Why should not batteries of different age be connected in parallel?

Roil
Roil Solar Expert Posts: 25
Can anyone guide me to the final documented truth about this topic.

It may be only me being a bad researcher, but I fail to find the tread disclosing this answer

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Why should not batteries of different age be connected in parallel?

    Batteries change their parameters with age/cycling/temperature/brand/model (talking here about lead acid batteries).

    More or less, you get optimum performance with matched cells. This is done quite a bit with RC Models and such.

    Say you have two batteries in series. If their capacity is different (unmatched due to age/cycling/condition/or even different brand/models) and you want to avoid discharging below 50% (for long life) and below 20% State of Charge (to prevent a cell from becoming reversed biased and instant cell/battery death)--You are limited to the capacity of the weakest member of the series string. So placing a "new" and "old" battery in series--It will (usually) be the old battery that limits capacity (and will probably eventually fail first).

    Another issue is that old batteries have higher "self discharge" than new. When you have two batteries in series, again old vs new, the old battery will self discharge faster and become "out of balance" with the new battery. The usual method to equalize a battery bank is to "over charge" the good battery (force excess current through the good cells to get charging current into the low cells). Equalization is hard (in general) on lead acid batteries. It erodes the plates, uses water (or catalyst in AGM/Sealed batteries) and causes positive grid corrosion (oxygen forms on the positive plates during equalization). And an out of balance battery bank--The low state of charge cells will again limit the series string capacity. And equalization will cause the newer battery to age faster (equalization also increases battery temperature. 10C increase in battery temperature, 2x faster aging).

    There are also problems with cells/batteries in parallel. The idea is to design your wiring such that you have equal current flow during charging/discharging. Older batteries have higher internal resistance (less charging/discharging current due to plate sulfation) and as they sulfate, lowering operating voltage (accept more charging current, supply less discharging current as battery voltage is proportional to specific gravity... Lower SG, lower battery voltage).

    And since during charging, low battery may gas more (less capacity, charges "quicker", lower SG and more gassing as more sulfur is locked into lead sulfate), old batteries can heat faster than the parallel new batteries. Higher temperature, lower charging voltage, accepts more current, lower charging current--repeat)--And you can have thermal runaway issues with your parallel bank. While the new batteries get cycled deeper and get less charging.

    Another issue is that you need to monitor cell/batteries in a battery bank for consistent operational voltages on each cell/battery (charging, discharging, resting), specific gravity, and use a DC Current Clamp meter to ensure that parallel strings are properly sharing current between strings. An open cell will kill the current in a string, and a shorted cell will "over charge" the other cells in the string--And discharge the other parallel strings.

    So--if you have a "mix" of unmatched cells/batteries in your bank--You will have to keep a very close eye on them as you will be running "more often" with some sort of aging/failing cells (i.e., 2/4/6 year old mix of batteries--You will have a battery/cell failure every ~2 years if you assume a 6-8 year battery life). So, a mixed bank of batteries with a good chance of cell failure every 2 years to catch/diagnose/replace vs a, hopefully, uneventful 6-8 year battery bank life--And replace all batteries at the same time (when one or two start to go, there will be others not too far behind).

    Not to say that people do not mix old and new batteries. There are a few people here that have large Forklift batteries. Say a 36 volt battery that they only need 24 volts from... They get a used battery, test each cell, and find 12 good cells out of the 18 in the battery bank--Rewire for 12 good cells (matching) and then get many years of extra life out of the "junk" battery. May even have a few "spare" cells that can be wired in when one cell does die.

    So--Mixing old and new cells/batteries together is not (usually) going to give you a pile of melted plastic and lead, with battery acid flowing down your floors in two weeks... It is more of an issue of overall battery bank reliability and how much is your time worth to keep a mixed bank operating (and close monitoring to avoid catastrophic failures--Which can happen at anytime with old or new batteries).

    Do people mix old/new and flooded cell with AGM--Yep, sometimes they do. It is the best they can with the limited funds that they have available.

    Make any sense?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • ToManyTrees
    ToManyTrees Registered Users Posts: 24
    Re: Why should not batteries of different age be connected in parallel?

    Thanks Bill

    TMT / Paul
  • TucsonAZ
    TucsonAZ Solar Expert Posts: 139 ✭✭
    Re: Why should not batteries of different age be connected in parallel?

    Or think about it metaphorically like a shot glass and a beer mug constantly trying to equalize their volume, I tend to like thinking of extremes to make points a little clearer in my head.
  • Roil
    Roil Solar Expert Posts: 25
    Re: Why should not batteries of different age be connected in parallel?

    Just small message to my friend Marwin in Norway
  • Roil
    Roil Solar Expert Posts: 25
    Re: Why should not batteries of different age be connected in parallel?

    Hi Bill

    Tried to send you a DM, but not sure if you got it, I just wanted to thank you for you answer.

    You sound very convincing with in depth knowledge.

    I used your answer in a debate regarding batteries on a Norwegian Boat forum, and get the "why should we belive something written on another forum", any help, your background?

    Thanks in advance
    Anders
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Why should not batteries of different age be connected in parallel?

    Those guys that don't believe, can just prove it to themselves easily enough. The basics of why doing this with batteries is easily demonstrated by simple AA batteries. Although obviously a different chemistry, the basic rules are easily demonstrated as to why this is a bad idea.

    Find a AA flashlight. Put fresh high-quality batteries in it, and note the runtime until the light from it is more or less unusable. Then, with another set of fresh batteries, replace just ONE with a discharged battery from the first test, simulating a large age difference. It doesn't even have to be dead flat, just half-way charged.

    Runtime isn't everything, as those three fresh cells will push that previously depleted one well into reverse-polarity and damage it.

    Then they can come back here and save a lot of time, energy, money, and frustration. :)
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Re: Why should not batteries of different age be connected in parallel?

    Such a test is "in series", not parallel. Two quite different things.

    I am available for custom hardware/firmware development

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Why should not batteries of different age be connected in parallel?
    jonr wrote: »
    Such a test is "in series", not parallel. Two quite different things.

    In this case it demonstrates the principal of uneven battery capacities causing drain from high to low.

    Batteries age and lose capacity as hey do. When the difference is slight, there is no issue. If the difference is great problems arise. It is not a case of "this will not work at all" or "extremely bad things will happen" just a case of not being the best idea in the world. In theory it could reach a point where the battery with low capacity (and thus resistance) gets too much current and overheats; thermal runaway. That would be worst-case.

    Doubters are free to do as they please; it's their money. But if you're after top performance and value over time it's best to listen to what others say and understand the reasoning behind it.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Why should not batteries of different age be connected in parallel?

    Hi Anders,

    I don't like to argue from authority (I designed small and large computer systems for many years as a systems/hardware engineer)--It does not do very much in terms of furthering education/exchange of ideas... I prefer to discuss the actual questions/issues and typically use the math+physics to show the problems/advantages of various design options.

    Is there some specific problem/issue/disbelief from your forum friends? Rather than shot gunning factoids, what is one issue with my overall discussion/observation that they disagree with? That would give us a more focused discussion were we can go into details.

    There is the "ideal battery" and how it works in parallel connected wiring/power systems (N+1 Redundancy, etc.).

    And there is the "real battery"--How it is temperature/state of charge/operational dependent in such a system. There are various negative and positive feedback mechanisms that will govern the behaviour of the overall system. Which effect governs the behaviour in the battery bank in any particular example is not always obvious). For example:
    • Negative Feedback: A lower state of charge battery will have a "lower voltage" (lower SG gives lower charging/resting/operating voltage) and accept more charging current (batteries will "equalized" unbalanced state of charge between parallel strings).
    • Positive Feedback: A lower state of charge battery will accept more charging current. Waste Heat = I2R heating. Note that 2x current = 4x amount of waste heat--Lower state of charge battery (or battery+wiring with lower resistance) will accept more current and could run hotter. Hotter battery has a -5mV per cell per oC charging voltage--Hotter battery, lower charging voltage, higher charging current--Possible thermal runaway.
    Which effect will "win" in the above example. Don't know. It all "depends" on system configuration, age of battery, miss-match of battery parameters (if any), air circulation/location in battery string, cabling issues, rate of charge, etc.

