AGM Battery Failure - Two Sets in Two Years

[sailjumanji]

Originally Posted by CarlF

Yikes - you’ve worked hard at this.

When totally stumped by something I revert to “When you hear hooves, think “horses” not “zebras”

Untold numbers of AGM batteries have had short lives on cruising boats because of PSOC. The only people I know who are really happy with their AGM’s are plugged in almost every night. While your failures seem way too fast for PSOC damage, the symptoms are right.

I finally gave up on AGM’s and switched to Fireflys. If your problem is PSOC, that will fix it.

Yeah, the first batteries, I could easily see PSOC as the culprit. The second batteries had less than a couple of weeks cumulative of being on anchor, and the rest was on MassCombi shore power. So that cant point to PSOC.

That said, today I did another solar controller test and everything went well. But at the end, I started both the fridge and the freezer, and put the MassCombi on to top off the batteries. The shore power charger stayed in absorption to satisfy the refrig load of around 10 amps, which is keeping tjhe charge at elevated level. This might be an artifact of bad batteries that continue to suck up charge. But if not, keeping the batteries at absorption voltage of 14.25 amps instead of float at 13.8 volts for long term, could be the culprit.

The hits just keep on coming!

[sailjumanji]

Originally Posted by Jammer

Wood doesn't conduct electricity and is relatively resistant to damage from electrolyte. That said, you may be overthinking it. A larger gap doesn't always mean better cooling due to the fluid dynamics of the moving air. If you have some gap, and don't have batteries getting hot, you may be trying to solve a problem you don't actually have. I've watched iterations of heat sink designs and have given up trying to predict whether it is going to help matters to add fins, remove fins, make fins deeper, make fins more shallow, etc etc. It's harder than trying to understand voodoo.

I agree. But at this point, I am trying to do anything and everything that might reduce our chances of a failure repeat.

I do a lot of projects with Starboard. Doesn't rot, and is pretty easy to work with. Electrolyte damage should never be an issue, as the AGM's are sealed. I can figure this one out. I also need to make slightly longer jumper cables between the batteries. The others ones are long enough to connect, but I suspect there is some spec for radius to keep cables distant from the posts.

[sailjumanji]

Originally Posted by sailjumanji

That said, today I did another solar controller test and everything went well. But at the end, I started both the fridge and the freezer, and put the MassCombi on to top off the batteries. The shore power charger stayed in absorption to satisfy the refrig load of around 10 amps, which is keeping tjhe charge at elevated level. This might be an artifact of bad batteries that continue to suck up charge. But if not, keeping the batteries at absorption voltage of 14.25 amps instead of float at 13.8 volts for long term, could be the culprit.

This am, the MassCombi is on float, and both freezer and fridge are running. So working correctly - or at least not stuck on absorption.

[Jammer]

The marine battery chargers used on yachts do not ordinarily use separate current sensors for the battery current and the load current; where a battery shunt exists it typically is used only for metering and does not inform the charge algorithm. Therefore, the ampere load that the battery charger "sees" is the sum of the battery current and the load current.

This complicates determination of the proper point at which the charge cycle should end, that is, the point where the charger transitions from absorption to float, since the end point of the charge cycle is typically determined by the charge current falling below a particular value. When there is a load present, the charge current seen by the charger may never fall below the chosen value or will do so later than desired, because of the effect of the load. A workaround for this is to limit the duration of the absorption cycle.

That's what you're seeing, and it's unremarkable.

If you're concerned that the batteries are being overcharged, my recommendation would be to reduce the absorption voltage, the time limit, or both. Batteries will continue to charge on float, and the use of a higher voltage/longer duration absorption state is just to expedite the completion of charging -- which accomplishes nothing useful if you're at the dock overnight since they'll be charged by morning anyway.


Single-stage chargers (that only have a "float" voltage) work just fine except that they take about 24 hours to charge a battery that is mostly discharged. People believe they eat batteries; some do, but it's because the float voltage is too high (or poorly regulated) or they aren't temperature compensated, not because of the simple charge algorithm.


A fact to consider is that charging voltages are specified by battery manufacturers to achieve a desired tradeoff between capacity and cycle life. Reducing the charge voltages slightly reduces capacity and increases cycle life, assuming that the battery chemistry (particularly the plate alloy) remains the same.

[Maine Sail]

I have not been impressed with the Mastervolt AGM batteries we've tested for customers. We've capacity tested about a dozen of them, all less than two years old, and they just don't seem to hold up like a known quality brand such as Odyssey, Lifeline, Northstar, Fullriver, Rolls or Firefly. Deka GEL batteries are exceptional but their AGM's don't hold up nearly as well.

There are quality differences between AGM batteries. Last I knew the Mastervolt AGM batteries are made in China, and this may not be a bad thing, as Fullriver makes very good batteries, but not all Chinese suppliers do. I don't base my assessment of the Mastervolt AGM's on country of origin but rather the actual testing of used batteries..

[Bycrick]

Since Maine Sail is here, I’d like to ask a question. Up thread there were comments about and ACR/VSR won’t cause "over charging" of a starter battery when connected to a bigger house bank. "Each battery in parallel will take what it needs." But I remember reading one his posts a long time ago that pointed out that when a battery is REALLY fully charged, it may only be drawing milliamps of current, but all of that would be going into outgassing or heat. Leave it like like for a long time and the smaller battery will lose electrolyte. What have I missed?

[s/v Jedi]

Originally Posted by Bycrick

Since Maine Sail is here, I’d like to ask a question. Up thread there were comments about and ACR/VSR won’t cause "over charging" of a starter battery when connected to a bigger house bank. "Each battery in parallel will take what it needs." But I remember reading one his posts a long time ago that pointed out that when a battery is REALLY fully charged, it may only be drawing milliamps of current, but all of that would be going into outgassing or heat. Leave it like like for a long time and the smaller battery will lose electrolyte. What have I missed?

I’m not Rod but think I can answer that: this is why it is recommended to use the same brand and model batteries from the same batch for being connected in parallel in a battery bank.
So yes, absorption times can be too long for a smaller battery in a parallel bank, causing unwanted outgassing.

[sailjumanji]

Originally Posted by Maine Sail

I have not been impressed with the Mastervolt AGM batteries we've tested for customers. We've capacity tested about a dozen of them, all less than two years old, and they just don't seem to hold up like a known quality brand such as Odyssey, Lifeline, Northstar, Fullriver, Rolls or Firefly. Deka GEL batteries are exceptional but their AGM's don't hold up nearly as well.

There are quality differences between AGM batteries. Last I knew the Mastervolt AGM batteries are made in China, and this may not be a bad thing, as Fullriver makes very good batteries, but not all Chinese suppliers do. I don't base my assessment of the Mastervolt AGM's on country of origin but rather the actual testing of used batteries..

Yep, I am not sure where they are made either. It's funny that you mention Deka AGM's. We had three Deka Group 31 AGM's that held up for 10 years, recharge with three-stage solar, running refrigeration. Then one of them crapped out. Different boat, but in the Texas heat.

Thanks for your post.

[sailjumanji]

Originally Posted by Jammer

The marine battery chargers used on yachts do not ordinarily use separate current sensors for the battery current and the load current; where a battery shunt exists it typically is used only for metering and does not inform the charge algorithm. Therefore, the ampere load that the battery charger "sees" is the sum of the battery current and the load current.

This complicates determination of the proper point at which the charge cycle should end, that is, the point where the charger transitions from absorption to float, since the end point of the charge cycle is typically determined by the charge current falling below a particular value. When there is a load present, the charge current seen by the charger may never fall below the chosen value or will do so later than desired, because of the effect of the load. A workaround for this is to limit the duration of the absorption cycle.

That's what you're seeing, and it's unremarkable.

If you're concerned that the batteries are being overcharged, my recommendation would be to reduce the absorption voltage, the time limit, or both. Batteries will continue to charge on float, and the use of a higher voltage/longer duration absorption state is just to expedite the completion of charging -- which accomplishes nothing useful if you're at the dock overnight since they'll be charged by morning anyway.

That all makes sense. And I suspect the reason that the MassCombi was on float this am was because it timed out on absorption phase sometime last night. Both the MassCombi and Solar Chargemaster have absorption time limits. (The spec is 4 hours, but can be adjustable, with some other equipment.)

[Bycrick]

Jedi — That’s why I’ve never liked diode splitters (there’s other reasons too, of course) or ACRs for charging asymmetric sets of batteries. Even voltage followers (eg Duocharge) if you leave them on all the time will eventually cook the target battery. It will be interesting to actually look at the Orion someday and see if it really shuts off when the target battery is "full" or whether it takes some external control. Thanks.

[sailjumanji]

Originally Posted by Bycrick

Jedi — That’s why I’ve never liked diode splitters (there’s other reasons too, of course) or ACRs for charging asymmetric sets of batteries. Even voltage followers (eg Duocharge) if you leave them on all the time will eventually cook the target battery. It will be interesting to actually look at the Orion someday and see if it really shuts off when the target battery is "full" or whether it takes some external control. Thanks.

Could you clarify what you are saying here? We have a VSR that commingles the house bank with the start battery, whenever voltage sensed is above 13.2. And then it cuts out at 12.8 (or 12.9, I dont remember exactly) volts. It does work as per spec. So if we have two 225 AH house batteries as house bank, and the 130 AH as start, which would be "cooked?" Which is the "target" battery?

Note that we also had a VSR on the previous cat, and similar house bank and start battery. But our house bank there was two 105 AGM's, and the start was a single 105 AGM. THose batteries lasted over ten years.

[Bycrick]

In the specific case of the "cooked battery," the fellow put a windlass battery forward, charged by a Duocharge. He installed a manual switch on the Duocharge, but never turned it off. The battery ended up dry, swollen and dead. Since he didn’t use the windlass for many moons, the battery was eventually fully charged send whatever small current it drew was enough to eventually "boil out" the electrolyte.

While Maine Sail and a lot of other people use an VSR/ACR in their system, but I don’t like them. I don’t like caviar either. While it’s true that batteries connected in parallel will only accept the charge current that they’ll accept, that only means that you can’t "over-charge" them with "too much current."

My position is that when one parallels different sized batteries, the smaller battery will end up being fully charged first. At that point, while it won’t accept much current, but it will still accept some current. Since it’s eventually "fully charged" whatever current it accepts will either "boil" the water out or be lost as hear.

This might not have been a problem when the primary charge system was the main engine which ran intermittently. But now, simply paralleling a relatively small starter battery with a much bigger house bank, while providing lots of solar power for many hours, which will hopefully keep your 1000Ah house bank above 13.2 volts, is going to be charging the smaller starter battery for way too long, albeit at low current.

That appears wrong to me. I’ll admit that I’m apparently in the minority as lots of people swear by these things. I’m also willing to be convinced that my thesis is incorrect. But to me, there’s a difference between "Well, it worked, didnt it?"" and "best practice."

[rslifkin]

ACRs have their uses, but they do have to be used carefully. Otherwise they can end up holding a fully charged battery at absorption voltage way too long while the other bank is still charging. It may not be a big issue in a lot of cases, but it's not ideal by any means.

Personally, I solved that problem on my boat by using ACRs with an enable/disable wire. When on shore power or generator, all 3 banks charge independently. The ACRs are interlocked to the engine ignitions, so they're only used to parallel each engine's start bank to the house bank (to allow house bank charging from the engine alternators) while the respective engine is running. As soon as the engines are turned off, the start batteries are isolated from the house batteries.

[er9]

Originally Posted by rslifkin

ACRs have their uses, but they do have to be used carefully. Otherwise they can end up holding a fully charged battery at absorption voltage way too long while the other bank is still charging. It may not be a big issue in a lot of cases, but it's not ideal by any means.

Personally, I solved that problem on my boat by using ACRs with an enable/disable wire. When on shore power or generator, all 3 banks charge independently. The ACRs are interlocked to the engine ignitions, so they're only used to parallel each engine's start bank to the house bank (to allow house bank charging from the engine alternators) while the respective engine is running. As soon as the engines are turned off, the start batteries are isolated from the house batteries.

isn't that why its recommended to send all charge sources to the large house bank first or the largest bank? the smaller start bank on the other end of the acr should be up to disconnect voltage, fully charged and disconnected before the house bank comes up to full charge. at least thats how i understand it. i could understand if the charge source went to the smaller bank first your scenario happening but maybe im misunderstanding?

[Bycrick]

Rslifkin — I agree: they’re hardly a panacea. They need some intelligent external control, not beyond that the voltage someplace is more than something. I like your idea of interlocking them to the engine, but I’d have the alternators (and all the other charge sources, connected to the house bank and only use the ACRs to charge the starter bank. Different horses for different courses.