-
so, finally i decided for this machine, as that seems to be better price/value ratio. however, i expect you may not agree
-
I understand what you are trying to do. But lithium ion batteries do not work like you think.
Example you are stuck thinking discharging a cell to 2.5 volts will damage the cell which is a half truth. I have already proven to you that number is a manufacture CYA claim to cover their butts as the real number is 2.0 volts. Granted their is no significant energy left below 2.5 volts, but 2.5 volts is the 0% SOC value. You are not going to hurt the batteries if you Bottom Balance them occasionally. Quite the opposite, you eliminate the chance of over discharging them if Bottom Balanced as that is the whole point of Bottom Balance. Nor does it hurt them to Top Balance once in a while when needed. Either way has to be done initially. Take your pick of Bottom or Top. I really do not care what you do and I am not going to keep repeating myself.
As for discharging no deeper than sat 80% DOD again I understand. However there are a couple of things you do not understand, nor does the pretender. It is done by design, and selecting the right battery capacity and to a much lessor degree LVD value. A LVD is a Fail Safe only mostly for cloudy days and expected loads outside the design range. Think of it like a wing on an airplane G-Load where you design the wing for 150% G-Load. They do that to cover their arse for that unexpected turbulence or crash avoidance high G turn.
The first step is to determine how many days of Autonomy to size the batteries. Example say you only need 320 Watt Hours per day. You would size the battery to a minimum 4 days of capacity or 1280 kWh. That gives you roughly 3 days of run time for cloudy days without a recharge. On a 12 volt system is a 100 AH battery. Under normal operation you would never get close to 2.5 volts or even 3.1 volts. It is by design, not luck or guessing.
The other thing you and the pretender do not understand is you do not use an OCV (open circuit voltage) to select LVD. OCV means a rested battery disconnect from everything. When any battery is under discharge, there will be voltage sag and wire voltage loses. You can literally have a fully charged battery, apply a heavy load, the Inverter will trip off-line from under-voltage if you set LVD to high. Why because you did not allow for voltage sag and loss under load. So if you set LVD for say 12.4 volts (3.1 vpc), you are going to have LVD operation from time to time. It will have you scratching your head because when you shut down the Inverter and look at cell voltages they will be greater than 3.1 volts. The LVD is NOT there to necessarily limit discharge to some DOD percentage, it is a FAIL SAFE to prevent damaging your batteries. An occasional discharge to 2.5 volts is not going to hurt the batteries unless you do not recharge quickly.
To limit DOD is a design of sizing the batteries, not selecting a LVD voltage. On a 4S LFP a good number is 11 to 12 volts which is significantly greater than 8 to 9 volts of danger zone. Even the default 10.5 volts most Inverters can work with a Bottom Balanced battery, but asking for trouble on a Top Balanced battery.Last edited by Sunking; 01-09-2018, 01:13 PM.Leave a comment:
-
sunking, thanx for coming back. im not sure why you try to see the things in worst possible angles. reason why im thinking about presented scenario isnt lack of time as i have so much time for this as i want. also my lazyness is out of discussion. if i was lazy, i would just leave it on default lifepo4 values of the charger and not even open this thread, not writing and reading a lot of texts on different sites in foreign language, watching videos, googling... dont you think so?
the main reasson is the basic law ive got imprinted into my mind regarding cycling of batteries, doesnt matter if lead acid or lithium. only a final number off cycles changes, but what stays is depth of cycling. lets say lifepo4 has 1000 cycles at 90dod, 2000 at 80 and 8000 at 50%dod. smaller discharge, higher cycle life.... therefore im searching for scenario, which doesnt need to discharge the battery to zero = 2.5vpc. to get some solution, which will balance the battery and prevent future disbalance related risks, but in the same time not to push battery to zero, max discharge, max cycling depth...Leave a comment:
-
I understood the first time. Comes down to this. You are trying to cut corners, risking your batteries, and just being lazy. There is no excuse for that. If you have the time to Discharge to 3 or 3.1 volts, then you have time to go to 2.5 to do it right. Same goes for 3.65 volts. Once you are Balanced, you are done and will not have to do it again for quite some time if ever.
Example the Reactor is a Chi-Com knock-off of iCharger 1010B+. Take a look at both and you will see what I mean. At least the iCharger has regenerative Discharge so if you had a few days you could either Top or Bottom Balance. [moderator note - iCharger is made in China too. Reaktor and iCharger may be the same, just badged differently]
The 6 amp unit you are looking at I do not know if it can Discharge or not, nor do I care. I assume your cells are 100 AH and you have 4 of them? If that is a true statement with all 4 in parallel is 400 AH. If the cells were say at 50% SOC is 200 AH. Well there is some math you need to do.
Amp Hours = Amps x Hours
Amps = Amp Hours / Amps
Hours = Amp Hours / Amps.
Makes no difference if you Top or Bottom with a 6 amp Discharge/Charge you are looking at 200 AH / 6 A = 33.3 hours. Additionally if you were to Top Balance how the hell are you going to detect when they are fully charged? To initially Top Balance a Lithium battery you charge at some reasonable rate like C/10 (10 amps on a 100 AH battery) and terminate at C/20 or 5 amps on a 100 AH battery.
One last thing. Disconnecting at 3.1 will work, but not like you think it will. When you apply a load to a battery, you will have voltage sag on the batteries. Couple that with voltage drop on the wires, and your Inverter will prematurely trip of line from under voltage. When it happens will leave you scratching your head, because when you go look at the cell voltages, you will see they are higher than 3.1 volts. Say 3.2 volts. The mistake in your thinking, is you are using Open Circuit Voltage to determine SOC on a battery under load. It will not work espeicially on Winston cells which have a high Internal Resistance which is why battery voltage sags under load. Davide website clearly discusses the subject. Look at the drawing he uses to determine voltage per cell of 2.0 volts under load. Myself I would not go quite that low because you are not running an EV. Somewhere between 2.5 and 3 volts. Seeing how most 12 volt Inverters default LVD is 10.5 volts works. Myself I would use 11 volts or 2.75 volts per cell. 3 volts is ultra conservative.
I sincerely do wish you good luck, but I cannot help you do a half arse job.
Hint. 100 Watt .01 Ohm power resistors are cheap.
[moderator note - please leave comments on other forum members out of your posts, unless you are addressing something specific that was posted]Last edited by sensij; 01-09-2018, 01:09 AM.Leave a comment:
-
I did think about that, even if you use the lower 13.75V (~3.44V/cell) if you look at my charge/discharge curves at a current of 12A (~0.13C) you get an SOC of ~90%, if the voltage had stayed at 13.75V and the current tapered down to 0.05C (4.5A) the SOC would be up around 97%.karrak, one think before i go through your answer - you may missed. the voltage shown during charging process on solar controller was 0.2v higher, than the voltage measured by hand multimeter and few other voltmeters. so in fact i believe the solar charger just measures more then real. also the charging curve of used/stored watts seems to be confirming it just wasnt so high. just the inaccuracy of solar charger controler - btw = scc
That alternative does not have regenerative discharge, discharge limited to 5A. The Reaktor is around the same price, has been around for a while, was designed by a well respected company and there is a large community of hackers that are doing some amazing things with it including working out how to calibrate it.im just checking the reactor 300w and this alternative
Doing the initial balance with a charger like the Reaktor will make it highly unlikely that you will have any unpleasant surprises when you first charge you battery from your solar controller but as you have already charged your battery with the solar controller without incident in makes the need for a charger like the Reaktor less important. I would still recommend you get an accurate multimeter, a cellog 8 or some other cell voltage monitor and an SOC meter that uses a current shunt to measure the current.
