BrentAPerry
Member
for everyones sakes, this is a good YouTube video about the risks that need to be managed properly in a lithium battery setup, in order to safely use this amazing tech. It is really relative, and is good not only for planning installation but understanding some fundamental management issues. Regarding the fires you see here, keep in mind that these are low Ah cells, not the same level of concentration and capacity we are all using now.WARNING: Long and rambling post
My present lead-acid battery bank consists of 4 6V batteries in two banks. They are deep cycle 260Ah on a 1-2-Both switch. In theory, I have 560Ah at 12V, although I set my shunt monitor to show only 245 usable amp hours, figuring that is probably 50% of what is remaining on my old batteries. I haven’t done a legitimate load test on them, but my readings with the hygrometer are not inspiring.
So I started looking at replacement. Simple drop in lead-acid, in fact the identical batteries, are $300 each, and I would have to drive to Seattle. I can’t complain about the service I got out of the lead-acid, including using them as the start battery, but they do require maintenance, they have spit out a little acid into the battery box, and they are monstrously heavy. You know, lead-acid.
I looked into alternatives like 6V AGM. No need to water. Not quite as heavy (3# lighter). Available as start/deep cycle. Some said they got 10 years out of them, others much less. Cost was going to be similar or maybe a little more money for only 440Ah (220 usable at 50% SOC). They would be true drop in, including using the same battery cables. So a simple install, never need watering, and maybe a little faster charging was the only benefit.
Lithium was interesting to Google research. Some threads reminded me of the doomsday predictions back in the day about microwave ovens, pushbutton phones, LED lights, induction stoves, etc. Lots of issues to learn about, be confused about, and/or be frightened about. Since my first mate is my wife, my first issue is always safety. Therefore, I had to look at the cases where lithium batteries caused uncontrollable fires. However, as with the “radiophobia” once associated with microwave ovens, I think that the “lithiophobia” surrounding lithium batteries has now receded and need not be addressed when considering LFP batteries.
There are still legitimate concerns about switching to lithium. The first is the possibility that a battery management system (BMS) in a lithium battery can turn itself off. This would be the same as turning off a 1-2-Both-Off battery switch to the Off position (that is probably the historical source of this danger). The alternator doesn’t have anywhere to send its production, causing a voltage spike. The voltage spike can damage the diodes in the alternator (requiring service) and also some of the boat’s electronics. Turning the battery switch to the Off position has happened, but it is sort of like a fighter pilot accidentally yanking on the ejection seat handle. Just don’t do that. Switching to Off has certainly happened and from that we know that a BMS shutdown could cause damage to the alternator (under certain circumstances).
I wondered how to assess the risk. While it is hard to come up with exact numbers on the possibility of an out-of-the-blue BMS shutdown, it might still be possible to do some kind of a risk analysis. Several purported methods of reducing the likelihood of an alternator failure when installing LFP include leaving a lead-acid battery somewhere in the system, often as a start battery. I understand this line of thinking, but I was having difficulty in accessing the risk, or even if there was a risk, other than the “accidental ejection seat.”
It just happens that I live in a house that has a combination wood/electric stove. Woodburning on one side and electric coil elements on the other. When Roosevelt instituted the Rural Electrification Administration in 1935, people were really wary of this new-fangled electricity inside the house. And what if an electric power line broke? How would one cook or heat bath water? So, the Monarch Stove Company came up with the combination wood/electric stove that is in my kitchen. The stove still works fine, but it is now kind of a curiosity. Maybe lead-acid batteries will be the same in a few decades. Maybe sooner.
Leaving lead-acid in the system didn’t seem to be necessary. The work-around sometimes includes the use of a gadget that still allows the alternator to send a charge somewhere to a lead-acid should the BMS shuts down. For instance, one could send the alternator charge to the LFP battery and include a DC/DC transfer to a lead-acid battery. Then, if the BMS shut down, and the DC/DC gadget didn’t fail, it might transfer the current and save the alternator. I say might because I think that all the various proposed protections against BMS shutdown are based on the laws of probability, and nothing is 100%. I didn’t think it made sense to examine the various protections against BMS shutdown without first examining the probability of a BMS shutdown.
I was never intending to have a single LFP battery, either for house or start. My present setup has two banks, two batteries each, and I wanted to re-use as much of the system as possible, including the connection cables. Therefore, four batteries made sense. Also, most single LFP batteries didn’t have as many amp hours as I wanted. Some of the larger ones could work, but they have some problems besides size and weight. The problems tend to be discharge current and redundancy.
Fortunately, both are addressed with the same solution, i.e., more LFP batteries. A 100Ah LFP battery generally has a 1C (100A) maximum continuous discharge rate allowed by the BMS. In theory, one could run a 100A draw (for less than one hour). My engine starter requires about 500A for less than one second, but then still has a substantial draw for another second. Using two 100Ah batteries in parallel would mean less stress per battery for starting. Three would be better. Four would be even better, but the bank would still exceed each individual battery’s continuous discharge rate. Better yet would be batteries that had a larger allowable BMS discharge current available. 280Ah LFP batteries are available that each have a BMS with a 200-300 continuous draw (some also have an 800A maximum draw for 10 seconds). That means that the starting “stress” on a bank of four 280Ah LFP batteries used as a house/starter bank could be <50% of their rated continuous current. And >14% of the bank’s maximum current.
But will discharging LFP batteries (even at less than their designed current rating) damage them? Probably. Just like all batteries, including lead-acid starter batteries. A lead-acid battery is only good for a certain number of starts over a certain amount of time before it wears out. Some get 500 starts, some get 1,000. The 280Ah LFP batteries I bought are rated at 15,000 cycles. That is likely an exaggeration. Maybe with my usage I’ll only get 2,000 cycles. I can live with that, especially since the 4 LFP will give me 6 times the available house amperage in each cycle and twice the number of starts compared to lead.
Then there is the redundancy issue and the law of probability. Instead of a single battery shutting down, I am planning on a bank of four LFP batteries. There is not enough information to support actual numbers, so I had to make something up for the purpose of risk analysis. Let’s say that I have my alternator connected directly to a single LFP battery and the chance of that BMS shutting down is 1:100, meaning that for every 100 times I go cruising there is one possibility of the BMS shutting down. From the reports of those who have cruised for years with the alternator direct to LFP (with a properly adjusted external regulator), we know that the probability of BMS shutdown is much less. Nevertheless, we can use that number. This includes the possible shutdown modes of over-heat, over-charge, under charge, and battery cell imbalance.
The lead acid “fixes” that are proposed generally include a lead battery to accept the charge should the LFP BMS shut down. But lead-acid batteries can also fail and cause damage to an alternator. Common failure modes for lead-acid batteries include overcharging, undercharging, sulfation of the negative plate, acid stratification, water loss, thermal runaway, short circuit, contamination, and premature failure (among others). When a lead-acid battery explodes, that obviously would result in the same voltage spike that a BMS shutdown would cause. But just for fun, let us say that that the possibility of a lead-acid failure is one in a million. LFP alternator damage is one in a hundred and lead-acid is one in a million. Looks like the lead-acid “fix” is the better bet.
But here is where redundancy and the law of probability work their magic. I decided that a single LFP has a 1:100 (.010) chance of a BMS shutdown. The probability of two shutting down at the same time is (.01) X (.01) or a one in ten thousand. For a three-battery bank, simultaneous shutdowns would be (.01) X (.01) X (.01) or a one in a million possibility. Four LFP batteries would be one in a hundred million. That’s just the way the math works. Four LFPs is much safer than the lead-acid battery “fix.” Maybe four LFP are even safer than relying on some kind of a DC/DC gadget. I don’t know how often they fail. Maybe also one in a million chance? Instead of spending $400 on the gadget, simply get another LFP battery to increase the safety factor to one in a billion (if I’m doing the math right). And instead of a lead-acid starter battery, for the same money one could buy still another LFP and increase the safety factor to one in 100 billion (that’s just two banks of 3.) Most LFP can be put in to banks of 4. You can do the math for that one.
The above calculation is based on a BMS shutdown being an independent events, of course. If they are dependent events, meaning that there is some correlation between shutdowns, the math changes. We can pretend that one battery overheating presupposes another battery is soon to overheat (which may also presuppose that the alternator battery heat sensor fails at the same time). Or, if a cell in one battery overcharges, then a cell in each of the other batteries will immediately overcharge. Those scenarios are imaginable, but I think most would likely be premised on a common external fault that isn’t related to whether the battery bank is LFP or lead-acid.
For instance, let’s assume that the alternator suddenly starts putting out 28V. All of the LFP batteries shut down, which causes a voltage spike that blows the diodes in the alternator, but not before burning up most of your sensitive electronics. Now let’s run that scenario with lead-acid. The lead-acid accepts the overvoltage until it melts or explodes, which causes a voltage spike that blows the diodes in the alternator, but not before burning up most of your sensitive electronics. The damage is the same except that all the LFP batteries saved themselves, while the lead acid battery was destroyed and spread acid all over the locker.
I bought LFP batteries with Bluetooth, but I’m not sure how much Bluetooth can save one from a catastrophic failure. It has proven interesting when charging them up for the first time. They were quite low (29-30%) based on what I had seen others say when receiving LFP batteries. I have a motley crew of old lead acid chargers (a 6, 8, and 10A charger) that I put on them and then watched them closely (the next day, as each took overnight). It was interesting watching the individual cells because they were sometimes off by a tenth of a volt, but later the lagging ones might be ahead of the ones that earlier had the higher voltage.
At one point, I was surprised to see an “overvolt notice” on one battery that was at 98%. One cell was at 3.6V and the other three were at 3.5V. While I was trying to understand what this meant, I figured out that although the notice was still on my phone, the event seemed to have passed. The battery was charging normally and soon at 100%. I was unable to tell if the battery had ever actually shut down, and if so, for how long. It was like driving a car that notifies of a flat tire, but before one can pull over, the system fixes the tire and you are on your way. I’m not sure what to think of that.
Besides installing the four LFB batteries, the only other change I plan on making is replacing each bank’s 250A ANL fuse with a 250A T class fuse (my entire electrical system, including the starter cable, is fused as per MarineHowTo). I may also buy a Balmar Alternator Protection Module (APM). I’m still uncertain as to just how much protection it provides and what are the actual chances of four BMSs shutting down at the same time.
Speaking of probability, as part of my research, I looked up the chances of being struck by lightening. A person has a 1:15,300 of being struck by lightening in their lifetime. It is more likely that one will be struck by lightening than four LFP BMSs failing at once and hurting an alternator. What have you done to protect yourself? Fortunately, I’ve already had my lightning strike (unfortunately, it doesn’t work like that.)
Mark
Last edited:
