question

alexander-arnold avatar image
alexander-arnold asked

Victron Super Cycle AGM sizzle and smoke

Hello Community,

Our domestic battery bank almost blew up on a 17 hour motor sail.
We are really happy that the Victron battery chassis are very very strong.

This is the setup, domestic battery, solar, and MPPT installed in Q3 2020:

  • 2x170Ah Victron AGM Super Cycle domestic bank
  • Victron 100/50 MPPT Charge Controller
  • Victron BMV 712 Battery Monitor
  • Victron 12/2000 230V Inverter Q3 2021
  • 400W LG Bifacial Solar
  • Volvo Penta MD2030 diesel with alternator
  • Bosch 90Ah starter battery 2019
  • Victron Argofet Battery Isolator
  • 70mm2 marine grade wire on batteries to bus bars and to inverter

Turn of events (It may be TMI but I will give location too, to make sense of distances and daytimes):

  • We left the boat on a mooring ball for 3 months in Grenada, the MPPT has been topping up very consistently just enough to keep the batteries full
  • We came back and used the boat on anchor and did short 1-3 hour sails under engine around the bays, all was well, batteries nice and full
  • Then we did a big jump under engine that blew up the batteries:
    • Monday morning we leave Tobago Cays to Bequia, a 5 hour motor trip
    • Monday noon we stay on anchor in Bequia, it is sunny
    • Monday evening we start our next leg after sunset, through the night we motor and get 10 hours of alternator power
    • Tuesday all day, we keep motoring and arrive in Martinique in the afternoon around 2:30 PM
  • After dropping anchor and stopping the engine we hear sizzling noise and smoke from the battery compartment. The batteries are smoking hot and bulging. We switch off ALL power and disconnect the batteries, take them out into the dinghy where they leak and melt the dinghy non-skid due to heat. We dispose them later that day at the marina.
  • We noticed no high voltage. After disconnecting, the batteries were at 13.5V
  • The starter battery is at 12.9V and has no damage
  • The MPPT is set to the Victron AGM deep cycle preset
  • The MPPT has bulked 100% and put 1.7kWh into the bank during Tuesday!

I am suspecting the following:

  • The alternator kept all batteries topped up, starter and domestic, but somehow the MPPT kept pushing all day as well, into domestic only. This is my only explanation to as why the starter is fine but the domestic overcharged.

Why we didn’t notice before a long full day motor trip:

  • On anchor there is no alternator and we use solar only to top up batteries. It is a well balanced system and we get and use between 1.5kWh and 2kWh per day
  • The alternator provides around 25 to 30 Amps when it detects a medium full battery, it dials down to 5 to 0 Amps when full, it is working as expected
  • The only time where both systems were in full use was on this full day long motor trip. I believe there is something wrong in the detection of needed current from the MPPT, misinterpreting as to what the battery voltage really is, when the boat is being used in conjunction with the alternator charge current.


Now, we installed Mastervolt 2x160Ah AGM, the only available option for us on Martinique. I am monitoring charge cycles on anchor as well, screenshots provided. You can see time of day at the top.

mastervolt charging screenshots after fresh install.zip


I attach all data I could gather in this post. “Yesterday” in the MPPT screenshots is the day of passage. “Today” is on the new Mastervolts. Even on the days before the incident, the MPPT should not have charged that much in my opinion, as we motored as well. I am confused this happened because the MPPT setting was set on the exact battery preset and the batteries are 2 years old, being actively used for about 1 year. We rarely go below 12.3. I am very happy that the Victron battery chassis held up strong and didn’t catch fire. But there's quite another mix of emotions when I think about how this could happen with an all-Victron charging and storage setup.

img-5530.png

On our next daytime motor passages we will monitor the MPPT charging patterns closely and when in doubt switch off the MPPT to prevent overcharging.

SmartSolar_ChargerMPPT_100_50 Victron AGM.vcsf.zip

SmartSolar_ChargerMPPT_100_50 new Settings mastervolt .vcsf.zip

Battery_monitorBMV-712_Smart mastervolt.vcsf.zip

Battery_monitorBMV-712_Smart.vcsf.zip

img-5461.jpg



img-5456.jpg



img-5463.jpg

Your thoughts on my hypothesis? And more importantly, how can I prevent this in the future?

