Battery state of charge: specific gravity vs. voltage

Options
benthicbread
benthicbread Registered Users Posts: 10
I live in an off-grid, PV home in Vermont. We've been struggling with lead-acid batteries for 20 years and just about ready to throw in the towel... Our latest adventure is with a set of 24 Surrette S-1380's, configured in one string for a nominal 48V system. These are rated for about 1050 A-hr (more at the rate we use them...). We use about 120-140 A-hr/ day for two houses. This use is pretty consistent - up to about 200 A-hr/day parts of the year. We don't have enough solar for this large bank (2400 W - which is not much here in VT), but another 2800W just arrived and will go into play this week. That said, a big part of the year, there is plenty of solar generated to maintain this bank fully charged.

The big problem I am having is that my voltages don't seem to reflect the same state of charge that my hydrometer readings show. I use three different meters to check voltage and check specific gravity with a temperature compensated refractometer. Specific gravity readings tend to be off by as much as .03 between adjacent cells when I haven't equalized recently, but tend to show what I would expect for state of charge. Voltage I monitor under various load conditions, but will shut system down periodically for 15 minutes after a moderate load and check it for at-rest voltage.

My auto-start on my generator is kicking on more than daily right now. Yesterday, I had left auto-start off and did some monitoring. Specific gravity was all between 1.240-1.275. Voltage had dropped low enough to shut down inverters (44.0 V) under load. I turned on auto start, let generator charge for half an hour or so, shut it down, put a load on it for a short time, then shut system down for 15 minutes. Voltage stabilized at 47.3 V.

According to Surrette's literature, specific gravity shows about a 80-90% state of charge (which is consistent with what our pentametric meter had clocked). At-rest voltage should be 49.68V at 75% state of charge.

I'm at a loss here - we realize that we need more PV to keep our bank maintained properly, but now I wonder if that is going to solve our problem. The way we've been using propane, we might as well get rid of the whole system and run off of the generator! Any help would be greatly appreciated. Thank you very much.

-Eric Warren

Comments

  • Volvo Farmer
    Volvo Farmer Solar Expert Posts: 209 ✭✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Sounds like you might have a bad cell? Did you recently check every single cell with a hydrometer and for voltage, or did you just test a pilot cell?
  • offgrid me
    offgrid me Solar Expert Posts: 119 ✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    I hope you are not using distilled water to calibrate the refractometer. You will not get an accurate reading. Why is your inverter shut down voltage so low. I have mine set to 47v. It seems like your batteries need a very long absorb setting and probably have not been getting a full charge for some time.
    Ned
  • BB.
    BB. Super Moderators, Administrators Posts: 33,474 admin
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    You have a single string of batteries in series? (That is a good thing).

    Then I would check the voltage across each cell/battery and make sure they are all pretty much equal. If you find one battery/cell with higher or lower voltage, then start looking for reasons.

    Check the voltage when the battery is under "heavy" load and also under "heavy" charging.

    And, set the digital volt meter to 2 volt (or 200mV/0.2 Volt) full scale and check voltage drops across each length of wire and electrical connection. You want all drops to be low and pretty much the same--If you find something different (high drop), then check connection/cabling for corrosion, etc.

    What is the absorb voltage setting and how many hours do you typically get in absorb (like 58 volts for 2 hours, etc.)?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Welcome to the forum, Eric. How old are your batteries?
    offgrid me wrote:
    I hope you are not using distilled water to calibrate the refractometer.
    good point, Ned.

    Eric, sulfation is the most common cause of premature low capacity in a lead acid cell. If your SG are accurate and you don't have a shorted cell, then sulfation is not the problem. There are other several other ways a cell can fail... tell us more about your battery's age and charging parameters.

    --vtMaps (also giving my generator a workout these past few weeks)
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Couple of points:
    Fifteen minutes is not a long enough time to get a resting Voltage from. The batteries need to sit for at least two hours with no current going in or out.
    The relationship between resting Voltage and SOC is less than perfect to say the least. It is entirely possible to have the "correct" Voltage but not full charge if the Absorb time isn't long enough and/or the batteries are sulphated (in which case they may be "fully" charged but have lost capacity).

    Monitor the current going to the batteries during Absorb with no loads on. See if it drops below about 3% before the Absorb stage stops.

    You're correct that 1050 Amp hours of 48 Volt battery needs a lot of panel even in good, sunny conditions. About two 3275 Watt arrays with a charge controller each. As it is your peak charge rate with 2400 Watts of panel is less than 4%; well below the minimum.

    You should start be re-evaluating your power needs. That much battery could supply roughly 24kW hours of power, which is massive for off-grid. You say you are using less than 200 Amp hours per day? That's 9600 Watt hours or about 20% capacity. Really you'd need at least 4600 Watts of panel to supply that, with good weather. No wonder the generator starts so often. You need to either shrink your power requirements or expand your PV array significantly.

    And when you buy your new batteries, don't buy Surrettes.
  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    I see no mention of any EQ (equalizing ) charging having been done in the posts so far... If you haven't done it, it is needed...
     
    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
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    I did check all of my cells (quickly) and have done so several times in recent months. All cells look good - at this point they range from 1.240-1.275 (almost this range in some adjacent cells). I have not taken it apart to check individual cells for voltage. Thanks.
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    I am using distilled water (as it says to do in the instructions that came with it). What should I be doing?

    Inverter shut-down is set higher for longer time periods, but I am going with the default setting as voltage can drop over short periods of time when big loads are on.

    -Eric
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Yes, this is a single string in series. I will try checking individual cell voltages.

    My charger has a setting for "bulk", which is set to 58.8V and "absorb time" which has been 2 hours (I've typically changed this to 3 hours this time of year).
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    I did check all of my cells (quickly) and have done so several times in recent months. All cells look good - at this point they range from 1.240-1.275 (almost this range in some adjacent cells). I have not taken it apart to check individual cells for voltage. Thanks.

    This is Westbranch'es point: SG ranging from 1.240 to 1.275 is too wide a scope; a difference of 0.035 between any two cells is definitely indicative of a need for equalization.
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    This is Westbranch'es point: SG ranging from 1.240 to 1.275 is too wide a scope; a difference of 0.035 between any two cells is definitely indicative of a need for equalization.

    OK I did equalize a few days ago (63 V for 4 hours -Surrette claims that this voltage is needed). Batteries have been in service for 1 year and I have brought them up with equalizing charges several times (using sun and generator).

    It sounds like you are saying I should expect batteries with this big a spread in specific gravity readings to have very little capacity, even when the bottom end of the spread is indicative of a high state of charge? Should I be concerned with how fast the batteries are sulfating given that equalization happened regularly with sun until a month and a half ago (and they have charged very regularly with generator and sun since)? If this is just a matter of needing more frequent, long charges, than adding PV seems like it will help with this - my concern is that something else is wrong here.

    Thanks for your help.

    -Eric

    PS am I using the forum correctly? I responded to individual posts, but it looked like my replies didn't include quoted text and were therefore very confusing...sorry
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    An SG of 1.275 is 'full charge'; 1.240 is not. The purpose of equalization is to try and get all the cells to have the same SG, whatever maximum they may be capable of (as batteries age their SG will go down regardless). You should apply the EQ Voltage repeatedly (but not continuously: one or two hours at a time) until the SG on the low cells doesn't rise any further. A low SG in one cell in a battery will indeed reduce the capacity of the whole battery. You should try to get them all equal. Batteries or individual cells that stay below 75% SOC will sulphate faster, making correction impossible.

    If you want to include the text of a previous post you need to click "Reply With Quote" in the bar just below the post on the right side. :D
  • BB.
    BB. Super Moderators, Administrators Posts: 33,474 admin
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    OK I did equalize a few days ago (63 V for 4 hours -Surrette claims that this voltage is needed). Batteries have been in service for 1 year and I have brought them up with equalizing charges several times (using sun and generator).

    It sounds like you are saying I should expect batteries with this big a spread in specific gravity readings to have very little capacity, even when the bottom end of the spread is indicative of a high state of charge? Should I be concerned with how fast the batteries are sulfating given that equalization happened regularly with sun until a month and a half ago (and they have charged very regularly with generator and sun since)? If this is just a matter of needing more frequent, long charges, than adding PV seems like it will help with this - my concern is that something else is wrong here.

    Thanks for your help.

    -Eric

    PS am I using the forum correctly? I responded to individual posts, but it looked like my replies didn't include quoted text and were therefore very confusing...sorry

    Eric,

    If you hit "Reply with Quotes" you should see the above (quoted quotes in the previous posts are stripped out).

    The problem is that as batteries discharge, they start sulfating faster and faster--the longer time they spend at the lower levels of charge. So if you have one or two cells that are already somewhat discharged, they will hit that area first and sulfate faster--and eventually die.

    Also, if you ever pull the battery bank down to ~20% State of Charge (average)--Any cells that started below 100% SOC will hit 0% SOC first, and can actually get reverse polarity during discharging and those individual cells will begin to "reverse charge". For almost any rechargeable battery chemistry, "reverse charging" is the kiss of death for a cell.

