Charging efficiency LifePO4

[createthis]

There is no such thing as a fixed full point at a certain voltage.

I think you're wrong about that.

[Sunking]

Creathus, with Solar there is absolutely no reason to go to 100% and there is absolutely no problem Floating LFP batteries on a Solar System. You do not have to worry about what SOC the batteries are at as long as they are not 100% or less than 10%.

GBS cells charge at a slightly higher voltage than say CALB. As long as you do not go over 14 volts, you are OK. 13.6 is just fine and when floated is roughly 80%. Your GBS batteries are 100% charged up at 3.5 vpc saturated. By Saturated I mean the cells held at 3..5 vpc (14 volts) until all charge current stops. So you can float at 14 if you so wish without any real problems as long as you draw some power. At 3.4 vpc relax. So set the charger to 13.6, and your Invert er LVD to 12 volts and relax, you are fully protected.

Here is a very good article on FLOAT CHARGING Lithium batteries. Think of Float Charging the exact same way as paralleling any lithium cells. Parallel as many as you want, an dlet them set as long as you want. All current stops. As long as the voltage is less than 100% Open Circuit Voltage, you can float a lithium infinitely. 100% OCV on GBS is 3.48 to 3.50. At 3.4 vpc you are well below 100%.

Createthis is you and even Karrak does not understand there is more then one-way to charge a lithium battery. One thing you do know, which is a good thing, there is absolutely no reason to fully charge a lithium battery. In fact you will double the cycle life by not charging to 100%. Limit to 80% SOC and limit discharge to 90% DOD and you can quadruple cycle life. That is exactly what commercial do to enable them to offer long warranties. Hell they use battery types with no more than 500 cycles.

So how to charge to say 80/90% is the question. How do you do that. Well the answer is simple, there are two good ways to do that.

1. If you use commercial AC Power, you charge at a Constant Current of .3C to 1C until the cell voltage reaches 3.65 volts, and terminate the charge. If you want to go to 100% then you hold 3.65 volts until the charge current tapers to 3% of C to saturate the battery. So if you have a 100 AH cell that would be 3 amps. But that is a terrible way to use Solar. If your system is sized correctly, in summer months you are charged up by noon or mid afternoon and you disconnect the panels and stop charging. Well hell stop and think about that for a moment. You turn off the solar and go on batteries while there is still good daylight left. Pretty stupid huh?

2. There is another way. Use a Constant Voltage method and set the voltage to 80 to 90% SOC OCV aka Open Circuit Voltage. What voltage is that? Depends on who the manufacture is. For GBS around 3.4 to 3.45 vpc. How do you do it? Simple you buy a Charge Controller where you can set the Bulk = Absorb = Float. On a GBS 4S system about 13.6 to 14.0 volts. Initially in the morning hours your controller operate sin Constant Current mode charging your batteries as fast as the panels can charge them. When they reach the st point, say 13,6 volts, the voltage holds and charge current tapers to Zero amps when the batteries Saturate. At 0-Amps is Open Circuit Voltage at 80/90% SOC. The panels do not shut-off. If your equipment demands power after the batteries are charged, the power comes from the panels rather than the batteries assuming the power demand is equal or less than what the panels can supply. You conserve battery power for the night and less cycling, thus increasing battery cycle life. What is not to like? You are mimicking what the pros do like EV manufactures.

To protect your batteries on the discharge side could not be easier. Set your LVD for 12 volts or 3.0 vpc. The danger zone for LFP batteries is 2.5 vpc or 10 volts on a 4S system. At 12 volts you eliminate all risk of over discharge.

[Sunking]

Originally posted by karrak View Post

There is no such thing as a fixed full point at a certain voltage.

I think you're wrong about that.

You are correct Createthis. It is called Open Circuit Voltage or OCV.

As I stated in the previous reply, there are two ways to fully or partially charge a Lithium battery. It is dome exactly like lead acid. Using full charge as an example there are two fully documented ways to do this.

1. Use a CCCV (Constant Current, Constant Voltage) algorithm. You charge ate a Constant Current of 0.3C to 1C until you reach 3.6 to 3.7 volts per cell depending on manufacture. Once 3.6 to 3.7 vpc is reached you hold the voltage at 3.6 to 3.7 vpc until the current tapers to 3% of C. Terminate the charge. Example 3 amps on a 100 amp hour cell. This is the exact same way and voltage you would use to charge a lead acid battery (Bulk-Absorb phase). The only difference is is witth Lead Acid when the current tapers to 3%, the end of the Absorb phase, you lower the voltage to 13.8 volts aka Float Phase.