MPPT ControllersAGM Battery
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

7 Answers
pwfarnell avatar image
pwfarnell answered ·

I have had a read through but can not read zip files on this device. The key data missing here is the voltage generated by the alternator when the batteries are full. Because what you seem to have had is absolutely fully charged batteries that you them subjected to a 5 hour and then a 10 hour charge at full alternator voltage which represents a lot of over charging of AGM batteries which is not good unless you have an external controller on the alternator to set it to float when full.

10 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

lawrence-craig avatar image lawrence-craig commented ·

Have a look at the Vmax. It’s only 14.07V vs the prior days of 14.2V. This suggests something a miss in terms of either the voltage being delivered by alt or the voltage being seen by the MPPT. It also suggests that the voltage wasn’t sufficient to reach ABS hence why it was in Bulk for the whole day.


The other thing to consider is that chargers don’t “push” charge, they only let charge flow if the battery is able to receive it, particularly the amount of charge you show in your screenshots. Therefore this suggest that there was something wrong with the battery on that day, not that it was the MPPT.

2 Likes 2 ·
pwfarnell avatar image pwfarnell lawrence-craig commented ·
Good observation about the lower voltage. A short in one of the cells could result in current continually flowing and not reaching the absorption voltage and overheating a battery. I could not understand both batteries simultaneously suffering a shorted cell, hence the damage to the other battery would be from overheating by its neighbour. With the batteries in parallel, the good one could still hold the voltage up at the end of charge before disconnection.


This may explain why the starter battery did not have an issue if it is not mounted next to the house batteries.


Do you have any temperature monitoring on the batteries.

0 Likes 0 ·
alexander-arnold avatar image alexander-arnold pwfarnell commented ·
Thanks for your thoughts! No, unfortunately no temperature data. Only thing I can tell you that one looked worse than the other, but both were too hot to touch when we tossed them
0 Likes 0 ·
alexander-arnold avatar image alexander-arnold commented ·

Thanks for the quick reply! We checked that and i wrote that it goes to 5 and then 0 when full, just didnt supply screenshots. The alternator is regulated and doesnt push in amps when the battery is full. It stops charging earlier than the MPPT does. An electric engineer checked our alternator after the incident.

0 Likes 0 ·
alexander-arnold avatar image alexander-arnold alexander-arnold commented ·
nonetheless, the MPPT shouldnt have put in 1.7kwh in pure bulk that day - especially not if the alternator is providing and voltage is high enough to switch to absorption
0 Likes 0 ·
pwfarnell avatar image pwfarnell alexander-arnold commented ·
Yes you listed the Amps, I am still asking about the voltage, you had a 10 hour motor run after sunset, the MPPT would not have been generating then, so the alternator would have been the only voltage source. Whilst the amps may be low, maintaining the battery at high voltage is not good.

Agree the long term bulk from the MPPT is a concern.

1 Like 1 ·
alexander-arnold avatar image alexander-arnold pwfarnell commented ·
makes sense. Will do some tests later today and post the data
0 Likes 0 ·
over-and-under-victron avatar image over-and-under-victron pwfarnell commented ·
Wouldn't the mppt history capture the highest system voltage regardless the source?
0 Likes 0 ·
kevgermany avatar image kevgermany ♦♦ over-and-under-victron commented ·
It will record battery voltage history, if something else is charging it will record the highest voltage, but not identify the creator of the voltage.
0 Likes 0 ·
alexander-arnold avatar image alexander-arnold over-and-under-victron commented ·
correct. so technically it's in the screenshots. minus the voltage drops from main switch and bus bars and length of wire. But it's not 20V, not 16, it's 14.x. The issue stays a mystery to me.
0 Likes 0 ·
alexander-arnold avatar image
alexander-arnold answered ·

Adding MPPT csv export:

SolarHistory.csv.zip


2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Alexandra avatar image
Alexandra answered ·

I have seen AGMS do that when they are overcharged.

AGM batteries dont like high charge currents. So really what were they receiving in amps during the charges? Remember they are c10/20 rated. So in 12v system with 2 thats a 30A charge at C10.

They also dry out when in hotter conditions and being overcharged.

(Or they froze?)

8 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

christern avatar image christern commented ·
But this was AGM, not GEL.
0 Likes 0 ·
Alexandra avatar image Alexandra ♦ christern commented ·
If AGMs are charged too fast they cant recombine the gas, the build up from pressure makes a balloon.


The other thing is a plate short that makes a huge gas/pressure build. But they get warm when doing it.


Gels bubble when charged too fast then it is finished since the bubbles dont move. And there is no contact between electrolyte and plate.

0 Likes 0 ·
alexander-arnold avatar image alexander-arnold Alexandra ♦ commented ·

Hey Alexandra,

the AGM Super Cycle battery says it handles a 40.5A max charging current. With one 400W solar and a standard alternator, my guess is that I can never surpass this with two batteries in parallel. The alternator provides around 25 amps, the solar averages between 15-25, but only for a few peak hours, around noon. But please let me know if I'm mistaken and a 340Ah AGM bank can't deal with 40-50 amps input when empty - and then when MPPT and alternator regulators kick in of course a reduced charge (which didn't happen from MPPT side on that day, for mysterious reasons)

I am not knowledgable enough to understand the C10/20 rating but my take away is that 2x170Ah AGM can only handle a 30A charge?

img-5459sm.jpg

Victron data sheets tell me that AGMs are perfectly suited for high load output and I expected that it can also deal with at least my kind of input (to be honest I don't consider that high, again please correct me).

We are talking about BAT412117081

Datasheet-AGM-Super-Cycle-battery-EN.pdf


edit:

(or they froze?)

no, we are in the Caribbean

0 Likes 0 ·
img-5459sm.jpg (523.8 KiB)
christern avatar image christern Alexandra ♦ commented ·
But first you wrote GEL, hence my comment.

Also AGMs easier takes charging than open lead batteries and, in this case, the charging was most likely under specified limits.

Getting back to the actual problem my guess is either a short within one of the batteries that overheated and brought the other with it, or an overvoltage during several hours.
I have seen AGMs exposed to 17-18 V during 5-6 hours without taking this kind of damage so the voltage must have been a lot higher. My suspect would be the regulator/alternator on the engine so keep a good watch on the house battery voltage during the coming days.

0 Likes 0 ·
alexander-arnold avatar image alexander-arnold christern commented ·

Hey Chris, my first guess was alternator regulator as well, but the starter is at 12.9 and did not take damage. This rules out a regulator fault IMHO.

After reading all your answers my conclusion is a short in one of the cells, that took the rest of the bank down. Then alternator AND mppt didn't see the voltage to regulate down to absorption and thus kept pumping bulk. Does this make sense?

The new Mastervolt bank is at 13.2 at the moment at 3pm local time. I didn't see any spikes higher than 14.2 over the past days. Currently on anchor thus MPPT only.

edit: the screenshot from MPPT on incident day shows a max of 14.07V. No high voltage detected

-1 Like -1 ·
pwfarnell avatar image pwfarnell alexander-arnold commented ·
Makes sense to me as a working hypothesis, per my comments earlier in the day.
2 Likes 2 ·
Alexandra avatar image Alexandra ♦ pwfarnell commented ·
On a DC toDC charger though is there not bulk protection to prevent overcharging due to being in bulk for too long? So maybe add one next time to prevent that.

There may not have been fault per-say as in over voltage from the alternator, just charging too long in bulk and then back to the drying our gas pressure issue.

Excessive Bulk prevention as a feature is important.

0 Likes 0 ·
alexander-arnold avatar image alexander-arnold Alexandra ♦ commented ·
  • 1.7kwh solar in bulk is normal when the battery is empty. It only becomes a problem when the battery is already full from another source (the alternator charge through the night), but then dies, and drops voltage, making the MPPT want to charge. What kind of bulk protection feature could notice the issue in this scenario? What exact box should I install?


edit: the only thing that would have helped IMHO is a temp sensor. Anything that gets too hot no matter the cause would be detected and could've been disconnected. My final conclusion is that a battery cell shorted and made every charger charge as if it was empty.

edit2: misunderstood what you said about the alternator, deleted my response in regards

0 Likes 0 ·
angusbeau avatar image
angusbeau answered ·

Interestingly i have had a similar problem but not quite as extreme as you but the circumstances are remarkably similar. We have a bank of 4 of those batteries. We left the boat for 3 weeks in Colombia and didn't disconnect the solar. When we got back to the boat I noticed on the app that the batteries had been hitting 50C after going into float. We then had a similar trip to you where they were being charged by the alternator for an extended period with the MPPT's also putting in charge and they got over 50C. luckily we monitor the temperature so we're able to cut off the charging. i have removed what I think is a problem battery from the bank and they seem better.

I have seen other comments online to a similar effect about these batteries and wonder if there is something going on with these in hot tropical countries.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

rostyvyg avatar image
rostyvyg answered ·

I know it is the old post but I have had experienced this exact issue TWISE already. I have battery temperature sensors used by the Victron charging system (MultiPlus 3000 and 2x400W solar panels with Victron MPTT controllers, BMV-712, Cerbo GX ) and a temperature sensor connected to my Balmar alternator regulator and this helped to prevent disaster in both cases. I had 4x170Ah Victron SuperCycle house bank batteries installed in December of 2022 and in September 2023 one of them started overheationg when charged either by alternator or by MultiPlus. After I disconnected it the remaining 3 batteries would charge normally but their capacity was severely degraded. I submitted a warrany claim and Victron replaced all 4 batteries. Now in November 2023 one of the new batteries started overheating again! A new warranty claim is submitted to Victron and they still have not processed the claim by January 2024. All dealer calls are answered with "please wait a few more days, we are inundated with battery warranty claims". So while all other lines of Victron products seem to be well designed, quality built and user friendly it appears that their AGM batteries leave much to be desired in the quality control realm. Which really poses a fire safety issue. They really should have those batteries recalled. I would not be surprised if some shrewed layer filed a class action suite.

5 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC avatar image JohnC ♦ commented ·
@rostyvyg

That you can reproduce this failure means you probably have 4x batteries in parallel. Likely pushing them hard too. Victron branded or no, charge imbalance will keep killing your batts.

One warranty claim acceptance is for goodwill, but no supplier can keep doing that for a poor installation.

Sorry to be blunt, but this is far too common. Easier to blame the batts than the real cause.

2 Likes 2 ·
rostyvyg avatar image rostyvyg JohnC ♦ commented ·
OK, please elaborate on "pushing too hard" and "charge imbalance". My absorption voltage is set to 14.6V, float to 13.6V. The average discharge current is around 10A. I never discharge my house bank to less than 80% (most days it is 85%). I make sure they are charged 100% every day. The installer used equal-length negative and positive leads from the bus bars to each battery (not from the battery terminal to terminal as many do). Anything, in your opinion I could do differently to avoid battery overheating while charging? And, BTW just today yet another battery started overheating while charging. So now I am effectively down to 2 batteries in parallel. Never saw this with any other AGM battery brand I used earlier on my boat, BTW, and each previous set served at least 4 to 5 years while living aboard full-time.
0 Likes 0 ·
JohnC avatar image JohnC ♦ rostyvyg commented ·

@rostyvyg

Good Pb batteries have low resistance. Like it's down at the level of the wiring itself, so it's good you at least have equal length wires. There are always manufacturing differences, and they can change with age too. Li's are even lower, but they come with a bms to manage it, and can handle fast charging without heating anyway.

Your 14.6V Absorb may well be a recommended value, but it's too high for paralleled batts. Reduce it to as low as you can tolerate, yes even below 14.0V.

The big problem with heating is that it reduces resistance even further, so results in a 'thermal runaway'. It also means you can't actually use average charge/discharge values for calcs, as you don't know if one batt is taking near all on it's own.

You must use Temp Compensation too. It's important.

This guy's calcs might open your eyes.. http://www.smartgauge.co.uk/batt_con.html

And of course none of all this would be relevant if your 4x batts were wired as 48V. Imbalance drove me there long ago, weren't Victron batts either.

0 Likes 0 ·
rostyvyg avatar image rostyvyg JohnC ♦ commented ·

Well, my installation uses "Method 3" from the article you mentioned since it was done to possibly switch to lithium batteries in the future. The charger is Victrom MultiPlus 3000 with a temperature sensor although CerboGX is set up to take battery temperature readings from the BMV-712 temperature sensor. Yesterday I reduced the Absorption voltage to 14.5 and limited absorption time to 3 hours as a safeguard against runaway overheating. mp.jpg

When charging all batteries consume roughly the same current (I used a clamp-on multimeter on each battery lead) until the charging current drops to about 12 Amp per battery, then it continues to drop on all but one battery, and the offending battery starts to consume progressively more current and overheating. It appears that MultiPlus gets fooled by this and stays in Absorption until I switch off the charger manually or until the temperature limit on the offending battery is reached (50C) and it shuts down.

Any advice?


0 Likes 0 ·
mp.jpg (120.7 KiB)
JohnC avatar image JohnC ♦ rostyvyg commented ·
@rostyvyg

Ok. It's good that you can measure the currents and confirm a fairly good balance. That may not always be the case though.

Not so good that one is gassing, heating and probably venting. Classic thermal runaway (especially if the Temp Comp probe is on another battery). Early death expected.

I say 'overcharging'. You must reduce Absorb Charge V. Victron default is 14.4V, but you can go much lower as needed to avoid this happening. It may take a little longer to charge them, but a slower charge will have no adverse effects. And make sure your T Comp is doing what it should.

I'm not really a big fan of the Multi's Adaptive algorithm. Not for it's technical merits, but it's a bit difficult to understand and tune it. Much better a 'Tail' termination of Absorb, like in the mppts, but that requires a Battery Current (devoid of loads).

For you, I'd go down to 14.0V Absorb for now and only increase it if the heating issue stays away and you can't wait for the slower charge. Float to 13.6V, that's enough. Fixed Absorb time of say 2-3 hours. Maybe adjust this when you see the Tail current levelling. That should happen at ~3-4% A/Ah for 14.4V, but will be a little lower at 14.0V.

This all may go against your grain, but aggression is killing your batts. Gentle does it, they'll thank you for it.

0 Likes 0 ·
rostyvyg avatar image
rostyvyg answered ·

After some Internet research, I see that there were quite a few instances where owners (all of them on sailboats that have multi-battery hose banks) reported about Victron SuperCycle overheating and, if no charger temp sensor was used, bulging and leaking. So this does not appear to be an isolated incident. If as you state the batteries are to be charged with the lower absorption voltage or absorption time limited to 3 hours then Victron must do some tests and come out with updated recommendations on how to set up chargers for SuperCycle batteries used in parallel. After all, 4 batteries in parallel is a typical house bank on an average sailboat and SuperCycle batteries at least from marketing materials seem to be the best candidates for sailors who can't or don't want to switch over to lythium. Also, lowering the absorption voltage recommendation while acceptable for boats in the marinas works against long-term cruisers who charge their batteries via a generator that they prefer to run the shortest time possible.

3 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

klim8skeptic avatar image klim8skeptic ♦ commented ·

@rostyvyg "cruisers who charge their batteries via a generator that they prefer to run the shortest time possible."

Sorry to butt in, but, lithium does have the advantage of taking in large currents, with minimal time required for absorption / cell balancing. You could probably only use a single 300Ah battery (or 2 paralleled 150Ah) battery for better than equivalent Ah capacity.

If you want to stick with LA and 12v system, a string of 2v cells will be easier to deal with vs paralleled 12v batteries.

1 Like 1 ·
rostyvyg avatar image rostyvyg klim8skeptic ♦ commented ·
Unfortunately, lithium is not an option for me since 1. My insurance company won't cover my boat if lithium batteries are installed and 2. My bow thruster consumes 600A so the lithium batteries with built-in BMSs will shut down on me the first time I use it.
0 Likes 0 ·
JohnC avatar image JohnC ♦ commented ·

@rostyvyg

All battery makers supply info for individual batteries. What the buyer does with them after that is out of their control. Victron have this.. https://www.victronenergy.com/media/pg/The_Wiring_Unlimited_book/en/battery-bank-wiring.html

1704772365845.png

'Tricky' it is. Yet you've chosen to use the max 14.6V recommended Absorb charge. What motor vehicle is set to charge that high? Sure they don't have Temp Comp - but nor do you on every battery. 14.2V would have been a far better choice until you saw the results. I've suggested reversion to 14.0V and subsequent increases if justified.

I live on a boat myself. Cheapo Pbs, no shore power, but a genset I have to restrict charge from. It never runs for more than 1.5 hours at a time. 48V but still have an old aux 12V (3x parallel) bank that gets a couple hours a day from the 48V bank at 13.8V. Some of those oldies have a >12 year pedigree - wouldn't have happened at 14.6V.

Your aggression is killing your batts.

1 Like 1 ·
1704772365845.png (43.1 KiB)
rostyvyg avatar image
rostyvyg answered ·
1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC avatar image JohnC ♦ commented ·
@rostyvyg

Indeed. I posted there too. Nearly 4 years ago and the advice hasn't changed. Daniel Boekel there is Level 3 Victron support, he knows..

0 Likes 0 ·

Related Resources

MPPT Product Page

MPPT Error codes

MPPT 150/60 up to 250/70 Manual

Additional resources still need to be added for this topic