what happens between OVD and CLV values of a controller?

**Sunking** · 01-02-2018, 02:16 PM

Power goes no where. Lost forever. This is the problem with LFP batteries. It makes no since to turn off the controller in mid day, forcing the battery to go on discharge while the sun is still shinning. Instead of using Panel Power when th esun is still up and shinning you are now on battery power. Not smart.

You can work around the problem real easy. Set Bulk = Absorb = Float = 13.6 volts. You are making the charge controller a simple CC/CV Float Charger. It will charge the battery up to 13.6 volts and float. Once floating any loads that demand power will come from the panels, not the batteries just like a solar system is suppose to do.

Another way is to set Bulk = Absorb = 13.8 volts if you want more SOC, and set Float = 13.6. Works as above but gets you a little higher SOC. You just have to find what voltage set points work to get you where you want.

Set your Inverter LVD to 11.0 to 12 volts so you eliminate the risk of over discharged.

To run without a BMS, you would be much better off if you Bottom Balanced the batteries as that eliminates any chance of over discharging and turning your cells into bricks and/or boat anchors. BMS is the number 1 cause of LFP failures. So many failures and fires, they can no longer be shipped on airplanes.

**racmaster** · 01-02-2018, 05:25 PM

thanx for ideas. some unclear, but in general definitely usefull.

bulk-absorb-float strategy undesrtood, will try, sounds good.

setting inverter u mean inverter or controller? however, for lifepo4, isnt it too low? my values suppose lvd at 12.80, which should be 20-30% soc. at 11.0 by my simple knowhow its around the dead of the battery... isnt it? why not higher then...?

bottom ballancing, good idea, i will go for...

what do you mean by lifepo fire? by my info lifepo4 doesnt explode, fire, or anything as liion with cobalt. as shown for example on this brutal force teste video

https://www.youtube.com/watch?v=aQs7L5LmEss

one last question not answered, if charging limit value is set to 13.85 and overvoltagedisconnect value to 14:00, what does it mean? at 13.85 battery is not being charged anymore, but whats next? from where is the next voltage measured now? from battery or pv? in case battery, this stage will last forever and never reaches 14.00. if pv, then it will last milisecond and immediately switch off everything as soon as charging limit 13.85 value will be set... which means, that as soon as battery will be charged based on defined values, entire system will be switched off?? pv and load also...? not a nice perspective

**Sunking** · 01-02-2018, 07:44 PM

Originally posted by racmaster

thanx for ideas. some unclear, but in general definitely usefull.

What is not clear? I will try to be more specific if I know what you need.

Originally posted by racmaster

setting inverter u mean inverter or controller? however, for lifepo4, isnt it too low? my values suppose lvd at 12.80, which should be 20-30% soc. at 11.0 by my simple know how its around the dead of the battery... isnt it? why not higher then...?

No sir the Inverter. LFP batteries claim 2.5 volts is critical low so with 4S would be 10 volts. That is a CYA number, the real number is 2.0 volts. I have mine setup for 3 volts per cell which is about 10% SOC, and on a 4S is 12 volts. So at 12 leaves you some room. But note I bottom balance, which is radically different than Top Balance. However set the Inverter LVD as high as you like as that is no problem. The problem with Top Balance is setting yourself up for over discharge. Very possible when the cells are getting low, th ewek or lowest capacity cell can be setting at 2.5 volts while other are above 3 volts indicating a good pack voltage. Not a real threat with 4 and 8S, but once you get up to 16S and above is a concern. Bottom Balance eliminates that threat because all cell voltages will be equal at the bottom. Top balance is only equal at the Top

Originally posted by racmaster

What do you mean by lifepo fire? by my info lifepo4 doesnt explode, fire, or anything as liion with cobalt. as shown for example on this brutal force teste video

Don't believe a word of it. While True LFP is more tolerant to overcharging, they can still go into thermal runaway when over charged. However when over discharged, all Lithium batteries have the same risk of fire. When a cell over discharges, it goes through polarity reversals, and the adjacent cells with energy left will force current through the discharged cell causing a massive thermal runaway and fire. BMS are number one cause of failures. They use Bypass Boards and they fail shorted which discharges the cell it is connected to. When that happens and you try to charge through a discharged cell cause it it go into runaway, or if discharge it will runaway. Either way is bad new.

Gotta go for now. Will answer last question when I get my chores done.

**Sunking** · 01-02-2018, 09:13 PM

Originally posted by racmaster

one last question not answered, if charging limit value is set to 13.85 and overvoltage disconnect value to 14:00, what does it mean?

It is technically called High Voltage Shutdown. It means if the charger malfunctions, there is a circuit that disconnects the charger from the batteries to prevent an over charge. You do not see that much of HVD in a solar charge controller. Would be useful if you are using more than one source to charge like a generator with its own charger.

Originally posted by racmaster

At 13.85 battery is not being charged anymore, but whats next? from where is the next voltage measured now? from battery or pv?

Nothing is next, charging stops. What is tripping you up is not understanding Ohms Law and how a battery charges up. For current to flow in any circuit, a battery is no exception, is there must be a difference in potential (voltage) for current to flow. That's Ohms Law. Makes no difference what battery type we are talking about. When the Battery Voltage = Chargers Voltage current stops. At that point the battery is FLOATING which means it is neither charging or discharging.

Read this Sticky where I go through the math. Not going to go all the way through it here again. To charge any battery, again does not matter what type, they all charge the exact same way. What changes is how you terminate the charge, and the voltage. Example to charge a 4S LFP battery with a BMS and charger is you set the Charger voltage to 14.4 volts. Lets say you have a 50 amp charger charging a 100 AH Battery and the battery voltage is 12 volts are discharged for argument sake. When you connect the charger, the voltage drops to about 12 volts plus the Current Resistance voltage drop in the battery. The charger limits current to 50 amps. As time goes on and the battery charges up, the voltage goes up and up toward 14.4 volts. At about 14.3 or so volts the battery is nearing the charge set point and current tapers off toward 0 amps. When charge current tapers to 3 to 5% of C, you terminate the charge because the battery is fully charged up, and if you keep the charger connected will start heating the battery up because all the power is now just generating heat. At that point you must do one of two things. Disconnect aka terminate the charge, or lower the voltage to a FLOAT Voltage. 3 to 5% of C on a 100 AH battery is 3 to 5 amps.

You charge a FLA battery exactly the same way. You charge to 14.4 volts until current tapers to 3 to 5%. Again you can terminate or lower the voltage. But there is more than one-way to charge a battery. Th emethod I used above is what every commercial battery charger does. Well almost all, telecom and utilities use a Float Charger. The charger works the same, but there is no termination or lowering the voltage. They use a lower voltage called Float Voltage and on a Lead Acid battery is 13.2 to 13.6 depending on the alloy used. Works just like I described. As the battery voltage reaches set point of say 13.6 volts, current will start to taper as the battery voltage and charger voltage equalizes. What is different is the charger does not change voltage or disconnect when the current tapers off. The battery is allowed to SATURATE, and when that happens, current stops. The battery is fully charged and FLOATS at full charge and stays that way. Any power required for the load comes from the charger, not the batteries. If there is no load, no current. If a load demands current comes from the charger. If power fails or the charge ris turned off, power comes from the batteries with no interruption in power.

Exactly what you want your solar system to do, and you can FLOAT CHARGE lithium batteries. In fact there are Lithium battery chargers that do exactly that with a twist. They first charge the battery to 14.4 volts by setting Bulk/Absob to 14.4 volts. When charge current tapers lowers the voltage to FLOAT of 13.8 volts. Exactly what a 3-stage charger does for lead acid. But there is a catch. No solar charge controller can determine when the current has tapered to the appropriate voltage in order to lower the voltage back to 13.8 volts.

Once you get your head wrapped around how the battery charges, you realize there is more than one way to charge a battery. FLOAT CHARGE IT. A LFP battery at 100% SOC that is allowed to rest is 3.45 vpc or 13.8 volts. But you do not want 100% SOC you want something less like 13.6 volts. The battery will Saturate at 13.6 volts and stop charging, but your solar panels are still on line to supply power to the loads until sunsets. Then the battery takes over for the night. Exactly what you want. So if you want 100% and risk over charge, set Bulk = Absorb = 14.4, and Float to 13.6. Not me, I would set Bulk = Absorb = Float = 13.6 volts and relax. Batteries will charge up and saturate up around 90% SOC, panels will supply power until dark.

There is one Charge Controller made for lithium called Genasun. Actually they work on any type of battery, they just set the algorithm when you order as you specify the battery type. If you order LFP is 14.2 volts from sunrise to sunset. For Pb uses 3-stage Bulk/Absorb = 14.4 volts and Float to 13.5 volts. Same package.

If you want a Lithium Charger with Float Option is made by Battery Tender. Charges to 14.4, then lowers to 13.8 volts. Here it is, scroll down to 5-Step Lithium Ion Phosphate Charger Algorithm.

**PNjunction** · 01-03-2018, 02:20 AM

racmaster - I'm not familiar with your controller. In addition to the good advice above, maybe we can simplify a little:

While not exactly precise, (good enough for our low-current needs) just remember this:

ANY individual cell voltage charged between 3.45v to 3.6v will *eventually* fully charge the cell to 100% when the current has tapered to .05C. We normally don't want to go to 100% charge on a regular basis, but we need to know just what that is!

At 3.6v / cell, (14.4v) that natural taper will be VERY quick.
At 3.45v / cell, (13.8v) that natural taper will be pretty slow. Like 8 hours or more. Unless you have that much time/sun, you are likely to never get fully charged - PROVIDED you are using this in say a daily cyclic service.

What I'm getting at is that if you were to set your charge to 13.8v, and never discharge the battery, it WILL become fully charged eventually and it is not good to be left that way.

Much of our talk assumes a steady constant supply of charge current like an ac charger in the garage. That's one thing.

With solar, the strategy changes a little depending on conditions:

1) Sure we like conservativism so we set our absorb cv voltage to 13.8v right? But now you have very cloudy conditions. Heck, you'll never even get 4 hours of absorb time in to reach 85% or so. You'll actually start tapering between every passing cloud. Might not be enough. If set to 14.4v, then those fleeting moments of sun will be blasting current in as fast as the batt can gulp it down in little chunks from the array.

Under these conditions, some might actually opt for the higher CV voltage rating to blast as much as they can in between passing clouds. I believe that Genasun's nominal 14.2v cv voltage was to help account for this while sailing, where that could mean the difference between reaching the dock or not.

What I'm really saying is that it all depends, and you'll have to just experiment to find out what is right *for you* in regards to recharge capacity.

Options include the following:
1) Setting a very high CV voltage like 14.4v. Useful for bad weather conditions, will actually reach 100% quickly when left to taper/absorb to the canonical .05C. Just don't leave it like this for long. It's not that you actually want to reach 100%, but without the taper taking place like it would at a lower cv voltage, you'll be able to pump in more current faster.

2) Set a lower conservative level like 13.8v absorb. You may never reach the .05C taper current fully charged condition in a single day, and is not really necessary as that indicates 100% recharge conditions - But, that might be TOO conservative during bad weather days if the lower CV taper is just not providing you enough charge.

3) Just stop charge completely (no absorb) at some pre-defined voltage. Anywhere from 3.45 to 3.6v per cell is fine (remember it is the amount of taper current that truly defines the fully charged condition, not voltage alone. - but yes, 3.45 with 8 hours or more of absorb to go, and 3.6v with very little absorb to go, means the difference between say 80% and 95% SOC.

The SOC when you just *stop* at a predefined voltage is obviously different than one that is left to taper naturally. If your controller actually does just stop, and not taper, then maybe this might be one of the easier solutions.