2. Again with both Lithium and Lead Acid you can fully charge with a Float aka Constant Voltage. Voltages are slightly different. Depending on which lithium battery manufacture your are talking about 3.45 vpc give or take .05 volts. Open circuit voltage of a fully charged lthium is approx 3.43 to 3.5 volts. Depends on manufacture.

Now what you do not want to do is hold a Lithium battery at 100% SOC like you can a Lead Acid. But there is absolutely no problem holding a lithium cell at 80 to 90% SOC OCV. The key to understanding is Open Circuit Voltage and what that voltage is with Lithium. Th eissue with Lithium is the discharge curve is very FLAT from 90% to 10% SOC. Roughly 3.4 to 3.0 volts. or 13.6 to 12 volts on a 4S battery.

[PNjunction]

I think we're all on the same page, just coming at it from different angles.

I'll distill what I'm saying from a 12v perspective, and without any balancing issues ...

Anything from 13.8v to 14.6v will result in a fully 100% charged battery, the only difference being that at lower voltages, the CV absorb period takes longer to reach .05C, which is the canonical place to stop *absorb* for a fully charged battery.

You can test this, by charging to any voltage and absorbing, then resting the battery for 12 hours, and taking an OCV measurement. 3.38v/cell resting for a fully charged GBS will be the result either way.

The only caution I mention is that when using very low charge currents, if you try to reach 14.6v - you will never get there as the battery will have already absorbed as much as it can.

If you have already absorbed as much as you can, even down to zero amps (although not recommended), if you wait long enough, then YES you may see a sudden rise in voltage after a lengthy stable period due to electrolyte heating, and not charging, since a full charge has already been reached.

Large cells may hide the fact that after you have charged to 100% that now all you are doing is promoting parasitic reactions if you don't remove the charge - even though no current is flowing. Fortunately, most chargers cut off at a certain amount of end-amps, or go to a lower float.

I mention this if someone uses a bench-supply, rather than an automated charger as something to watch out for - perpetual charge once the battery is full - whether that was reached after a slow 13.8v charge-and-absorb, or a fast 14.6v charge and absorb.

From a practical standpoint, most of the serious users try not to go above 14.0v absorb, although bikers who need quick recharges to get back on the road may be very happy with a 14.4v charger / alternator output since time is limited.

In the end, to NOT reach a 100% charge, you just use whatever is convenient for you. Stop at a certain voltage. Don't allow for a full absorb. Basically, just do a load test to find whatever is comfortable for your application capacity wise.

See the problem with trying to pin down an exact voltage with LFP?

[Sunking]

Anything from 13.8v to 14.6v will result in a fully 100% charged battery, the only difference being that at lower voltages, the CV absorb period takes longer to reach .05C, which is the canonical place to stop *absorb* for a fully charged battery.

PN I don't have any real issues with anything you have said except with this statement. There are a couple of ways to get to 100% on a LFP. Fastest way is to charge CCCV at .3 to 1C at 3.65 vpc or 14.6 volts on 4S. Even at the 1C rate is up to two hours on a fully discharged cell. At that fast of a rate you get to 14.6 volts in just under an hour, but you only get to around 60/70% when you hit 14.6 volts and takes a longer Absorb time till you taper down to 3% of C. Slow down to C/2 and you reach 80% at 14.6 volts, and spend less time in absorb. There is plenty of documentation that details that out and I can pass it along if you would like. If you use this method to get to 100% you have to terminate the charge and quit applying 14.6 volts. Great if you are charging an EV. Power Tools, Flashlights. etc. Not really a good way for Solar because you do not want to turn off th esolar and miss out on what remains in the day.

The other way to get to 100% or something less like say 80% to 90% is use Open Circuit Voltage set Point and use the same CCCV algorythim. That varies a bit depending on which manufacture you use. But for GBS 90% is around 3.4 vpc. You can float at that voltage and the SOC and voltage cannot possible go up when it saturates unless you have a Malfunction with your charger. You can charge as fast or slow as you want. For a Solar system if you follow reccomended practice of 3 day autonomy, you are going to be charging fairly quickly in excess of C/6 up to C/2. Using OCV Float charging works best for solar as it allows you to utilize all the power possible from the Sun in a day rather than cut-off solar at mid day.

Point to make here is a BMS is pretty much useless and not needed in this application. Not needed at all in 4, 8 and to some degree 16S systems. If you are only charging to OCV at 90% SOC, the dang BMS would never turn itself on. All you gotta do is an Initial Balance at whatever point you like, Top, Middle, or Bottom and forget about it.