    At least, that is my reasoning.

    -Bill

    PS: Note that the above quote has tags:
    [HTML]
    ...
    ...
    [/HTML]

    You edit anything in the tags, and use the short [QUOTE tag if you don't have the "=mkldmkakd;124434" thread information.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    It sounds like from what you are both saying that I should just get my new PV online and keep these batteries more fully charged. It is true that I pretty much never clock more than 25% of their capacity before recharging (I keep pretty good tabs on how much energy we use and confirm them with an amp-hour meter that shows net state of charge), and it does seem like between sun and generator, I've been giving them enough juice frequently enough that they should be staying at a high state of charge. But, maybe holding them at 58.8V for 2 hours regularly as the generator does this time of year is not enough. I'll get them equalized fully again and get the new PV on the roof and see if they do any better in the coming year.

    On another note, it was mentioned that this battery bank is far to large for our useage. My thinking on this was that sun can be infrequent in this part of the world for a big part of the year and having 4 days of capacity would be a good thing if we could charge the bank up in one day of sun. This battery bank is a total capacity of between 7-8 X our daily use, though there are times of the year when we use more like 1/5 of the capacity. Problem is, I've never seen even the fully charged and equalized battery bank maintain voltages for more than 2 days before dropping low enough to kick on the autostart on the generator. Is this an unrealistic expectation?

    Thanks again.

    -Eric
    BB. wrote: »
    Eric,

    If you hit "Reply with Quotes" you should see the above (quoted quotes in the previous posts are stripped out).

    The problem is that as batteries discharge, they start sulfating faster and faster--the longer time they spend at the lower levels of charge. So if you have one or two cells that are already somewhat discharged, they will hit that area first and sulfate faster--and eventually die.

    Also, if you ever pull the battery bank down to ~20% State of Charge (average)--Any cells that started below 100% SOC will hit 0% SOC first, and can actually get reverse polarity during discharging and those individual cells will begin to "reverse charge". For almost any rechargeable battery chemistry, "reverse charging" is the kiss of death for a cell.

    At least, that is my reasoning.

    -Bill

    PS: Note that the above quote has tags:
    [HTML]

    [/HTML]

    You edit anything in the tags, and use the short [QUOTE tag if you don't have the "=mkldmkakd;124434" thread information.
  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    ...
    ...Problem is, I've never seen even the fully charged and equalized battery bank maintain voltages for more than 2 days before dropping low enough to kick on the autostart on the generator. Is this an unrealistic expectation?

    -Eric

    There are other threads going over the same problem, here is one recent one almost identical starting point..
    http://forum.solar-electric.com/showthread.php?14333-Specific-Gravity&highlight=equalize

    The sudden drop in SoC is indicative of a problem, you just have to sleuth it out...


    hth
     
    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
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Eric;

    The larger the battery bank the longer the Absorb time needed. You may well not be getting full charge in a mere two hours of Absorb. Further, the Absorb Voltage of 58.8 may be a tad low; try upping it to 59.2 In addition to that, the longer the batteries can be held at Float Voltage the better.

    Because batteries aren't linear, trying to store up four days worth in one set can be difficult; once the Voltage starts to slide it goes down rapidly. Since you have a generator, following the pattern of keeping two days worth of power in the batteries and recharging on the third day will probably work better. Especially if it means you can get a complete charge in one good day of sun. Running a generator for full recharging is very expensive in fuel, but once you get past the Bulk stage and use some heavy loads so you're not wasting fuel it can be okay on an occasional basis.
  • Ralph Day
    Ralph Day Solar Expert Posts: 1,019 ✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Definitely not enough absorb time. I have Surrette CS type batteries and they require a lot of charge. Put your charge controller to max time and check the SG after it's through. If not full SOC trip it to run again.

    I found that distilled water for calibration of a specrometer would result in low sg readings. Can you go to a battery supplier (fork lift truck service place, NAPA store) and purchase a small amount of known sg electrolyte? Have them check it with their hydrometer as 1.265 and use that to calibrate your spec. That's what I do. Keep the sample stuff in a glass jar sitting on the batteries so it's at the same temperature, or very close at least. One drop to calibrate, the sample should last years. When you know your batteries are at 1.265 you could replenish your stock from your own batteries. That what I do too. Needless to say, I usually check sg's with the hydrometer now, too much futzing around with the optical one.

    Remember to charge robustly, they are very robust batteries you have.

