Hello to everyone, this is my first Topic and I hope someone can help me.
I made a lithium installation is a 12V sailing boat, with the next configuration:
5xLithium Smart 12,8 V / 200 Ah NG.
2x MultiPlus-II 12/3000/120-32 connected in parallel.
Lynx Smart BMS 1000 NG.
Cerbo GX. v3.55 (Venus OS).
SmartSolar MPPT VE.Can 150/100 rev2.
2x SmartSolar MPPT 100/50.
smart battery protect 200A.
Orion XS smart dc-dc charger 12-12V.
2x Lynxs IN.
2x Linxs distributor.
The connection between the lyxs distributor and the multis is made by 4x95 mm2 cable per 6m due the complicated space in the boat, was imposible to make it smaller.
the average charge of solar power is 10-13 Kw and the average consumption in the boat is 15-19 Kw. the owner starts the generator sometimes because of the deficit of power. They use a lot of 220 stuff, in the kitchen, the watermaker, the washing machine so thats why I decided to install 2 multis in parallel and my maths I think was ok choosing 1000A of batteries, at nigt they usually drop at 50 or 40% but an 14:00 h usually they are full.
Is not the first time i found a higth baterry temperature pre-alarm and because of this the BMS decrease the discharge A and make me a fault on the inverters for low power. I’m leaving a list of the things that happened that day with the readings of the vrm if anyone can help me whats wrong when they use more than 3800W in 220 from the inverters. Is suposed to give at least 5000 W in 220 with the 1000 A/h batteries.
Here is a timeline made by chat GPT from my .xls file from my VRM page, with all the data received. Shame I can’t upload the file here, or at least I don’t know how to do it.
Timeline of the 28-Jun-2025 fault (all times = Atlantic/Canary, taken straight from the VRM log)
15:37:00 Cloud cover – PV production collapses from ≈ 1 kW to ≈ 70 W in <45 s.
15:37:45 Load spike – House demand rises to ≈ 3.2 kW.
15:38:20 Heavy battery discharge – 12.99 V / -420 A / 33.5 °C.
15:38:55 BMS “High-Temp pre-alarm” at 34.9 °C → Charge Current Limit (CCL) forced to 40 A.
15:39:10 Multi issues “Low-battery” Dynamic Cut-off warning (12.89 V @ -460 A).
15:39:20 Owner manually starts the generator.
15:39:43 AC-in detected (230 V); Multi switches to Charger and immediately draws the full 40 A CCL.
15:39:45–15:40:45 PV almost zero – MPPTs limited to 0–7 W because the Multi is already using the whole 40 A CCL.
15:40:45 Battery recovers – 13.55 V / +40 A; temperature peaks at 35.2 °C then starts to fall.
15:41:10 BMS clears the pre-alarm (temp ≤ 34 °C) → CCL raised to 200 A → PV production instantly jumps back to > 900 W.
15:47:00 Generator stopped manually – System returns to normal inverter mode, PV + battery supply the load.
18:32:30 Minor AC sag to 193 V during a 2.7 kW evening load while charging at +241 A; no alarms, recovers in 1 s.
That sounds like a normal operation…
Personally, at this power level, I would have moved to a 24 or 48V system to avoid the thick cabling required by the 12V inverters. This also assists in keeping things cool, as currents are lower and all losses are I^2R…
Thanks Mike for taking your time reading my problem. Unfortunately, The boat was already 12 V everything, and was my first “big” installation, I never expect to have a battery temperature problem that causes the system to shut down. My concern was battery capacity, battery power instantly, inverters power and time of battery capacity at certain time. But this battery temp alarm, was unexpected. And was not a big temperature 34 centigrade, y cool waters of the Atlantic, I don’t know how it will behave in caribean with hot clima and hot water.
Any solution, I ask Chat gpt and he told me to upgrade the cable double it. and update the cerbo software to 3.62, here is a setting that allows you to take the voltage in the BMS instead the terminals of the inverters, and it won’t cut for low voltage. What do you think?
is it possible that one or more connections between cables and battery terminals get hot due to resistance in the connection?
If so, can you try to lightly sand the terminals (with a very fine sandpaper), clean and apply some grease (petroleum jelly for example, but never use copper grease). clean cable lug also and re-connect.
Please don’t rely on Chat GPT for anything technical. These LLM Ai things don’t know right from wrong.
Upgrading the Cerbo software is not the problem: Your system should already be taking the BMS voltage. Also upgrading the cable will not cool either the battery or the inverter.
Checking to see if contact resistance at terminals is causing unexpected temperature rise is a valid exercise, a quick feel of battery terminals after a few minutes of heavy load should give you an indication if you don’t have an IR thermometer.
Having a high temperature pre alarm at 34.9C means that these batteries are NOT suitable for Tropical applications, Ambient temperatures can easily reach that level. Hull cooling works ok in a steel boat, but less so in wood or GRP. Current trend of lightweight balsa cored hulls are very good thermal insulators too, so heat dissipation will be a problem: 10% loss on a 10kW load is 1kW, which does plenty of heating.
Ok, thank you for the advice, I will check the temperature in the battery bolts in higth load after a few minutes, maybe I have a bad battery 200A NG victron, and they are brand new.
Is my first big installation, and I’m scared, hopefully security systems in victron take care of a big failure.
In your data you quote the battery temperature which will be the average and 34.9degC is in the working range of the battery and should not alarm. What you need to look at is the min and max cell temperatures, one of the individual cell temperatures has gone high and caused this. If a cable is not connected to a terminal very well it will heat up and affect the cell temperature sensor close to that. Here is a plot with a slightly lose terminal bolt.
When a high current flows the max cell temperature increases rapidly and the min cell and battery temp do not respond as much. Also when the high current stops, the max cell temperature cools off quickly showing that the heated area has little thermal mass. This was fixed by retorquing the bolts.
@MikeD see explanation above, in my case the max cell temp reached 36degC, no alarm was raised, the batteries are rated for 50degC, from the datasheet
Thank you very much, I will take an eye on that point, now the boat starts another 10 days charter, so is good to make tests, and I will warn the capitain because I´m not near by. I will look the VRM and I will ask him to check the temperature bolts and in victron connect the cells temperature, because I use 200 A NG and a BMS 1000 NG, and only have the individual cell temperature in the Lynx smart BMS victron connect app, not in the VRM sadly, hope will be in next update.
Thank you to take your time to help a victron installer in Canary Islands.
The min and max cell temperatures (and min and max cell.voltages) ate sent from the Lynx NG to the GX device and to VRM. My graph was from VRM, advanced graphs, it will need to be plotted with a custom widget. These values are also in the Excel data download so you should be able to look at your event data again. You should deal with this as soon as possible, it may just be a loose bolt.
Hello, but not the individual, this only in the victron connect in the app, as long as I know, thank you for the reply, i´m sending you the data of the fault in the vrm 24 hours.
Yes it is like mine but much worse, the average temperature and minimum temperature stay very close so it is only 1 of the many temperatures that is high. And it increases and decreases very rapidly as power goes up and down.
I will do, and also I´m afraid is one battery faulty, that has internal problems, because I know the bolts are well tights, but anyway we will check, during a big discharge 300A few minutes, touch every bolt and touch every battery, some other advice? and once again thank you very much.
