Have I found a bug in Victron's DC-DC charger software? On an Orion-Tr Non-isolated 30a charger configured to charge our start battery from house bank.
It did rebulk 3 times in one day even though it *never* hit the rebulk threshold (12.8v). And completed absorption on all 3 rebulks, so it's not trying to finish an incomplete bulk.
I do have input voltage lockout in use which is what causes the target battery to drop below float when float charging gets temporarily stopped. But it never dropped 0.7v below float (13.5), as shown by the attached Victron SmartShunt per-minute chart.
I considered that perhaps the DC-DC charger is reading lower voltage due to greater voltage loss than the SmartShunt. But I really don't think I have more than 0.1v voltage loss (the DC-DC wires are larger gauge than the SmartShunt, and a shorter run).
Excessive absorption cycles isn't a major problem (AGM battery) since they are short. But it is still irksome that the DC-DC charger (30a non-isolated) is disobeying its own programming.
I believe that input voltage lock-out causes an automatic rebulk when the voltage cut-off ends (upper threshold resumed). This seems like a clear bug to me. It means that the rebulk threshold is simply ignored, which defeats the purpose of having a rebulk threshold.
Anyone from Victron reading this? Can you file a bug or see if there's an existing one? 