If you have a cell go bad in a 4 or 8S system you are going to notice it right away. Either when charging when the voltage never reaches its normal high value, or immediately during any discharge with abnormally low operating voltages. 1 in 4 or 8 cells is going to catch your attention real quick especially when your Inverter trips off-line form under-voltage. Missing 3 volts on a 12 or 24 volt system is an attention getter. Guess what? If you have a bad cell guess what that means? Replace it. A BMS does not stop that from happening, chit happens.

Edit Note:

As for further proof no BMS is required for solar 12, 24, and 48 volt solar systems look no further than Genasun Li MPPT controllers. They are Float CCCV controllers with only two ports. One for Panels and one for batteries. You could not interface a BMS if you tried. Well you could if you disconnect the panels or batteries. They use the same concept. they lower the output voltage. For LFP they float a little higher than I an comfortable with at 14.2 volts, but that is an easy fix, just order it to use a lower voltage if you want. The only protection they offer is HV shutdown in the event the controller fails. Other than that they stay on until the sunsets so you can utilize all the Sun Power as possible to conserve battery power. .

[createthis]

I doubt anyone who has participated in this thread so far really wants to watch this, but for posterity, if anyone stumbles upon this thread in the future, here's my LifePO4 battery capacity testing results: https://youtu.be/gMEbXG94vPc

And here are the preceding videos in the series:

• Part 1 is about draining the battery to 12V: https://youtu.be/kFLFqpB-DAU
• Part 2 is about bottom balancing the cells: https://youtu.be/rZGLbAv2MCA
• Part 3 is about installing the battery: https://youtu.be/tLQ7TY5xeIs
• Part 4 is about charging via solar for the first time: https://youtu.be/dI36N79bnOE

[Sunking]

Well I hate to say it; But I told you so. You stated inconsistent answers being all over the place on Solar Panel Talk right? Who was right? It was Sunking. Now you know who to ignore. I watched your video and knew exactly what you were going to observe before you observed it.Been down that road years before you.

You do have a couple of wrong assumptions. No big deal, but you need to know.

1. Your Victron SOC meter is pretty much meaningless using voltage to determine SOC. Your Victron is made for Pb batteries. Secondly the discharge curve is so flat on LFP, there is no resolution. Only way you can really tell is with Coulomb Counting or counting AMP Hours in and out. However there is even inaccuracy counting AH and requires constant re-calibration. So use your Victron Coulomb Counter as your fuel gauge as an ball park indicator

2. This one is your most glaring error in your assumptions. You said if you draw more power like 1000 watts, to raise the LVD Set Point for higher current drains. Wrong Sir, you lower it, not raise it. Your own observations should be telling you that.

All batteries have Internal Resistance. If you know Ohm's Law then you know Voltage = Current X Resistance. So when you draw more power means you draw more current right? Of course it does. If you draw more current through a resistance then you develop more voltage drop. You seen this with your very own eyes and noticed it because you talked about it. You said what the voltage was before you started your discharge test, and noticed the voltage sagged as soon as you turned on the Inverter. You also noticed the battery voltage rises as soon as you turned off the Load.

You can take your battery, fully charge it, set your LVD to 12 volts, and if you draw say 1C (100 amps), or even C/2 (50 amps) your LVD would operate immediately despite the battery being 100% charged up. Why the Internal Resistance of the battery and your wiring/connector resistance accumulated voltage loss. The GBS batteries have the highest Internal resistance of all the Chi-Coms. 100 AH cells are roughly .004 Ohms per cell. So 4 in series is roughly .016 Ohms. Add in wiring and connection resistance, and you are up close to .020 Ohms. So do some simple math. If you are running 50 amps with an Open Circuit Voltage of 13.6 volts - [50 amps x .02 Ohms] = 12.6 volts. At 100 amps drops to 11.6 volts on a fully charged cells

You do not want to take your batteries below 2.5 volts per cell open circuit voltage. If using a LVD you would set it lower than 2.5 for high loads. A good quality BMS like an Orion has some smarts. Its LVD defaults to 2.5 vpc for more than 30 seconds under load. That allows an EV to accelerate at some 5 to 10C or 500 to 1000 amps. The cell voltages will all drop well below 2.5 volts under heavy load of accelerating , but once up to speed, the current backs way off and the voltage recovers back up above 2.5 volts up to around 3+ volts. That is just the nature of the beast and physics. If you were to set a EV to 3.0 vpc as you do with a solar system, you would never get it out of the garage.

So let's go back to what I told you originally Float your batteries at 13.6 to 14.0 volts, and set your LVD to 12 volts. 12 volts works great on a Solar system because you are not drawing high current. You need to understand the difference between Open Circuit Voltage and Loaded voltage. Your LVD is passive, it does not know the difference between OCV or Loaded voltagge. All it knows is disconnect at 12 volts.

At 12 volts leaves you lots of room becaus danger is at 10 volts and less. Only thing you need to check periodically is the voltage of the cells when discharged. and look for a low cell voltage. When you Bottom Balance the voltage should be equal at the bottom, not the top. So you need to look out for a cell dropping to low because it is possible, although extremely remote chance, you could have a cell go bad and never detect it from pack voltage. Example one cell at 2.5 volts and the other 3 cells at 3.2-volts you would read a pack voltage of 12.1 volts. I ran my first EV with the same cells you did for 6-months and never had any drift. I now use Nissan Leaf cells going on 8-months and no drift. I know other EV guys going on 4 years and no drift.

So going forward:

• Keep an eye open and check cell voltages at the Top and Bottom and measure OCV, don't get hung up on Loaded or Charging voltages because that does not tell you much of anything.
• Set your LVD to 12 volts. You can even go lower if you want, but I would not go below 11 volts. 10 volts is pushing it.
• Calibrate your Coulomb Counter ever 10 cycles or so because it will drift due to charge efficiency. More goes in than comes out and over a few cycles induce inaccuracy.
• Set you Midnite Solar Bulk = Absorb = Float = 13.6 to 13.8 volts
• Relax

[Sunking]

Just watched you Bottom Balance video. Your Cell pro does not require you to discharge into another battery. It can do regenative charging, but that is not how you should be using it. Look at the Specs

Output battery discharge current: Internal discharge 10mA to 10A, 100W max Regenerative discharge 10mA to 40A, 1344W max**

The way to Bottom Balance with the unit is real simple.
1. Do the initial discharge like you did to get the cells near 2.5 volts. You got that part right.
2. Then connect all the cells in Parallel.
3. Then set your PL8 to discharge at 10 amps to 2.5 volts, connect it and walk away. You keep doing this until the cell voltage rest at 2.5 volts with NO LOAD aka OCV. Very important to get this right.

Additional If you have time you could have tested each cell capacity. The PL8 has a Cycle test. It can cycle 2 to 5 times fully charging each cell, and fully discharging each cell howevery many times you tell it to do that. Charge it at 40 amps, discharge it at 10 amps, or you can do it from a battery and regenerate. at 40 amps charge and discharge.

[createthis]

Well I hate to say it; But I told you so. You stated inconsistent answers being all over the place on Solar Panel Talk right? Who was right? It was Sunking. Now you know who to ignore. I watched your video and knew exactly what you were going to observe before you observed it.Been down that road years before you.

I had to chuckle out loud at that. I think everyone else had good feedback too, for the record, but I do tend to find your explanations a little easier to parse. In everyone else's defense, I found myself referring to the battery as "full" a lot in that video, when I damn well know it's not really full unless I charge it to 14.4v. I understand why everyone describes it that way at 13.8 or 13.6 volts when the battery stops accepting amps.

You do have a couple of wrong assumptions. No big deal, but you need to know.

1. Your Victron SOC meter is pretty much meaningless using voltage to determine SOC. Your Victron is made for Pb batteries. Secondly the discharge curve is so flat on LFP, there is no resolution. Only way you can really tell is with Coulomb Counting or counting AMP Hours in and out. However there is even inaccuracy counting AH and requires constant re-calibration. So use your Victron Coulomb Counter as your fuel gauge as an ball park indicator

I think I was starting to come around to that realization. The voltage didn't seem very accurate and the SOC percentage seemed out of sync with the AH display.

2. This one is your most glaring error in your assumptions. You said if you draw more power like 1000 watts, to raise the LVD Set Point for higher drains. Wrong Sir, you lower it, not raise it. Your own observations should be telling you that.

All batteries have Internal Resistance. If you know Ohm's Law then you know Voltage = Current X Resistance. So when you draw more power means you draw more current right? Of course it does. If you draw more current through a resistance then you develop more voltage drop. You seen this with your very own eyes and noticed it because you talked about it. You said what the voltage was before you started your discharge test, and noticed the voltage sagged as soon as you turned on the Inverter. You also noticed the battery voltage rises as soon as you turned off the Load.

You can take your battery, fully charge it, set your LVD to 12 volts, and if you draw say 1C (100 amps), or even C/2 (50 amps) your LVD would operate immediately despite the battery being 100% charged up. Why the Internal Resistance of the battery and your wiring/connector resistance accumulated voltage loss. The GBS batteries have the highest Internal resistance of all the Chi-Coms. 100 AH cells are roughly .004 Ohms per cell. So 4 in series is roughly .012 Ohms. Add in wiring and connection resistance, and you are up close to .015 to .020 Ohms. So do some simple math. If you are running 50 amps with an Open Circuit Voltage of 13.6 volts - [50 amps x .02 Ohms] = 12.6 volts. At 100 amps drops to 11.6 volts.

You do not want to take your batteries below 2.5 volts per cell open circuit voltage. If using a LVD you would set it lower than 2.5. A good quality BMS like an Orion has some smarts. Its LVD defaults to 2.5 vpc for more than 30 seconds under load. That allows an EV to accelerate at some 5 to 10C or 500 to 1000 amps. The cell voltages will all drop well below 2.5 volts, but once up to speed the current backs way off and the voltagg egoes way back up above 2.5 volts up to around 3+ volts.

I remember asking you what happens if I pull the battery past a certain (I don't remember exact voltage, but it was intended to be battery EMPTY voltage, so... low) voltage under high amp load. You gave me a slingshot analogy and told me once I let it go I could never get it back.

The Victron BP-220 LVD has a 90 second delay at a given set point voltage. I stated I would raise my set point because I'm not sure what 90 seconds at high amp draw will do to the battery's resting voltage. I figure I'll start high and work my way down until I'm as low as I can get while still being sure it won't damage the battery.

So let's go back to what I told you originally Float your batteries at 13.6 to 14.2 volts, and set your LVD to 12 volts. 12 volts works great on a Solar system because you are not drawing high current. You need to understand the difference between Open Circuit Voltage and Loaded voltage. Your LVD is passive, it does not know the difference between OCV or Loaded voltagge. All it knows is disconnect at 12 volts.

Well, yes and no. The BP-220 can accept an external signal to disconnect the battery. I was thinking I might try using the Victron BMV-700's AH trigger alarm relay output to send this signal. But yeah, if I use the BP-220 purely in voltage set point mode, it's going to operate off of the voltage sag, not the resting voltage. I understand that.

At 12 volts leaves you lots of room becaus danger is at 10 volts and less. Only thing you need to check periodically is the voltage of the cells when discharged. and look for a low cell voltage. When you Bottom Balance the voltage should be equal at the bottom, not the top. So you need to look out for a cell dropping to low because it is possible, although extremely remote chance, you could have a cell go bad and never detect it from pack voltage. Example one cell at 2.5 volts and the other 3 cells at 3.2-volts you would read a pack voltage of 12.1 volts. I ran my first EV with the same cells you did for 6-months and never had any drift. I now use Nissan Leaf cells going on 8-months and no drift. I know other EV guys going on 4 years and no drift.

So going forward:

• Keep an eye open and check cell voltages at the Top and Bottom and measure OCV, don't get hung up on Loaded or Charging voltages because that does not tell you much of anything.
• Set your LVD to 12 volts.
• Calibrate your Coulomb Counter ever 10 cycles or so because it will drift due to charge efficiency. More goes in than comes out and over a few cycles induce inaccuracy.
• Set you Midnite Solar Bulk = Absorb = Float = 13.6 to 13.8 volts
• Relax

I have no idea how to calibrate my coulomb counter or view that info on the victron bmv-700. I'll make it a point to acquire this knowledge. Thanks!

[createthis]

Just watched you Bottom Balance video. Your Cell pro does not require you to discharge into another battery. It can do regenative charging, but that is not how you should be using it. Look at the Specs

Output battery discharge current: Internal discharge 10mA to 10A, 100W max Regenerative discharge 10mA to 40A, 1344W max**

The way to Bottom Balance with the unit is real simple.
1. Do the initial discharge like you did to get the cells near 2.5 volts. You got that part right.
2. Then connect all the cells in Parallel.
3. Then set your PL8 to discharge at 10 amps to 2.5 volts, connect it and walk away. You keep doing this until the cell voltage rest at 2.5 volts with NO LOAD aka OCV. Very important to get this right.

Additional If you have time you could have tested each cell capacity. The PL8 has a Cycle test. It can cycle 2 to 5 times fully charging each cell, and fully discharging each cell howevery many times you tell it to do that. Charge it at 40 amps, discharge it at 10 amps, or you can do it from a battery and regenerate. at 40 amps charge and discharge.

I followed this guy's procedure: https://youtu.be/J2WvQre8sAQ
I think it's probably fine, but I really should create a 4S balance connector for the PLCP8 so I can charge/discharge on the bench.

[Sunking]

I had to chuckle out loud at that. I think everyone else had good feedback too, for the record, but I do tend to find your explanations a little easier to parse. In everyone else's defense, I found myself referring to the battery as "full" a lot in that video, when I damn well know it's not really full unless I charge it to 14.4v. I understand why everyone describes it that way at 13.8 or 13.6 volts when the battery stops accepting amps.

No Sir that is not accurate, you can fully charge the cells at 3.45 vpc or 13.8 volts on 4S. What is getting in your way is misinformation and misunderstanding what you are being told. This is Karaks issue and passing it on to you. There are a few ways to fully charge a Lithium battery, but you only need to know two of them to understand what is going on.

1. Is the Only way you and Karak knows and is spread around the Internet. It is the basis of every BMS system out there, and every Charger made specifically for lithium battery. It is the CCCV (Constant Current - Constant Voltage) method. Exact same method you use for lead acid batteries with 1 modification. You charge at Constant Current, say C/5/ C10, 1C whatever rate you want as long as it is greater than C/10 and 1C or less. You apply constant current until the cell voltage reaches 3.6 to 3.7 volts per cell (rather it is 3.6 to 3.7 depends om manufacture and model). Once you reach 3.6 vpc you hold 3.6 vpc until the charge current tapers to C/33 or 3% of C. For a 100 AH Cell would be 3 amps. Terminate charge and disconnect. Once you disconnect and allow the cell to rest the voltage will bleed off to roughly 3.45 volts which is 100% SOC OCV for LiFeP04. Let me say that again: Once you disconnect and allow the cell to rest the voltage will bleed off to roughly 3.45 volts which is 100% SOC OCV for LiFeP04. You need to understand if the resting Open Circuit Voltage is 3.45 volts the cell is 100% Got it?

2. Method 1 sucks for Solar. To use that method with solar means if you are charged up by noon, you disconnect the panels and they will not come back on till the next day. You are on battery power from that point on til next day. What a waste right? You could be utilizing solar power. You could use Solar power if the rest of the day instead of batteries if you FLOAT the batteries. Well you can Float the batteries, no problem at all. It is dome at OCV.

So the second method is choose a SOC OPEN CIRCUIT VOLTAGE point you want to operate at. There is that word again, OPEN CIRCUIT VOLTAGE. Remember? So if you want 100% set your voltage to 13.8 volts and forget about it. All FLOAT means is it is a CCCV algorythim, but we do not terminate the charge at any point. We just let it FLOAT. When the batteries reach 13.8 volts, current will taper toward 0 Amps. When it reaches 0 Amps the batteries are fully charged and Float. Panels are not disconnected. If you use power, the panels supply the power assuming the demand does not exceed what the panels can generate. You conserve battery power until the sunsets.

Now should you charge to 100%. Not IMO or any other person who knows about lithium batteries. We can double the cycle life by limiting to 90% SOC. Can you do that. Will heck yes you can silly, lower the damn voltage from 13.8 to 13.6 volts. Your GBS cells are fully charge with a OCV of 3.45 volts or 13.8 on 4S. 90% is roughly 3.4 vpc OCV or 13.6 volts.

I remember asking you what happens if I pull the battery past a certain (I don't remember exact voltage, but it was intended to be battery EMPTY voltage, so... low) voltage under high amp load. You gave me a slingshot analogy and told me once I let it go I could never get it back.

That is you not understanding exactly what you are being told and my failure to communicate. So let's see if we can get on the same page. You can take a lithium battery to 0 Volts with little or no damage. That is not what kills them. What kills them is being reversed polarity. So how does that happen should be your question?. Glad you asked.

The answer is simple, when adjacent cells still have capacity left in them to drive discharged cells negative below 0 volts. That is the fatal flaw of Top Balance systems. If you were to measure each cell capacity on the PL8, they are not equal. Tolerance is 10%. So your cells can range from 90 to 110 AH. You need to make sure you understand that point.

So we Top Balance your batteries and fully charge them to max capacity. You have 3-cells at 110 AH and one cell at 90 AH. OK now we discharge and we have discharged 90 AH. You 90 AH cell is now 2.5 volts, and the rest of your cells still have 20 AH capacity left in them so pack voltage is say 12.1 volts, and you are unaware you have a flat cell. It falls off the cliff and now have reverse current flowing through it and reverses polarity and is destroyed.

Realistically that is highly unlikely in a 4 S system, or even 8 and sometimes up to 16S. Reason is simple you have 1 cell with reverse polarity, and even if th eother 3 cells were fully charged at 3.4 volts only give you 10.2 voltts on the pack. Your Inverter turned off at 10.5 volts. You system will notice 3 volts missing on a 12 volt system. If you charge up and only see 10.2 volts you know something is WRONG.But what about say a 32, 45, or 100S system. 3, 6 or 9 volts low is not a problem at much higher voltages. On a 32S or 96 volt system full charge is 110 volts and discharged is 80 volts. So if you see say 100 volts is perfectly normal, but you could easily have 2 cells reversed and dead.

However what if we change the reference point to 0% SOC at 2.5 vpc? We know on a 4S system 10 volts is fully discharged. All cells have 0 AH at 2.5 vpc. When we charge them in series they all have the exact same capacity. Now when we discharge all cells arrive at 2.5 volts at the same time. No cell has the energy to drive the adjacent cell into reverse polarity. You eliminated the possiblity altogether. You would not even need a disconnect device, your Inverter shut off automatically at 10.5 volts or 2.625 vpc. Heck you could hook a load up to the cells and take pack voltage to 0 volts if you wanted.

So if you Bottom Balance, use an Inverter, and a external LVD you have 3 lines of deffence. First is th eLVD set to a extremely conservative 12 volts or 3.0 vpc with 10 to 15% capacity left in the cells. Second line of Defense is your Inverter shuts off at 2.625 vpc. If those two modes fail you Bottom Balanced and they cannot over discharge themselves anyway.

One caveat. Over a period of time, the cells age and can drift. That can be months or years. To protect yourself routinely check cell voltage at the Bottom and Top. Re-balance only as needed.

The Victron BP-220 LVD has a 90 second delay at a given set point voltage. I stated I would raise my set point because I'm not sure what 90 seconds at high amp draw will do to the battery's resting voltage. I figure I'll start high and work my way down until I'm as low as I can get while still being sure it won't damage the battery.



Well, yes and no. The BP-220 can accept an external signal to disconnect the battery. I was thinking I might try using the Victron BMV-700's AH trigger alarm relay output to send this signal. But yeah, if I use the BP-220 purely in voltage set point mode, it's going to operate off of the voltage sag, not the resting voltage. I understand that.

OK but it does not have Lockout or reconnect logic. Not that it matters. If you had somethin glike an Orion BMS, it is a PLC or Program Loop Controller.

By default it will operate LVD if one of two conditions are met:

If any cell drops below 2.5 volts form more than 30 seconds.
If any cell drops below 2 volts for more than 2 seconds.

If the LVD operates, it LOCKS OUT the LVD until every cell is 3.3 volts or greater. Meaning it will not reconnect until you recharge the batteries. However the Orion BMS i sof the Top Balance mindset. But you can programm Top Balance to OFF and reprogram anyway I want. I do not use it. I use a PL8.

[Sunking]

I followed this guy's procedure: https://youtu.be/J2WvQre8sAQ
I think it's probably fine, but I really should create a 4S balance connector for the PLCP8 so I can charge/discharge on the bench.

Well you do need to build a Balance Plug, But if you intend to Bottom Balance, Dot not use any Balance Charging method or you effed up your bottom balance. The Balance plugg wil just monitor the cell voltaes and shut down if any cell goes outside a safe range.

Bu tif you intend to Bottom Balance, you need to nail it perfectly and if you have the PL8 is the right tool to do it correctly, But it can only be dome one way to nail it. Connect your Inverter up and turn on as much load as you can for a fast discharge. Monitor the cell voltages. If one drops to 2.5 volts or less stop. If not let it run until your Inverter LVD operates at 10.5 volts. Your cells should rest at around 2.6 to 2.7 volts when th eLVD operates.

Pull the straps off and connect all 4 cells in parallel. Set the PL* to discharge at 10 amps to 2.4 volts. When the cells reach 2.4 volts, the PL8 will disconnect and the voltage will recover. Reset to 5 amps and 2.4 volts and repeat. Keep doing this until the cell voltages rested for a hour or two are 2.4 to 2.5 volts. Let them sit over night connected in parallel but nothing else connected.

THE DO NOT CROSS LINE is 2.0 Volts. Once the cells set rested at 2.4 to 2.5 volts is perfect Bottom Balance and ready to go.

[createthis]

No Sir that is not accurate, you can fully charge the cells at 3.45 vpc or 13.8 volts on 4S. What is getting in your way is misinformation and misunderstanding what you are being told. This is Karaks issue and passing it on to you. There are a few ways to fully charge a Lithium battery, but you only need to know two of them to understand what is going on.

1. Is the Only way you and Karak knows and is spread around the Internet. It is the basis of every BMS system out there, and every Charger made specifically for lithium battery. It is the CCCV (Constant Current - Constant Voltage) method. Exact same method you use for lead acid batteries with 1 modification. You charge at Constant Current, say C/5/ C10, 1C whatever rate you want as long as it is greater than C/10 and 1C or less. You apply constant current until the cell voltage reaches 3.6 to 3.7 volts per cell (rather it is 3.6 to 3.7 depends om manufacture and model). Once you reach 3.6 vpc you hold 3.6 vpc until the charge current tapers to C/33 or 3% of C. For a 100 AH Cell would be 3 amps. Terminate charge and disconnect. Once you disconnect and allow the cell to rest the voltage will bleed off to roughly 3.45 volts which is 100% SOC OCV for LiFeP04. Let me say that again: Once you disconnect and allow the cell to rest the voltage will bleed off to roughly 3.45 volts which is 100% SOC OCV for LiFeP04. You need to understand if the resting Open Circuit Voltage is 3.45 volts the cell is 100% Got it?

2. Method 1 sucks for Solar. To use that method with solar means if you are charged up by noon, you disconnect the panels and they will not come back on till the next day. You are on battery power from that point on til next day. What a waste right? You could be utilizing solar power. You could use Solar power if the rest of the day instead of batteries if you FLOAT the batteries. Well you can Float the batteries, no problem at all. It is dome at OCV.

So the second method is choose a SOC OPEN CIRCUIT VOLTAGE point you want to operate at. There is that word again, OPEN CIRCUIT VOLTAGE. Remember? So if you want 100% set your voltage to 13.8 volts and forget about it. All FLOAT means is it is a CCCV algorythim, but we do not terminate the charge at any point. We just let it FLOAT. When the batteries reach 13.8 volts, current will taper toward 0 Amps. When it reaches 0 Amps the batteries are fully charged and Float. Panels are not disconnected. If you use power, the panels supply the power assuming the demand does not exceed what the panels can generate. You conserve battery power until the sunsets.

Now should you charge to 100%. Not IMO or any other person who knows about lithium batteries. We can double the cycle life by limiting to 90% SOC. Can you do that. Will heck yes you can silly, lower the damn voltage from 13.8 to 13.6 volts. Your GBS cells are fully charge with a OCV of 3.45 volts or 13.8 on 4S. 90% is roughly 3.4 vpc OCV or 13.6 volts.

Hold it right there. In my video, I charged to 13.6v and discharged to 12.7v, withdrawing 48.5ah. Then I charged to 13.8v and discharged to 12.7v, withdrawing 50.0ah. Are you trying to tell me 1.5ah is 10% of battery capacity? It's not. It's 1.5% of battery capacity. According to your logic, I charged to 98.5% at 13.6v and 100% at 13.8v.

[Sunking]

Hold it right there. In my video, I charged to 13.6v and discharged to 12.7v, withdrawing 48.5ah. Then I charged to 13.8v and discharged to 12.7v, withdrawing 50.0ah.

Your test is invalid with the LVD set at 12.7 volts, nor are you using meter quality equipment to measure Amp Hours. Use your PL8 and fully discharge to 2.5 vpc.

The only possible way to measure LFP or any battery capacity is a FULL DICHARGE TEST. Any other method is INVALID. 12.7 volts on your system is meaningless under any load or charge rate. You keep forgetting OCV.

By your logic you can take your fully charge 100 AH cells, put a 100 load on them and the voltage will fall instantly below 12 volts and conclude they are discharge when in FACT they are not even remotely close to being fully discharged. At 100 Amps would take 36 seconds to drop from 100 AH to 99 AH.

If you really want to measure the Capacity of the cells you are going to have to use your PL8. Fully charge the cells, discharge at 10 amps until cell voltage reaches 2.5 volts. Any other method is Invalid.

You keep getting stuck that battery voltage is an exact science of SOC. It is NOT, it is a Ball Park indicator only on a fully rested OCV. 3.4 vpc is roughly 90%, 3.45 is roughly 100%. You need to get past that and quit getting stuck.

[createthis]

Your test is invalid with the LVD set at 12.7 volts, nor are you using meter quality equipment to measure Amp Hours. Use your PL8 and fully discharge to 2.5 vpc.

The only possible way to measure LFP or any battery capacity is a FULL DICHARGE TEST. Any other method is INVALID. 12.7 volts on your system is meaningless under any load or charge rate. You keep forgetting OCV.

By your logic you can take your fully charge 100 AH cells, put a 100 load on them and the voltage will fall instantly below 12 volts and conclude they are discharge when in FACT they are not even remotely close to being fully discharged. At 100 Amps would take 36 seconds to drop from 100 AH to 99 AH.

If you really want to measure the Capacity of the cells you are going to have to use your PL8. Fully charge the cells, discharge at 10 amps until cell voltage reaches 2.5 volts. Any other method is Invalid.

You keep getting stuck that battery voltage is an exact science of SOC. It is NOT, it is a Ball Park indicator only on a fully rested OCV. 3.4 vpc is roughly 90%, 3.45 is roughly 100%. You need to get past that and quit getting stuck.

I'm not opposed to doing this. Would I need to have the balance connector attached for the full discharge test on the pl8? Or can I just run the pack in series through the banana jacks?