SLA Batteries in series. . voltage issues

Shooter Magaven
Shooter Magaven Registered Users Posts: 16
First, I have no doubt that this has been asked and answered a hundred times. But I cant find it anywhere. So here it is again.. sorry.

I am not that new to solar off-grid, but I am new to 48v series battery bank. I thought I was told that batteries in series must be identical in every respect or the whole bank will die a horrific death (not really but might as well). And that parrallel strings can be more forgiving with different batteries. Then i read just the opposite....aaaaah! Im so freaking confused.

So I just sent two 3 year old 100Ah 12v AGMs too the battery warehouse in the sky because they would not play nice with the two newer (same type/model/manufacturer) batteries I put in series with. The voltages during absorb were all over the place. Never had this issue when they were in parallel. So I though that the difference in age was the problem and exchanged them.
Now, the all new bank is doing the same. I knew I needed to charge them first, so I wired the 2 new in parallel on a 12v charger for a day and a half. Then afterwards, connected in series with the others, I still get voltages of 15.2 to 13.6 when absorb charging at 48v (58.6v) from battery to battery.

am I being anal? Or is this a problem that will sort Iitself out. Im real tired of nursing batteries all weekend not knowing if i'm messing them up or correcting them.

Any help would be greatly appreciated.

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Series banks--They are strong as their weakest link. If you mix size of battery (200 AH and 100 AH cells) in series--The bank will only have 100 AH of useful storage.

    If you put AGM in series with flooded cell... AGM's have very low self discharge and do not like "over charging". Lead Acid have higher self discharge and can be over charged to a degree--So we "equalize" the string by over charging... The good cells are over charged and "leak current' to the under charged cells--All gets balanced in a few hours and all is well.

    Over charging an AGM will cause gassing and over heating of the catalyst--And ruin (or age faster) the catalyst. Excessive heat will cause cell to Vent.

    GEL battery in Lead Acid string, equalizing will cause gas pockets to form and ruin the cell.

    In a series string, if one cell fails (shorted or open), the rest of the cells are OK (if the failure is caught soon enough).

    In parallel strings, an open (failed) cell will stop current flow in the rest of the cells (in that string). Cells will not support loads (reduced battery capacity) and if not caught soon enough, cells in open string will sulfate and age/fail.

    In parallel strings, if shorted cell... That string will get most of charging current. Causing cells to over heat, "equalize", and age faster, use more water (until boiled dry). Also, shorted cell will discharge the rest of the parallel strings and ruin them too.

    All "good cells"--Variations in wire resistance, cell construction/age, cell temperatures, sizes will cause different strings to accept different levels of charging current/provide load current. Out of balance charge/discharge currents will age/cycle battery cells differently, causing weak cells to fail sooner.

    If I had 4x 6 volt batteries of different ages/life... I would put the matched batteries in series to make a "matched" string. And put the them in parallel strings.

    If you have AGM and Flooded cell strings in parallel, then your maximum charging voltage is limited to that which the AGM can withstand (14.4 volts max for AGM, 14.8 volts max for flooded cell).

    With out a battery monitoring system--There are all these various types of failures that can occur in a battery bank... You have to go out (for example) once a week and measure charging current per string with a DC Current Clamp meter, and measure cell/battery voltage with a volt meter. You are looking for "differences" that need further investigation.

    For me, the most difficult battery bank to "monitor" per the above is a 12 volt battery bank made from 12 volt batteries... you cannot measure cell/battery voltage to see differences.

    And once a month check/log specific temperature of each cell... A messy/tiring procedure (if you have lots of cells). But the gold standard for understanding the health of your battery bank (and you cannot do this with AGM/GEL/Sealed batteries).

    Secondary signs (high cell/battery temperature, buldged cases, boiling/hissing sounds, melted insulation on wiring, etc.) are other signs of issues.

    Put 24 cells in series (i.e., 48 volt lead acid battery bank), The failure of one cell becomes more difficult to "see".
    • 48 volt - 2 volt cell = 46 volts
    • 24 volt - 2 volt cell = 22 volts
    • 12 volt - 2 volt cell = 10 volts
    • 6 volt - 2 volt cell = 4 volts
    The higher the battery working voltage, the more difficult it is to "spot" a weak cell from the overall battery bank working voltage range.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Shooter Magaven
    Shooter Magaven Registered Users Posts: 16
    Thanks Bill,
    I would use flooded but I have no practicle way to vent the bank.

    Also, I see the voltage for AGMs being lower alot. But the battery mmanufacturer of these AGM state a bulk charge voltage of 2.45v/cell (14.7v). And float at 2.25-2.30v/cell. I have cut back the voltage a little to 2.4v/cell just incase.

    But are the different votages per battey that I'm getting normal? Will they ever match in series charging?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    You go with whatever the battery manufacturer recommends--At least as a starting point. And then check voltage/capacity/specific gravity/water usage/etc. results.

    Also, you have to look at the charging profile used... Many battery mfg. assume constant current charging... and when the voltage rises, then the charging is stopped.

    Others recommend a constant voltage charging and when the current falls below ~1% current flow (100 AH battery, 1 amp charging current), then terminate charging.

    With solar charging, we really do not have the ability to have constant current charging or "unlimited" constant voltage charging time--Plus you may have loads and/or even other charging sources.

    Off Grid Solar realities are frequently missing from battery mfg. literature.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    can you give us a table with the cell / battery voltages and show these in a bit of a chart that shows their configuration with voltages of the first battery and then at the start of the next one in the string to show V loss of the cable if any...??? how long are the cables, are they crimped fittings? etc
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • Shooter Magaven
    Shooter Magaven Registered Users Posts: 16
    westbranch wrote: »
    can you give us a table with the cell / battery voltages and show these in a bit of a chart that shows their configuration with voltages of the first battery and then at the start of the next one in the string to show V loss of the cable if any...??? how long are the cables, are they crimped fittings? etc

    The cables are custom 2AWG interconnect cables from Lawence Marine. They are all identically crimped and exactly 14 inches long. I think the batteries are at different states of charge, but not by much. Maybe one at 100% and another at 98% etc... Would this cause the voltages to very. Also 2 of the batteries were warm from transporting them home in 90 deg heat and tne other 2 were in airconditioning. I let them cool a bit before top-off charging. The initial SOC on both new batteries were off too. One was 12.7 ocv and the other was 12.9. Both were date coded Jan15.

    just a major pai trying to get them to exact SOC with each other and then with the existing bbatteries too. Should not have to go through all this.

    as far as a graph....the voltages fluctuate from one battery to the other dependeing on how I hold my mouth I guess. So that would be chasing my tail.

    Just want to know if this is normal or not with 48v strings. It is looking more and more like it is not by the responses so far.
  • Shooter Magaven
    Shooter Magaven Registered Users Posts: 16
    The Charger is a Morningstar TS-45 MPPT charge controler. It is a 3 stage Bulk/Absorb/Float charger. The literature apears to describe a constant volt charge profile for Absorb and float at least.


    BB. wrote: »
    You go with whatever the battery manufacturer recommends--At least as a starting point. And then check voltage/capacity/specific gravity/water usage/etc. results.

    Also, you have to look at the charging profile used... Many battery mfg. assume constant current charging... and when the voltage rises, then the charging is stopped.

    Others recommend a constant voltage charging and when the current falls below ~1% current flow (100 AH battery, 1 amp charging current), then terminate charging.

    With solar charging, we really do not have the ability to have constant current charging or "unlimited" constant voltage charging time--Plus you may have loads and/or even other charging sources.

    Off Grid Solar realities are frequently missing from battery mfg. literature.

    -Bill
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Solar chargers are pretty much variable current (bulk, based on amount of sun and loads when battery is <80% SOC) and voltage (absorb 80-95% SOC once the battery voltage rises enough to absorb set point, sun is up, solar panel power > battery power+loads power). And Float voltage (>95% SOC as long as the sun is up).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Wxboy
    Wxboy Solar Expert Posts: 70 ✭✭✭✭
    Shooter, any luck getting the voltages to be more consistent?

    I too have a 48 volt series parallel configuration that I've been using for about 7 months. I didn't start checking voltages on individual batteries until a few days ago and I am also seeing large variances. I have 8 100 amp hour 12 volt batteries. I am seeing 15 volts at one end of a series string and 13.9-14.0 at the other. In each series string the battery that has positive or negative cable bringing current into the bank is the battery with the highest voltage and the battery at the far end has the lowest voltage. The batteries in the middle of the strings have the voltage I'm shooting for but not the batteries on the ends.

    For many months I was only checking the current between batteries to make sure it was pretty even but I had neglected to check the individual battery voltages.
  • Shooter Magaven
    Shooter Magaven Registered Users Posts: 16
    Wxboy wrote: »
    Shooter, any luck getting the voltages to be more consistent?

    I too have a 48 volt series parallel configuration that I've been using for about 7 months. I didn't start checking voltages on individual batteries until a few days ago and I am also seeing large variances. I have 8 100 amp hour 12 volt batteries. I am seeing 15 volts at one end of a series string and 13.9-14.0 at the other. In each series string the battery that has positive or negative cable bringing current into the bank is the battery with the highest voltage and the battery at the far end has the lowest voltage. The batteries in the middle of the strings have the voltage I'm shooting for but not the batteries on the ends.

    For many months I was only checking the current between batteries to make sure it was pretty even but I had neglected to check the individual battery voltages.



    Sorry Wxboy, I have been away for a while.

    This is what I did to help correct the issue BTW, I had a 100AH 48v bank- 4x100AH AGMs.
    First, understand that you will likely not get EXACT readings across each battery. But, you should stay within +/-0.05 volts of one another). My batteries were bought a few weeks apart and all had different SOC from storage, heat, and self discharge. They were connected first as a parallel 12v bank as they arrived and later re-cabled as a 48v bank. This was the original problem but I didn't know it at the time.

    So anyway,
    I pulled all my Batteries and individually (or 2 batteries paralleled if voltage at rest was close) charged each one for 48 hrs @ 13.8v (75 DegF) with a 12v 2-stage golf cart-type charger.
    Next, I reconnected my bank and ran the inverter at 10Ah draw for 5 hours (50% SOC) so as to get a decent discharge (then recharge) to the whole bank.
    Next I charged the bank back to 100% SOC with 58.2v bulk/absorb charge at a 12 Ah current/rate.
    Next I floated the bank at 54.0v for 48 more hours.
    So far my readings are MUCH more consistent.

    My next 4 Batteries (arriving today actually) will be put on individual charge for 48 hrs like above before i ever introduce them to the bank. This is battery commissioning and is what was needed.

    The voltages I used were for 12v 100Ah SLA AGM UPS type batteries recommended from the manufacturers website for max charge rate, float voltage and bulk charging. Please use the ratings for YOUR specific batteries.

    Good luck.

    BTW, my voltage differences were in the middle, at the Pos and sometimes at the neg. It moved around. It seems to be very important that the inter-connecting cables be of the EXACT same length and the EXACT same crimp quality at the lug so you don't get different resistance between your batteries. I bit the bullet and bought some 14" marine grade factory crimps inter-connect cables from Lawrence Marine (no affiliation). If you own a $200 crimper and have quality cable and lugs you can make your own.... but I did not.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    That all sounds about normal. What you/we are doing there is a top balance, and is important for series connected cells. Ideally yes, all series cells would be from the same production batch, or be otherwise factory binned, but this rarely occurs in lead chemistry. So youll get natural variation in SOC and internal resistance that can at times lead to terminal voltage variance. You did the right thing bringing them all to 100% using a consistant workflow.

    Of course voltages vary by different amounts at differnet times. At rest, ie first thing in the am, with no or trivial overnite loads, a some millivolts per cell is considered averge. At start of absorb you will see more and at the end of absorb can be quite frightning to watch. Thats just the variation in cell performance starting to show. With flooded cells, the EQ process helps to bring the weaker cells into line with the others. With AGM, we have to rely on occassional longer absorbs, and decent afternoon floats help too.

    BTW the longer the string the worse this phenomona gets.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar