Not Discharginging Batteries Enough at Night?

KJ7YM
KJ7YM Solar Expert Posts: 30 ✭✭
I have a dedicated solar array that just runs my refrigerator/freezer. I run a log of battery voltage every night, and of course it looks a lot like a square wave when graphed each morning. With a higher flat when the battery is at rest and the lower flats when under load by the refrigerator. My concern is that the final at rest portion in the morning before the panels get sun and start everything climbing back up is never less that about 24.9V. I am worried that it is unhealthy to just lightly discharge the batteries like this. Do you think I should add a further load to this thing? I guess this might work in my favor as the batteries age. The system has only been in service a few months. But still I wonder if I am harming them now?

Comments

  • Mountain Don
    Mountain Don Solar Expert Posts: 494 ✭✭✭
    What kind of batteries? 24 VDC I assume. How many watts of solar panel?

    Flooded lead acid batteries do need enough input to churn the electrolyte and the taller the battery case the more important that becomes.
    Northern NM, 624 watts PV, The Kid CC, GC-2 batteries @ 24 VDC, Outback VFX3524M
  • CALLD
    CALLD Solar Expert Posts: 230 ✭✭
    In general lead acid batteries last longer the less they are cycled & the shallower they are cycled, so in your case your batteries will last longer than they would if they were going down to 24v or lower. Of course the most important thing to ensure is that they are reaching their absorb voltage regularly and spending a few hours there at least once or twice per week.

    I could understand if your concern was that you are not fully utilizing available solar production - but that is unavoidable with an off-grid system. It is far more cost effective to lose say 20-30% of solar production during the absorb phase and keep your batteries happy than to try and use all the available solar power and risk not letting your batteries get a full charge. Replacing batteries that have died a premature death is far more expensive than having to run a generator or use grid power.
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Hi KJ7YM - Larry,

    I am in the camp that says that Lead Acid (LA) batteries should be discharged below 90% SOC in each cycle.

    Often, if one looks at manufacturer's Cycles verses DOD, one notes that these charts usually begin at 90% SOC, and descend from there.

    It is often generally said that LA batteries will develop Lead Dioxide clumping, which can damage the plate, and cause bulging, etc if they are not discharged below 90% SOC.

    The MidNite Classic CC allows a settable number of days of full charge to be "Skipped". On Skip days, the CC just Floats the battery. This approach can also result is less heating of the battery bank, which is also generally good.

    So, IMO, it can be good to avoid recharging LA batteries that are fairly often lightly-cycled. Used to avoid the recharge by simply switching the PV input to the CC for a couple of days. This was a manual system, but allowed for estimation of the approximate SOC of the bank, with the knowledge of the forecast for weather for the following few days.

    But, as CALLD mentioned, lighter discharge cycles are generally better than deep, deep discharges, within limits, as I read it. However, Forklift batteries are really designed to be hammered every day to about 20% SOC, but recharged quickly, immediately after the discharge
    FWIW, 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.
  • KJ7YM
    KJ7YM Solar Expert Posts: 30 ✭✭
    I just tested my Specific Gravity after a nights discharge. It is showing 1.245. I cannot really tell much detail from Crowns documentation except that this is higher than a 75% State of Charge.
  • KJ7YM
    KJ7YM Solar Expert Posts: 30 ✭✭
    Further Study into the subject has taught me that State of Charge can not really be determined from testing Specific Gravity. So my taking a reading of the SG after a nights discharge really gave me no valuable insight into what the State of Charge was just before the Sun began the charge cycle all over again.

    I now understand that Voltage at Rest an indicate the State of Charge, but that voltage has no bearing on the health or condition of a battery.

    Specific Gravity however can show the overall condition of a battery,such as when it is reaching end of life, as well as pinpoint imbalanced cells. But S.G. cannot be used to determine the State of Charge.

    I'm glad I got that cleared up in my head. :-)

    Now if I can just get a more refined State of Charge voltage chart for my Crown Batteries!!!! The Data Sheet for my particular battery does show that 2.05 volts per cell is a 20 DOD. Which would make it 24.6 Volts on my system. Still leaving me in the dark as to what my 24.9 readings each morning represent? A Crown representative did indicate that it is "approx. 15% DOD; at 24.9 volts" So I guess will have to live with that. If it is accurate I guess I am pleased and will not add any further load to my system.

    But does 24.9 seem a long way from 24.6 to represent a measly 5 percent difference from 20% DOD to anyone else?

    Larry
  • waynefromnscanada
    waynefromnscanada Solar Expert Posts: 3,009 ✭✭✭✭
    KJ7YM wrote: »
    Specific Gravity cannot be used to determine the State of Charge.
    Larry

    Actually SG is an excellent method of determining a flooded lead acid battery's state of charge. Actually it's the BEST way.
    1. The best way to determine when a battery is fully charged is to check specific gravity of each cell with a hydrometer. (background--As the battery discharges, the sulfuric acid content of the electrolyte solution is reduced, leaving water. Therefore, the specific gravity decreases as the battery is discharged).
    2. Specific gravity measurements show the state of charge of a battery filled with 1.265 electrolyte at the factory, according to the following chart.




    100%
    1.265


    75%
    1.225


    50%
    1.190


    25%
    1.155


    Discharged
    1.120
    http://support.rollsbattery.com/supp...ecific-gravity



  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    And, you should give the electrolyte a chance to equalize concentration (I usually guess around 3 hours or more).

    There is supposed to be an equivalent between battery SG and battery resting voltage:

    http://forum.solar-electric.com/forum/solar-electric-power-wind-power-balance-of-system/off-grid-solar-battery-systems/16512-battery-voltage-and-specific-gravity-question?p=209119#post209119
    Re: Battery Voltage and Specific Gravity Question

    SG factory levels are "variable":
    BB. wrote: »
    Here is a thread on Hydrometers and how best to use them.

    I understand that some tropical locations will use a bit weaker electrolyte (batteries will have a longer service life?).

    From this thread, the range of Electrolyte Specific Gravity seems to be:
    Specific Gravity vs Applications
    1.285 Heavily cycled batteries such as for forklifts (traction).
    1.260 Automotive (SLI)
    1.250 UPS – Standby with high momentary discharge current requirement.
    1.215 General applications such as power utility and telephone.

    As mentioned earlier, the specific gravity (spgr.) of a fully charged industrial battery, or traction battery, is generally 1.285, depending on the manufacturer and type. Some manufacturers use specific gravities as high as 1.320 in an attempt to gain additional Ah capacity, but at the cost of a shorter cycle life.

    ...

    Higher Gravity = vs Lower Gravity =
    More capacity / Less capacity
    Shorter life / Longer life
    Higher momentary discharge rates / Lower momentary discharge rates
    Less adaptable to "floating: operation / More adaptable to "floating" operation
    More standing loss / Less standing loss
    So, where do your readings fall in the above scale... And I would guess that a difference of 0.030 SG or less is nothing to panic over.

    And while there is a formula for converting from SG to Cell voltage:
    Specific gravity = single-cell open-circuit voltage - 0.845 (example: 2.13v – 0.845 = 1.285)
    Or
    Single-cell open circuit voltage = specific gravity + 0.845.

    Battery temperature, additives to battery plates, voltage measured when battery has not "rested" for 3 or more hour, etc. can all affect the readings.

    New poster "leaf" has a really nice set of charts that compare battery voltage against different rates of discharging and charging (as well as resting voltage readings). From here:
    vtmaps wrote: »
    I believe those charts are from Home Power #36, August- September 1993. Lead-Acid Battery State of Charge vs. Voltage ©1993 Richard Perez.
    Here is a link: http://www.scubaengineer.com/documen...ing_graphs.pdf

    vtMaps
    leaf wrote: »
    Am trying to upload the charts I am using...

    Attachment not found.Attachment not found.

    attachment.php?attachmentid=3655

    attachment.php?attachmentid=3654

    I don't quite a agree with the resting voltage line (I think the voltage is a bit low)--But it shows how to estimate a battery's state of charge while operating.

    Note, where the charts "flatten out"--the room for error estimating state of charge is pretty high.

    -Bill

    Note that as batteries sulfate, the full charge specific gravity drops (sulfur is permanently attached to lead).

    Something to pay attention to is how you are charging the batteries. The voltage that is being held (absorb) at something like ~29.0-29.6 volts (roughly) and for how long (2-4-6 hours, the deeper the discharge, the longer the absorb time).

    Also, watch the charging amps... The batteries are typically near 100% full if the charging voltage falls to ~2% to 1% of the 20 Hour Capacity (i.e., a 200 AH battery bank charging current falls to 4 to 2 amps)...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jcheil
    jcheil Solar Expert Posts: 722 ✭✭✭
    I was always under the impression that SG was the way to go always.

    Voltage is a "guess" because technically, you must measure the voltage with the batteries "at rest", meaning no load or charging for several hours.
    Nobody can leave their batteries "at rest' for several hours, just to check the SOC via battery voltage. That would be foolish, plus you would be in the dark (unless running a generator, but again a waste). SG is the way to go.
    Off-Grid in Central Florida since 2005, Full-Time since June 2014 | 12 X Sovello 205w panels, 9 X ToPoint 220w panels, 36x ToPoint 225w panels (12,525 watts total) | Custom built single-axis ground mounts | Complete FP2 Outback System: 3 x FM80, 2 x VFX3648, X240 Transformer, FLEXnet-DC, Mate-3, Hub-10, FW500 AC/DC | 24 x Trojan L16RE-B Batteries 1110ah @ 48v | Honda EU7000is Generator and a pile of "other" Generators | Home-Made PVC solar hot water collector | Custom data logging software http://www.somewhatcrookedcamp.com/monitormate.html
  • Mountain Don
    Mountain Don Solar Expert Posts: 494 ✭✭✭
    KJ7YM wrote: »
    Further Study into the subject has taught me that State of Charge can not really be determined from testing Specific Gravity.


    Where did you see that?? I have always believed in my hydrometer above all else.
    Northern NM, 624 watts PV, The Kid CC, GC-2 batteries @ 24 VDC, Outback VFX3524M
  • waynefromnscanada
    waynefromnscanada Solar Expert Posts: 3,009 ✭✭✭✭
    Check post #7 for info from Rolls Battery - - - - SG IS the way to check state of charge.
  • KJ7YM
    KJ7YM Solar Expert Posts: 30 ✭✭
    Me to Mountain Don, me to.
    Well I thought this might spark some interest. This is the site that kind of put it all into perspective for me. It turns out I was a little confused about the S.G.
    As the paper points out about the Specific gravity goes down a little bit after every charge. So what was a useful number whebn your battery was new is not really useful many months later. The article was excellent and after reading so many out there I had been getting more and more confused. Try it guys, it is fairly short, it does start out a little too deep for my liking, talking about formulas but that part passes quickly. Then he starts talking English, and explains things very well. What I really got out of this, is that you definitely, absolutley need both tools, an S.G. tester, and a Volt meter.

    http://www.smartgauge.co.uk/sg_soc.html

    I'm gonna hop over now and read post #7 for info from Rolls Battery to see what else I can learn this week.

    Larry
  • KJ7YM
    KJ7YM Solar Expert Posts: 30 ✭✭
    Oh guys! I almost forgot to say thank you to each and every one of you, I love learning and I appreciate every single comment.
    A special thanks to you Bill, your Some Guy all right? That had to take a while to put together. I found the different S.G. types particularly interesting.
    Sadly I think that my Crown CR-185's must be more of a fork lift battery, and I am a little depressed about that. But good to know.
    Again thanks everyone, I love the learning on the journey.

    Larry
  • Thom
    Thom Solar Expert Posts: 196 ✭✭✭
    jcheil wrote: »
    I was always under the impression that SG was the way to go always.

    Voltage is a "guess" because technically, you must measure the voltage with the batteries "at rest", meaning no load or charging for several hours.
    Nobody can leave their batteries "at rest' for several hours, just to check the SOC via battery voltage. That would be foolish, plus you would be in the dark (unless running a generator, but again a waste). SG is the way to go.

    We turn off all loads to the battery every night . The battery's rest 6 to 8 hours every night. Voltage does work for me
    Off grid since 1984. 430w of panel, 300w suresine , 4 gc batteries 12v system, Rogue mpt3024 charge controller , air breeze windmill, Mikita 2400w generator . Added 2@ 100w panel with a midnight brat 
  • CALLD
    CALLD Solar Expert Posts: 230 ✭✭
    Voltage is the most convenient way to check state of charge, but you need to "know your batteries". Agree with previous posts on the difficulty of taking resting voltages as it just isn't feasible to let the batteries rest for long enough - in fact even 8 hours is not enough as I've witnessed batteries in storage taking a full week or more before the voltage truly settles.

    Get to know the behaviour of your batteries under charge and under load. For my AGM batteries I have got to know how they perform under various loads and as such can estimate their state of charge quite accurately. You obviously need to know what is going in and out of the batteries at any given time - I've installed a shunt and a digtal millivolt meter to do that.

    For my 24v AGM's I use the following table:

    Discharging voltage:
    SOC; C/20; C/10; C5; C3
    100%; 26.0; 25.4; 25.0; 24.6
    80%; 25.2; 24.8; 24.5; 24.0
    60%; 24.8; 24.5; 24.1; 23.8
    40%; 24.3; 23.9; 23.5; 23.0

    Under Charging conditions its much harder because the voltage spread is much wider and charging current with solar is erratic sometimes - but it's safe to assume SOC is near 80% when the voltage reaches 28.8v. It's also safe to assume SOC is near 100% once the current tapers off to less than 2% of rated capacity at 28.8v (less than 2A per 100Ah at 28.8v for a 24v battery). If they did not reach 100% SOC or battery damage occurred you will notice it in the discharging performance.

    SG is handy for checking SOC on large FLA batteries but for AGM's, sealed or large banks of smaller FLA batteries is can be too time consuming to do regularly. Also taking SG readings before the batteries have started gassing wont give you an accurate result as the density won't be uniform yet.
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Hi Larry,

    It looks like your Crown batteries have a target SG of 1.275 for full charge.

    IMO, SG IS the Gold Standard for SOC. It will not be possible to measure exactly with most Hydrometers, and each Hydro will have its own Calibration and Repeatability errors. But I personally feel that I can bet within 2 - 3 points when looking for differences between cells during a specific SG measurement session ... not necessarily that close to absolute correct SG.

    I do forget weather you have good data on Charger set-points when using PV as the power source, but would expect that with the fill SG with your batteries, that you may need to set Vabs up at least a few tenths of a volt verses a battery with 1.265 SG electrolyte.

    One of the many reasons that using battery voltage, even on a battery that has been rested for a number of hours, is that the battery temperature has a fairly large effect on terminal voltage - - about negative 5 millivolts per cell per degree centigrade for many/most Flooded batteries. And it is a good idea to have an electrolyte thermometer, or at least a good readout of what the CC believes the battery temperature IS, so you can manually do the compensation. SG measurements on Hydrometers need to also be compensated, but, the compensation value is relatively small, and can usually be ignored unless the battery is HOT, or COLD..

    That Surrette SG vs SOC chart that wayne posted, should be close, it probably has a few percent error on the high-side.

    FWIW, 73 - Good Luck, 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.