Monitoring lithium batteries

[WotEver]

2 minutes ago, TheBiscuits said:

Only if you believe your voltmeter calibration is accurate enough. 

I only quoted the voltages. How they are measured is your problem

 

[MtB]

8 minutes ago, WotEver said:

13.8V is 3.45V/cell. 14.0V is 3.5V/cell.

 

100% SoC is 3.6V/cell.

The minimum cut-off voltage is 2.8V/cell.

 

So there is a range of 0.8V from 0% to 100%.

 

80% of 0.8V is 0.64V. 2.8V + 0.64V = 3.44V.

90% of 0.8V is 0.72V. 2.8V + 0.72V = 3.52V

 

So... 13.8V (3.45V/cell) is nearasdammit 80%. 14V is nearasdammit 90%. 

 

Sufficient evidence?

 

Tony. 

 

Or so you say!

 

A statement is not evidence. I'm sure you're right but I only ever see unsupported assertions, like yours. 

 

[WotEver]

2 minutes ago, Mike the Boilerman said:

 

Or so you say!

 

A statement is not evidence. I'm sure you're right but I only ever see unsupported assertions, like yours. 

Which bit are you querying?  The cutoff voltage of 2.8V and the maximum cell voltage of 3.6V are well documented and available all over the place. The rest is just sums

[MtB]

1 minute ago, WotEver said:

Which bit are you querying?  The cutoff voltage of 2.8V and the maximum cell voltage of 3.6V are well documented and available all over the place. The rest is just sums

 

I suppose I'm querying whether the SoC is actually linearly proportional with the terminl voltage. And if there is any concept as surface charge. And No-one has addressed my repeated question about how SoC can possibly be measured during charging, as others here ave asserted can be done.

[WotEver]

6 minutes ago, Mike the Boilerman said:

And No-one has addressed my repeated question about how SoC can possibly be measured during charging, as others here ave asserted can be done.

I have said that it’s (properly) done by the BMS which reads the voltage of the cells. Your irritating Yank goes into detail about it. He also explains that those who don’t rely on a BMS to do so are the ones who come running to him wondering why their systems have failed them after a couple of hundred cycles. 

 

As for discharge curves, lots of graphs around on the web all showing a very straight line from empty to full. 

 

No, there’s no surface charge effect and no equivalent to sulphation. 

[WotEver]

 

Voltage/SoC curves for a LA battery are also pretty straight. 

[Dr Bob]

1 hour ago, rusty69 said:

Jeez Dr Bob, you read 1/2 the linked article last time it was posted in March. Maybe you just need to read the other half.  p9, post #203

 

 

Me? Remember things that happened last March? I can barely remember what I had for lunch. Did I have lunch? I'll ask the duck.

 

I think though I am on the same wavelength as Mike. Can you watch voltage rise WHILE charging so know when to pull the plug at 80% ? ....or do the batteries just show the charge voltage the charge device is putting in which may be just a constant voltage? 

 

Ill take another look at that link. Really, did I read it in March?

 

 

 

edit to add......Oh yes Rusty. You're right as usual. Now I Rembrandt it. What a mistaka to maka.

Edited August 8, 2018 by Dr Bob

[MtB]

32 minutes ago, Dr Bob said:

Me? Remember things that happened last March? I can barely remember what I had for lunch. Did I have lunch? I'll ask the duck.

 

I think though I am on the same wavelength as Mike. Can you watch voltage rise WHILE charging so know when to pull the plug at 80% ? ....or do the batteries just show the charge voltage the charge device is putting in which may be just a constant voltage? 

 

Ill take another look at that link. Really, did I read it in March?

 

 

 

edit to add......Oh yes Rusty. You're right as usual. Now I Rembrandt it. What a mistaka to maka.

 

Me too. What is 'lunch'? I'm busy learning a new song on ze geetar so all this stuff is just a trivial distraction. 

[cuthound]

10 hours ago, TheBiscuits said:

Only if you believe your voltmeter calibration is accurate enough. 

 

If you can afford lifepo4 batteries, then you can afford a decent Fluke multimeter and annual calibration ?

[WotEver]

10 hours ago, Mike the Boilerman said:

No-one has addressed my repeated question about how SoC can possibly be measured during charging, as others here ave asserted can be done.

The ‘others’ either don’t understand the subject or are deliberately obfuscating. A charger that is JUST a charger will not know what the SoC is. They just throw current at the battery until a certain voltage is reached, then clamp the voltage until the charge current falls. Exactly the same as a lead acid charger running bulk then acceptance. However ‘proper’ Lithium batteries and chargers are designed to run together as a pair and combine a BMS integrated in them. This works as an amphour counter with all its inherent advantages and problems. Without that amp hour counter, the charger doesn't have a clue what the SoC is.c

[WotEver]

I have a couple pieces of G-Tech equipment. You charge their batteries just like any other piece of equipment, plug the charger in the mains and plug the output into the battery. HOWEVER, you can also plug in a USB cable to connect the battery to a PC and it will show you the number of cycles the battery has performed, its charge status, its capacity etc. All that information comes from the internal BMS. 

[peterboat]

My charger [midnite] is designed for the job I have a over voltage cut off if something goes wrong, James/John/me never charge to 100% yet the batteries remain balanced James has over a year of this with no issues. as Tony says if my charger only goes to 13.8 volts and its timed for 50 mins then goes to 13.6 volts it cant over charge, I know because its what I see everyday

[WotEver]

9 minutes ago, peterboat said:

then goes to 13.6 volts it cant over charge...

Oh yes it can. Continuous float at 13.6V will most certainly overcharge a LifePo4 battery given enough time. After the battery is full (or as full as you want it) the charger should switch OFF, completely. 

[peterboat]

4 minutes ago, WotEver said:

Oh yes it can. Continuous float at 13.6V will most certainly overcharge a LifePo4 battery given enough time. After the battery is full (or as full as you want it) the charger should switch OFF, completely. 

It does I see the amps go to zero what I never see is volts over 13.8 and thats only for 50 mins after that 13.6 max. This has worked through this very sunny hot summer with no issues, no cell inbalence  so lets ee wht its like next year when I have my drive batteries in place as well

[Robbo]

21 minutes ago, peterboat said:

It does I see the amps go to zero what I never see is volts over 13.8 and thats only for 50 mins after that 13.6 max. This has worked through this very sunny hot summer with no issues, no cell inbalence  so lets ee wht its like next year when I have my drive batteries in place as well

You may have answered this before but don't you need to charge at least above 14.2v to balance the cells at least monthly?

[Dr Bob]

1 hour ago, WotEver said:

The ‘others’ either don’t understand the subject or are deliberately obfuscating. A charger that is JUST a charger will not know what the SoC is. They just throw current at the battery until a certain voltage is reached, then clamp the voltage until the charge current falls. Exactly the same as a lead acid charger running bulk then acceptance. However ‘proper’ Lithium batteries and chargers are designed to run together as a pair and combine a BMS integrated in them. This works as an amphour counter with all its inherent advantages and problems. Without that amp hour counter, the charger doesn't have a clue what the SoC is.c

 

53 minutes ago, peterboat said:

My charger [midnite] is designed for the job I have a over voltage cut off if something goes wrong, James/John/me never charge to 100% yet the batteries remain balanced James has over a year of this with no issues. as Tony says if my charger only goes to 13.8 volts and its timed for 50 mins then goes to 13.6 volts it cant over charge, I know because its what I see everyday

Ok, we are making progress. Peter is saying the solar charges at 13.8 volts then after 50 mins goes to float. Lets ignore the float for now. Why 50 mins? If you start charging after 'full' discharge (so the lithiums are at say 30% SoC), or if they are at 60% SoC, is the 50 mins always the same? I know you dont need to get to 'full' ( ie 80% SoC) but why 50mins?

The american guy  in Mike's link says that the BMS activities on over or undercharging ie HVC and LVC, are for emergency only and the BMS should not do the switching off of the charge source. He says this should be done by the charge source itself..ie you just cant use a conventional alternator or charger. So he is saying you need a charge source that will make the decisions to switch itself on and off. I assume Peter's solar charger is designed specifically for lithiums hence the 50 mins....but then the decision to turn off when the selected charge point is reached is based on time...not voltage. What happens at 49 mins if Mrs Bob puts the Nesspresso machine on and the voltage drops? Does the solar charger reset itself to another 50 mins? (I guess the answer here is that there is no voltage drop as lithiums dont vary voltage enough with heavy loads??????).

Is the American guy wrong in what he is saying here?

Picking up on Tony's point then, if an amp hour counter is needed to predict SoC, then this HAS to be done by the BMS and cant be left to the charge device as the charge device cannot measure amp hours out of the battery. That seems to contradict the American.

 

The take away from all of this is that you CANNOT estimate the SoC and so switch off your charge source based on voltage (as Tony says). I had hoped that you would see a voltage rise during charge so could 'turn off' when it got to a set point. That doesnt seem to be the case........or is the BMS doing this by measuring voltage of individual cells in some way.

 

The American guy also says that float charging is bad and even 13.6V will wreck the bank.

[MtB]

13 minutes ago, Dr Bob said:

 

[WotEver]

13 minutes ago, Dr Bob said:

The take away from all of this is that you CANNOT estimate the SoC and so switch off your charge source based on voltage (as Tony says).

Picking up on this last point first, it depends on your chosen solution. If you have a gentle charge source that acts like a dumb (so-called ‘intelligent’!) charger, it will current-limit itself by lowering the charge voltage. Think LA in ‘Bulk’. Once the voltage finally raises to 13.8V (as an example) then it is switched off. This is entirely based on voltage. 

 

17 minutes ago, Dr Bob said:

Ok, we are making progress. Peter is saying the solar charges at 13.8 volts then after 50 mins goes to float. Lets ignore the float for now. Why 50 mins? If you start charging after 'full' discharge (so the lithiums are at say 30% SoC), or if they are at 60% SoC, is the 50 mins always the same? I know you dont need to ge

Because he has a dumb charge source, presumably. So 50 minutes is chosen as a safe time limit. As LiFePO4 don’t have to be at any particular charge state to be used then he’ll get away with it for now. Once 50 minutes is only bringing his charge state up to 30% then it won’t be so good. 

 

20 minutes ago, Dr Bob said:

The american guy  in Mike's link says that the BMS activities on over or undercharging ie HVC and LVC, are for emergency only and the BMS should not do the switching off of the charge source. He says this should be done by the charge source itself..ie you just cant use a conventional alternator or charger

I think you’ll find that he says that HVC and LVC shouldn’t be used to switch off the charge source. They are last-resort settings. Certainly the BMS and charger should talk to each other and decide when to stop charging. Once again we’re back to a ‘system’ such as Victron’s solution. 

 

22 minutes ago, Dr Bob said:

Picking up on Tony's point then, if an amp hour counter is needed to predict SoC, then this HAS to be done by the BMS and cant be left to the charge device as the charge device cannot measure amp hours out of the battery. That seems to contradict the American.

Nope. Once again, a complete system where the various pieces of equipment communicate with each other. That’s what I’m saying and that’s what the Yank is saying. 

 

24 minutes ago, Dr Bob said:

The American guy also says that float charging is bad and even 13.6V will wreck the bank.

Indeed. You should NOT float charge lithiums. They don’t need it and it will damage them long term. 

[WotEver]

1 hour ago, peterboat said:

It does I see the amps go to zero what I never see is volts over 13.8 and thats only for 50 mins after that 13.6 max.

So are you saying that the charger is not actually going into float but is switching off. So the 13.6V you’re seeing is the 3.4V/cell of the battery?

1 hour ago, Robbo said:

You may have answered this before but don't you need to charge at least above 14.2v to balance the cells at least monthly?

He’s said that (occasionally?) checking the cells shows that they’ve remained in balance, which isn’t so unusual aiui. 

[MtB]

 

It seems to me that there is no consensus on even the mot basic facts.

 

1) Is there a terminal voltage 'droop' or not when a high current (or any current) is drawn?

 

2) If there is no 'voltage droop' on load, why can SoC not be determined with certainty using terminal voltage at all times?

 

3) Some sources seem to say that if a charger set to say 13.6v is attached to a battery with a terminal voltage of 13.8v, it will still charge at a low rate. How can this be possible?

 

4) Some sources seem to say that if a charger at 13.80v is attached to a battery with terminal voltage of 13.80v, the full output current of the charger will commence. How can this be, with no potential difference present?

 

5) No-one seems to be drawing a distinction between single 'large cell' batteries comprising one large cell and 'multi-cell' batteries, comprising dozens or hundreds of little cells in a series-parallel grid inside. The latter presents huge difficulties in charge measuring and individual cell balancing within the battery.

[peterboat]

5 minutes ago, WotEver said:

Picking up on this last point first, it depends on your chosen solution. If you have a gentle charge source that acts like a dumb (so-called ‘intelligent’!) charger, it will current-limit itself by lowering the charge voltage. Think LA in ‘Bulk’. Once the voltage finally raises to 13.8V (as an example) then it is switched off. This is entirely based on voltage. 

 

Because he has a dumb charge source, presumably. So 50 minutes is chosen as a safe time limit. As LiFePO4 don’t have to be at any particular charge state to be used then he’ll get away with it for now. Once 50 minutes is only bringing his charge state up to 30% then it won’t be so good. 

 

I think you’ll find that he says that HVC and LVC shouldn’t be used to switch off the charge source. They are last-resort settings. Certainly the BMS and charger should talk to each other and decide when to stop charging. Once again we’re back to a ‘system’ such as Victron’s solution. 

 

Nope. Once again, a complete system where the various pieces of equipment communicate with each other. That’s what I’m saying and that’s what the Yank is saying. 

 

Indeed. You should NOT float charge lithiums. They don’t need it and it will damage them long term. 

Tony you are wrong, I am on many sites dealing with this, some of the off grid livers do exactly what we are doing to protect and make sure the batteries have long lives, We are no problems the batteries are staying in balance if I boil a kettle the amps going back into the batteries goes up and then back down, James uses an Outback John uses a Tracer and I use a Midnite solar controller all are doing the job and we have no problems. Charging these batteries to the max shortens their life, doing what we do extends them. General Motors with its Ampera does something similar on its charging regime to protect the batteries. People on Navitron do the same, this is real people doing it for real, and I would rather copy them than some of the tripe that is on some forums, from people that have never even seen one of our batteries, never mind used one

[WotEver]

5 minutes ago, Mike the Boilerman said:

Some sources seem to say that if a charger set to say 13.6v is attached to a battery with a terminal voltage of 13.8v, it will still charge at a low rate. How can this be possible?

It’s possible because a fully charged LiFePO4 nominal 12V battery won’t have a terminal voltage of 13.8V. The natural cell voltage is 3.2-3.3V, so 12.8-13.2V for a 12V battery. 

1 minute ago, peterboat said:

Tony you are wrong,

I made a lot of points. Would you care to be a little more specific?

[peterboat]

11 minutes ago, WotEver said:

So are you saying that the charger is not actually going into float but is switching off. So the 13.6V you’re seeing is the 3.4V/cell of the battery?

He’s said that (occasionally?) checking the cells shows that they’ve remained in balance, which isn’t so unusual aiui. 

Exactly this Tony, as we speak I am looking at my BM2 it is showing 13.4 volts [I have just boiled a kettle] its shows 15 amps going into the batteries which given the day is correct, if it stays like this at the end of the day it will read 13.6 volts and 0 amps going in [unless the fridge kicks in

4 minutes ago, WotEver said:

It’s possible because a fully charged LiFePO4 nominal 12V battery won’t have a terminal voltage of 13.8V. The natural cell voltage is 3.2-3.3V, so 12.8-13.2V for a 12V battery. 

I made a lot of points. Would you care to be a little more specific?

Later I have to drain down my heat exchanger it hold 140 litres of pink antifreeze deep joy?

[WotEver]

15 minutes ago, Mike the Boilerman said:

Is there a terminal voltage 'droop' or not when a high current (or any current) is drawn?

Very little as the internal resistance of the battery is extremely low. One of the signs of approaching EOL of a lithium is greater voltage drop under load. 

 

16 minutes ago, Mike the Boilerman said:

why can SoC not be determined with certainty using terminal voltage at all times?

At ‘all times’? Even when charging? For the same reason that you can’t do it with LA Batteries - you’d be reading the charger not the battery. 

 

17 minutes ago, Mike the Boilerman said:

Some sources seem to say that if a charger at 13.80v is attached to a battery with terminal voltage of 13.80v, the full output current of the charger will commence. How can this be, with no potential difference present?

Because, as above, a LifePo4 battery won’t have a terminal voltage of 13.8V. 

[WotEver]

11 minutes ago, peterboat said:

Exactly this Tony, as we speak I am looking at my BM2 it is showing 13.4 volts [I have just boiled a kettle] its shows 15 amps going into the batteries

So, because it is a low-current source, it’s currently voltage limiting to keep maximum power output and then switching off on voltage, exactly as I described earlier. Once the voltage reaches 13.8V it will switch off. 

So it has nothing to do with 50 minutes, that’s simply how long it’s taking to raise the voltage to 13.8V.