One Dead Cell out of 72 - Now What

Lance
Lance Registered Users Posts: 3
I have a system that is 7yrs old, consisting of three strings of 8 Trojan L16RE-B's.  Performance has slowly degraded over that time, as expected.  I feel like I got more life out of them than I deserved :)   Not good with keeping up with water...

Today I was checking the SG of each cell and I found one that is very "rusty" looking down inside and I can't even read the specific gravity.  The bulb won't even float.  I would assume the voltage on that one cell is zero?  I haven't isolated the battery to check it's overall voltage.

The bank works, but it can't handle a heavy load (3500 watts) for more than about an hour.  Small ~1kw loads are still not an issue. It'll handle it for hours if not days...

My question is: What should I do?  
I don't plan to replace the whole bank with LA - probably would go LiFePO4's next.  But the current banks isn't all use up yet.
I could try to buy a replacement L16RE-B, but they are a little hard to find since manufacturing of those stopped.  I think Trojan makes a drop in replacement though?
I can't pull the one offending battery, but I *could* pull out a whole string?  
Or I just leave it alone and wait?

Thoughts?




Comments

  • jtdiesel65
    jtdiesel65 Solar Expert Posts: 242 ✭✭✭
     Looks like you can buy one of those from NAWS for $414. It says there is a 7 yr prorated warranty.


  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Normally I'd think 7 years is good life out of L16 batteries, but I've heard so many good reports of the longevity of the Trojan RE batteries I wonder if you've been equalizing regularly? and have you considered trying some corrective equalizing now?
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    Lance said:
    I have a system that is 7yrs old, consisting of three strings of 8 Trojan L16RE-B's.  Performance has slowly degraded over that time, as expected.  I feel like I got more life out of them than I deserved :)   Not good with keeping up with water...

    Today I was checking the SG of each cell and I found one that is very "rusty" looking down inside and I can't even read the specific gravity.  The bulb won't even float.  I would assume the voltage on that one cell is zero?  I haven't isolated the battery to check it's overall voltage.

    The bank works, but it can't handle a heavy load (3500 watts) for more than about an hour.  Small ~1kw loads are still not an issue. It'll handle it for hours if not days...

    My question is: What should I do?  
    I don't plan to replace the whole bank with LA - probably would go LiFePO4's next.  But the current banks isn't all use up yet.
    I could try to buy a replacement L16RE-B, but they are a little hard to find since manufacturing of those stopped.  I think Trojan makes a drop in replacement though?
    I can't pull the one offending battery, but I *could* pull out a whole string?  
    Or I just leave it alone and wait?

    Thoughts?




    Welcome to the forum 

    Making the switch from flooded lead acid to LiFePo4 was the best decision I made, my suggestion is use the time the existing bank allows to study the requirements of the chemistry in detail, the equipment needed along with environmental limitations, freezing temperatures being the archillies heel. Building a DIY bank  is the most cost effective but requires more knowledge than pre manufactured but having the right equipment to properly charge is , in my opinion, essential.

    Maintenance wise there is virtually none, other than observation, with a good BMS the individual cells are protected, if one cell  goes out of parameter the battery is disconnected to preserve the ballance of cells.


    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • Lance
    Lance Registered Users Posts: 3
    edited March 2021 #5
    Photowhit said:
    Normally I'd think 7 years is good life out of L16 batteries, but I've heard so many good reports of the longevity of the Trojan RE batteries I wonder if you've been equalizing regularly? and have you considered trying some corrective equalizing now?
    I probably fall into the category of 'going too easy on them' during EQ.  Combined with not EQ charging frequently enough.   Fully charged, virtually every cell reports 1.275-1.285.  I haven't checked a delta, but I will.  Over the last 7 years I probably checked on them about every 3 months.  2 or three times, I caught a few of the cells with exposed plates.  I travel quite a bit and am away from the system for sometimes more than a month at a time.  Now with Covid, I haven't traveled in a year and keep a much closer eye on the system.  
    When I EQ, it's at 60v and only for a couple hours.  I only EQ on a full charge.  

    Can I "corrective EQ" with that one bad cell in the mix?  Can you suggest parameters?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,433 admin
    If you can do it--Shut down the system and disconnect that one battery--And run a 6 volt battery charger on a genset (or utility power) and try EQing that battery.

    My guess (no battery expert here)--That cell is dead (near 1.000 specific gravity). And I would not suggest you take the next few days playing with that battery.

    My suggestion--Look at the voltage across all the 8 batteries. If the rest of the batteries hold voltage well under load (3 kWatt)--I would be very tempted to replace the one battery with a new (or used if you can find a good replacement) and get another year or two out of the bank (assuming everything else is OK). Always a throw of the dice--But if the rest of the batteries are "in balance" (nearly the same voltage charging and discharging--May be a good bet).

    I believe you are in the Rochester NY area (?)--If so, how cold does your bank get. LiFePO4 batteries do well in 40-100F+ conditions. Much better than Lead Acid performance wise (price wise, not cheap, but hopefully cycle life with Lithium bank).

    The draw back with Li Ion (rechargeables) is that they generally should not be charged below ~40F and especially below 32F (always check specifications). Lead Acid are pretty rugged below 0F... If you run Li Ion, you need to keep them warm when cycling in winter (insulated box, a heater to warm the bank with a genset if cold soaked, etc.).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    Lance said:
    Photowhit said:
    Normally I'd think 7 years is good life out of L16 batteries, but I've heard so many good reports of the longevity of the Trojan RE batteries I wonder if you've been equalizing regularly? and have you considered trying some corrective equalizing now?
    I probably fall into the category of 'going too easy on them' during EQ.  Combined with not EQ charging frequently enough.   Fully charged, virtually every cell reports 1.275-1.285.  I haven't checked a delta, but I will.  Over the last 7 years I probably checked on them about every 3 months.  2 or three times, I caught a few of the cells with exposed plates.  I travel quite a bit and am away from the system for sometimes more than a month at a time.  Now with Covid, I haven't traveled in a year and keep a much closer eye on the system.  
    When I EQ, it's at 60v and only for a couple hours.  I only EQ on a full charge.  

    Can I "corrective EQ" with that one bad cell in the mix?  Can you suggest parameters?


    Monoblock flooded batteries, in my experience, almost without exception suffer from the demise of a single cell, which is masked by the remaining cells being at elevated voltage. This is difficult to detect without regular SG readings, once the cell becomes inactive it's extremely difficult to target it as an individual without affecting the remaining cells.

    The higher the cell count the less obvious a single cell has gone bad because everything on the surface seems to be normal, until the overall capacity comes into question. With early detection it may be possible to perform a corrective equalization, but this comes with the stress on the ballance of cells in the string.

    In an ideal situation the single cell would be isolated and charged with a 2V nominal charger, even removing a 6V monoblock then charging with a 6V charger subjects the remaining 2 cells to undue stress.

    Given the age of the batteries in question, it's unlikely there would be much  benifit in attempting recovery, as the effort required wouldn't justify the reward, if any. Having said that I did spend extensive time and effort doing a water treatment recovery (diluting the electrolyte with distilled water to encourage desulfation)  with success, albeit on a 1 year old battery with early detection. Life expectancy for FLA  is  3 years in 35°C temperatures, thus the switch to LFP which tolerate the temperatures where I'm located, in the tropics, they do a lot better due to their inherent efficiency, not self heating to the same degree during charging.

    Start thinking of possible replacements, I had considered tuber 2V cells, both flooded and AGM, but economics, geographic location  and suitability favored LFP, your situation may have better results with different chemistries.


    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • Lance
    Lance Registered Users Posts: 3
    Thanks all for the input.  Very helpful!
    I'm grid tied, so I could shut the system down.  But taking the offending battery out just to try to correction EQ it, etc.. doesn't seem practical at this point.  If I look down into the cell, the view is very different than all the others.  It's like looking into a rusty mess, while the other cells are all clear.  

    I may purchase a single Trojan L16 to replace it.  
    But otherwise, I guess I'll just limp along until other failures creep in and LiFePO4 prices continue to slowly decline.

    Now, thoughts on how to build a bank out of LiFePO4s that would roughly match my 3-strings of 8-series L16s?  I know I can go deeper with the lithium, so I shouldn't need to match watt for watt.  I never let the LA bank go more than 50% depth of discharge.  

    My bank is in an insulted crib/bin in the basement and never goes lower than 55F..




  • BB.
    BB. Super Moderators, Administrators Posts: 33,433 admin
    Look at the bad battery voltage string current.
    If the cell is open? The bad string does nothing.
    If the cell is shorted?  The bad string is discharging the other two strings.

    You may be better off with the bad string disconnected.
    Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    edited March 2021 #10
    Lance said:
    Thanks all for the input.  Very helpful!
    I'm grid tied, so I could shut the system down.  But taking the offending battery out just to try to correction EQ it, etc.. doesn't seem practical at this point.  If I look down into the cell, the view is very different than all the others.  It's like looking into a rusty mess, while the other cells are all clear.  

    I may purchase a single Trojan L16 to replace it.  
    But otherwise, I guess I'll just limp along until other failures creep in and LiFePO4 prices continue to slowly decline.

    Now, thoughts on how to build a bank out of LiFePO4s that would roughly match my 3-strings of 8-series L16s?  I know I can go deeper with the lithium, so I shouldn't need to match watt for watt.  I never let the LA bank go more than 50% depth of discharge.  

    My bank is in an insulted crib/bin in the basement and never goes lower than 55F..
    There are many claims that lithium batteries can be cycled to 100% of their respective capacity, although they can, it's recommend not to do so on a regular basis, but to discharge to 20% SOC maximum, to extend cycle life expectancy. Unlike lead acid they don't respond well to being fully charged, particularly being  held at that state for extended periods, so the general concensus is to charge to 90% SOC. This gives a useful 70% of nominal capacity, extending cycle life to > 4000, double the  100% value.

    Considering the useful capacity a 3 parrallel L16 bank is 645 Ah (50% of 1290) a comparable LFP  bank should be approximately 900 Ah. 
    As all systems are different the choice of overall capacity should be based on demands as well as charging ability and wether alternative charging sources are available, such as grid or generator. In my case,  being offgrid, I chose to replace a 400Ah LA bank, which worked well whilst it lasted, with 400Ah LFP, bank because I wanted to keep generator useage at a minimum.

    The beauty with LFP, unlike LA is the batteries or cells are assembled in parallel first, then series, so no matter what capacity is required, there is always a single string, one of the many benefits

    There are many sources of information, don't focus on one, in particular I've found Australian forums particularly useful as they seem to be ahead in the game. There are Web sites which provide a wealth of information,  I personally avoid the battery suppliers as their information may be biased, have to take everything with a grain of salt, so to speak.

    Avoid EV sites as the information is generally tailored to a different useage, which can confuse matters, they are high demand for limited periods, whereas in home storage it's generally low demand for extended periods.

    In conclusion, study as much as possible before making any decisions, I divided the pros and cons of LFP vs LA  on a sheet of paper, the results were clearly biased in favor of LFP, based on my tropical location, where heat is enemy number one, everyone has different priorities and I respect that, do your homework.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.