Testing capacity of AGM batteries

Matty
Matty Registered Users Posts: 36 ✭✭
edited February 2022 in Off Grid Solar & Battery Systems #1

I have 6 x 130Ah AGM batteries that I purchased 3-4 years ago that have been used infrequently between annual camping trips. They have not been discharged to less than 50% state of charge, they have been recharged quickly after use, have been stored in a cool room and have had a top up charge at least once per month. I noticed that the resting voltage when fully charged had dropped from around 13V to around 12.85V. I wondered whether this was a sign of expected deterioration. So I decided to test a battery so see whether I could determine its current useable capacity. I connected an amp meter between a fully charged battery and my 12V fridge and I monitored the voltage and the number of Amp Hours consumed (which the meter tracks) until the resting voltage dropped to 12V. Then I took the reading for the number of Amp Hours consumed to determine useable capacity. I discovered that the battery consumed 38Ah to drain from fully charged 12.85V to a (resting) voltage of 12V (see image attached).

These are my questions:

Is this a sound methodology to measure usable battery capacity?

Are there special devices that provide this information without having to discharge the battery under a load for 24 hours or more?

My findings determined that the battery has a usable capacity of 38Ah (based on a maximum of 50% depth of discharge) whereas the rated usable capacity is 65Ah (half of 130Ah). Does this sound like typical and expected deterioration of a well maintained AGM battery over a 3-4 year period?

Could I expect better performance and longevity if I were to upgrade to a Lithium battery which is the primary imperative of this enquiry?

· Share on Twitter

Comments

  • softdown
    softdown Solar Expert Posts: 3,898 ✭✭✭✭
    #2
    What temp is the battery at? Battery capacity drains precipitously as temp. falls.
    It would be easier to measure aH with a battery meter. This looks similar to the one I use: https://www.amazon.com/MOTOPOWER-MP0515A-Automotive-100-2000-Alternator/dp/

    Though mine measures aH and I'm not sure if that one does.

    I find resistance is a valuable measurement of battery health. Though I'm not sure how heavily resistance is determined by battery temperature.

    All batteries seem a lot weaker when they are cold.

    First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 165(?) watt BP Solar with Renogy MPPT 40A controller/ and Xantrex C-35 PWM controller/ and Morningstar PWM controller...Cotek 24V PSW inverter....forklift and diesel locomotive batteries
    · Share on Twitter
  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,167 ✭✭✭✭
    edited February 2022 #3
    Your approach is good common sense but how do you know that you were at 50% SOC? Generally, that voltage number is on a curve based on load, and temperature.
    The only capacity test measurement accepted and endorsed by Concorde battery is a controlled discharge from 100% to 0% SOC. The number of amp-hours measured is the battery capacity. They consider anything else to be a guess. With Concorde AGMs, 10.5v under load (large or small load) is 0% SOC. Yes, the test must be compensated for temperature and time.
    Last year, I had an Army base wanting to return 60 new batteries because sample testing showed that they were faulty according to their super-duper $3,000. test rig. I brought my calibrated load bank down there and ran capacity tests. They used their machine to verify the actual load being applied. The 125 ah batteries provided 127-129 ah and they had not been cycled enough to reach maximum capacity. They still rejected the shipment because their tester said "bad." 
    Depending upon the brand/quality/construction of your battery, you should be able to take a true deep cycle AGM down to 0% SOC 250-800 times. So an occasional test that deep is not going to hurt it. In fact, it may help your capacity assuming you charge them back up at their maximum rate. Exercise is good! Again, all of the above is based on decent quality true deep cycle batteries.


    I always have more questions than answers. That's the nature of life.
    · Share on Twitter
  • Matty
    Matty Registered Users Posts: 36 ✭✭
    edited February 2022 #4
    softdown said:
    What temp is the battery at? Battery capacity drains precipitously as temp. falls.
    It would be easier to measure aH with a battery meter. This looks similar to the one I use: https://www.amazon.com/MOTOPOWER-MP0515A-Automotive-100-2000-Alternator/dp/

    Though mine measures aH and I'm not sure if that one does.

    I find resistance is a valuable measurement of battery health. Though I'm not sure how heavily resistance is determined by battery temperature.

    All batteries seem a lot weaker when they are cold.

    I’m in Australia where the temperature is currently in the mid 20’s Celsius and I was testing in my garage which is quite cool.

    Does your battery tester provide an immediate estimate of battery capacity in amp hours? If so, I would be interested to find out the make and model. it’s not apparent that this feature is supported with that Motopower product.

    How does the resistance reading correlate with battery health?
    · Share on Twitter
  • softdown
    softdown Solar Expert Posts: 3,898 ✭✭✭✭
    edited February 2022 #5
    My meter is a red Ancel. It does measure aH capacity and yields a digital read out.

    As battery quality deteriorates, internal resistance measurements increase. High quality batteries may start with a resistance reading as low as 1.5, Normal batteries seem to start around 2.5 and may measure as high as 5.5 when new. Quality manufacturers often reject batteries measuring over 5. They will not sell a new battery measuring over 5. I return new batteries if they measure over 5. I find an alarming amount of new Interstate (auto) batteries sold via Costco measuring over 5.

    I will typically find a resistance around 6.5 in a battery that is several years old. Sometimes I replace them at that stage. A junk battery that barely starts a vehicle, or a bad cell, may have resistance over 8.

    CCA measurement is my favorite number for evaluating an auto battery. I have 14 cars and trucks (business). Battery maintenance is a vital monthly chore.

    Of course aH is the vital measurement for off grid energy storage.
    First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 165(?) watt BP Solar with Renogy MPPT 40A controller/ and Xantrex C-35 PWM controller/ and Morningstar PWM controller...Cotek 24V PSW inverter....forklift and diesel locomotive batteries
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,613 admin
    edited March 2022 #6
    Not speaking for Marc, but I would guess his test is to pull current at a 20 AH rate (or whatever rate is appropriate for the customer) and time until the battery is empty. For example:
    • 125 AH battery at 20 hour rate
    • 125 AH / 20 Hours = 6.25 Amps
    • Run test at 6.25 Amps and time out:
    • 129 AH / 6.25 Amps = 20.64 Hours of "Testing Time" (not quick 1 minute capacity test here)
    Note that loads vary... A simple resistive load draw more current at 12.8 volts vs 10.5 volt battery bus test (constant resistance load)

    Then there are very nice constant current testers--Basically electronic loads that adjust current draw to maintain set current with variable battery bus voltage (constant current test).

    And there are the "real loads" of today... AC inverters. These are "constant power" loads. I.e., draw less current at 12.8 volts and more current at 10.5 volts:
    • 6.25 amps * 12 volts = 75 Watt load
    • 75 Watts / 14.4 volts = 5.21 Amps (battery under charge)
    • 75 Watts / 12.8 volts = 5.86 Amps (full battery)
    • 75 Watts / 10.5 volts = 7.14 Amps (empty battery)
    Each will method will give somewhat different answers.

    Battery resistance--Marc can answer better than I... In general, you have batteries that have higher resistance (flooded cell deep cycle), batteries with less internal resistance (such as AGM and GEL). And batteries with very low internal resistance (some Li Ion, NiCad).

    In general... I would use "battery resistance" as a quick "aging/capacity test". I.e., whatever the battery measures at voltage/load "new"--That is the starting point.

    What the battery resistance is when old/failing in the application (i.e., won't start the car, won't run the loads overnight)--Is the "no good for application" resistance.

    Whatever you measure in between good/bad is an indication of aging.

    Note that "battery resistance" is really multiple "things". Some of the major things are:
    • physical resistance: Resistance of internal connections, battery plate grid condition, active plate surface area (are some)
    • chemical activity: electrolyte specific gravity state of charge, how much sulfur has been locked into sulfated plates, active plate area vs sulfation, etc.
    • And if you use a tester that "pulses" the battery at higher frequencies--You have physical capacitance (plate to plate surface area) and inductance of internal wiring/grids/etc. This is technically impedance and use with capacitors in electronics, but applies to batteries too (i.e., a 60 Hz AC inverter hits batteries with 120 HZ "ripple current"--Looks very much like an electrolytic capacitor).
    In the end, no one measurement will say a battery is good or bad for the application. However, monitoring over time for degradation (such as XX.X volts @ YY.Y Amps -- Such as running the toaster in the morning before the bank gets charged) will show how the battery is aging.

    Or even just monitoring the end battery voltage first thing in the morning (for remaining capacity).

    Lead Acid batteries tend to be "complicated" (temperature, aging, state of charge, internal resistance, etc.). Li Ion batteries tend to be "simpler" (very low internal resistance on "good cells"--So State of Charge tends to follow battery voltage better and no temperature offsets, low current offsets).

    -Bill "a few thoughts" B.  :)
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • Matty
    Matty Registered Users Posts: 36 ✭✭
    edited February 2022 #7
    Marc Kurth said:
    Your approach is good common sense but how do you know that you were at 50% SOC? Generally, that voltage number is on a curve based on load, and temperature.
    The only capacity test measurement accepted and endorsed by Concorde battery is a controlled discharge from 100% to 0% SOC. The number of amp-hours measured is the battery capacity. They consider anything else to be a guess. With Concorde AGMs, 10.5v under load (large or small load) is 0% SOC. Yes, the test must be compensated for temperature and time.
    Last year, I had an Army base wanting to return 60 new batteries because sample testing showed that they were faulty according to their super-duper $3,000. test rig. I brought my calibrated load bank down there and ran capacity tests. They used their machine to verify the actual load being applied. The 125 ah batteries provided 127-129 ah and they had not been cycled enough to reach maximum capacity. They still rejected the shipment because their tester said "bad." 
    Depending upon the brand/quality/construction of your battery, you should be able to take a true deep cycle AGM down to 0% SOC 250-800 times. So an occasional test that deep is not going to hurt it. In fact, it may help your capacity assuming you charge them back up at their maximum rate. Exercise is good! Again, all of the above is based on decent quality true deep cycle batteries.

    I worked on the assumption that 50% SOC was a reasonable estimate as the voltage was exactly 12V, the battery was not under load (for some time) and the temperature was modest.

    Is 12V a good estimate of 50% SOC or could the SOC be much higher at 12V?

    If I only intend to discharge to 50% SOC and use 12V as an estimate of this then the methodology used gives me an indication of the capacity that I intend to use and then I can plan around having extra batteries if required though I am still curious to know what the current useable capacity is compared to the battery’s rated capacity. 

    I hadn't thought to drain the battery to 0% SOC as there are a lot of conservative views suggesting against that but I believe the key is to immediately recharge to prevent sulfation? Thanks for clarifying that this is a reasonable thing to do.

    Do you think that it could be that a lot more amp hours will be consumed to discharge from 12V to 10.5V than the 38Ah I observed consumed between discharging from 12.85V to 12V? Is it possible that the battery could actually still be close to its rated capacity of 130Ah?
    · Share on Twitter
  • Matty
    Matty Registered Users Posts: 36 ✭✭
    #8
    softdown said:
    My meter is a red Ancel. It does measure aH capacity and yields a digital read out.

    As battery quality deteriorates, internal resistance measurements increase. High quality batteries may start with a resistance reading as low as 1.5, Normal batteries seem to start around 2.5 and may measure as high as 5.5 when new. Quality manufacturers often reject batteries measuring over 5. They will not sell a new battery measuring over 5. I return new batteries if they measure over 5. I find an alarming amount of new Interstate (auto) batteries sold via Costco measuring over 5.

    I will typically find a resistance around 6.5 in a battery that is several years old. Sometimes I replace them at that stage. A junk battery that barely starts a vehicle, or a bad cell, may have resistance over 8.

    CCA measurement is my favorite number for evaluating an auto battery. I have 14 cars and trucks (business). Battery maintenance is a vital monthly chore.

    Of course aH is the vital measurement for off grid energy storage.
    This page on Amazon has a nice chart showing the capabilities of Ancel testers:

    https://www.amazon.com/ANCEL-Battery-100-1100-Automotive-Analyzer/dp/B07Q6NBXQG

    It looks like I need a BST200.

    Do you know whether there is a similar product that also tests Lithiium batteries or is a separate tester required?
    · Share on Twitter
  • Matty
    Matty Registered Users Posts: 36 ✭✭
    edited February 2022 #9
    softdown said:
    My meter is a red Ancel. It does measure aH capacity and yields a digital read out.

    As battery quality deteriorates, internal resistance measurements increase. High quality batteries may start with a resistance reading as low as 1.5, Normal batteries seem to start around 2.5 and may measure as high as 5.5 when new. Quality manufacturers often reject batteries measuring over 5. They will not sell a new battery measuring over 5. I return new batteries if they measure over 5. I find an alarming amount of new Interstate (auto) batteries sold via Costco measuring over 5.

    I will typically find a resistance around 6.5 in a battery that is several years old. Sometimes I replace them at that stage. A junk battery that barely starts a vehicle, or a bad cell, may have resistance over 8.

    CCA measurement is my favorite number for evaluating an auto battery. I have 14 cars and trucks (business). Battery maintenance is a vital monthly chore.

    Of course aH is the vital measurement for off grid energy storage.
    I purchased a red Ancel BST200 battery analyser. Do you know which value to use for the "battery standard" when testing an AGM battery? I can't see anything obvious on the battery itself that has this information. The possible values are:

    CCA
    CA
    MCA
    JIS
    SAE
    DIN
    IEC
    EN

    When selected, most of these request the rated cold cranking amps of the battery being tested which is only relevant to a car battery.
    · Share on Twitter
  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,167 ✭✭✭✭
    #10
    I am not familiar with that unit. I do not see how a few seconds of testing can tell you how many amp-hours of storage capacity you have remaining in your batteries.
    CCA? Sure, that is easy. Perhaps someone can explain to me what I am missing. Seriously, I am always learning!

    I always have more questions than answers. That's the nature of life.
    · Share on Twitter
  • Matty
    Matty Registered Users Posts: 36 ✭✭
    #11
    Marc Kurth said:
    I am not familiar with that unit. I do not see how a few seconds of testing can tell you how many amp-hours of storage capacity you have remaining in your batteries.
    CCA? Sure, that is easy. Perhaps someone can explain to me what I am missing. Seriously, I am always learning!

    I was thinking exactly the same thing. I am wondering how accurate the test is with this unit and whether there is a more robust device available that puts the battery through more rigorous testing.
    · Share on Twitter
  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,167 ✭✭✭✭
    #12
    Matty said:
    Marc Kurth said:
    I am not familiar with that unit. I do not see how a few seconds of testing can tell you how many amp-hours of storage capacity you have remaining in your batteries.
    CCA? Sure, that is easy. Perhaps someone can explain to me what I am missing. Seriously, I am always learning!

    I was thinking exactly the same thing. I am wondering how accurate the test is with this unit and whether there is a more robust device available that puts the battery through more rigorous testing.
    Matty, What are the deep cycle specs for your batteries? What is the brand name on the label? (You may have posted it and I missed it)
    Marc

    I always have more questions than answers. That's the nature of life.
    · Share on Twitter
  • Matty
    Matty Registered Users Posts: 36 ✭✭
    #13
    Marc Kurth said:
    Matty said:
    Marc Kurth said:
    I am not familiar with that unit. I do not see how a few seconds of testing can tell you how many amp-hours of storage capacity you have remaining in your batteries.
    CCA? Sure, that is easy. Perhaps someone can explain to me what I am missing. Seriously, I am always learning!

    I was thinking exactly the same thing. I am wondering how accurate the test is with this unit and whether there is a more robust device available that puts the battery through more rigorous testing.
    Matty, What are the deep cycle specs for your batteries? What is the brand name on the label? (You may have posted it and I missed it)
    Marc

    I have attached a spec sheet for my 130Ah AGM 12V battery
    · Share on Twitter
  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,167 ✭✭✭✭
    #14
    Not much detail is available without knowing who the actual manufacturer is, but I do see that they are rated for 350 charge/discharge cycles at 100% DOD. So I would refer back to a capacity test that counts the amp hours when draining the battery from 100% to 0%, using 10.5v under load as the 0% point. I note that they indicate 40 amps is the max charge rate and I would use that rate after taking it down to to 10.5v,
    I always have more questions than answers. That's the nature of life.
    · Share on Twitter

Categories