AGM Battery Test, Six Month Report Card

imageSix months ago, we started a test of AGM batteries from LifeLine to see if, by implementing recommendations from that manufacturer, we could solve the problem of short life that we, and many other live aboard voyagers, have experienced with AGM, and in fact all, lead acid batteries.

It’s still early days. But on the other hand, our last set died in a year, so if we have improved things it should be showing by now.

Results of AGM Battery Test

Drum roll…can I have the envelope please… A discharge test indicates that the batteries, after some hundred 50% discharges, are performing at near-new levels.

Discharge Test

A true discharge test as documented in the LifeLine Batteries manual requires pretty sophisticated test equipment that we don’t have. However we were able to get a satisfactory measurement for our purposes simply by leaving our computers and lights on for a total draw of between 10 and 20 amps (depending on what we were doing) on each battery until the voltage dropped to 10.5 volts and then reading the number of amp-hours used off our Link 2000 digital meter.

Both batteries (tested separately) were able to supply 230 amp-hours, give or take 5 amp-hours, before turning their little lead toes to the sky. Their rated capacity is 255 amp-hours over 20 hours—a current (amperage) draw of 12.75 amps—or 206 amp-hours at a draw of 15 amps, so our result was pretty respectable and in line with what we got from the same test when they were new.

It is also encouraging that the performance of these two house bank batteries is near identical, indicating that our technique of  keeping the combiner switch on “both” (parallel) except when equalizing, is working out well.

So what do we attribute the dramatically better performance of this set when measured against our last three house battery banks? Well, you can read the whole series of posts on the changes we made. But in the next post in the series we will summarize all that mind numbing detail, as well as revealing a couple of other things we changed but have not posted on yet.

Like what you just read? Get lots more:

Meet the Author


John was born and brought up in Bermuda and started sailing as a child, racing locally and offshore before turning to cruising. He has sailed over 100,000 miles, most of it on his McCurdy & Rhodes 56, Morgan's Cloud, including eight ocean races to Bermuda, culminating in winning his class twice in the Newport Bermuda Race. He has skippered a series of voyages in the North Atlantic, the majority of which have been to the high latitudes. John has been helping others go voyaging by sharing his experience for twenty years, first in yachting magazines and, for the last 12 years, as co-editor/publisher of AAC.

13 comments… add one
  • Chris Jan 13, 2011, 8:45 pm

    In the “mind-numbing detail” to follow could you consider including some graphical exhibits on charge, discharge, equil. cycles, etc?

    • John Jan 14, 2011, 10:49 am

      Hi Chris,

      Sorry, I guess I was not clear. The mind numbing detail is already up in the form of the six posts written over the last six months on what we changed.

      The next post will be a summary of what we did.

      I won’t be getting into graphs on charge and discharge characteristics for batteries. First off, I’m not qualified to do so and second, that stuff is generally well understood and well documented. (The LifeLine Battery site is just one good source.) Therefore I have and will confine myself to the actual real-world application of those principals on a live-aboard voyaging sailboat.

  • Viv and Mireille Jan 14, 2011, 8:57 am

    Interesting subject. We tend to think, or maybe not think, enough of batteries. Just looked at a Grand Soleil 46 in Greece that had fire damage in the aft cabin caused by a dead battery (or a battery with a dead cell) being overcharged and catching fire. (Broker didn’t tell us about this but that’s another story.) This fire, although small, cooked the cockpit sole resin enough that it will all have to be cut out and re-built at a cost of about $10K. So batteries need care and attention, should not be taken for granted and should be considered as an essential piece of boat gear.



    • John Jan 14, 2011, 11:09 am

      Hi Viv,

      Really good point. I am always amazed by how many boat owners are willing to leave battery chargers on and unattended for weeks at a time. All it takes is for one cell to go bad and there is a big potential for a fire.

  • Steve & Rebecca Jan 30, 2011, 12:06 pm

    Hi folks,
    Very late to the conversation I know but I have read it all with great interest and have in fact delayed our electrical refit pending some conclusions!
    We are about to splurge on an (almost) complete system (we have some good bits that we wish to retain) and, having listened to all that has been said, here are our thoughts…

    1. 100% agree with John’s ‘keep it simple’ philosophy. Having worked in the marine business for nearing 40 years we have seen no end of ‘must-have’ gizmos that simply make life more difficult & costly. Good quality in simple design is our philosophy.
    2. AGM is the way to go IF it can be made to work.
    3. John has illustrated (beyond reasonable doubt in my mind) that modern gear can do amazing things but NOT all at the same time! (e.g. Correct absorption charging whilst the batteries are also on high load duty.)
    4. It would seem to be the conclusion of most contributors that the final stages of charging (i.e. beyond ‘bulk’) is best carried out on ‘no load’ batteries.
    5. The above naturally leads to at least two battery banks.
    6. ‘DOD’ and the number of cycles are of supreme importance.

    But this is where I am tending towards a difference of opinion (whilst fully respecting everyone else’s opinion of course!).

    It has become the norm to split two sets of batteries into ‘House’ and ‘Engine Start’ – and I’m beginning to wonder if this is now the best approach. Bear with me just a little.

    Lifeline 4Ds / 8Ds and similar AGMs have extremely high CCA (Cold Cranking Amp) ratings – they are more than capable of starting a well maintained 4 or 6 cylinder diesel – especially as they are started from cold so infrequently on a typical cruising boat compared to say a delivery truck.

    The DOD during engine start on a set of these batteries (especially in 24volt pairs) is absolutely minimal compared to the energy expended in say a typical 50% DOD on a house bank.

    So why does it make sense to buy two high capacity banks of AGMs and then hammer the house banks with daily 50% DODs whilst letting the engine bank off so lightly?

    I am giving serious consideration now to installing two equal AGM banks (probably 2 x 4Ds in each bank) and using banks for full duty (i.e. House AND engine start) on alternate days.

    The bank NOT on duty would be the one to receive absorption (or any post bulk) charging (i.e. charged to =/- 100% every other day).

    There would be no attempt to charge beyond bulk the duty bank – it would get its 100% charge every other day anyway.

    This could be achieved with a simple 3-way C.O.S. on both the input (charger or alternator) and the output (DC Distribution panel + engine start).

    When input is selected to Batt 1, output is selected to Batt 2 and vice versa. Smart regulators then only have to deal with isolated (offload) battery banks with temperature detection.

    If, in the unlikely event that you discharge the duty bank below 50% inadvertently, you still have the other bank to swap to or combine as normal.

    50% cycling duty is shared between two banks rather than one, common batteries are in all banks, and charge monitoring is much, much, simpler.

    We are even considering doing away with any ‘Smart’ regulation on our old (heavy case) 180A/24V CAV alternator once we can confirm a reliable / trustworthy mains charger – the idea being to let the alternator bulk charge the isolated batteries, then swap it over to bulk charge the duty bank + running load, AND, when sufficient juice is available, let the alternator ‘top off’ the isolated bank via the inverter and (separate) charger running off the duty bank – this is ’round the houses’ I know but it means again only 100% charging isolated batteries accurately and you have the ‘get you home’ bank up of simply bulk charging each bank (and no more) until you sort the problem.
    Have I gone mad, am I missing something?

    Ps. The Lifeline charging regime promoted by Justin and now trialled by John is only a smidgen away from how submariners have been caring for batteries since WW2 – we only need ‘Quarterly Discharges’ and the circle is complete!

    • John Jan 31, 2011, 6:06 pm

      Hi Steve and Rebecca,

      Um…I have to admit that I only got about half of that. Having said that, perhaps the following will help.

      I would be nervous about not having a dedicated engine start battery. On Morgan’s Cloud we have a 4D (probably bigger than we need) to start our Perkins M92B that is completely isolated from the rest of the system so that it can’t be discharged flat by accident. The only time it is connected to anything else is when the engine is running through a relay to the alternator for charging, which is energized by the engine room blower switch…dead simple.

      Our house bank is split in two because it makes it easier to equalize, see this post.

      There are some other fairly small advantages to a split bank, which I will cover in a future post, but they are, at least in my opinion, different from the ones you list.

      Maybe I’m being thick here (it would not be a first) but I just can’t see making DC (alternator) then converting to AC (inverter) then converting to DC (charger). Not only would the efficiency loss be in the order of 75% but there are way too many things to break for me.

      In a week or two we will be doing a summary post on what we have learned in the last six months. Since you have waited this long, you may wish to see what you think of that post before pulling the trigger on a lot of gear.

  • Eric Feb 1, 2011, 10:39 pm

    This is a very interesting thread. I would like to attract some attention on a critical point here however. John indicates that his sealed AGMs are gassing quite significantly during equalisation, to the point of requiring ventilation. Gassing, i.e. the generation of hydrogen and oxygen by electrolysis of water in the battery, consumes water by definition. In a sealed AGM, this water cannot be replaced, hence the recommendation of most AGM manufacturers not to equalise like standard flooded cells.

    I would say that John’s AGMs will fail at some point as a result of reduced electrolyte and excessive acid concentration, just like a flooded battery that has been low on water for long enough. He is trading off plate sulfation for electrolyte depletion. How many such equalisation cycles will be needed to reach failure is an open question of course, but he has clearly been better off so far. No argument there.

    AGMs are supposed to recombine H2/O2 generated during charging back into water through an internal catalytic reaction. This is why they don’t normally lose water or vent like ordinary flooded cells. There will be a limit to the capacity of this recombining process and intense overcharging would overcome it. It seems to be the case already here.

    Equalisation of AGMs is probably possible and most certainly beneficial, but only as long as the battery does not start venting. The acceptable equalisation voltage will therefore vary between manufacturers, models, etc and should be determined by observation. Here, in my opinion, John should really consider reducing his equalisation voltage, or possibly “pulsing it” so the battery venting valve does not open on over-pressure. This might give the gases enough time to recombine, dropping internal pressure, between pulses.
    The idea behind pulse-charging is getting the benefits from a power level that would be entirely unsustainable due to gassing and heating, without these harmful side-effects.

    This brings up the subject of pulsing. Equalisation and desulfation are two different things. Equalisation means fully recharging all cells and is normally achieved by increasing charging voltage for a time at intervals. This basically causes the “best” cells to gas and the “worst” cells to fully charge. This is very important, because if left partially charged, these “worst” cells will sulfate, develop high resistance and ruin the performance the whole battery very quickly.
    Desulfation on the other hand means breaking down the insulating lead sulfate crystals that have formed on the plates. This is best achieved using short-duration current pulses at a relatively high frequency.

    After six months only, I think it would be hard to say whether the equalisation scheme used has been effective at truly reversing the sulfation process; it may have mainly controlled its rate down dramatically.

    Typical desulfation schemes use micro-pulses at higher voltage while the battery is under no load (in order not to damage the loads; alternatively, you can filter them out before going to the loads and desulfate almost continuously). There are very simple circuits you can build that provide desulfation pulsing however, see:

    There are a few (more recent) commercial products that now implement this technique as well, but to my knowledge none seem to outperform the one above, and several can’t match it.

    Sulfation is a slow, gradual process. So is desulfation. The best is working at it gently over time and such a circuit drawing a few watts only may be the very best answer. A little more research can provide interesting feedback and more information about the way it operates.
    Remember that equalising a battery that is not sulfated will take much less voltage and time and won’t be required as often. A decent solar system will easily do that on a good day if you shed (some of) your consumption for a while and boost your charging voltage adequately.

    • John Feb 2, 2011, 11:23 am

      Hi Eric,

      Great comment and all very good points, thanks. Here is our thinking:

      • You are of course right that equalization is boiling off irreplaceable electrolyte from an AGM battery. However, what we are trying to do on Morgan’s Cloud is extend our batteries’ life to the point where AGMs become cost effective for our usage, not create the ultimate and best charging system possible. Justin, principle at LifeLine Battery, designed our protocol to do that and we are testing that protocol.
      • We did talk to Justin about pulsing. However, he felt that in our use where a constant power source for such a machine is not available for the very long periods required to de-sulfate a battery, together with the impracticality of not loading the battery being pulsed, made such systems impractical for an offshore live-aboard voyaging sailboat.
      • As you say, it would probably be possible, with some fairly complex DIY design and building, to get around the above objections, but for us and I suspect most other live-aboard voyagers, this would violate our commitment to keeping things simple.
  • Chris A. Feb 3, 2011, 4:35 pm

    Maybe this is silly and wouldn’t work but… Could you weigh the batteries at the end of six months? Subtract that from the factory spec weight. Any wt loss in grams should equal water loss in cubic centimeters. Plot the original volume of electrolyte per factory specs and use the water loss calculated by weight to get a trend line and at least make a rough prediction as to when the batteries might fail due to water loss.

    • John Feb 4, 2011, 9:43 am

      Hi Chris A,

      Might work, I guess. But if I even suggested such a regular task to Phyllis, or for that matter to my aging back, there would be mutiny.

      I really don’t expect the batteries to fail without warning from either sulfation or water loss. We should see slowly diminishing capacity well before they actually die.

  • Chris Feb 4, 2011, 10:10 am

    Also, engineer that I am, I did some math. These HOH reductions on the equalize cycle under discussion would be measured in fractions of grams from a ~62 kg mass. This would require a lab sensitivity measuring device. A 1 gram change would be 1/62000th requiring at least e-5 precision.

    While you are running an experiment, I don’t think you want to pony up the $ for that kind of equipment. It would be interesting data to see, though.

  • Garry Jan 18, 2012, 6:55 pm

    Hi John,

    No pressure, but 12 months on since the last test report how are your batteries now, Has the charging regime worked to date?

    I am just about to install 2 x Lifeline 8D and a Balmar alternator and reg on my launch (I know a dirty word) and your reports so far are the reason for choosing the Lifeline product way down here in NZ where they are even more expensive so I would love the reinforcement that what I a planning has worked for you.


    • John Jan 19, 2012, 10:35 am

      Hi Garry,

      I have a post in the hopper on this—the news is good.

Only logged in members may comment: