AGM Battery Failure - Two Sets in Two Years

[sailjumanji]

Am looking for input on what else to check and possible causes. Have a 2019 Seawind 1160 Lite, with two 225 AGM Mastervolt 12-volt battery house bank, and a 130 AGM Mastervolt start battery. Recharge is via shore power (Mastervolt MassCombi charger) or solar (Mastervolt SolarCharger 60 MPPT tied to 800 watts solar panels). There are also two 25 hp Yamaha outboards, each with rectifier/regulators on the alternators.

We took delivery in February 2019, spent a month in a marina doing commissioning, then three months in Bahamas - mostly at anchor with occasional marina stay. Everything worked well until about end of May we started having issues with low battery voltage in the am, and by the time we reach Miami in mid June, the batteries couldn't maintain refrigeration overnight. The boat is all LED, we don't use the inverter, and at max refrig draw we are at around 10 amps (less when both compressors are not running). Overnight usage is 60-90 amp-hours, out of 450 amp-hr house bank.

There is also a VSR that commingles the batteries when in charge condition, but then drops out around 12.9 volts. So the start battery covers some of the house load until mid evening, when voltage hits 12.9, and it cuts out.

So first set of AGM's lasted from Feb to mid June. We were anxious to get home to Texas before hurricane system, and chose just to change them out with identical batteries and get on our way. Second set of two Mastervolt 225 AGM's go in. We spent about a month in Florida, most of it at marina, getting crew swapped out and provisioning and prepping for Gulf jump. We had a rough crossing, and the batteries showed some weakness on the fourth day, but we attributed it to overcast and a hard-working autopilot, and radar at night. We had a Honda generator that we used to charge them up one morning, and no other issues. Made the crossing from Ft Myers, Florida to Freeport, Texas, had a couple of days motoring down the ICW, and the boat was parked in July 2019. Since then, due to extended family issues, we've only done a few day sails. Batteries have been on MassCombi charger - on float - when on dock.

A couple of months ago, we decided to get away for a week. Motored ICW about 8 hours, then anchored. Batteries fully charged. Overnight, around 4 am, the refrigerator started kicking off due to low voltage. The freezer was turned off, so we were using about 5 amps - so less than 50 amp-hrs overnight. Recharged the next day, and the following night did the same thing. It all looked like an exact repeat of a year ago, with AGM failure.

Since then, have run capacity check on all three batteries. One 225 AGM had 30 amp-hrs capacity, the other 20 amp-hrs remaining. The start battery had about 100 amp-hrs remaining, out of 130 amp hour initial rating. (BTW, the start battery is still original.) Mastervolt has replaced them all under warranty, and we opted to upsize the house bank to two 270 AGM's. I have not installed them as I want to get the system fixed first, as its unlikely our issues are purely battery.

We've also tested the output of the MassCombi charger, both with and without temperature calibration, and its all within Mastervolt spec. The Solar Chargemaster appears to be about 0.10 volts high on absorption, and 0.10 volts low on float.

Haven't found any other issues except we have one post on the negative buss that was corroded to the point of removing the plating and exposing the copper beneath. The three cables attached to it were tight, but also their terminal ends had corrosion. One of the wires was from the solar controller, the other might have been from the engine alternator. The attachment was tight. Everything else was pristine, as it is inside of a breaker box, under a set in the saloon.

Motoring, as our first thought is we are cooking the batteries with the outboards. The max voltage we've seen during motoring with full batteries is 14.1-14.2 volts, and 1-2 amps. So it appears the rectifiers are working. The first batteries had 185 hours of motoring before failure. The second set, 60 hours.

Seawind is working closely with Mastervolt on this, reviewing systems design, installation, etc. Covid has prevented factory visit for in-person review, but there has been lots of meetings. One thing they plan to do is removed the VSR so that the house batteries cannot experience the outboard motor charge current. There is also additional monitoring planned to go in. Right now we have the BEP DCSM and shunts on both the house and start batteries. So we can see amps, calculates amp-hours, etc. The shunts have both been calibrated.

Have also measured all of the loads, and all are reasonable. The background load with everything off is about 0.6 amps - not excessive. We've also done drawdown and recharge checks, and the solar panels have put out a little over 700 watts on occasion, so all four are working. Chargers all go thru bulk, absorption and float, as per the specs (except for slight voltage difference as mentioned before.) House batteries have 2/0 gauge wire, and voltage drop is not excessive. When we were in Bahamas, anchored out, we would usually be on float by early afternoon.

Finally, as far as we've been able to identify, this has been an issue with only one other Lite. (He's not been able to find the issue, and just chose to change his batteries out annually!)

Thoughts? Anything else to check? Oh, and I know LiPo or whatever is a better battery, weighs less, etc. But I suspect whatever system issue I am having wouldn't care, and would kill that more expensive battery just the same!

[sailorboy1]

You spin a nice story, but it is low on tech details of volts, amps, power in/out of the system really. To me it all sounds that your batteries are not charging fully and they died from this. But you don't provide voltage and acceptance amps to know and ot would need to be on good batteries.

[sailjumanji]

Originally Posted by sailorboy1

You spin a nice story, but it is low on tech details of volts, amps, power in/out of the system really. To me it all sounds that your batteries are not charging fully and they died from this. But you don't provide voltage and acceptance amps to know and ot would need to be on good batteries.

"Spin?" Really, there is no spin? I'm not trying to prove anything to anyone, to get something or whatever. Actually, I was just trying to present facts. Mastervolt has already replaced all of the batteries under warranty. I'm just trying to figure out how to prevent a third set of batteries from failing. I've tested the MassCombi, the Solar Chargemaster, and both roughly within spec. Just wondering where else to look. All equipment has normal amp draw.

Mastervolt says heat, and excessive discharge are typical failure issues. With 60-80 amps usage overnight, and sunny Bahama skies and 800 watts of solar, we were usually on float by around 1-2 pm. (And charger would go thru the three stages, and go into float because of low acceptance amps, not because it was timed out.) When we did check into marina and plug in, the MassCombi would quickly cycle thru bulk, absorption, and then into float - which also confirmed that we were already full. When motoring, the batteries would get full and then hang at 14.1-14.2 volts and 1-2 amps going in, which is probably overcharged.

Again, the first set of batteries was on MassCombi shore power charger for a month, and then Solar Chargemaster with occasional MassCombi for three months, before they crapped out. Even if they were undercharged by the solar, would you expect this level of failure - not able to maintain 8-10 amp load overnight - in three months? From 450 amp-hour bank.

The second set of batteries was on solar charge for less than a week. Then marina stay, a four-day Gulf crossing, motoring the ICW for two days, and then on MassCombi for about a year. Hard to envision how they could be not charged fully during that time.

So considering that, what other factors could cause premature failure of an AGM?

[Captain Bill]

I fried a set with my solar charger. The default agm setting was misprogrammed at the factory and severely overcharged the batteries. My controller automatically goes into bulk/absorption anytime it goes through a dark period. At the dock even when the batteries are fully charged and on float the unit would go to absorption mode for 3 hours. This is chronic overcharging which eventually kills the battery. I made a custom program for when I'm at the dock that sets the absorption voltage only .01V above float (it won't accept them being the same) so that I am not overcharging with the solar panels while the batteries are being maintained by the shore power charger. I don't know if your solar controller is doing the same thing mine does, but it could be the source of your problem.

[DeValency]

LFP batteries will not help you and their loss will just be more expensive.
As it looks, something (a partial shortcut, moisture, electrical device, wiring issue) is sucking your juice. Once AGM battery discharges below 60% or so, you notify the low voltage but that’s almost too late. Few discharges like that and your batteries are permanently damaged.
You need to continuously measure the current out of the battery in different loads and situations. Make sure nothing else is connected directly to the battery. If you’re lucky you’ll find the leak right away. If not, you better install a must-have battery monitoring device (such as Victron BMV) that should tell you what and when the discharge is going on and then you’ll have to connect and disconnect circuits one by one until you identify the bandit...

Best of luck

[Jammer]

I would suggest measuring absorption voltage at each battery several times during charge cycles on different days, also measuring the temperature of each battery with an IR thermometer (on the middle of the side of the case). It is possible/likely that the voltage is too low at the battery even though it is correct when measured elsewhere in the system, possibly due to the corroded connectors you describe or other unwanted high-resistance connections.

[Richard Harper]

I installed a set of gell batteries and the 9.9 yamaha outboards will take the voltage well above the recommended maximum charge voltage of 14.1v when the batteries are fully charged. I added an overvoltage alarm and turned on a bunch of lights when the voltage gets too high. I how have a load box that turns on when the voltage goes above 14.1v.

[sailjumanji]

Here is some additional information, that addresses some of the comments. THANK YOU SINCERELY FOR YOUR INPUT. Do any of these issues look like they could be failure cause?

We've run recharge tests using the Solar ChargeMaster and the MassCombi on the batteries in various stages of depletion. Have tested each both with, and without the temperature probes connected. In addition, the MassCombi has the ability to do a test run and cycle thru the settings to confirm voltages. Note that Mastervolt charge voltages - BEFORE TEMPERATURE COMPENSATION - are:

Bulk 14.40 volts
Absorption 14.25 volts
Float 13.80 volts

Here are specifics summarized for each (below). If you could provide your thoughts on whether any of these voltage shortfalls, etc., could have damaged the house batteries so quickly. Please. If it’s not the MassCombi or Solar Chargemaster, then we need to look elsewhere. Remember that first set of batteries failed after almost exclusive life on the Solar Chargemaster. Second set failed while being maintained by the MassCombi.

Solar Chargemaster - Bulk
1. Both the temp probe disabled, and the temp-corrected bulk stage voltage (albeit limited data) are at Mastervolt specifications for AGM battery.

Solar Chargemaster - Absorption
2. With temp probe disabled, the absorption stage voltage is 0.10 volts higher than the 14.25 volt spec
3. With temp probe installed, the voltage was 0.05-0.14 volts higher than the temperature-corrected specification.
Median was 0.10 volt higher than temp-corrected spec, established by longest absorption stage on solar-specific test of depleted start battery.

Solar Chargemaster - Float
4. With temp probe disabled, the float stage voltage is 0.10-0.12 volts lower than the 13.80 volt spec
5. With temp probe installed, the voltage was 0.14-0.19 volts lower than the temperature-corrected specification.
Longest float period (10.5 hours) with temperature-controlled (air conditioned) conditions had charge voltage 0.19 volts lower than spec.
In four tests, the temp correction of 0.32 to 0.40 volts exceeds the Mastervolt correction limit of 0.30 volts (in part because of the base voltage shortfall)

MassCombi - Temperature Probe Disabled
6. With temp probe disabled, the bulk, absorption and float voltages are all within 0.01 volts of Mastervolt specifications.

MassCombi - Bulk w/Temperature Correction (Probe Connected)
7. Three of the tests were in bulk stage for 3 minutes or less. Maximum voltage was 0.05-0.06 volts lower than temp-corrected specification.
8. Two tests with the longest bulk stage period - 8 minutes and 1 hour, respectively - exhibited max voltage that was 0.17 volts lower than the temp-corrected spec.

MassCombi - Absorption w/Temp Correction
9. All of the absorption stage measurements have voltage lower than Mastervolt temperature-corrected specification.
Voltage shortfalls range from 0.05-0.14 volts, with an average of 0.09 volts

MassCombi - Float w/Temp Correction
10. All of the float stage measurements have voltage lower than Mastervolt temperature-corrected specification.
Voltage shortfalls range from 0.05-0.15 volts, with an average of 0.09 volts
In two tests, the temp correction of 0.31 and 0.35 volts exceeds the Mastervolt correction limit of 0.30 volts (in part because of the base voltage shortfall)

To sum it up, the Solar Chargemaster is off spec, as even with the temperature probe disconnected, the absorption and float voltages are 0.10 volts high and 0.10 volts low, respectively. The MassCombi has issue with temperature correction, resulting in bulk, absorption and float voltages all too low vs Mastervolt specification.

[rslifkin]

Keep in mind, spec-ed voltage is at 77*. With temp compensation, if it's hotter, the voltage will be lower. That's normal (and desired).

[sailjumanji]

Originally Posted by Richard Harper

I installed a set of gell batteries and the 9.9 yamaha outboards will take the voltage well above the recommended maximum charge voltage of 14.1v when the batteries are fully charged. I added an overvoltage alarm and turned on a bunch of lights when the voltage gets too high. I how have a load box that turns on when the voltage goes above 14.1v.

Hi Richard,

Our previous boat was the Seawind 1000. We have three Deka group 31 AGMs, that lasted 12 years before one failed and we replaced all three. Same concern with overcharge from the outboards. However that boat wasnt as efficient on lighting, fans, etc., as the Lite. We would turn on fans, lights, stereo to keep the overcharge down, but sometimes it would still run up to 14.4 volts.

It seems that the direction that Seawind is taking on changes to our current system, is to isolate the outboard charge to the start battery alone. That protects the house load, but OTOH will probably trash the start battery. You comment gives me a thought that I could easy simulate a load box, by merely turning on the inverter with a trouble lamp plugged in. That said, the inverter is hooked to the house battery side.

[Pete7]

I am just looking at the second and third photos of the bus bar. Looks like one post has severe corrosion. Is that copper in the third photo sitting on top of the silver metal? If so that has taken some power. What was that connection?

Also do all the existing batteries now have a similar capacity, or is one substantially lower than the rest? They will need to be tested individually.

Pete

[DeValency]

Better focus on the power consumption not the charging. Charging is rarely an issue and it can be tracked right away. Looks like you are ok on that side anyway.

The whole short battery life story suggests it is a consumption issue that you need to identify. On most boats there are 2-4 charging sources but 10-100 power consumers- each one can drain your batteries without even noticing that.

[sailorboy1]

I still dont see info that verifies that the batteries ever get fully charged. All i see are voltages and that the system went through bulk/absoption and float. That means nothing far as fully charged.

In one post the the OP says he is fully charged by 12-1. How many times does it need to be posted how unlikely that is.

The question being asked is:

Why do my batteries only last 1 year?

Well the answer is:

Because they aren't getting properly charged.

But i got dismissed and am going to stay out of it.

[sailjumanji]

Originally Posted by Jammer

I would suggest measuring absorption voltage at each battery several times during charge cycles on different days, also measuring the temperature of each battery with an IR thermometer (on the middle of the side of the case). It is possible/likely that the voltage is too low at the battery even though it is correct when measured elsewhere in the system, possibly due to the corroded connectors you describe or other unwanted high-resistance connections.

Good suggestion. We have done this, except for the IR thermometer. When we run charge tests, I have a clamp meter on the battery (amps), plus a meter on the battery. Then I moved the clamp meter back and forth to get charger voltage. Initially we did this to determine if we had any excessive voltage drop.

We took measurements every 2-5 minutes, trying to track exactly what is happening, from the initial charger turn on, to a few minutes after it went into float. The maximum voltage drop observed - from the charger to the battery - was:

Start battery, 0.29 volts (2.1%) at 98 amps peak. But down to 0.24 volts (1.8%) at 80 amps. The start battery has 2 AWG wire.

House battery, tested individually, not the two as a house bank. 0.18 volts (1.3%) at 61 amps for highest of the two batteries. The house bank has 2/0 gauge wire.

The house battery tests were done before before the corrosion at the buss terminal was noted.

Can you tell me more about your comment on the battery temperature? We've run tests with both the temperature probe connected, and disconnected. That said, it is freaking hot in Texas in August, and much of the testing was in 84-89 deg F. (Luckily we live on the water, and that keeps temps moderated.) As posted separately, I dont think temp compensation is working exactly as designed, but its hard to know when we were just measuring air temp, and not battery temp. The charge voltages were being compensated (lowered) vs the 25 deg C standard. Are you concerned with really high battery temps? Or just charger voltage not being adjusted for temp?

[sailjumanji]

Originally Posted by rslifkin

Keep in mind, spec-ed voltage is at 77*. With temp compensation, if it's hotter, the voltage will be lower. That's normal (and desired).

Yep, understood. That is why we ran tests with both the temp probe connected and disconnected. Wanted to make sure the unit was putting out voltage at spec without the temp correction issue. And then see what the temp correction was that it applied. The data shown previous has results from both tests.