-
MSEE, PE -
I don't do it every cycle. Mine is initiated manually and is for 15 minutes at a time to push the voltage up to 3.575v per cell ( 14.3v ) to allow for shunting on the balance boards ( 3.55v ). I'd be guessing that it may be necessary do maybe once a year, but it's been 6 months and the cells are still in perfect balance.
Karrak, Glad your here, you get it.Comment
-
Did you get anywhere with this, the more I think about it the more I think that it is that the controller wakes up in float mode unless the battery drops below the 'Rebulk Voltage'. Have you tried the Outback forum?. I am sure there is a way to force the controller to start with a bulk cycle in the morning.
SimonOff-Grid LFP(LiFePO4) system since April 2013Comment
-
If I had an EV I would probably bottom balance and only put as much charge into the battery as needed to keep the battery centred on 50%SOC. Under this scenario it would be very rare that one would need to fully charge the battery. Come to think of it I never charge my electric bike to more that around 80% for my standard commuting.
I would love to keep my off-grid power supply battery centred at an SOC of 50% to maximize its life but unfortunately I don't have control over the weather. As the old saying goes "make hay while the sun shines". To keep my battery centred on 50%SOC would probably mean doubling its size and of course doubling the cost without a doubling in the lifespan.
To maximise the lifespan of my battery I only charge my battery to 3.45 volts/cell at most once a week, the other days I charge to 3.375 volts/cell which gives me an SOC of between 80% to 90+% depending on the current from the solar panels. The 3.45 volts/cell resets my coulomb counter and in cloudy weather can take several days or even weeks to get there. The coulomb counter (a JLD404 meter) is great, by putting a small negative offset on the current measurement to account for the coulomb inefficiency of the battery it is only ever out by a few percent even after weeks of not being reset. I can tell the state of my battery at a glance.
Others in Australia have taken a different approach and while keeping the full charge voltage to 3.45-3.50 volts keep the float voltage high at 3.34 volts/cell which means that they end up with a full battery at the end of the day. Although I would think this will decrease the lifespan of the battery it means that they can get by with a smaller battery. Interesting to know which method works out more cost effective in the long run.
I am pleased we both agree that having some form of individual cell monitoring and a coulomb counter is a good thing.On the Charge side I monitor every cell voltage, charge at C/2, and terminate when th every first cell reaches 3.5 volts. Any Cell. The second Fail Safe on the charging side is I set the charge voltage to 56 volts on my 16S battery. That would be 3.5 volts per cell. If I topped balance would have to be at a minimum of 57.6 volts and use Vampire boards to bleed off weaker cells.
On the discharge side Fail Safe 1 is I use Battery Monitor to disconnect the pack if ANY CELL dips to 2.9 volts for more than 15 second, or immediately disconnect if ANY CELL touches 2.5 volts. Second line of defense is to use the Motor Controller Low Voltage Disconnect if the pack voltage touches 46.5 volts. Across 16 cells that is 2.9 volts per cell or 10% SOC. Well above the panic 2.25 volts or point of no return cliff.
Lastly for a gas gauge I use a calibrated Coulomb Counter. It measure Amp Hours In, and Amp Hours Out. I know at any point in time what the AH capacity is in the batteries. WhenI see 10 to 20 AH remaining on my 100 AH battery, I go home and recharge.
SimonOff-Grid LFP(LiFePO4) system since April 2013Comment
-
Do you know how many systems I have ?? I thought not. The LiFep04 is a system I am playing with to see if I want to switch from FLA in the future. If you took the time to read you'd have seen it's connected to a outback GFX 1312 Inverter, hardly a house sized system. You know what you can do with the name calling.
Back to your Golf Cart
Did I mention I am not financially challenged and my boat is 78 feet long and I have a big house and a sports car in the garage ??
Wow, the most dangerous man with LiFep04. I daily charge to 3.45v per cell. I discharge to 3.15v. I have a BMS and boards on every cell, I balance at 3.575 per cell. I have a Tyco relay that engages at 2.6v to cut the bank off. I have a back up charger on a VCS that is set at 2.7 v and disconnects at 3.1 volts. I think I am safer than others.Comment
-
Simon I have never said to keep LFP around 50% DOD. 50 to 60% DOD is Storage voltage, not operating voltages. I do understand perfectly not going to 100% which is what I have been saying all along. For maximum cycle life one wants to operate within the range of 10 to 90% SOC.
One of the unique characteristics of a LFP battery is they use very simple charging algorithm. You can charge with a constant current charger until the voltage the cell voltage reaches 3.4 to 3.5 volts and terminate the charge. That gets you to 90% quickly because that constant current can be as high as 1C or whatever current your panels can supply. In a Solar system you want to get the batteries charged as fast as possible. If you use a conventional off the shelf controller it has to be one that allows you to set all 3 stages to one voltage of; 14 ,28, or 56 volts to force the controller into a 1 stage CC/CV charger algorithm required for Lithium batteries.
The problem with Vampire Boards discovered by the DIY EV crowd are two fold. The Bleeder Boards can only bypass X amount of current. X = 1 amp or less depending manufacture or model. However you do not charge at 1 amp unless you are talking about a 1 or 2 AH battery. Typically you charge at C/5 to C/2. On a 100 AH battery like I have would be 20 to 50 amps. FWIW I use 50 amps. So say you are charging at 25 amps, first Vampire Board turns on when it is fully charged. That cell now is bypassing 1 amp and still being charged with the remaining 24 amps. Next Vampire board turns on with same 24 amps left flowing. Process continues until the last Vampire Board turns on just as the charger hits the CV mode and stops current. In say a 16S system you now have 15 over charged cells and 1 properly 100% charged cells. That is how cheap passive BMS works using Vampire Boards.
The other problem and the bigger of the two is Vampire Board Failure/ More times than not they fail shorted which can completely drain a cell. Unless you have a monitor that monitors cell voltages you are not going to know it as the pack voltage still appears to be good and everything still works until you notice that weird burning pine tree tar smell coming from your batteries followed by heavy smoke.
I am not saying your top balance method will not work, I know it will as there are many doing that way. All I am saying is there is another way, less expensive, and a whole lot less risk without having to give up any capacity. Top Balance is a Pb mentality carried over. It is also driven by the manufactures to up sale products as Top Balance is expensive to implement. Today manufactures are now starting to respond to consumer demands who prefer to Bottom Balance.
Now to answer one of your earlier questions when you asked how do you perform Bottom Balance in an operational system. Same way you Top Balance with a Vampire Board. You Bleed it once a year if needed.MSEE, PEComment
-
I and I think many others after their own independent research from many sources including battery manufacturers have arrived at the conclusion that keeping the charge in an LFP battery centred around 50% maximizes the lifespan. How practical it is to do this is another matter
Balancing boards are not designed to balance a battery that is badly out of balance, they are there to keep a battery in balance. They should only be working when the charge controller has started to limit the charge current going into the battery. At the end of the balance you have all the cells at the same voltage with the current being supplied by the charge controller equalling the current going through the balancing boards and equal current going into the individual cells until the current from the controller equals the controller cutoff current or the controller charge timeout.The problem with Vampire Boards discovered by the DIY EV crowd are two fold. The Bleeder Boards can only bypass X amount of current. X = 1 amp or less depending manufacture or model. However you do not charge at 1 amp unless you are talking about a 1 or 2 AH battery. Typically you charge at C/5 to C/2. On a 100 AH battery like I have would be 20 to 50 amps. FWIW I use 50 amps. So say you are charging at 25 amps, first Vampire Board turns on when it is fully charged. That cell now is bypassing 1 amp and still being charged with the remaining 24 amps. Next Vampire board turns on with same 24 amps left flowing. Process continues until the last Vampire Board turns on just as the charger hits the CV mode and stops current. In say a 16S system you now have 15 over charged cells and 1 properly 100% charged cells. That is how cheap passive BMS works using Vampire Boards.
Do you have any stats on how many disasters have been caused by shorted transistors on balance boards compared to other reasons? I doubt it is a big problem. I think that the balance boards should be designed to fail safe. I have a simple circuit design in my head that only has two extra components that would stop the problem of the shorted transistor. Anyway, I thought we had agreed that you should have individual cell monitoring which would detect this problem.The other problem and the bigger of the two is Vampire Board Failure/ More times than not they fail shorted which can completely drain a cell. Unless you have a monitor that monitors cell voltages you are not going to know it as the pack voltage still appears to be good and everything still works until you notice that weird burning pine tree tar smell coming from your batteries followed by heavy smoke.
Less expensive? You do not need balance boards to do a top balance. I don't have any balance boards in my system, I do the balancing manually with some simple 5W resistors and my charge controller.I am not saying your top balance method will not work, I know it will as there are many doing that way. All I am saying is there is another way, less expensive, and a whole lot less risk without having to give up any capacity. Top Balance is a Pb mentality carried over. It is also driven by the manufactures to up sale products as Top Balance is expensive to implement. Today manufactures are now starting to respond to consumer demands who prefer to Bottom Balance.
Top balancing has nothing to do with Pb mentality. If you want to balance on a regular basis it is far easier and I think less dangerous that bottom balancing.
Ah, I see it is an evil conspiracy.
SimonOff-Grid LFP(LiFePO4) system since April 2013Comment
-
That is not how SK charges, he apparently just cuts the charge off at full charging amps and has no short period of CV and tapering amps. So with 50 amps going into the Balancing board I could see where it would do nothing or burn up.Balancing boards are not designed to balance a battery that is badly out of balance, they are there to keep a battery in balance. They should only be working when the charge controller has started to limit the charge current going into the battery.
A problem with not doing a short period of tapering amps ( absorb ) is that the charger voltage and the actually battery voltage is different. Battery voltage is a lagging voltage, the charger has to have a over voltage to be able to raise the Battery Voltage and charge. At 50 amps that could be as much as .3v + /- . If you chop 50 amps off main stream you'll end up with a under charge. Using a short period of CV ( absorb ) corrects this and allows the charger and battery voltages to converge.
Since the text Book of Solar LiFeP04 has yet to be written, the terms of Bulk, Absorb, Float, and EQ may be different than the way they are used for pb. Please use the term in the way you choose. Example : EQ one could say " BB " ( Board Balance ), Absorb, " VC " ( Voltage Convergence ) a short period of Tapering amps. Ending Amps ( EA ) could still be ending amps, if in doubt ask, don't assume.
Karrak, thank you for sharing what you'v learned in the 18 + months on PV Solar LiFeP04 in a intelligent way.Comment
-
Two kids playing with toys sure do know a lot - NOT![SIGPIC][/SIGPIC]Comment
-
I love the kid thing, at 68 I don't get that complement often. Thank you. I have yet to succumb to Cerebral atrophy and am still willing to learn. I guess you could call my learning system a toy, a $4,000 coffee making system is a better description, since thats all I have on it. Maybe a golf cart would be a better test of a PV solar system, but how many panels could I install ??Comment
-
There is absolutely no need for saturation, that is where you destroy your cycle life going from 90% to 100% saturation. I can charge as slow as C/100 if I wanted and use the exact same algorithm. There is only 1 algorithm used to charge Lithium called CC/CV which is nothing more than fixed bench DC power supply with current limit. I chose current limit to 50 amps and can fully recharge in less than 2 HOURS, typically 1 hour. Depends on how deep discharge.
Point here is going above 90% DESTROYS CYCLE LIFE. TOP BALANCED SYSTEMS are designed to go to 100%, thus why EV manufactures DO NOT ALLOW their batteries to go to 100%, and they use batteries that are matched in Capacity so essentially they are Bottom Balanced. All cells have the exact same capacity making it impossible to drive any cell into reverse polarity. The ONLY way you can have Balanced Cells, meaning equal capacity between all cells in a Solar System or a DIY EV is to BOTTOM BALANCE. The only other possible way to have Balanced Cells is to buy them quantities of 10's of thousands and match them up to get within 1% tolerance.
In a Solar System you are fighting one HUGE BATTLE every day. Time, you don't have enough of it to mess around with any SATURATION. You DO NOT NEED and Saturation as it destroy CYCLE LIFE. Why would any fool want to do that? Lithium batteries are extremely easy to charge. To get to 90% SOC all they require is a Constant Current applied until the cell voltage reaches reaches 3.45 volts. So back to TIME and SOLAR. You do not have TIME on your side. The fastest way to 90% SOC is CONSTANT CURRENT, as much power as your panels can generate.
What is so damn hard you cannot understand Willy. Even a child can understand that. EV manufactures understand that. Most of the DIY EV understands that. Why can't you?. Yes you are stuck in a Pb World because you still use the 4-stage Pb mentality. Lithium Charge is clearly defined, has been for 10 years and you still do not know it. I will tell you what it is, CC/CV. It is that simple. If you only want to go to 90% to maximize battery life, Constant Current to 3.4 to 3.5 volts per cell and you are DONE. Saturate them like you are doing and you cut cycle life 30 to 50%. I am not going to allow you to tell people to destroy their batteries because of your ignorance and incompetence.MSEE, PEComment
-
Please share how....
It has everything to do with Pb batteries. TOP BALANCE systems are designed to go to 100% SOC. Last thing you want to do is take your batteries to 100%
Two GOLDEN RULES for Lithium Batteries.
1. Never Go ABOVE 90% SOC
2. Never Go Below 10% SOC
If you use a Commercial Manufactured Top Balance System you go to 100% SOC.MSEE, PEComment
Tweet
Comment