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 controllerLeave a comment:
-
sunking, theres no need for such a reaction. i would appreciate your answer to the presented scenario example.Leave a comment:
-
karrak, one think before i go through your answer - you may missed. the voltage shown during charging process on solar controller was 0.2v higher, than the voltage measured by hand multimeter and few other voltmeters. so in fact i believe the solar charger just measures more then real. also the charging curve of used/stored watts seems to be confirming it just wasnt so high. just the inaccuracy of solar charger controler - btw = scc
im just checking the reactor 300w and this alternative
Leave a comment:
-
You made a huge mistake buying two of those chargers and listening to a pretender. I know you bought them already. I am out of here. Karrak is your man. He will teach you how to destroy your batteries.im considering something cheaper - like this:
classical imax b6, 6A any battery charger/discharger/balancer dc to dc
also one question with that - will connecting 2 chargers like this in parallel to one battery result in double charging current? or it will not work because they will be changing the measured voltage/amperes to each other and fake themselves?
Leave a comment:
-
Yes that should work with a couple of reservations.- The one being advertised is a copy of the original, what is the accuracy like and how well built is it?
- If you want to do discharge testing the discharge current is severely limited
- There is no data logging
I can't see any reason why you couldn't use two of these units in parallel
I would still recommend the 20A Reaktor 300W as it has regenerative discharge and logging. It is what I did the charge/discharge tests with.
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 controllerLeave a comment:
-
I base the 95% on the fact that from the computer screenshot the solar controller stated the maximum charge voltage was 13.95V (3.49V/cell). If the 3.49V/cell occurred with the maximum charge current of around 12A (~0.13C) using my graphs gives an SOC of ~94%-95%. If the current dropping off from 12:05 is still at 3.49V/cell the SOC will be even higher.KARRAK - why do you mean i have reached 95%soc? based on that graph you cant see detailed voltages curve...
and ok, if yes, then maybe we found a safe way throught the mined field by luck, as we allready went throught without knowing its mined there... if that was really 95, then it is safe to charge it until then, right?
If the cells are like the ones in my system and the system I set up for a friend I doubt that the difference in the capacity of the individual cells will be greater than a couple of % so charging them to 95% and even a little higher should not have caused a problem. You are lucky that you got away with it though. I think it is really important to have some sort of cell monitoring at all times to warn you if a disaster is just about to happen, that is why I suggested the cellog 8. An accurate multimeter is also important.
Not sure what a scc is.?of course, give your idea how would you do that balancing based on existing equipment/limits of scc. i am interested, more info to consider the better...
OK,- Set the charge parameters back to what you have in your first post
- Let the battery charge, check the individual cell voltages while it is charging to make sure none go above 3.6V, this still gives you plenty of safety margin. The cell voltage has to get well above 4.0V before bad things start to happen
- When the controller has switched to float and the charge current has reduced below C/50 (~1.8A) measure all the individual cell voltages. Lets say the highest cell reads 3.45V and the lowest reads 3.40V.
- You have to remove some of the charge from the 3.45V cell to drop the voltage down to 3.40V. To do this you have to connect a resistor across the cell. A 3.3 Ohm 5W resistor works well or you could use the high beam filament of a tungsten car headlight which should be 65W @12.8V (~5A) which would be ~1.3A @3.4V.
- I would remove a fixed amount of charge from the cell by timing the amount of time that the bulb is connected. For each 0.01V above the lowest cell reading I would connect the bulb/resistor up for 1 minute so in the case of our 3.45V cell I would connect it up for 5 minutes. It is important to set an alarm timer when you do this as it is easy to forget and leave the resistor on for too long.
- After disconnecting move onto the next highest cell etc.
- The time I have specified will not be enough so you will have to do this several times. This can be done over several days
This is pretty much the procedure I use to keep my battery balanced with a charging voltage of 3.45V/cell. I use the daily maximum cell voltages logged by my voltage monitor as a guide as to when to to the balancing. I had to remove about 0.4%SOC (~1.5Ah) from 2 cells in my battery last year to keep it balanced.
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 controllerLast edited by karrak; 01-08-2018, 10:05 AM.Leave a comment:
-
im considering something cheaper - like this:
classical imax b6, 6A any battery charger/discharger/balancer dc to dc
also one question with that - will connecting 2 chargers like this in parallel to one battery result in double charging current? or it will not work because they will be changing the measured voltage/amperes to each other and fake themselves?
SUNKING - im not sure, if it was easy to understand, what i proposed... il try on example
ive got your point with top/bottom balance. that means 100% accuracy and safe 90-100%soc cycling of battery, which will provide me with more usable capacity in - repeat - safe conditions. my question was, whether it is not possible to balance them by connecting to parallel somewhere NEAR to top/bottom, but not so far as 2.5/3.65vpc.?? by my understanding this could be possible - BUT with decreased accuracy of SOC determination. determination of what SOC currently measured voltage is - is it right, or absolutely not?
EXAMPLE 100ah lifepo4 battery:
i know that for example at 3.1vpc it still can mean, that REMAINING capacity at this voltage can be for example
C1=20ah
C2=30ah
C3=10ah
C4=22ah
which should not be a problem at this moment, but if i will then consider this 3,1vpc to be 20%soc, then the problem will occure as soon as i will charge this 100ah battery with +80ah, as cell 2 and 3 will be overcharged then. (or even less than +80ah, as there is the problem of dividing the current between unbalanced cells on the end of the chraging curve, i know, i know. lets forget that for this example and hold this scenario values)
my point is, that i will NOT charge it in such a manner. instead of - i will charge it slowly, attended, by small amperes until they come to 3,4vpc, which will take lets say 50AH of charging. then the same cells should be (based on previously choosed start-charging state of charge/capaciity)
C1=20ah + 50ah=70ah
C2=30ah + 50ah=80ah
C3=10ah + 50ah=60ah
C4=22ah + 50ah=72ah
still much far enough from 100ah(100% soc in this example).
than, the cells if being charged slowly, slowly on and on with another lets say safe +1ah will continuously start becoming more and more disballanced, as they will be moving out of the flat part of the curve. CELL2 voltage will NOW raise faster than other. THIS is the point where i could consider tha battery to be charged ENOUGH (not fully), measure the actual voltage at lets say it will be 3.50vpc and set the charging voltage limit to 4x3.5=14.00v
so we would have - by this scenario - lov voltage disconnect at 3.1x4=12.4v and charging limit voltage 14,00v. because we will know, that here they start to become disbalanced and HERE they are getting out of the flat part of the curve. NEXT, to be even more safe and to include the inaccuracy of this process, we will change these values even more to conservative, safe area.
lvd=12.6
clv=13.8
which could give us a lot of reserve to compensate the inaccuracy. ALL future cycling will be between these values. in fact cell 2 will be cycled in about 30-90%soc, while cell3 in 10-70%soc as highest and lowest cell, but it still should be safe, if really kept there.
i hope this concept is easier to understand now.
KARRAK - why do you mean i have reached 95%soc? based on that graph you cant see detailed voltages curve...
and ok, if yes, then maybe we found a safe way throught the mined field by luck, as we allready went throught without knowing its mined there... if that was really 95, then it is safe to charge it until then, right?
of course, give your idea how would you do that balancing based on existing equipment/limits of scc. i am interested, more info to consider the better...Leave a comment:
-
You do not need to go down to 2.5V or up to 3.6V to balance the battery but doing so will balance your battery to an accuracy better than 1%. I balance my battery at 3.45V which will balance it to within 1%, if you picked to bottom balance at 3.1V you could get accuracy to within 2%. At a voltage any higher than this the accuracy decreases very rapidly as you are getting into the flat part of the charge/discharge curves.
The question I have for you is how do you reliably charge your battery up to 80% using solar as your charge source?
You are correct, as long is you make sure that the individual cell voltages don't diverge too far apart and stay within the range 3.6V to 2.8V within the SOC range that you are operating in you won't have any problems.by my understanding we will find where are the 20-80%soc limits and fullfill them in the future, we will not find the real capacity, but we will find where the safe interval is with some reserve of course, as this in not exact. doesnt matter, if that interval will mean 40ah, 50ah or 70ah of real capacity... just safe 20-80%soc of this one particullar battery...
From the information and graphs you supplied in an earlier post I would say you have already had your battery up to an SOC of ~95% which did not cause any problems, if you are prepared to do this again and are interested I can tell you how to balance your battery using the same or similar charge settings.
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
Leave a comment:
-
You cannot do Middle Balance without unacceptable risk because the capacity tolerance of your batteries falls outside acceptable limits. You would need cells within +/- 2%. Chi-Coms are +/- 10%. You have to pick Top or Bottom. You cannot define what 3.1 volts capacity is with LFP. You have to pick a known REFERENCE POINT of either 2.5 volts or 3.65 volts.
To be frank the real issue here is you do not have the equipment, skills, and knowledge necessary to use any lithium battery type which will most likely lead to destroying one or more cells. LFP cells are somewhat tolerant to over charging. But like all lithium batteries will not tolerate over discharge. For equipment you will need to have a means to discharge to 2.5 volts, or a DC power supply with a precision regulator to charge to 3.65 volts.
My suggestion since you already have your cells is to buy a Revolectrix Power Lab 6 Charger and use your 12 volt battery charger to supply the PL6 with Power. It will charge any battery of today and tomorrow of 6S Lithium (any lithium), 19S Nickel, and 6,12, and 24 volt Pb from 10 ma up to 360 AH. Additionally will allow you to Bottom Balance worry free. Just wire the cells in parallel, set disconnect voltage to 2.5 volts, discharge current to 10 amps, and go to bed.
Leave a comment:
-
Guys, please keep this discussion on topic. I'm cleaning out the tangent from the past couple posts, op is not engaging in that exchange at all.
Sunking, the NASA paper you linked is for Li-Ion, not LFP, irrelevant to the OP. Please start a new thread if you want to continue on that point.
Lastly, if you are going to cite someone else's work, please have the courtesy to get their name correct. (Davide vs David)Last edited by sensij; 01-06-2018, 10:27 PM.Leave a comment:
-
ok, so i went throu the site and downloading video, which will take a while as the internet is only 2g here. also, im currently midle-balancing the cells based on studied. after i disconnected them from serie, 3 of them sit at 3.27, one is 3.26 which seems not bad.
the question im thinking about is, whether we really need to discharge it to absolute 0%SOC = 2.5vpc, if i plan to cycle the battery between 20-80 SOC. wouldnt it be enough, to consider it discharged at lets say 3.1 vpc, which we would consider to be 10%soc with another 10% reserve... in fact it doesnt matter, as we will not go in the future down to 10%soc, as we will set the charge limit on 20%soc. where exactly this 20%soc limit is? we will measure that by capacity that will be charged from this point to up to lets say 90soc, which means point, where the curve will no longer be flat and the vpc will start to increase much faster under similar charging current...
so in fact it would mean we will not bottom balance, not even top balance, but we will instead notice when the charging curve starts to turn faster and immediately at those break-points middle balance on 10% soc +/-10%soc reserve then midle ballance it at 90%soc +/- 10%soc reserve with the same way. therefore +/-10%soc reserve, because the final cycling interval will be set to 20-80%soc... got my point? wouldnt it be usefull, if we just do not plan to use full capacity of the battery? and charging limit values will be monitored and set by measured voltages at this stages between charging process from cca 10 to 90%soc - as mentioned...
by my understanding we will find where are the 20-80%soc limits and fullfill them in the future, we will not find the real capacity, but we will find where the safe interval is with some reserve of course, as this in not exact. doesnt matter, if that interval will mean 40ah, 50ah or 70ah of real capacity... just safe 20-80%soc of this one particullar battery...Leave a comment:
Leave a comment: