I have a Victron system (Multiplus II and Cerbo-GX, DVCC enabled) with WatchmonCORE and a Cellmate K9, software is 2.17.51.0. The battery is a 16s1p LiFePO4. When discharging the battery to almost empty (about 8% SoC) I have the following problem.
If one cell falls below “Cell Low Cutout” (CV3 = 3.05 V) during discharge, the Victron system shows a low battery alarm and disconnects the battery from the system via a shunt-trip. Batrium itself is not in a critical fault state.
The critical limits in WatchmonCORE are “Low Cell Volt” CV1 (2.50 V) and “High Cell Volt” CV12 (3.7 V). The limits in the Victron system are the same.
My understanding is that “Cell Low Cutout” is the limit at which Batrium should stop discharging the battery and not to trigger a low battery alarm on the Victron system. This should be done when “Low Cell Volt” (2.5 V) is reached. Or am I wrong?
To rule out that a limit at the Victron system triggers this alarm I increased the “Cell Low Cutout” limit to 3.35 V during discharging when the minimum cell voltage was 3.3 V. Immediately after doing this the Victron showed the low battery alarm. So it seems that the only thing that triggers this low battery alarm is when the ‘Cell Low Cutout’ limit is exceeded. I suppose that no Victron limit is involved in this situation when all cell voltages are at healthy 3.3 V.
Can someone confirm this behavior? Is this the intended behavior? At the moment I get this low battery alarm when the battery becomes empty. I would expect that in this situation only discharging is stopped, at least as long as no critical limit is exceeded.
(Uppps, sorry, no more pictures allowed in the first post. The important critical settings are - as mentioned already in the text above - “Lo Cell Volt” 2.5 V and “Hi Cell Volt” 3.7 V)
Yes, Victron (in detail Cerbo-Gx, relay 1) triggers to switch off the battery due to a low battery alarm. But the situation is a normal “battery becomes empty” situation which - in my opinion - shouldn’t be treated as an alarm as long as no critical limits are reached. Which isn’t the case here. Or I oversee something…
Thank you for posting your question! I think I know how to fix this issue.
The Victron remote profile is special in that it can convey extra info like a low voltage warning, but this is not intended to trigger a shunt trip, just disable discharge and log the event. If it throws a shunt trip, there’s no way to charge it up to the Resume value.
The BMS should have the final say on whether the battery remains connected to the inverter. For example, if a CAN communication error occurs, the inverter won’t receive that warning and will damage the cells by discharging too low.
In other words, you should have your shunt trip triggered by the relay on the WatchMonCORE.
Thank you for answering my question. I appreciate it very much. Here my comments.
The Victron remote profile is special in that it can convey extra info like a low voltage warning, but this is not intended to trigger a shunt trip, just disable discharge and log the event. If it throws a shunt trip, there’s no way to charge it up to the Resume value.
I’m aware of this and I ensured in the Victron configuration that only the “low battery alarm” is taken into account and not the “low battery pre-warning” (Alarm-Setup: Low battery voltage: “Alarm only” and not “Alarm & Warning”). So, the pre-warning shouldn’t trigger the alarm relais.
The BMS should have the final say on whether the battery remains connected to the inverter. For example, if a CAN communication error occurs, the inverter won’t receive that warning and will damage the cells by discharging too low.
I totally agree. In fact, the WatchMonCORE relay A is in parallel with the Victron alarm relay (both will trigger the shunt-trip to disconnect the battery from the system). The idea behind this was to have the Victron system as backup in case the WatchMonCORE fails and vice versa.
My first thought after having this problem was to disable the Victron alarm as you suggest. But I hesitated to accept this as a solution. I have read many articles and watched endless videos about Victron, including the special combination of Victron and Batrium, but so far I have not noticed anywhere that the ‘Low Battery Alarm’ should be deactivated to avoid a “low battery alarm” when the battery becomes healthy empty. So, I decided to ask my question in this forum. But it seems that you confirm this idea. I think I will follow your advice.
But I still don’t understand where this alarm comes from I will observe this further.
In general, you may want to expand your battery if you are reaching 8% SoC regularly.
This was the first time after about 9 month of usage. But I’m building already a second LiFePO4 14 kWh battery
Anyway, thanks James for your feedback and also thanks for the informative graphic regarding the charge/discharge limits.
I will chat to our software engineer about whether we can send the pre-warning rather than alarm, but we may be limited by the protocol Victron has provided us with.
I’ll respond here if we make any progress on that front
Good hint, Daniel.
I checked the Low Battery value in VEConfig and it is slightly higher than the limit I set in Batrium. This could indeed be the reason for the early shutdown.
I will observe this further.