Unfortunately, everyone is looking for the magic-wand formula when it comes to LFP.

What I would suggest knowing your interest in taking care of your system, is to set your OVD to 14.6v or so. This is an over-voltage hardware failure catch, and is not catastrophically damaging to the LFP's. Normal CV absorb - try conservative at 13.8v at first. If your conditions and depth of discharge calls for it, perhaps raise it to 14.2v and see where you stand. Or, if your controller actually does just stop at a predetermined voltage with no absorb, then experiment.

Can't disable float? Yeah, then adjust it lower than 3.4v per cell. While not necessary, it *may* act as a catch for unintended small parasitic loads / tiny shorts etc from totally discharging the battery down to the real low-voltage disconnect.

In short - there's no magic wand setting other than not letting your cells sit at 3.45 to 3.6v all day every day. Nor letting them get too low in voltage and staying that way.

As long as you've got your LVD, and your controller has the catastrophic OVD set right, you'll find out soon enough what's right for you and conditions. As much as we like to do our due diligence, these cells are not made of glass.

**racmaster** · 01-03-2018, 05:50 AM

sunking, pnjunction, this was exciting reading. thanx!

lot of things starting to make sense for me now. gonna play with it today and refer back if my understandings were confirmed by test and how much...

the only thing which im not sure now is if i understood your charging process explanation right. based on my understanding of sunking i understood that it doesnt matter too much on bulk/boost/float phases because battery will just be charged based on possible generated current vs some empty space in capacity. simplified. voltage just tells when this "automatic" reaction process will be turned off by policeman - charger, or controller... then it will be switched to float and no charging happens, because ohms law - voltage is not downhill... as long as float voltage is set lover, then battery voltage. right?

based on my understanding of pnjunction - even in float voltage the charging process will go on until 100% soc or until charging limit policeman turns it out mechanically...

a little bit contraversial. seems like i didnt get something...

however, pnjunction, if my charging limit voltage is 13.85, then should no float happen, isnt it? float in my scenario based on those values should go like this:

until 13.75v - bulk
13.75 - absorb starts and lasts 10 minutes based on set counter RaisDur: 10
then float - charging voltage goes down to specified 13.65v
if even afterward SOC rises, charging will be completly shutoff at specified charging limit 13.85v set.

this is how i understand these settings vs process flow now, after your explanations. is it right?

as i said, going to test it right now, will refer back with a result. thanx again

**karrak** · 01-03-2018, 07:53 AM

I hope the following Charge/discharge curves will help.

If you haven't come across charge discharge curves before, they show what the cell voltage will be if you charge or discharge the cell with a particular current. In this graph the blue line shows what the cell voltage will be as the cell charges with a charge current of 0.2C (18A for this 90Ah cell), the green line is with a charge current of 0.1C, and the red line is with a charge current of 0.05C.

So if we charge at 0.2C we will see the voltage slowly rise from an SOC of 20% to nearly 90% and then go up rapidly. If the charge controller is set to a bulk/boost charge voltage of 3.45V and terminates the charge when the voltage reaches 3.45V by following the blue line we see that the cell will achieve an SOC of around 88%. If the charge rate is 0.1C following the green line shows that the cell will achieve an SOC of around 94%, A 0.05C charge rate results in an SOC of around 98%.

By setting bulk/absorb(boost) and the float voltage to 3.4V/cell (13.6V for a 12V battery) you can see that the final SOC of the battery will be greater than 95% (the point that the red curve crosses 3.4V) as the charge current will keep getting smaller and smaller and will be less that 0.05C at the end of the day.

I know that by bulk/boost charging my battery to 3.45V/cell and terminating the charge at 0.02C and floating at 3.35V/cell I end up with an SOC at the end of the day around 98%-99%. If I bulk/boost charge and float at 3.35V/cell I end up with a final SOC of around 90%

As PNjunction has said the charge rate from a system charging from solar is not constant. If you set a solar controller to bulk/boost charge to 3.45V/cell (13.8V for 12V battery) and terminate the charge when it reaches 3.45V/cell with no absorb time, on a sunny day you might get a charge rate of 0.2C and an end SOC of 88% but on a very cloudy day you might get a charge rate of only 0.05C and a final SOC of greater than 98%.

Simon

Off grid 24V system, 6x190W Solar Panels, 32x90ah Winston LiFeYPO4 batteries installed April 2013
BMS - Homemade Battery logger github.com/simat/BatteryMonitor/wiki
Latronics 4kW Inverter, homemade MPPT controller

**karrak** · 01-03-2018, 08:18 AM

What SOC you charge and discharge your battery to to get the best bang for your buck is very dependant on how you are going to use the battery. If you are only going to use it intermittently it would be better keeping it charged below 70% . If you are going to be using it full time as part of an off grid power supply like my battery you are better charging to around 99%SOC when there is enough sunshine to get you through the cloudy days. If you are using it full time for small periods of time and then storing it it would be worth storing it at an SOC below 50% and charging it to 99% on the days you will be using it.

Charging an LFP battery to around 99%SOC if it is in use all the time does not reduce its lifespan appreciably. My battery is nearly 5 years old and I can't detect any change in the performance of the battery.

Simon

Off grid 24V system, 6x190W Solar Panels, 32x90ah Winston LiFeYPO4 batteries installed April 2013
BMS - Homemade Battery logger github.com/simat/BatteryMonitor/wiki
Latronics 4kW Inverter, homemade MPPT controller

**racmaster** · 01-03-2018, 08:28 AM

very usefull graph, detailed and under both conditions. and whats more, exactly for the very similar battery. nice, very nice, simon! i was searching for something like that for long, but was able to get just the standard graphs where just the thickness of the curve was difference between live and death...

so your point is, that based on my limit values, i can get out of proposed interval 30-80 soc, especially when the battery will be without load still being charged for long time under minimum cloudy current... which is nice to know, even if i think in my conditions rare scenario. however, 5 or 10 points of lovering parameters would do, right?

here we are getting to another technical problem, i noticed that there is too big difference between voltage value on controler display, pc sw, digital led voltmeter and standard multimeter in my case. if we are talking about so small diferences, then 1 - which to believe, 2 - how to implement the diference into limit values without going into risk, but keeping at least 50% capacity (planned 30-80 soc) . in detail im thinking about those 3 scenarios - charge, sitting, discharge + charge and load at the same time variations... there will be different value for each scenario, but there is only 1 value to be choosed from them and then using as a limit. i think without lot of testing and monitoring - no chance... or is there some trick?

btw, what do you power from 32 batteries like this? small city of yours?

**karrak** · 01-03-2018, 09:51 AM

Originally posted by racmaster

so your point is, that based on my limit values, i can get out of proposed interval 30-80 soc, especially when the battery will be without load still being charged for long time under minimum cloudy current... which is nice to know, even if i think in my conditions rare scenario. however, 5 or 10 points of lovering parameters would do, right?

Yes, with your current parameters you will could be charging the battery to greater than 95% under cloudy conditions and discharging to around 20%.

I have my LVD on my inverter set to 24V (3.00V/cell) and my BMS individual cell voltage set to 2.8V. My battery very rarely gets below 30%SOC. If my memory is correct the lowest it has been in the last three years is around 8%SOC.

What exactly are you using the battery for and what are your charge and discharge currents?

here we are getting to another technical problem, i noticed that there is too big difference between voltage value on controler display, pc sw, digital led voltmeter and standard multimeter in my case. if we are talking about so small diferences, then 1 - which to believe, 2 - how to implement the diference into limit values without going into risk, but keeping at least 50% capacity (planned 30-80 soc) . in detail im thinking about those 3 scenarios - charge, sitting, discharge + charge and load at the same time variations... there will be different value for each scenario, but there is only 1 value to be choosed from them and then using as a limit. i think without lot of testing and monitoring - no chance... or is there some trick?

Yes, this is a problem. I would get a good multimeter and use this as your reference. If you have lots of money a Fluke with an accuracy of at least 10mV would be OK. For the amount of use I have for a multimeter these days I cant justify the price of a Fluke and have gone for a UNI-T UT61E. If you want to be sure of the accuracy of the multimeter check it with a calibrated and certified voltage reference something like this

I would be very surprised if the solar charge controller will be able to regulate the battery voltage any better than 0.100 volts.

btw, what do you power from 32 batteries like this? small city of yours?

This powers my home, allot of people on this forum would think my power system with its 10kWh battery is tiny. This system gives me about 3kWh/day in winter and 5kWh/day in summer. Electricity is used for all our cooking in summer.

Simon

Off grid 24V system, 6x190W Solar Panels, 32x90ah Winston LiFeYPO4 batteries installed April 2013
BMS - Homemade Battery logger github.com/simat/BatteryMonitor/wiki
Latronics 4kW Inverter, homemade MPPT controller

**racmaster** · 01-03-2018, 10:55 AM

battery is used on my cottage as offgrid, to power some led lights, notebook, small 5amp water pumps and so on... nothing big. charging current about 25a max, but currently its snowy weather, so up to 1 amp

)

btw, next week no sun predicted, so i plan to charge it with classic leadacid 20a "smart" charger as there is nothing better. what i suppose is the charger recognizes it as almost fully charged, so maybe bulk part will be skipped and some absorption or float will occure. i plan to sit there and monitor it by hand to prevent overcharging, hoping to get some 10-20% higher soc and then turning it off. have no better idea currently and some power would be usefull to have. charging will go from petrol generator for about a hour or so. any hint?

was also thinking to use the old lead acid charger as an input to the solar controller mppt 30a, which could be theoreticaly nice as it could take care for the charging process without my fulltime attention, but as im not sure what that could make, im not going to risk that. would be nice if that would be working, but things will not be probably so easy... do you have any experience or ideas regarding charging with ac leadacid charger/solar controler and lifepo4?

**Sunking** · 01-03-2018, 11:24 AM

His graph is telling you how bad of shape his batteries are in, and and nothing about capacity. What is laughable is he does not even realize what the graph is telling him. That is show clueless he is, he is hanging himself with his own rope. . All his graph is telling you is the Internal Resistance which is very high and the batteries are due for replacement. at 25 milli-ohms or roughly 10 times higher than when new. They should be around 2 to 3 milliohms.

[Moderator note -

Calculate Internal resistance = (delta V) / (delta I).

Evaluate at 50% SOC on the chart
At 0.2 C (18 A), V = 3.26 V
At 0.05 C (4.5 A), V = 3.22 V

(3.26 V - 3.22 V) / (18 - 4.5) = 0.003 Ohm.

Note that SOC is very difficult to determine in the flat portion of the curve, so this may not be entirely accurate, but it looks like Sunking has made an error here]

**Sunking** · 01-03-2018, 11:28 AM

Originally posted by racmaster

btw, next week no sun predicted, so i plan to charge it with classic leadacid 20a "smart" charger as there is nothing better. what i suppose is the charger recognizes it as almost fully charged, so maybe bulk part will be skipped and some absorption or float will occure.

You have absolutely no control of that. you can use any charger made for lead acid if you want to go to 100% and use a BMS to control it which is the trick because there is no way for the BMS to communicate with the charger. Stay away from 3.6 volts on any given cell.

**racmaster** · 01-03-2018, 01:03 PM

hmm, intreresting...

so to get back to conclusion, my limit voltages could be fine for my purpose 30-80soc? no matter of what charging current will flow from panel? (0-25a)

please check my answer to your previous reply #7 . thats where our discussion ended before karak entered

what happens between OVD and CLV values of a controller?

what happens between OVD and CLV values of a controller?

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment