What is the optimal SOC range for cycling lead acid batteries?

124

Comments

  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    westbranch wrote: »
    The only difference I can ferret out is that it would be the length of time, maybe > 4 days, until the battery sees enough Ahr's to 'fill' it to a useable level, >80%, so deficit charging results over a month or 2 in summer and the battery would start to build hard sulphation. The start of a slow irreversible death.

    Chronic deficit charging, long term like due to insufficient PV to get the job done, is quite a bit different than having sufficient PV to get the job done once every week to 10 days or so. Surrette says that 50-80 cycling is optimal as long as the battery is returned to 100% state at least once every 30 days. These systems where there is insufficient PV probably don't have enough grunt to get the job done in that amount of time, so the batteries sulfate and start to die. It takes a long time for "hard" sulfation to accumulate to where a good absorb on a periodic basis can't put it back into solution. It's not an overnight thing, or even something that can happen in a week or two weeks. And I think that's a common misconception. These systems we hear about where the batteries are "flat" because of chronic deficit charging most times have been running a year or more and the owner suddenly decides there's a problem so they post it here. And the solution is to usually put the generator and inverter/charger to it and try to bring 'em back.

    Vic,
    What I was referring to was GB Industrial's claim that "opportunity charging" is bad for a forklift battery because of the heat generated in the battery. But again, RE systems are different than high-capacity three-phase industrial IUIa profile chargers. Those industrial chargers charge at extremely high rate during bulk, then switch to Vreg mode for a long absorb, then switch to a C x 5% constant current stage for finish. They do this to bring a big forklift battery from 80% DOD back to 100% SOC before the next shift. And they get the battery VERY hot - so hot that most forklift batteries have to be watered every 6-10 charging cycles.

    This is the life that a forklift battery lives. And they usually last 5-6 years in forklift duty, where the same battery on RE duty will last 12 years or more. After doing a bunch of research on this, and asking people in the know, the reason they last longer on RE duty is because the RE chargers don't create as much heat in the battery. Few people have enough RE power to charge a big forklift battery at C/5. That would be a 240 amp bulk charge rate on a 1,200ah battery. Most RE installations are lucky to make C/10 bulk rate. So the effect of so-called "opportunity charging" with a RE system is not going to be the same as a big 50 kW three-phase industrial charger.

    On the stratification issue, I have just not seen it. Our batteries are 2 feet tall and stratification has to show up as low SG because of high concentration of water in the top of the cell where the sample is pulled from. When we first got our batteries I think we maybe had some stratification issues - and that due to the fact that the batteries were 2 months old before we got them - they sat in a warehouse that long. I could not get proper SG on them. It took close to 2 months, and extremely high voltages (2.58 VPC), to get them up to 1.255. I've been using the high voltage ever since and never had another problem.

    For a tall cell battery that is "pampered" - shallow discharges and low voltage absorbing etc - I could see where maybe you might get some stratification. But it's going to show up as low SG. And anybody that gets that says, "Oh - I need an EQ cycle" and it fixes it. So I really think, like 'coot says, it's a non-starter of a problem. For somebody selling an Electrolyte Mixer, it can probably be made to be a bigger problem than it really is.
    --
    Chris
  • NorthGuy
    NorthGuy Solar Expert Posts: 1,913 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    On the stratification issue, I have just not seen it. Our batteries are 2 feet tall and stratification has to show up as low SG because of high concentration of water in the top of the cell where the sample is pulled from. When we first got our batteries I think we maybe had some stratification issues - and that due to the fact that the batteries were 2 months old before we got them - they sat in a warehouse that long. I could not get proper SG on them. It took close to 2 months, and extremely high voltages (2.58 VPC), to get them up to 1.255. I've been using the high voltage ever since and never had another problem.

    What you have seen is stratification. You're pumping energy into batteries at 58V for hours. Where do you think it goes? No heat, almost no bubbling. It charges batteries, creates acid. You measure SG - it is low - meaning there's very little acid in your hydrometer sample. Where's the acid? It's on the bottom. This is stratification.

    You increase voltage, it then starts bubbling, mixing, and voila - your SG is up. Stratification is cured.

    As you know, I had the same thing.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    NorthGuy wrote: »
    I have a question to the people who say they never seen the stratification. What would you need to see to say: "look, this is a stratification"?

    Quite correct: with some technical analysis you won't see it in normal use. It merely appears as low SG after what should be full charging (because insufficient mixing has occurred). This is because the SG is normally read "off the top" where the solution is least dense if it is stratified.

    In the old days when we had some clear-case batteries it could be measured at different levels refractively. Now you'd have to have a way to access the bottom of the battery (without disturbing anything) to check the SG variation. Over time it will show up as heavier sulphation on the lower parts of the plates, which you have to open the battery to see.

    Batteries are very much "black boxes" in that we have to judge what's going on inside by the limited amount of information available to us outside. With flooded cells you can at least check SG and Voltage. With sealed batteries you'd down to Voltage. If you've got the equipment to do load testing all the better, as you can get a good idea what shape the battery is in. But if it's bad you won't actually know why without opening it up, and that's not such an easy thing to do.
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    NorthGuy wrote: »
    What you have seen is stratification. You're pumping energy into batteries at 58V for hours. Where do you think it goes? No heat, almost no bubbling. It charges batteries, creates acid. You measure SG - it is low - meaning there's very little acid in your hydrometer sample. Where's the acid? It's on the bottom. This is stratification.

    You increase voltage, it then starts bubbling, mixing, and voila - your SG is up. Stratification is cured.

    As you know, I had the same thing.

    ALL of the Surrette batteries here bubble vigorously at 58-ish V. The amount of bubbling is a bit of a subjective thing, but 58 - 58.2 V on a 48 V battery is sufficiently high to allow full recharge in a reasonable length of time, from about 75% SOC. To try to recharge from a lower SOC only from PV in a single day requires a fairly high charge rate, and might require increasing the Vabs to a higher value to accomplish this recharge in a single day.

    Charging the largest banks here at C/10 causes very little heating in a two hour Bulk. As the Bulk V approaches the Absorption V, the heating rate rises significantly (based on measuring the electrolyte temperature - it takes a bit of time for the BTSes to Heat Soak to actual bat temp).

    Yes vtmaps, I am a bit of a Stratification "doubter". We do see an effect that is probably caused by a small bit of stratification, which is when using End Amps at the abs terminator, the first day of a full recharge does not quite bring the SG to correct nominal value of 1.265, it needs a second identical Absorption on the following day for SG to reach correct values. This is on a recharge from about 75 - 80% SOC. It probably means that for these recharges, the Vabs needs to be a bit higher to achieve full recharge in one day.

    The EA value and Vabs had been determined from doing daily full recharges from 90-ish percent SOC. Rather than change the Vabs from its correct nominal setting for daily full charges, just allow two full recharge days in succession to get the SGs to target values, and then back to 5-7 days of Vabs @ Vfloat +0.1V, and so on.

    Just the experience here. Doubting Vic
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    If you've got the equipment to do load testing all the better, as you can get a good idea what shape the battery is in.

    Yes, and I have that here (Sun Battery Tester with a 200A fan-cooled carbon pile in it). I load test them on every service.

    I think NG is right in that what I initially saw was more than likely stratification. But the batteries had been idle since manufacture. After I got them straightened out I have been doing what many people consider "abusing" them. Charging them at excessively high voltages (on Surrette's recommendation, though), and sometimes not charging them fully for 7-10 days. They have never been inactive since I bought them.

    I have no problems getting the SG up to 1.255-1.265 on a 7-10 day charge interval to full 100%. And they still test at 100% new capacity on a load test the last time I serviced them (two weeks ago) at 3+ years. I don't do a full 20 hour test - I load them at the 5 hour rate (40 amps) for 10 minutes and measure voltage drop under load, then measure the recovery voltage after setting for 10 minutes. A fully charged and healthy battery should not drop below 12.25V (our batteries are six cell 12V) during the 40 amp load test, and should recover to 12.55 after 10 minutes at rest. Every single one checks out brand new yet.

    So whatever I have done, I have not hurt them by "abusing" them. From fully charged, and just on battery power, we can load the inverter to 6 kW for an hour and the voltage at the inverter studs remains above bank nominal. If there was even one that was starting to get weak the battery bank wouldn't be able to do that either.

    People want to see "studies" and so on. My "study" has only been going three years, and so far so good. I'll let you know at 7 years how it panned out, or at the first sign of failure.

    One thing I am going to do in the next year is install battery monitors in each parallel string to keep tabs on amp-hours in vs amp-hours out. I figure that will be useful to provide early detection of a failing battery. I fully expect to have at least 1 or 2 fail before it's time as they age. That's pretty much normal with batteries when you buy a big bank.

    My next battery will be a couple big 2 1/2 ton forklift batteries. I will not buy little ones again.
    --
    Chris
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    Vic wrote: »
    Charging the largest banks here at C/10 causes very little heating in a two hour Bulk. As the Bulk V approaches the Absorption V, the heating rate rises significantly

    Vic,

    Our batteries can only be bulked in 2 hours @ C/10 if we get two nice days in a row and on the second day the battery starts out at maybe 80% SOC. Otherwise, on two good days in a row I have seen our batteries remain in bulk stage all day on the first day and get to gassing voltage but never quite make absorb. On the second day, they have been recovered enough on day 1 to get them to absorb within a couple hours. The absorb will last a varying time, depending on how fast they were bulk charged.

    But yes, definitely, does the temperature of the battery not start to rise significantly until it gets to absorb stage! I have never seen much of any temp rise at all during bulk - not ever. It's the absorb that gets 'em hot.
    --
    Chris
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    But yes, definitely, does the temperature of the battery not start to rise significantly until it gets to absorb stage! I have never seen much of any temp rise at all during bulk - not ever. It's the absorb that gets 'em hot.
    --
    Chris

    Because this is when the most activity occurs. Molecules in motion = heat.
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    Because this is when the most activity occurs. Molecules in motion = heat.

    It is difficult to break the bond between Hydrogen/Oxygen. This requires a lot of WORK which generates heat -- these are those molecules that are forced to move, and are reluctant to do so.
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • NorthGuy
    NorthGuy Solar Expert Posts: 1,913 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    Batteries are very much "black boxes" ....

    Not all of them are black you know. Mine have fancy brownish color. Surrettes are even red boxes :D
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    Vic wrote: »
    It is difficult to break the bond between Hydrogen/Oxygen. This requires a lot of WORK which generates heat -- these are those molecules that are forced to move, and are reluctant to do so.

    If that were the only thing occurring, the work would go entirely into chemical energy, not heat.
    I believe that what is happening is that energy goes into separating H from O and then some of that energy is released again when H + H --> H2 and O + O --> O2. That released energy has nowhere to go but into molecular motion, i.e. heat.
    SMA SB 3000, old BP panels.
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    inetdog wrote: »
    If that were the only thing occurring, the work would go entirely into chemical energy, not heat.
    I believe that what is happening is that energy goes into separating H from O and then some of that energy is released again when H + H --> H2 and O + O --> O2. That released energy has nowhere to go but into molecular motion, i.e. heat.

    Isn't it just because batteries have internal resistance? And when you feed electricity into something that has resistance you get heat? The higher the resistance goes (like during absorb), the more heat you get?
    --
    Chris
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    Isn't it just because batteries have internal resistance? And when you feed electricity into something that has resistance you get heat? The higher the resistance goes (like during absorb), the more heat you get?
    --
    Chris

    No; lower resistance causes greater current flow (for a given Voltage) and thus higher heat.
    As a battery charges its resistance goes up and the current through it goes down.
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    Isn't it just because batteries have internal resistance? And when you feed electricity into something that has resistance you get heat? The higher the resistance goes (like during absorb), the more heat you get?

    Current flow certainly causes heat. The most heating, however, occurs when the current is low... during late absorb. This is because of (as inetdog pointed out) exothermic chemical reactions related to gassing.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • northerner
    northerner Solar Expert Posts: 492 ✭✭✭✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?

    Lead acid batteries are still susceptible for heat production even when bulk charging. That is why a limit for the charge rate is adhered to, so that the batteries don't get too hot. The same holds true for rates of discharge. Of course, most off gridders stick to a rule, limiting the charge rate to about c/10.
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    northerner wrote: »
    Lead acid batteries are still susceptible for heat production even when bulk charging. That is why a limit for the charge rate is adhered to, so that the batteries don't get too hot. The same holds true for rates of discharge. Of course, most off gridders stick to a rule, limiting the charge rate to about c/10.
    Interesting, how much heat do they produce ?? I generally will charge a 1000 amp hr bank at 250 amps in bulk @ 50 % dod and I don't see much heating, maybe 4-6 degrees on a complete cycle ( 1-1 1/2 hours ). Once you reach gassing voltage ( absorb ) it's a different story, you can then drop to c/10 or so.
  • northerner
    northerner Solar Expert Posts: 492 ✭✭✭✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    Interesting, how much heat do they produce ?? I generally will charge a 1000 amp hr bank at 250 amps in bulk @ 50 % dod and I don't see much heating, maybe 4-6 degrees on a complete cycle ( 1-1 1/2 hours ). Once you reach gassing voltage ( absorb ) it's a different story, you can then drop to c/10 or so.

    Not my personal observations, but just going by what others report. I am getting conflicting info about the effects of heat when bulk charging. My question is then, why is the charge limit for lead acid batteries recommended between 5 and 13 % of the 20 hour rate, if heat or excess bubbling is not a factor? Maybe some batteries are more susceptible to over heating when undergoing a charge, or perhaps it has to do with the temperature the batteries are operating in. When the batteries go into absorb, the voltage is held constant, and current flow reduces over time, so the charge rate is not a factor then.

    Or is the reason for the restricted charge rate to reduce grid corrosion over time?
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    northerner wrote: »
    My question is then, why is the charge limit for lead acid batteries recommended between 5 and 13 % of the 20 hour rate, if heat or excess bubbling is not a factor?

    Usually it's 10-20% of the amp-hour capacity of the battery. I was told that it's primarily due to the optimum rate at which the chemical reactions can take place in the cell during the bulk charge.

    The same thing happens during discharge. If you discharge at low amp rates the battery has more amp-hour capacity than it does if you discharge at high amp rates.
    --
    Chris
  • NorthGuy
    NorthGuy Solar Expert Posts: 1,913 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    Usually it's 10-20% of the amp-hour capacity of the battery. I was told that it's primarily due to the optimum rate at which the chemical reactions can take place in the cell during the bulk charge

    I think the heat is important too. The amount of heat is proportional to the square of the current. If you charge at 10%, the heat may not be detectable without special means. If you start charging at 50%, you get 25 times more heat and this can damage the battery.
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    NorthGuy wrote: »
    I think the heat is important too. The amount of heat is proportional to the square of the current. If you charge at 10%, the heat may not be detectable without special means. If you start charging at 50%, you get 25 times more heat and this can damage the battery.
    All you have to do is prove that theory with some real documentation. My tests prove to me that I can double the current in Bulk with very heat rise with the temperatures recorded at plate level.
  • NorthGuy
    NorthGuy Solar Expert Posts: 1,913 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    All you have to do is prove that theory with some real documentation. My tests prove to me that I can double the current in Bulk with very heat rise with the temperatures recorded at plate level.

    I once turned on the charger when sun was on (not intentionally), and the charge rate was 23%. The battery temperature rose 7C (or so, forgot the exact number) in an hour. My regular bulk charging only rises temperature by 1.5C/hour - I recorded many of these. This is inline with Ohm's law. I belive that if I charged at 46%, the temperature would rise 28C (must discharge to 20% SoC first to make sure it goes for the full hour).
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?

    Which battery? :D

    The higher current means higher heat. This is physics; you can't get around it.

    But heat is a relative thing. Higher than what? Blackcherry was getting a few degrees temp rise at 25% charge rate. But a few degrees above what? Before a battery is going to suffer damage from heat it has to get hot enough to be damaged; the temperature rise alone means nothing.

    How long the battery is subjected to the heat will make a difference too. First you'd have to warm the rather significant internal mass up near temps that would cause damage. All batteries will be able to radiate some heat, based on mass to surface area and conductivity of construction. In normal applications even the 25% current BC was using is not going to be present long enough to do any real damage.

    So there it is: apply the current long enough to raise the internal temp to damaging levels and you can get damage. But Bulk stage doesn't last that long, and as it progresses the current will come down and the Voltage will go up. When you get to Absorb the process of chemical separation comes into play, and that sustained higher Voltage causes a lot of molecular activity inside: heat. This is why extended equalization can be damaging: even higher Voltage = even more heat and even more potential for damage.

    And it is always only potential as every battery is not the same. Some can take the heat better than others.

    There are a lot of theories about batteries, most of them formulated in a lab under controlled conditions. We don't live in a lab under controlled conditions though. It is the real-world variables that will 'get you' more so than any scientific technical point.
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?

    I'll have to wait till tomorrow, for some accurate readings, I am already back up to 12.8 @ 250 charging amp's and 78.4 f internal battery temperature. I think the temperature was 76.xx when I started, didn't pay that much attention. Under normal conditions it'll get to 88.x f by time I finish absorb. The 12.8 V is over voltage from the charger, when the actual battery voltage catches up, it's been @ 12.8 since I started the generator, it'll go up faster my normal temperature rise is about 10-15 deg F from start to finish. Most all published numbers use 110-125 F as the danger zone on temperatures.
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    We don't live in a lab under controlled conditions though. It is the real-world variables that will 'get you' more so than any scientific technical point.

    With the Schneider Conext ComBox you can turn your RE system into a lab and do as detailed data analysis as you want. This was an absorb charge cycle that I logged at 1 minute resolution on Sept 2 with our prime diesel generator operating with the inverter in Gen Support mode. There was limited power available from the XW inverter for battery charging. We had cloudy condition with the sun popping in and out. But it gives you an idea of the relationship of net battery current at various stages of the charge cycle. This absorb cycle lasted for 1 hour, 18 minutes.

    Although I did not graph it because it was really boring, the batteries started out at 26C at the beginning of the morning, were at 27C at the start of absorb, got to 28C by the end of absorb, and continued to rise to 29C after absorb was done. At 7:30 that evening, the batteries were still at 29C. 4,800 lbs of lead does not change temperature readily.

    AVPageView+972013+93617+AM.jpg

    --
    Chris
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?

    That's not exactly what I meant.

    Collecting data is easy. But that doesn't mean the data proves anything other than the system works (or doesn't) in a range of conditions.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: What is the optimal SOC range for cycling lead acid batteries?

    The rules of thumbs we use are to get a generic solar RE power system design quickly and easily to a workable solution. Once a system is roughly sized, you can select the components and then read the manuals for those units to confirm the actual limits (and adjust your design to those "actual" limits--if needed).

    More or less, this is the reasoning for the limits I have seen:

    5% * C20 rate of charge. Several reasons.
    • A couple mfg. of batteries recommend a range of 2.5% to 5% for equalization current.
    • For older flooded cell batteries (particularly forklift types), their self discharge can be as high as 1-2% per day--If you have 6 hours of sun, then 5% rate of charge will barely maintain the battery "old" capacity.
    • The "nominal" off grid design of 2 days of storage (1-3 days normal suggested range) means that you are discharging 25% of capacity per day. A 5% (or C/20) rate of charge means that it takes about 5 hours of "full" sun (~1 day of charging during summer) to fully recharge a battery bank. Add self discharge and less than ideal weather, this is about 2 days to recharge a 25% discharged bank. That seems to be a reasonable minimum rate of charge. People that do less than 5% seem to get into the deficet discharging trap unless they are willing to use a generator to help recharge the battery bank (fuel costs and noise keeps many people from using the generator as much as they should in poor weather/during heavy loads).

    10% * C20 rate of charge.
    • Trojan recommends a 10% (C/10) rate of charge for (at least some types) of their batteries.
    • With solar panels as cheap as they ever have been (and perhaps ever will be), This will recharge a 50% discharged bank in 1-2 day during most clear days all year long for most people (short of Canada/Alaska/etc.).
    • Generator usage is usually limited to the three months of winter (or monsoon seasons for others) for most people.

    13% * C20 rate of charge.
    • This has been the recommended maximum rate of charge (C/8=112.5%) for many manufacturers without thermal management (i.e., cooling fans). Realistically, this is probably based on industrial chargers that pump rated current in until batteries are nearly full. The Solar RE chargers probably pump out "rated current" for only a few hours per day... So, a solar RE charger is less likely to overheat a battery (exceed 120F/50C) unless way over the C/8 rate of charge.
    • This is probably also the maximum rate of charge I would recommend for using a charge controller without a remote battery sensor.
    • Solar off grid is also a lot about energy management and efficiency. Flooded cell batteries become less efficient at charging/discharging the higher the rates of current... I try to design a system with C/8 maximum rates to keep batteries "happy", wiring/cabling reasonable sizes, heat down.
    • Batteries (and most everything else) ages faster the hotter they get. The rule of thumb (based on Activation Energy) is for every 20C increase in temperature, the "thing" will age 2x faster. A battery bank kept at 0C will last, very roughly, ~2(50C/20C)=5.7 times longer vs a 50C lead acid battery. So keeping bank temperature lower will help with longer battery life.

    20% to 25% * C20 rate of charge.
    • I have used this number as the maximum I would recommend for generator charging... Typically people use generators during winter/poor weather and, for solar types, at the lower state of charging (50-80% state of charge). Batteries are in a cooler climate and much less charging current goes to heat/electrolysis when below 80% state of charge. Near 100% SOC, from what I recall, approximately 40% of the charging power goes into heat (and the balance goes into creating hydrogen+oxygen gases).
    • For such high charging currents, I highly suggest a remote battery temperature sensor to help reduce the chance of thermal runaway.

    40% * C20 rate of charge (discharge).
    • For flooded cell batteries--Going beyond 40% rate of charge/discharge (C/2.5)--You run the risk of taking the battery out of "regulation". Normal charge controllers that we use are not really designed to hold stable DC voltages. They actually use the battery bank to hold a stable voltage range (many chargers/alternators/etc. will not hold a stable voltage with no battery or a "very small" battery).

    100 AH @ 48 VDC per 1 kW of charging/discharging
    • Note that this rule of thumb would be 200 AH @ 24 VDC or 400 AH @ 12 VDC battery bank per 1 kW of inverter/charger.
    • This was based on several suggestions... This basically the 40% maximum discharge rate for a 1 kW AC inverter with 2kW of maximum surge support.
    • Was originally created for Hybrid AC Inverters where you have a solar array "floating" a battery bank and a Hybrid Inverter feeding power back to the grid. A single phase AC inverter actually draws an "I2" current wave form from the batter bank to supply the AC inverter's voltage/current output to feed into the grid. This limit does two things--First keeps batteries from overheating because of "ripple current" and also reduces the chance of "micro cycling" (where peak current pulls battery voltage below 12.7 volts and actually discharges/recharges the battery at 100/120 times per second).
    • Also a good limit for MPPT type charge controllers/array sizing upper limit. MPPT controllers (some/most/all?) seem to draw ~100% of the arrays power/current every few minutes to figure out Vmp-array. This is a normal operation, but all that current has to go "somewhere". That somewhere is into the battery bank. If the battery bank is too small (and/or aging and getting higher internal resistance), this can take the battery bank out of regulation. A couple of systems here have reported Inverter Faults with battery bus voltage >72 volts (for a 48 volt battery bank).

    2% of C20 maximum float current:
    • This is not one that we have discussed much before here.... But I have found it to be interesting. In doing research on why batteries fail--I ran across one website that mentioned battery fires in large installations (UPS/Telecom/etc.). And their failure analysis has shown that in most of those cases, the float current was >~1-2% rate of charge (C/100 to C/50)... That also would agree pretty well with a 2.5% maximum equalize current--Any more current runs the risk of "boiling" a battery bank dry and starting a fire.
    • Most battery banks will be well below 1% rate of charge (except pretty much at end of battery bank life)... And AGM/GEL batteries are usually below ~0.1% rate of charge.

    The above are the reasons I uses for our various rules of thumbs around battery banks. They are specifically for Flooded Cell batteries.

    Some brands/models of AGM batteries can supply/absorb much more current (Concorde are rated 4*C maximum rates or 15 minute to discharge). Cabling has to be much heavier. And I would not assume that such a design would have very long cycle life (most UPS batteries seem to have about 1-2 year life in a typical UPS application and will probably only support 10-20 such cycles before "dieing").

    The above are my personal observations/readings--I am not a battery engineer... Discussions are welcome.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    That's not exactly what I meant.

    Collecting data is easy. But that doesn't mean the data proves anything other than the system works (or doesn't) in a range of conditions.

    Collecting data and analyzing it (exactly what is done in a lab) provides you with a baseline. So then you make a change to the system (like absorbing every 7 days instead of every 1 day). Then pull your new data out and compare it to the old baseline. What did it do to average and peak battery temp with the new settings?

    Real world data analysis on real world systems is always much more useful than lab controlled conditions.
    --
    Chris
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    Real world data analysis on real world systems is always much more useful than lab controlled conditions.
    --
    Chris

    Which is closer to what I meant. :D

    Once you get a system working though all that data analysis is so much busy work.
    You'd be amazed at how infrequently I look at my system's data.
  • NorthGuy
    NorthGuy Solar Expert Posts: 1,913 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    ChrisOlson wrote: »
    With the Schneider Conext ComBox you can turn your RE system into a lab and do as detailed data analysis as you want. This was an absorb charge cycle that I logged at 1 minute resolution on Sept 2 with our prime diesel generator operating with the inverter in Gen Support mode. There was limited power available from the XW inverter for battery charging. We had cloudy condition with the sun popping in and out. But it gives you an idea of the relationship of net battery current at various stages of the charge cycle.

    That's off topic, but since you posted these graphs, I can't help some analiyzing.

    1. See how at 12:45 XW system depressed solar and kept XW charger at full blast. Would be nice if they could reverse that. May be they post a software update ...

    2. For the period from 12:20 to 12:50, you could've shut down the generator. There was some solar power available for sure - you can see how when generator output goes down at 12:25, solar fills in the gap. Similarly, when generator power goes up at 12:40, the solar drops down.
  • ChrisOlson
    ChrisOlson Banned Posts: 1,807 ✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?
    NorthGuy wrote: »
    1. See how at 12:45 XW system depressed solar and kept XW charger at full blast. Would be nice if they could reverse that. May be they post a software update ...

    2. For the period from 12:20 to 12:50, you could've shut down the generator. There was some solar power available for sure - you can see how when generator output goes down at 12:25, solar fills in the gap. Similarly, when generator power goes up at 12:40, the solar drops down.

    No, there may have been a little excess solar available during the absorb stage. But not enough to carry system loads plus get the absorb stage done, that hadn't been done for close to two weeks and the battery needed it pretty bad. The system spent all day for three days in a row in bulk stage, and kept gaining on amp-hours. I assumed we were finally going to get a "free" absorb on 9-2. When it became apparent it was not going to happen I made the decision to start the prime generator and run it at 9 amp load to take some of the load off the RE system to get the job done.

    The solar output was varying wildly in the morning due to clouds from lifting fog. The only thing that got the system to absorb, with the genset running, was the fact that it cleared for a bit. But then it clouded over and it steadily deteriorated. Between 12:30 and 12:45 the solar dropped dramatically due to a rain shower so the system bumped the XW charging current up to compensate for it. The rain shower went thru and the solar came back, the system detected that there was sufficient solar power available, and it stopped using the XW inverter/charger altogether and let the solar carry it, waiting to stop the generator on the Stop On Float AGS trigger. The clouds from the morning rain cleared up after lunch and the solar maintained float all afternoon, plus carried loads.

    For people that only periodically absorb their battery, this is the way it has to be done sometimes. I could go back and look in the ComBox to see how many hours I ran the little diesel that day, but I think it was around 4.8 hours.

    The thing is, most people with a large system are going to do this differently. They'll run the system until the battery is depleted, then start a 12 or 14 or 20 kW generator, or whatever, and finish what the RE didn't do with the XW inverter/charger putting out 90-100 amps. By having this data and being able to analyze how to do it more efficiently, I can do it with a tiny 3 kW diesel set at 2.2 kVA peak load on a 4.8 hour run. So how much fuel do you think I burned (hint: ~1.3 gallons) compared to the "traditional" way of doing it?
    Once you get a system working though all that data analysis is so much busy work.
    You'd be amazed at how infrequently I look at my system's data.

    Like you said, collecting data is easy. It doesn't take any time at all, doesn't require any effort, and it's fully automatic. But at some point when you decide to make a change to your system you can either go seat-of-the-pants or pull your data out and analyze it to see if what you're doing is the right thing to do. It's better to have the data and not need it than it is to not have it at all.

    My goal is to make off-grid living more affordable by making the right decisions on when to run generators, and reducing the costs of the consumable items like batteries. Without the data to see if what you're doing is working, all you got is the proverbial WAG.
    --
    Chris
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: What is the optimal SOC range for cycling lead acid batteries?

    I spend much of my time fixing systems that don't work; getting them to work with as few changes and little addition investment as possible. In those circumstances the relevant data comes down to "works/doesn't work" and there is no vast collection of logs to analyze.

    Some people will spend far too much time/effort/money getting a system that runs an average 58% efficient to run at 59% efficient. As an engineer of my acquaintance was fond of saying "talk to me about 10% or don't talk to me". :D