    Ralph
  • Rybren
    Rybren Solar Expert Posts: 351 ✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    Ralph Day wrote: »
    . Keep the sample stuff in a glass jar sitting on the batteries so it's at the same temperature, or very close at least. One drop to calibrate, the sample should last years. Ralph


    It is very important to have the lid off of the calibration solution for as short a time as possible AND to keep it tightly sealed when not in use. Evaporation will cause the SG to drift over time.


    As a side note, the refrac doesn't actually measure Specific Gravity, it measures the Refractive Index of the solution. The scale on the refrac is based on a Refractive Index to SG conversion table that is based on standard conditions which will likely differ from your testing conditions. (I know that this is semantics, but I do believe that it is important to understand what we are measuring and potential areas where we may unwittingly introduce errors)
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Hi again all. I have found out more about my battery bank and Surrette is sending me some replacement batteries. As it looks now, I have some bad cells - I was looking at total bank voltage and finally looked at voltages across the bank. What I found were several cells which show reasonable voltage soon after charging, but drop off to basically 0 after some hours under load. Specific gravity in those cells remains high while the rest of the bank looks to be tracking voltage and s.g. in a reasonable way. There seem to be 4 bad batteries and a different one that has a piece of metal turned sideways visible under the cap - the batteries on either side of this one have shown problems (2 dead ones on one side, low S.G. readings on the other).

    I've attached a document showing data with readings taken 12 hours apart.

    A few things:
    Surrette is sending me three replacements (we discovered 2 bad cells plus the one with the dislodged metal at first - don't know if they are going to want to send me more...). Seems like I need to get the bad ones out of there quickly...however, I have one string of batteries in series. I can lower the voltages on my equipment accordingly, except for my SolarBoost charge controllers. I have a new bank of PV with a more adjustable Outback controller. Do I take my old PV off-line, lower the voltage and remove old batteries until I can get new ones?
    Are the bad batteries dead for sure? Should I try to resurrect them?
    What could I be doing to these batteries? The bank is one year old. One battery went bad immediately after going in and they replaced that. Now there are 5 more! I have been watching temperature and have had temperature correction on charge controllers(pair of SolarBoost 3048) and inverter (-2- Trace SW5548). I replaced the previous bank of Surrette 12V batteries that lasted me about 9 months after they did a similar thing. Surrette claimed that plates on the test cell I sent back to them from that bank was disintegrated. All they could come up with was that they were "overcharged". Could I be cooking these things somehow? I am staying well within parameters I am supposed to and they have looked like they are under-charged. There is a desulfator hooked up to the bank - could that be doing something?

    I can't be replacing $8000 battery banks regularly... Nor do I want to replace my inverters and charge controllers without knowing that they are a problem. Could these just have been some defective batteries? Seems unlikely given the history of two bad banks in the last couple of years.

    Thanks for your help.

    -Eric Warren
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    Hi again all. I have found out more about my battery bank and Surrette is sending me some replacement batteries. As it looks now, I have some bad cells

    I wish I could say that's unusual. Unfortunately with Surrette it doesn't seem to be.
    A few things:
    Surrette is sending me three replacements (we discovered 2 bad cells plus the one with the dislodged metal at first - don't know if they are going to want to send me more...). Seems like I need to get the bad ones out of there quickly...however, I have one string of batteries in series. I can lower the voltages on my equipment accordingly, except for my SolarBoost charge controllers. I have a new bank of PV with a more adjustable Outback controller. Do I take my old PV off-line, lower the voltage and remove old batteries until I can get new ones?

    I take it you mean you won't have enough batteries to make system Voltage, but want to keep the ones you've got charged until the new ones arrive? By all means do it if you can.

    Are the bad batteries dead for sure? Should I try to resurrect them?

    There's no practical way of resurrecting dead batteries. These could have shorted or broken plates inside; merely applying Voltage to the terminals hasn't done anything for them and there's nothing else you can do from outside. Don't waste your time.
    What could I be doing to these batteries? The bank is one year old. One battery went bad immediately after going in and they replaced that. Now there are 5 more! I have been watching temperature and have had temperature correction on charge controllers(pair of SolarBoost 3048) and inverter (-2- Trace SW5548). I replaced the previous bank of Surrette 12V batteries that lasted me about 9 months after they did a similar thing. Surrette claimed that plates on the test cell I sent back to them from that bank was disintegrated. All they could come up with was that they were "overcharged". Could I be cooking these things somehow? I am staying well within parameters I am supposed to and they have looked like they are under-charged. There is a desulfator hooked up to the bank - could that be doing something?

    They're scrap metal. Scrap them. Learn from this event. That's all you can do. A desulphator, if it works at all (consensus around here is they don't) would only help if the problem was sulphation. Even then that can be "hard" which does not reverse no matter what. Don't waste your time, and money.
    I can't be replacing $8000 battery banks regularly... Nor do I want to replace my inverters and charge controllers without knowing that they are a problem. Could these just have been some defective batteries? Seems unlikely given the history of two bad banks in the last couple of years.

    At the risk of starting a major flame war ... don't buy Surrette batteries. A lot of money for seems to be increasingly poor quality in my opinion.

    Once you get the system balanced you should not have to replace the batteries for years. This is one reason why we recommended getting the least expensive bank of power for the first set, as it is easy to wreck batteries. After that, if you keep the DOD to 25% or so and the peak charge current around 10% and get the batteries recharged fully and quickly and don't let them sit about discharged for days they should last for years.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,474 admin
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    Eric,

    My two cents... I have not seen the batteries and I am not a battery engineer--So take my comments with a grain of salt.
    Hi again all. I have found out more about my battery bank and Surrette is sending me some replacement batteries. As it looks now, I have some bad cells - I was looking at total bank voltage and finally looked at voltages across the bank. What I found were several cells which show reasonable voltage soon after charging, but drop off to basically 0 after some hours under load. Specific gravity in those cells remains high while the rest of the bank looks to be tracking voltage and s.g. in a reasonable way. There seem to be 4 bad batteries and a different one that has a piece of metal turned sideways visible under the cap - the batteries on either side of this one have shown problems (2 dead ones on one side, low S.G. readings on the other).

    Cells may fail with high resistance (open) or shorted... Open cells would prevent charging current from fully charging the rest of the cells and could interfere with equalization too (they take all of the voltage drop and limit current flow).

    Shorted cell(s) would cause excessive current to flow during charging even with normal charge voltage set points and could "over charge/over equalize" the other cells. The Battery Bank would appear to discharge very quickly (voltage drop by "x" number of cells discharging, then hold that lower voltage while the rest of the cells do their job.

    Over charging/lots of equalization is bad too.... Can cause plates to shed material, use excessive water, and cause battery positive grid corrosion (O2 forms at the positive plate during charging).

    Did you notice how much water you need to add to each cell? Where all cells using water equally? Very roughly, if you add water every two months (always prevent plates from being exposed to air)--you probably are charging about right... If you never add water or need to add water every month or less--You probably are under/over charging the cells (although--I think--Surrette 4000 series batteries have calcium added to reduce water usage and need higher charging voltages for proper recharging).
    I've attached a document showing data with readings taken 12 hours apart.

    Note it probably takes more than 3 hours for batteries to diffuse the electrolyte concentrations through the cell to measure proper resting voltage. The partially discharged readings, however, tell the story of the dead cells.

    Battery #10 probably needs replacement too... The rest of the cells discharged by an average of ~0.020-0.025 SG units. That cell discharged by ~0.038 -- Not good.

    When looking at your cells, most vendors recommend equalization if the cells are around 0.015 to 0.030 or more SG units apart. At this point, spending money on generator fuel to equalize before you get the new cells is probably not the best use of fuel. When you get the new cells equalize the battery bank (high voltage charging, a maximum of ~5% rate of charge, measure measure/log the temperature compensated SG for each cell--When all of the cells stop SG rising between readings--Log the final values and stop equalization).

    Have a container of distilled water you can rinse the hydrometer out after each set of readings--If the hydrometer dries out with electrolyte/gray dust from the battery inside--the float can become sticky and give inaccurate readings.
    A few things:
    Surrette is sending me three replacements (we discovered 2 bad cells plus the one with the dislodged metal at first - don't know if they are going to want to send me more...).

    I am not the guy filling out the RMA's... You have 5 bad batteries--Not three (in my humble opinion). The four that show very low/zero voltage after nominal discharging, and the one with the debries inside (the 0.038 specific gravity drop and the very high starting SG does not look right (unless there was an inaccurate reading on your part).
    Seems like I need to get the bad ones out of there quickly...however, I have one string of batteries in series. I can lower the voltages on my equipment accordingly, except for my SolarBoost charge controllers. I have a new bank of PV with a more adjustable Outback controller. Do I take my old PV off-line, lower the voltage and remove old batteries until I can get new ones?

    Check the batteries during charging. Very likely, the "bad cells" will limit charging current. You want to measure the voltage on the "good cells" and make sure you are not damaging them further. As long as they are charging--I guess I would just continue to run the bank "as normal".

    And on the failed batteries, as long as they are "normal" to high voltage, you probably are not over charging/damaging the rest of the bank.
    Are the bad batteries dead for sure? Should I try to resurrect them?

    Many of the "Surrette" problems we have read about here seem to have very low SG in failing/poorly operating cells (1.100 or even lower SG).

    Your "bad cells" all have normal full charge to higher SG readings when reading zero/low voltage. There is probably nothing you can do for them to even temporarily get them functional again. Although--If they are highly sulfated--Charging/Equalizing the cells A BUNCH may cause the insulating sulfates to shed and expose fresh material--However, because your SG is already high--It would not seem you have a lot of sulfation (which locks up the sulfur in the electrolyte--leading to low SG readings over time, not normal to high readings).

    If you try anything--talk with Surrette first--You don't want your extraordinary efforts to cause the Mfg. to deny warranty claims by your attempt to get the cells functional again.
    What could I be doing to these batteries? The bank is one year old. One battery went bad immediately after going in and they replaced that. Now there are 5 more! I have been watching temperature and have had temperature correction on charge controllers(pair of SolarBoost 3048) and inverter (-2- Trace SW5548). I replaced the previous bank of Surrette 12V batteries that lasted me about 9 months after they did a similar thing.

    It is all too easy to "murder batteries" with under charging / over discharging (guest, kids, etc. using power like there is no tomorrow during bad weather, something left on during a trip, just not quite fully recharging every time--but slowly declining state of charge over time--"deficiet charging", etc...

    But we have a few reports here of Surrettes going bad soon after installation, or even in once case, appeared to have a cell/battery filled with distilled water instead of electrolyte.

    Normally, when installing a new bank, a mfg. will have a very specific set of instructions on charging to put the bank in service--obviously, follow those recommendations.

    We also suggest that you inspect/number the batteries on the pallet and log any lot numbers/etc., measure the resting voltage of each battery/cell, and log the temperature corrected specific gravity of each cell. And reject anything that does not look "right" (you have at least 24 cells--If you find any high/low voltage or SG readings, cells with low electrolyte/wet batteries or pallet, mixed date codes, "old date code batteries", batteries delivered with less than 75% SOC, cracked/split/leaking cases, etc. are all indications of "issues" that need to be addressed.
    Surrette claimed that plates on the test cell I sent back to them from that bank was disintegrated. All they could come up with was that they were "overcharged". Could I be cooking these things somehow? I am staying well within parameters I am supposed to and they have looked like they are under-charged. There is a desulfator hooked up to the bank - could that be doing something?

    Batteries are very sensitive to charging voltage... Nominally ~59 volts is a "vigorous" charging voltage and 62-64 volts volts is "equalization" (and equalization is very hard on a battery bank--so you want to do the minimum required for good battery life). The Surrette 4000 series require something like 63 volts for Absorb charging--Which is already a very high voltage--but within their specifications.

    With potentially shorted cells in a battery bank, they will "subtract" voltage and raise the charging voltage on the rest of the "good cells" which can cause them to be over charged/damaged. If you did not notice the "bad cells" in time--it is possible that the other cells were over charged. Checking/logging each cell voltage during charging (once a week?) will give you an early warning of issues.
    I can't be replacing $8000 battery banks regularly... Nor do I want to replace my inverters and charge controllers without knowing that they are a problem. Could these just have been some defective batteries? Seems unlikely given the history of two bad banks in the last couple of years.

    Failing/unreliable batteries has been a complaint we have seen in the last few years about Surrette batteries... Of course, none of us here on the forum work for Surrette or sell batteries--So we have no idea if this is a wide spread issue or just bad news traveling quickly while huge numbers of installations are working well with nothing to post here.

    Best of luck.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    At the risk of starting a major flame war ... don't buy Surrette batteries. A lot of money for seems to be increasingly poor quality in my opinion.

    Once you get the system balanced you should not have to replace the batteries for years. This is one reason why we recommended getting the least expensive bank of power for the first set, as it is easy to wreck batteries. After that, if you keep the DOD to 25% or so and the peak charge current around 10% and get the batteries recharged fully and quickly and don't let them sit about discharged for days they should last for years.

    Well, I guess at this point I'm hoping that I got a bad batch of batteries from Surrette, but given that I succeeded in killing two sets in the last couple of years, I am skeptical of that - even given that other people are experiencing problems with these batteries. So far, Surrette has been accommodating and making good on their warranty. We'll see when they get sick of sending me batteries...

    I do keep DOD to 25% and peak charge current briefly goes just above 10% when generator starts up, but quickly is below that. The first set of batteries on this system (12 - Surrette T-12-250s) lasted about 7 years but limped along for a lot of it. I thought that I would get a lot more years out of that set...I certainly expect to get quite a few years out of these more robust ones.

