Data is Beautiful

Were still working our Firefly issues, so this will probably be another data filled quest to find a solution. Since last week we’ve really just been spending a lot of time testing and running “Restore Charges.” Brent has been in contact with his supplier of the batteries, so all we can really do is wait them to respond, and collect some data.

First thing they suggested was to gather the serial numbers. When we grabbed them, it turns out they are all really close together. xxx11, xxx29, xxx37, xxx42. Not precisely sequential, but close enough to think they all came from the same batch. The chances of 4 individual batteries failing at the same time in the same way are astronomically small, so it helps a bit for me to see that it could be a batch issue.

According to the Firefly Manual, it is possible that over time they batteries lose some capacity, so they recommend a “restoration charge”. This is a full discharge do 10.5v resting, and then a hard charge of as many amps as you can push at them until they hold 14.4v with only a .66amp current. To start this process, I disconnected my solar charger so that it would not be trying to charge all the time, and disconnected the ACR (Automatic Charge Relay) that automatically connects/disconnects my house bank with my starter battery when needed. These are just 2 more variables I’m removing from the system.

The first issue with this is that once the batteries start dying, they drop voltage so fast that my fridge and portable freezer start shutting down. To prepare for that, I took my portable freezer and installed it on a friends boat. Then when the charge starts dropping below 12v, I just shut my main fridge off so that I don’t damage the compressor. The next issue is that with no fridges not much else pulls on the 12v system. I can use the inverter, but it draws so much (20+Amps) that while it is running, it draws down to 10v, but as soon as I stop the draw, the batteries recover back to 12v+. I need to draw slowly, enough to pull all the power out of the batteries such that they recover to 10.5v. It took me most of an afternoon, but I eventually pulled them down to the point where even with no draw, they sat at about 10.4v. The problem? I had only used about 60Ah. The bank us supposed to have over 465Ah in it.

With the batteries well and truly dead at 10.5v, I reconfigured my charger to charge at full power, 100amps. I had reduced this to 50% becasue in the heat of the summer, if my Air conditioner is running (9amps), and my water heater (15amps) is running, any more than 50amps (5amps @120v) can easily overload my 30amp (120v) shore power connection. I shut off the water heater, and let the charger rip. This is supposed to take hours, but almost immediately the batteries are back at 14.4v. Drawing 100Amps, but it still should have taken some time to get there. Just over an hour later, the charger is pushing only 2amps to maintain 14.4v, (.5 per battery so enough to satisfy the restore charge) and shuts down.

Hopeful, I start repeating the process. I make sure the charger is off, reconnect the fridge and power everything up. I run the whole boat off the inverter and pull 30amps happily. I’m feeling good as once the initial surface charge wears off at about 12.5v, they stabilize and hold there. I continue my work day for about 30 min, then I look over and the batteries are readying 10.2v. It didn’t work. I shut off the inverter and they recover back to 12.5v. I spend the rest of the weekend and nurse the batteries down to empty again, and run the restore charge twice more. No help.

Now I’m convinced they are dead. With that all we can do is wait and hope they get replaced.

Since then I’ve dug into the charger settings of both the solar charger, and the main charger. Nothing looks out of place, both chargers are set to the 14.4, and 13.4 absorb and float voltages as recommended by firefly. I’ve also done some very close watching with as while the solar charger is running, I can somewhat log the data via the Victron app. I find that with minimal draw (just my main fridge, ~4amps) they last for some time, then as soon as they hit some threshold around 12.5v the bottom drops out.

You can see that in the orange line when the fridge is running. Just as the voltage drops below ~12.5v it starts dropping much faster than normal. This is just as the sun is coming up so the solar had started helping a bit, but only while the fridge isn’t running. Once the batteries hit 12v, the main charger kicks in and does a full charge. I also see similar behavior on charge. Once the batteries hit 13.6v, the ceiling disappears, and the voltage climbs out of proportion to the charge rate.

I don’t know if this is normal or not, I just thought it was odd, and I wanted more graphs becasue Data is Beautiful.

P.S. I really wanted to put a picture of Data as the features image of this post, but that didn’t feel right. If you made it this far, here is your bonus Data!