    One thing that I learned over the years with paralleled cabling (distributing 10's - 100's of amperes of +5 volt power, etc.) was that you could have a wonderful design that worked well for months or even years, but then a failure mechanism would present itself and cause a failure.

    For example.... In computers (especially the older computers/test equipment) that used lots of low voltage current)--It is very common to run a bunch of parallel small gauge wire in parallel to carry the current from the main power supply to various cards/modules in the equipment.

    Many of the failures of my test gear (logic analyzer for example) would turn out to be one or more charged contacts in the DC power cabling.

    I would pop open a system and find a connector with (as an example) 5 browned to blackened connectors. At first, it did not make a lot of sense--Inevitably, it was one or two connectors that were completely burned/melted. And the rest went from lightly browned to perhaps no browning at all. Why would one or two cable connections burn and the rest not?

    After measuring voltages/resistances of various good and bad connectors/wiring--It became obvious... Gets back to the basic formula of Power = I2R.

    What was happening was not the "bad/high resistance" contact that was burning... It was the "best" contact/cable with the lowest resistance that was causing "the failure". The contacts could easily have a 2:1 or even 5:1 variation in resistance between "matched" contacts. If there was not much else in the way of resistance (i.e., short/heavy cables with very low resistance relative to contact resistance)--The "good contact" would carry the most current. For example, if the contact had 1/2 the resistance and carried 2x the current, because of the I2R term, the self heating would be 2x worse than the next connector. So, very quickly, the "good contact" would overheat and eventually fail, then the next lowest resistance contact would start carrying excessive current, fail, etc... And the power connections would "unzip" until the system failed due to low voltage or somebody smelled smoke.

    Similar issues with parallel battery banks. The "ideal" battery has very low internal resistance. If the cabling is not properly designed, or is made with "too heavy" of copper cable (i.e., cabling has lower series resistance than batteries themselves)--There is no "ballast" in the battery bank to "steer" current other than the battery resistance itself (which changes over time/temperature/state of charge/etc.). This can create an "unstable" system that, over time, may have an early life failure of some sort. What that failure will be (burned cable end, melted cable, failed battery, etc.)--Is hard to predict. But the eventual failure can, many times, be traced back to not following one design guideline or another.

    But it is pretty easy to see that "the string that carries the most current" (heavy charging/discharging) will carry much of the overall cycling current--And will age faster--Until the second best string takes over, etc...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Re: Why should not batteries of different age be connected in parallel?

    Due to internal resistance (or connector) variations, a battery bank with batteries of the same age and equal resistance cables is still NOT balanced. It could even have worse balance than one with unequal length cables.

    I am available for custom hardware/firmware development

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Why should not batteries of different age be connected in parallel?
    jonr wrote: »
    Due to internal resistance (or connector) variations, a battery bank with batteries of the same age and equal resistance cables is still NOT balanced. It could even have worse balance than one with unequal length cables.

    No two batteries or even battery cells will ever be in perfect balance. That is not possible.

    However two identical size/type age and use batteries with equal length cabling to the common points will come as close to this as is practically possible. Your reiterate position that this is not so has yet to be proven anywhere, whereas the opposite has repeatedly.

    If you want any better balancing you will need individual cell management such as used on lithium batteries.

    BTW when you connect two batteries in parallel you do in fact create a series circuit between them as they try to reach equilibrium.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Why should not batteries of different age be connected in parallel?

    That is true--And why "unmatched" batteries in parallel strings becomes such an issue.

    And one reason that a person may want to use smaller diameter/longer copper wire runs per string--For example, if the battery has 5 mOhms of resistance, and the wiring has 10-50 mOhms of resistance, then the wiring will "steer" the current more evenly and the battery variations in resistance will be "less important".

    On the other hand, high(er) resistance in wiring, higher voltage drop, more losses... 100 amps * 10 mOhms = 1 volt drop... Not good on a 12 volt battery bank. There is no "good" solution at times.

    In general, if you fuse each string to the rated capacity of the wiring+connectors--Then there should not be any wiring hot spot failures.

    You could still, for example, boil a cell/string dry if there is a shorted cell in the string (not enough current to pop a breaker, but more than enough current to "over charge" the rest of the cells in that series string--And discharge the rest of the bank over time).

    And why I, for sample, have my rules of thumb include:
    • Do not Parallel "12 volt" batteries in parallel for a 12 volt bus. You cannot easily measure battery/cell voltages of each battery/cell (most modern batteries, you cannot measure per cell voltages because the bus bars are buried under the battery housings).
    • Do use 2x 6 volt batteries (or single 2 volt cells) to make up a 12 volt string. You can quickly check battery/cell voltages for shorted/open cells.
    • Flooded cell batteries are easier to monitor because you can measure SG of each cell.
    • AGM/Sealed batteries are more difficult to monitor because you cannot measure SG at all.
    • When wiring up a battery bank, make it easy to measure per battery string current with a DC Current Meter:
      • http://www.smartgauge.co.uk/batt_con.html
      • Method 1 == No balancing of wiring resistance, not easy to measure per battery current flow
      • Method 2 == Not easy to measure per battery current flow
      • Method 3 == Very easy to use Clamp meter, and fuse/breaker each string
      • Method 4 == No easy clamp meter access, no easy breaker/fuse per string
      • Note, all of the above become "easy" to measure per string current and per battery voltage if you use (for example) a pair of 6 volt batteries vs a single 12 volt battery.
    Anyway--Just some of the details that can go into "picking" your battery bank.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Re: Why should not batteries of different age be connected in parallel?
    BB. wrote: »
    Method 3 == Very easy to use Clamp meter, and fuse/breaker each string.

    Exactly. Measuring net imbalance, at least at some selected load/soc points, is easy - not impractical as some suggest.

    I am available for custom hardware/firmware development

  • sotomar77
    sotomar77 Registered Users Posts: 5 ✭✭

    Hello, everybody.

    I have designed a solar system setup; however, I am still confused about the too many series parallel strings of banks.

    I need to know the logical electronic or scientific principle supporting the law referred that too many or more than two strings are more difficult to charge. As far as I understand, you will use more strings (series sets of batteries) connected in parallel to increase the amperage capacity of the system. Why is it bad to have 7 sets to supply the energy demand? Is it that critical? When I check my wiring resistance it is zero from one connection to the other. To be on the save side, I have connected the negative charge in the middle of the bank's parallel wiring and the positive at the beginning of the parallel bank's wiring connection. Would that help?? Is a seven string too many for my 24v bank using 6 volts cart batteries 215 amps.

    Can anybody explain the why it is wrong to have many strings?

    What would be my best option?

    Thank you for the help,

    Marcelino


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Short answer--With a "proper design" and everything working well--Yes you can run parallel multiple (more than 3) strings.

    More or less, the problem starts when real life hits theory. First, current sharing--Today, with relatively cheap DC Current Clamp Meters, it is relatively easy and quick to measure the per string current (under load, under charge, and/or even idling) to see that a) current sharing is "as expected" under load and under charge. Also, when there is no current flow, checking current flow to find if there is a shorted cell/failing battery in the parallel strings.

    I also suggest that you choose batteries that allow you to check battery terminal voltage directly (i.e., 2x 6 volt batteries in series vs 1x 12 volt batteries only in parallel--This is a very quick check with a volt meter that all batteries are reasonably balanced in charge (no very low or very high battery voltage). With 1x 12 volt batteries in parallel (in a 12 volt battery bank). You can only measure "bank voltage" and you cannot easily tell if there is an open/shorted cell (most modern deep cycle batteries do not let you check "per cell" voltages).

    So, you have to rely on a) clamp current measurements and b) specific gravity measurements of electrolyte (and SG measurements are not possible with AGM/GEL/VRLA type batteries).

    The "second level issues" mostly deal with extra costs (parallel cabling should have a fuse/breaker per string if >2 parallel connections) and maintenance issues (lots of cells to check electrolyte levels and specific gravity on). And failures--7 parallel strings means 7x more cells that can fail (and you have to "find" the failed cell/battery) and other "hidden" failures (poor/dirty/corroded connections/cables).

    And there are "third level issues"--Lead acid batteries are affected by cell temperature and minor (10's of milliOhms) differences between cells/connections/cabling. You can have a shorted cell which discharges other parallel strings (hidden failure that can damage other "good strings")--You can have an open cell which reduces capacity of one battery/string (but does not discharge other strings). You can have one battery (or string) start running "Hot" during charging (due to poor wiring, age of cell, cell starting to fail, poor airflow around a few batteries, etc.) and that string takes the majority of the charging current--leaving other strings "starved for charging" and subject to thermal run-away---T.R. is even more of a problem with AGM/GEL/VRLA type batteries).

    It is sort of the "issue" of why do you have 1x 8 cylinder car engine vs 5x very small 8 cylinder car engines in parallel. Both will work, but one has 5x more parts, 5x complexity for servicing, and 5x complexity for things to fail.

    In your case, 7x12=84 cells/connections to inspect/water/etc... Vs 1x12=12 cells to inspect/water/etc. (1x parallel string with larger batteries). Checking/logging water/SG level on 84 cells is not on the top of most people's fun list.

    I am not quite sure I understand your wiring with the center negative bus bar--The rule of thumb is that all (total wire) runs for each string should be the same gauge of wire and same overall length (i.e., each string adds up to 6 feet total, don't care if + wire is 2 feet and - 2 wire is 4 feet for one string and the other is +4feet and -2feet -- Just want total run(s) to be the same length).

    And unless you have a milliOhm meter, most Digital Volt meters cannot measure down to 0.001 Ohms accurately--Which is what you need to measure if you were to "Ohm out" your battery cables (~10-50 milliOhm total or so typical? Lots of "it depends" hand waving here--bank size, voltage, load current, etc.)

    So--If you have no other choice (in many regions, countries, the choice of deep cycle batteries is severely limited--In some cases, people cannot move large 2/4/6 volt high amp*hour cells--which can weigh over 300 lbs each)--Yes, you design the bank properly and keep on top of the maintenance/inspection--It will function for you (and may function well).

    However, if you do not keep up on maintenance/inspections, and as the bank ages--The "little hidden failures" can add up to a major problem down the road.

    The 1-2 String bank is probably "ideal"... Personally I would suggest not going more than 3x parallel strings if you can avoid it (again a personal recommendation). And I would avoid lots of parallel strings unless you enjoy "tinkering" around your battery bank (some people do enjoy weekly/monthly work on a large battery bank, others do not--That is your choice, I cannot help you there).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • sotomar77
    sotomar77 Registered Users Posts: 5 ✭✭
    edited June 2016 #17

    Thank you Bill.

     I do appreciate your help. I understand what you mean; there is a lot of maintenance to be done when you have that many batteries. As a matter of fact, have to add 3-4 gallons of distilled water on those batteries. It would cost a lot if I were to buy the water, so I designed a solar water distiller which gives me more than the water I need for the batteries.

    I wish I could send you a picture of my bank but there is no way to send the attachment here. If you had an email, I would gladly send you a picture so you get the idea of how the batteries are set up. You are right about old and new batteries mixing; the new batteries get over-charged and the old gets less until you get a burnt new battery or open old bat. Discharging of old and new batteries in series under load is indirectly proportional too; the old batteries discharge fast while the new keeps its charge; during the next charging event, the new prevent the old bat. from fully charging. The controller will see the string as a unit and will continue to charge because of the old bad batteries have a lower voltage so the good new battery will give an even higher voltage until the controller is satisfied. This goes on and the new one never gets more than 10% discharge its life cycle. Would this new battery be destroyed from overcharging? I have had this situation. I have disabled the new battery by disconnecting one of the series cables (not the parallel). I would have to buy 3 new to keep them the same. Also thinking about having 4 strings larger batteries, since I would have less batteries dying at the same time, and it would cost more or less the same to replace them.

    Thanks a lot for your information; have a great day,

    Marcelino

  • sotomar77
    sotomar77 Registered Users Posts: 5 ✭✭
    Hello, Bill.
    Here is the picture of my battery bank so you get the idea of what I have written.


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Some of your batteries are from late 2011--That is not too bad. 5 years of life is probably a good number for golf cart batteries in a warmer climate cycling.

    And, the old problem--replace one or two batteries that may have early life failures--But at what point to replace the whole bank (possibly even the "newer" batteries too?)--Or you end up replacing a battery or two at a time for the foreseeable future as each each hits its own failure time. (you will probably get really good at identifying failing batteries in your bank).

    What happens when you mix old and new batteries--It depends... If you replace a new battery in a string--Old batteries tend to have higher self discharge--So the new battery may "over charge" during Absorb/Equalization.

    Batteries also tend to increase in capacity (and lower internal resistance) over the first ~100 cycles or so--They are changing parameters--How that mixes with series/parallel strings--It all depends on how deep you cycle, temperatures, age of other batteries, incipient failures of other series/parallel batteries, etc.

    Fewer large batteries vs a lot of small batteries--Probably similar water usage (big batteries do use a fair amount of water--Although, there are some newer solar series batteries from Trojan, for example, that some folks have used here that do consume less water).

    Your home made solar still--I would be interested to see your design and how much water it produces for you--Sounds like a very good idea.

    Also--You have a large parallel bank--How has that worked out for you in your opinion? Are you happy with the results? Rules of thumbs are used to help guide people to what (we think) are optimum system designs--But learning from others that do push the limits helps us a lot to understand those limits (and if they are accurate or not).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • South Africa
    South Africa Solar Expert Posts: 295 ✭✭✭
    Am I blind or is there a cable missing there?

    5kVA Victron Multiplus II, 5.2kW array, 14kWh DIYLifepo4 bank, all grid-tied.


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    sotomar77 said:
    ... Would this new battery be destroyed from overcharging? I have had this situation. I have disabled the new battery by disconnecting one of the series cables (not the parallel). I would have to buy 3 new to keep them the same. Also thinking about having 4 strings larger batteries, since I would have less batteries dying at the same time, and it would cost more or less the same to replace them.

    Thanks a lot for your information; have a great day,

    Marcelino

    I think that is why.

    And to answer better to Marcelino--If you are replacing one battery until the rest of the bank "dies"--Then I probably not worry.

    If you are going to try and keep the rest of the bank for many more years--Perhaps it would be best to replace all 4 batteries in the string... But at this time, I really cannot give you any better information (I do not have an statistics to backup choice A or B--other than "just a feeling").

    Perhaps some others can help.

    At some time in the future Marcelino--What would your next bank be? Same batteries, same number of batteries? Reduce bank size (oversized for load)? Or would you look at larger AH batteries to reduce the number of parallel strings?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • sotomar77
    sotomar77 Registered Users Posts: 5 ✭✭

    Hello, Bill and everybody.

    Bill the reason there is a cable missing is to disable that particular string of battery; it just happened that there were 3 batteries with good specific gravity (green), and since I was replacing one of the banks I decided to buy 5  batteries to replace the one that was red.

  • sotomar77
    sotomar77 Registered Users Posts: 5 ✭✭

    Hello, Bill and everybody.

    The reason there is a cable missing from one of the string is to disable that bank; Since had to replace a one bank of battery I decided to replace a bad (red) battery from that string since the other three had a good specific gravity (green). I bought 5 batteries and added a new one to that bank. I used the other four to make the new bank. But unfortunately, it did not work; just like you had said: Old batteries don’t work with new ones unless they are in really good shape. I learned from mistakes; that new battery was fully charged, but the other 3 old batteries never got fully charged. This happened less than a month ago. I just replaced the three old ones today, and everything seems to be normal.

    I will wait until these batteries get old enough to get some bigger batteries with higher capacity.

    The missing wire is not a mistake for that is the way I test my banks rather than going around checking every single cell. I do it two different ways. First, once I have that wire disconnected, I can check each battery on that particular bank with a volt meter for correct voltage. If I find something unusual, I verify checking specific gravity. Another test I like to do is to see if that bank can handle the load without a drastic voltage drop; to do that, I just take a wire off each other strings and let that specific bank run the house. There should be a reasonable voltage drop because the load would be more for them. However, if they are no good there will be a drastic voltage drop; that bank has to be replaced. You may think: How about if there is a good battery there in that bank? Just one good battery will keep the voltage stable to a point. And you will double check voltage and specific gravity to make sure they're all bad. By the way, a good voltage and specific gravity doesn't mean they are good. If they are worn out and have very low capacity they have to go. The only way to know is with a heavy load.

    I don’t want to bore anyone, but I thought this may help someone who could run into a similar situation.

     Thank you for your help have a great day,

    Marcelino

  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Differences in battery internal resistance is a major reason to avoid parallel batteries AND mixing old/new batteries

    New batteries have lower internal resistance, and will therefor hog all the loads and charging.  This will rapidly age them till they perform just as bad as the rest of the bank, This degradation can happen in weeks to months. Then the whole bank approaches equilibrium and works for a while longer.

    Also, parallel cells should have fuse or breaker protection, so if a cell shorts out, the rest of the bank is prevented from dumping unlimited power into the short.   Adding all the fuses is expensive, and adds resistance and failure points to the system (each connection is a failure point)
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    edited July 2016 #25
    > Differences in battery internal resistance is a major reason to avoid parallel batteries AND mixing old/new batteries

    But why not fix this imbalance?  Not perfectly, but well enough that under the typical usage pattern, the charge in/out of each battery is much more equal.  Same question for batteries of the same age, but inevitably not perfectly matched.

    I don't understand "match the cable lengths and hope for the best".  Would also be interesting to see actual numbers - I expect that battery to battery variations can exceed the imbalance caused by quite a bit of cable length mismatch.

    I am available for custom hardware/firmware development

  • 2manytoyz
    2manytoyz Solar Expert Posts: 373 ✭✭✭
    I actually did have batteries of different ages, and manufacturers, in a series/parallel configuration.  I used T105 type golf cart batteries, and put two in series to make 12V.  I would add a pair every so many months, until I had 8 batteries.  There was about a 2 year difference between the first batteries, and the last.  This wasn't by design, but rather a tinkerer with an expanding system.

    They each lasted almost 6 years in the heat of central FL.  Of course that meant I'd have to remove a pair at a time, diminishing the overall capacity of my battery bank as they reached the end of their life.

    I didn't really notice any difference in the performance when using a digital ammeter on each pair of batteries.  They all supplied amount the same amount of current during heavy loads, and about the same charge rate afterwards.

    When those batteries were done, I bought 8 new batteries, same age, same brand.  Hope these last as long as the others.

    Some brands, or battery types, will be more sensitive to matching.  The golf cart flooded batteries are quite forgiving...