LifePo4 Batteries and Humility - Rod Caple
Submitted by Rod Caple
S/V SAWLEEAH
GYC Member since: 2000
Current Cruising Theme Song: "Beyond the Sea", Bobby Darin
A cautionary taleā¦.Rod shares the lessons learned from his adventures with his new lithium batteries: purchase, installation and performance.
How it Started
In early 2021 I needed to replace Sawleeah's house batteries and got interested in Lithium as a contender against lead acid. I was attracted by the idea that taken over the long haul, Lithium batteries could be cheaper and that they required no maintenance through their service life of about ten years.
A lithium iron phosphate (LifePo4) battery installation requires a battery management system (BMS) to protect against over or under charged conditions and to prevent excessive charge or discharge rates. These conditions can quickly cause damage. The BMS has a connection to the negative of the battery and to the positive of each cell or group of cells. It works to equalize the voltage of each pair of cells as well as the charge and discharge rates for the battery.
The installation I chose has four pairs of cells giving a total capacity of 640 Amp hours. This is about 1 1/2 times the rated capacity of the lead-acid installation being replaced. A temptation to increase the total capacity arises from the smaller size, about 1/4 the volume of lead-acid for a given rated capacity. Another feature of LifePo4 is that the full rated storage can be utilized. Lead-acid are generally restricted to 1/2 the rated capacity as the charge range. This would come to 225 Amp hours of usable capacity for my previous 450 A hour lead-acid installation. So my new usable capacity is about 2.8 times the previous useful capacity.
I ordered my cells direct from China on March 19, 2021. Eight cells from Aliexpress was $1,631. The BMS from Bangood $376, the same day. The little bars needed to connect the cells were included with the cells. It took about a month for delivery.
A feature relished by my aging body is the lighter weight of LifePo4 batteries. A core enticing benefit is that the heaviest element I now have to lift is a single 12 lb LifePo4 cell. Compare this with a single 6V battery weighing 60 lb in the lead-acid installation.
LifePo4 batteries are arranged differently than lead-acid. The practice is to form groups of cells connected in parallel to reach the desired storage capacity and then connect those groups in series to attain the desired voltage. The result is a single battery. A single battery only needs a single BMS. Another advantage is that each cell can be connected to the other cells in a parallel group through a fuse. Without a fuse all the other cells in a parallel group would provide current to a shorted cell and melt it to perhaps start a fire.
An example is a Tesla automobile battery that uses a pack of 6S74P, that is 74 cells in parallel as a sub group and then 6 of those sub groups joined in series to make a pack. Then they join packs in series to make a single giant battery of the desired Voltage. Imagine, a cell that becomes shorted becomes like a wrench dropped across the terminals of 73 cells wired in parallel. Molten wrench! The need for a fuse per cell is clear. My installation has fuses.
The Good News
Lets start with some good news about how my installation worked out. The shore power battery charger that served my lead-acid house bank works just fine for the LifePo4 installation. The charging choices available for various lead-acid battery types and ambient temperatures give adequate ability to select suitable charging for LifePo4. It's 20A output would take 30 hours to restore a full charge range, but this hasn't happened because the alternator and a single solar panel seem to keep it up. The large frame alternator used to charge the house battery is a 130A truck alternator with the original internal regulator.
The Voltage settings work just fine for charging the LifePo4 battery. The BMS will protect against a charge rate that might harm the battery. However, it does nothing to protect the alternator. Limiting the current to protect the alternator against overheating needs to be done. Actually this is true for any alternator that charges a large battery bank.
The better efficiency is palpable. What you put in is pretty much returned so less charging seems to be needed. The Voltage is mostly around 13.3V so things are more lively or brighter. The stories I have heard about the BMS abruptly cutting off charging and causing a Voltage spike from the alternator don't apply. It turned out that the BMS I bought limits charging by dropping Voltage gradually about 0.1V at a time, so no issue with causing power spikes.
The Bad News
On to the bad news. One day in July of 2021, motoring south in Georgia Strait, a small bump noise was noticed and a while later it was noticed that the VHF radio could no longer receive. I had an experimental circuit that regulated the alternator current and let me turn a small knob to limit the output in response to a digital display showing the alternator temperature. Searching for possible causes of a Voltage surge that could explain the damage to the radio, I came to the conclusion the experimental circuit had gone into oscillation and put out a burst of radio energy. So I removed the circuit.
However, months later in March of 2022, another event destroyed the radar and the navigation monitor.
The monitor clearly needed replacement, but the radar was not so clear. The manufacturer, Simrad, provided a detailed guide to diagnosis that lead to replacement of one component followed by further trials which concluded that the dome unit, a non-repairable component, was defective. To be sure, I took the dome unit to a shop that does navigation equipment and had a Simrad monitor device to connect it to. They confirmed it was done for.
Along with all this I had noticed that the house alternator was not putting out rated current. I took it to Brighouse Auto Electric for an assessment and repair or replacement. It turned out to be a rebuilding and when I picked it up they showed me a tiny 1/2 inch number 6 woodscrew that they found inside. They said it was the cause of my troubles, bouncing around inside and occasionally jabbing into the windings.
You could say I was screwed out of two radios, a monitor and a radar scanner.
Some of the blame can be laid on the LifePo4 battery. While the battery didn't create a power surge, it did fail to absorb it. A nice thing about lead-acid batteries is that they will absorb a surge of current and so prevent a surge of voltage that can cause damage. The BMS is designed to protect the battery. It is reasonable to expect that it would block a surge of current from reaching the battery, leaving it to look for another place to be absorbed.
I do have a start battery. It has a separate alternator and there was no connection between the start battery and the house battery. This was because I had come to the view that the BMS wouldn't make a quick disconnect of the running house alternator that would cause a pulse of voltage. I hadn't thought about voltage pulses caused in some other way.
Resolution
I now have a connection from the house system to the start battery through a pair of 150A full wave rectifiers. They are connected in a manner that employs them as a pair of four diodes in series (no connections on the ac terminals). A surge in the house system going through these diodes sees a voltage drop of about 1V due to the intrinsic semiconductor diode forward Voltage drop. This drop will vary only a small amount with variation of current so the effect is to allow a large pulse of current to flow to the start battery, preventing a rise in voltage that could cause damage, while the 1V drop is enough to limit overcharging the starter battery.
Another thing, added in April of 2022 to protect the instruments, is a dc to dc Voltage regulator that will convert an input of 9 to 36V dc to an output of 12V dc. These gadgets can be found on the Internet Amazon.ca ("COCAR/EKYLIN C121212 Heavy Duty DC 12V voltage stabilizer accept DC 9- 36V input, DC 12V output. Max Current: 12A. High Power: 144W, usually for car battery auto truck vehicle boat motor solar system protection. ") for about $70. A tiny cost compared to the value of the assets it protects.
Associated with all this were some improvements to my navigation equipment. In November of 2021 after the first power surge event I had installed an AIS transponder that shares the masthead antenna with the VHF radio. Fortunately it survived the subsequent surge. It is the Em-trak brand.
A complication in installing the new Simrad Halo radar was that it wouldn't connect with the navigation computer. The connection is through an Ethernet cable and there is a process of acquiring an IP address for the radar that should occur automatically, but wasn't. The failure to connect situation led me to think about a possible problem with the OpenCPN software. To rule this out I uninstalled it and then reinstalled a clean new up-to-date copy. That didn't improve anything, but it did make my up-to-date charts for western Canada disappear.
Long story short, the AIS data traffic was swamping something and preventing the hookup. Eventually, a start-up with the AIS turned off allowed the connection to be made. After that first connection the radar will now start up just fine with the AIS running. I don't understand it either.
Eventually, after wasting some of the chart suppliers time, it was found to be that I had skipped one small step in reinstalling OpenCPN. I went back to "Charts" and clicked on the "Prepare all ENC Charts" button. After that and invoking the monthly update, I had the latest charts working.