chasn]

I have a Victron setup using two 90Ahr batteries with the Victron BMS. The system works OK but I have one small concern. The use of a Votronic shunt on the negative battery terminal means that the wiring for the shunt bypasses the BMS enabling a small current drain regardless of the state of charge of the batteries. This is a very small current but it would have been nice if the BMS had total control of battery isolation in the event of very low battery state. I have tried to think of a way around this but cannot work out an alternative way of connecting it. Perhaps Geeky Philip has a solution?

 

[Admin]

I have a Victron setup using two 90Ahr batteries with the Victron BMS. The system works OK but I have one small concern. The use of a Votronic shunt on the negative battery terminal means that the wiring for the shunt bypasses the BMS enabling a small current drain regardless of the state of charge of the batteries. This is a very small current but it would have been nice if the BMS had total control of battery isolation in the event of very low battery state. I have tried to think of a way around this but cannot work out an alternative way of connecting it. Perhaps Geeky Philip has a solution?

I would use the BMS load disconnect output (normally high) via a MOSFET switch PCB (£3 from eBay) to supply the power for the shunt. Then on low cell voltage, the BMS would take this pin to float and the MOSFET switch would disconnect the power to the shunt.

DC 3V 5V Low Control High Voltage 12V24V36V E-switch Mosfet Module For Arduino | eBay

Find many great new & used options and get the best deals for DC 3V 5V Low Control High Voltage 12V24V36V E-switch Mosfet Module For Arduino at the best online prices at eBay! Free delivery for many products!

www.ebay.co.uk

 

[glenm]

Hi don't know much about lithium batteries why don't you talk to Roadpro they fit a lot of them and do know what they are on about, I have a EZA fitted by Roadpro and have no wild camping problems

 

[in h]

In my van it would also be quite complicated to arrange this, there are several current paths on the negative side that would not easily be re-routed and connected together at one point, to simply bolt onto the shunt, without stripping out the Electrobloc entirely.

I don't follow this. I had my meter shunt on the neutral side with none of these difficulties. It is wired in between the battery negative terminals and the vehicle earth, which all the wiring uses as a negative.
Don't have a shunt any more, because I replaced it with a hall effect sensor, but the principle is the same: it's round the cable that runs from the negative terminals' confluence point and the vehicle earth.

 

[alwaysared]

Just on another note, how do lithium react to being left on hookup when using my Victron Inverter/charger on charger only? Is it safe or should I disconnect the EHU every now and again. Since going full time I've not really used hookup until this lockdown where I'm now in the van on my son's drive until the 12 week isolation period is over and of course EHU is freely available.

Regards,
Del

 

[Admin]

Just on another note, how do lithium react to being left on hookup when using my Victron Inverter/charger on charger only? Is it safe or should I disconnect the EHU every now and again. Since going full time I've not really used hookup until this lockdown where I'm now in the van on my son's drive until the 12 week isolation period is over and of course EHU is freely available.

Regards,
Del

The charger should go into float once the batteries are charged which is fine.