My house bank consists of 4 Victron AGM SuperCycle 170Ah batteries which randomly go into thermal runaway with no rhyme or reason. Once in a while when in absorption one of the batteries starts consuming increasingly more power, heats up, etc. After the battery cools down it might go through multiple charging cycles with no issues and then another battery exhibits the same behavior. And then another one. There is currently no way to give Multiplus an early warning since if the temperature sensor selected as active in DVCC is on the terminal of a nearby battery the overheating battery continues to heat up until it heats the battery with the active sensor at which point the damage has already been done. I installed a temperature sensor on each battery in my boat house bank (4 sensors in total) and connected them to the temperature sensor inputs on CerboGX. Still, there is no way to make CerboGX report to DVCC the temperature of the HOTTEST sensor in this group for temperature compensation while charging. The best I could do was use the CerboGX relay to power up the fans to improve the battery box ventilation and sound the buzzer alarm. And that was after I fought Victron tooth and nail to have the battery bank replaced TWICE in TWO years under the warranty. So far the forced battery box ventilation is helping (fingers crossed), but…This is a ridiculous situation! We have a smart highly programmable device that is supposed to be able to use every available bit of information to make the charging process safer and more efficient but it can’t automatically select the temperature of the hottest battery in the charged bank for temperature compensation. Really, Victron?
How are the batteries connected? Series or parallel or group?
The answer to the question above “How are the batteries connected? Series or parallel or group?” is I have 4 batteries connected in parallel for a total of 670 Ah 12V house battery bank.
and how do you charge your batteries …
A little more information apart from the Cerbo and Multi would be important …
Charging only via Mppts or other WR ???
OK. First of all, we are speaking about cruising (liveaboard) sailboat. All charging equipment is Vitron (Multiplus 3000, 2 x 400 Watt solar panels regulated by Victron SmartSolar MPPT controllers, BMV-712, CerboGX) all installed and networked by a certified Victron electrician, the firmware on all devices updated to the latest versions over time, installation and settings triple checked by various Victron dealers during the warranty claims. Also, if the old Victron forum had not been taken down one could have easily found multiple references to SmartCycle batteries overheating there…
It would be worth looking at Node Red code to be able to select the highest temperature. What I do not know is if this could then be output to the battery monitor temperature or a virtual temperature that could then be used in DVCC.
You could connect 2 of the resistors in series and then connect them in parallel to the others.
then you would have the same basic resistance … You would only have to adjust the ratio of the temperature offset slightly as all batteries are now measured simultaneously and the characteristic curve would be slightly different.
It does not appear that Victron temperature sensors are simple thermal resistors…
There are dozens of battery warranty claims submitted to Victron yearly related to this situation. Why would an end user who knows nothing about software development have to deal with Node Red to have it addressed?
Hi @rostyvyg Remember me? Yeh, still about.
It’s been archived, not ‘taken down’. One of your posts which is relevant to @Netrange’s question here…
https://communityarchive.victronenergy.com/questions/173977/victron-super-cycle-agm-sizzle-and-smoke.html
Yes, there’s history with your issue. You at least owe it to the helpers here to advise what changes you’ve made to your charge profile. If any…
I have not made any changes to my charging profile. As I said earlier, I installed two 120mm computer fans on one side of my battery box, made an intake opening on another side of the battery box, and powered the fans from the CerboGX relay, which takes input from four temperature probes located each on a terminal of every battery. Then I programmed CerboGX to switch on the fans if ANY probe registered over 34C and stop the fans when the temperature of ALL of the probes drops below 32C. No instances of thermal runaway since! I noticed the relay kicking in a few times over the year by the end of the absorption stage but no batteries went into thermal runaway yet. Again, if an existing CerboGX relay can detect the temperature of the hottest probe and trigger the relay why can’t CerboGX report the temperature of the hottest probe to the charging device to adjust the charging voltage? It is a mini-computer after all and should not be that hard to program to get the desired behavior…