    Thanks for the reply - I won't mess around trying to do anything with the bad cells. Looks like Surrette is going to take some back and autopsy them.

    -Eric
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    I do keep DOD to 25% and peak charge current briefly goes just above 10% when generator starts up, but quickly is below that.
    -Eric

    I am assuming that your 10% charge current refers to an amp rate of C/10?

    Since the general rule for FLA batteries is that the charging system should be capable of delivering C/12 to C/8 during bulk and EQ (if needed), either you have a low capacity charging system or the batteries SOC is so high that the CC rapidly goes into Absorb and the current goes down as a result.
    Could be responsible for undercharging or electrolyte stratification.
    SMA SB 3000, old BP panels.
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    inetdog wrote: »
    I am assuming that your 10% charge current refers to an amp rate of C/10?

    Since the general rule for FLA batteries is that the charging system should be capable of delivering C/12 to C/8 during bulk and EQ (if needed), either you have a low capacity charging system or the batteries SOC is so high that the CC rapidly goes into Absorb and the current goes down as a result.
    Could be responsible for undercharging or electrolyte stratification.

    Seems like the charging system has plenty of power - C/8 is often reached initially, but tapers off to closer to C/10 after 20 min or so. I meant that it is just below that rate in my last post.
  • benthicbread
    benthicbread Registered Users Posts: 10
    Options
    Re: Battery state of charge: specific gravity vs. voltage
    BB. wrote: »
    Eric,

    My two cents... I have not seen the batteries and I am not a battery engineer--So take my comments with a grain of salt.



    Cells may fail with high resistance (open) or shorted... Open cells would prevent charging current from fully charging the rest of the cells and could interfere with equalization too (they take all of the voltage drop and limit current flow).

    Shorted cell(s) would cause excessive current to flow during charging even with normal charge voltage set points and could "over charge/over equalize" the other cells. The Battery Bank would appear to discharge very quickly (voltage drop by "x" number of cells discharging, then hold that lower voltage while the rest of the cells do their job.

    Over charging/lots of equalization is bad too.... Can cause plates to shed material, use excessive water, and cause battery positive grid corrosion (O2 forms at the positive plate during charging).

    Did you notice how much water you need to add to each cell? Where all cells using water equally? Very roughly, if you add water every two months (always prevent plates from being exposed to air)--you probably are charging about right... If you never add water or need to add water every month or less--You probably are under/over charging the cells (although--I think--Surrette 4000 series batteries have calcium added to reduce water usage and need higher charging voltages for proper recharging).



    Note it probably takes more than 3 hours for batteries to diffuse the electrolyte concentrations through the cell to measure proper resting voltage. The partially discharged readings, however, tell the story of the dead cells.

    Battery #10 probably needs replacement too... The rest of the cells discharged by an average of ~0.020-0.025 SG units. That cell discharged by ~0.038 -- Not good.

    When looking at your cells, most vendors recommend equalization if the cells are around 0.015 to 0.030 or more SG units apart. At this point, spending money on generator fuel to equalize before you get the new cells is probably not the best use of fuel. When you get the new cells equalize the battery bank (high voltage charging, a maximum of ~5% rate of charge, measure measure/log the temperature compensated SG for each cell--When all of the cells stop SG rising between readings--Log the final values and stop equalization).

    Have a container of distilled water you can rinse the hydrometer out after each set of readings--If the hydrometer dries out with electrolyte/gray dust from the battery inside--the float can become sticky and give inaccurate readings.



    I am not the guy filling out the RMA's... You have 5 bad batteries--Not three (in my humble opinion). The four that show very low/zero voltage after nominal discharging, and the one with the debries inside (the 0.038 specific gravity drop and the very high starting SG does not look right (unless there was an inaccurate reading on your part).



    Check the batteries during charging. Very likely, the "bad cells" will limit charging current. You want to measure the voltage on the "good cells" and make sure you are not damaging them further. As long as they are charging--I guess I would just continue to run the bank "as normal".

    And on the failed batteries, as long as they are "normal" to high voltage, you probably are not over charging/damaging the rest of the bank.



    Many of the "Surrette" problems we have read about here seem to have very low SG in failing/poorly operating cells (1.100 or even lower SG).

    Your "bad cells" all have normal full charge to higher SG readings when reading zero/low voltage. There is probably nothing you can do for them to even temporarily get them functional again. Although--If they are highly sulfated--Charging/Equalizing the cells A BUNCH may cause the insulating sulfates to shed and expose fresh material--However, because your SG is already high--It would not seem you have a lot of sulfation (which locks up the sulfur in the electrolyte--leading to low SG readings over time, not normal to high readings).

    If you try anything--talk with Surrette first--You don't want your extraordinary efforts to cause the Mfg. to deny warranty claims by your attempt to get the cells functional again.



    It is all too easy to "murder batteries" with under charging / over discharging (guest, kids, etc. using power like there is no tomorrow during bad weather, something left on during a trip, just not quite fully recharging every time--but slowly declining state of charge over time--"deficiet charging", etc...

    But we have a few reports here of Surrettes going bad soon after installation, or even in once case, appeared to have a cell/battery filled with distilled water instead of electrolyte.

    Normally, when installing a new bank, a mfg. will have a very specific set of instructions on charging to put the bank in service--obviously, follow those recommendations.

    We also suggest that you inspect/number the batteries on the pallet and log any lot numbers/etc., measure the resting voltage of each battery/cell, and log the temperature corrected specific gravity of each cell. And reject anything that does not look "right" (you have at least 24 cells--If you find any high/low voltage or SG readings, cells with low electrolyte/wet batteries or pallet, mixed date codes, "old date code batteries", batteries delivered with less than 75% SOC, cracked/split/leaking cases, etc. are all indications of "issues" that need to be addressed.



    Batteries are very sensitive to charging voltage... Nominally ~59 volts is a "vigorous" charging voltage and 62-64 volts volts is "equalization" (and equalization is very hard on a battery bank--so you want to do the minimum required for good battery life). The Surrette 4000 series require something like 63 volts for Absorb charging--Which is already a very high voltage--but within their specifications.

    With potentially shorted cells in a battery bank, they will "subtract" voltage and raise the charging voltage on the rest of the "good cells" which can cause them to be over charged/damaged. If you did not notice the "bad cells" in time--it is possible that the other cells were over charged. Checking/logging each cell voltage during charging (once a week?) will give you an early warning of issues.



    Failing/unreliable batteries has been a complaint we have seen in the last few years about Surrette batteries... Of course, none of us here on the forum work for Surrette or sell batteries--So we have no idea if this is a wide spread issue or just bad news traveling quickly while huge numbers of installations are working well with nothing to post here.

    Best of luck.

    -Bill

    Thanks Bill. These batteries seemed fine when installed a year ago - all S.G.s were close and the bank held voltage through its cycle the way it should. I am a bit concerned about the "good" cells having been over-charged as I have not been monitoring individual cell voltage until just recently. But, since the bad cells show a voltage across each of them that is normal when charging, it seems like they are doing their job on that end to keep other cells from seeing too high of a voltage...?

    I haven't been having to add water too often - probably every other month or a little less.

    Clearly I am having a different issue than low S.G. In fact, it seems like I've got the opposite problem. Could the high S.G. readings be indicative of something other than sulfuric acid concentration? Other stuff being suspended in the electrolyte? Water has been black after bubbling (which Surrette says is normal).

    Guess I'll wait for new batteries, send the bad ones in and see what they come up with.

    -Eric
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    I haven't been having to add water too often - probably every other month or a little less.

    -Eric

    Important to know, but hard to estimate unless you deliberately check it (which can be real pain) is the relative amount of water used by different cells.
    A cell in a string which uses either much more or much less water to get back to the same level it had before can be an early indication of problems.
    Specifically, it indicates that that cell was gassing either a lot more (overcharged) or a lot less (not getting completely charged) compared to the others.

    Since all of the cells in a string are handling the same charge and discharge current, their water needs should be very close. If they are not, there may be something about the odd cell which is responsible for the difference.
    If you have strings or batteries in parallel, a general trend of difference between one string and another is an indication of current matching problems.

    If the batteries get really warm/hot during charging, the position of a battery or cell within the bank may affect its temperature and therefore is relative charging efficiency and gassing compared to the other cells.
    SMA SB 3000, old BP panels.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,474 admin
    Options
    Re: Battery state of charge: specific gravity vs. voltage

    The first failure report from Surrette sure do not seem to explain what you are seeing so far... I agree that the failed cell, when charging maintaining "reasonable" voltages should keep the rest of the string healthy with functional relevant charging.

    The "higher than normal" SG does not make a lot of sense if your initial numbers were OK (you always add distilled water--not electrolyte or acid--right?)--Otherwise, you have bubbles on the float giving false high readings (or cell was very cold compared to other cell/reference temperature).

    I certainly am not a battery expert--But I cannot think of any failure mode in the battery that would have this problem (stratfication would have the high SG electrolyte at the bottom of the cell--and adding water would temporarily give you lower SG levels).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset